State Your Business! An Evaluation of World Bank Group Support to the Reform of State-Owned Enterprises, FY08–18 © 2020 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org ATTRIBUTION Please cite the report as: World Bank. 2020. State Your Business! An Evaluation of World Bank Group Support to the Reform of State-Owned Enterprises, FY08–18. Independent Evaluation Group. Washington, DC: World Bank. COVER PHOTO Graham Crouch / World Bank EDITING AND PRODUCTION Amanda O’Brien GRAPHIC DESIGN Luísa Ulhoa 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 bound- aries, 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 dissem- ination 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. State Your Business! An Evaluation of World Bank Group Support to the Reform of State-Owned Enterprises, FY08–18 December 3, 2020 Contents Abbreviations v Acknowledgments vi Overview vii Management Response xviii Management Action Record xx Report to the Board from the Committee on Development Effectiveness xxiii State-Owned Enterprise Challenges and World Bank Group Reforms��������������� 1 1.  Evaluation Purpose and Scope 3 Evaluation Approach 6 Five Types of Bank Group Support for SOE Reform 8 Where and How the Bank Group Delivers SOE Reform Support 22 Effectiveness of Bank Group Support to SOE Reforms�������������������������������������� 27 2.  SOE Reform Performance in the Portfolio and Literature 29 Factors of Success and Failure at the Country and Project Levels 36 External Factors of Success 40 Internal Factors of Success 44 Recommendations to Address Outstanding Challenges in SOE Reform��������� 59 3.  Enhancing Selectivity: Corruption and Competition 61 SOE Reform through Coordinated, Consistent Application of MFD and the Embedded Cascade Approach 62 Bibliography����������������������������������������������������������������������������������������������������������������64 ii Boxes State-Owned Enterprises in the Response to Box 1.1.  the Coronavirus Pandemic 3 Three Perspectives on State-Owned Enterprises Box 1.2.  in the Literature (Not Mutually Exclusive) 5 Financial Sector Assessment Programs’ Treatment Box 1.3.  of State-Owned Financial Institution Reform 14 Pendulum Swings on Privatization in Bank Group Box 1.4.  State-Owned Enterprise Reform 15 Bank Group Engagement in Extractive Industries Box 2.1.  State-Owned Enterprise Reform 33 Box 2.2. The Sonali-Hallmark Scandal in Bangladesh 39 Sequencing and Complementarity to Build Credibility Box 2.3.  and Capacity 50 State-Owned Enterprise Reform in World Bank Group Box 2.4.  Analytic Work 53 Figures World Bank Group SOE Reform Support by Type, Figure O.1.  FY08–18 (no ASA) xi Bank Group SOE Reform Support by Type, Figure 1.1.  FY08–18, and by Institution (no ASA) 9 Interventions Supporting State-Owned Enterprise Privatization Figure 1.2.  15 Success Rate of State-Owned Enterprise Figure 2.1.  Reform Projects Evaluated, FY08–18 30 Success Rate of SOE Reform Interventions by Type and Instrument Figure 2.2.  (Bank Group Evaluated by IEG) 31 Factors of Success and Failure for World Bank Figure 2.3.  Lending and IFC IS and AS 37 Frequency of Success and Failure Factors Figure 2.4.  by SOE Reform Type 38 iii Tables Table O.1. Projects and Commitments by Institution ix World Bank Group SOE Reform Projects, Table 1.1.  Commitments by Institution, FY08–18 (est.) 22 Appendixes Appendix A. Methodology 92 Appendix B. Portfolio Review Framework and Identification Methodology 107  tate-Owned Enterprise Reform Support: Appendix C. S Financial and Energy Sectors 122 Appendix D. Econometric Analysis 201  tate-Owned Enterprise Reform Deep Dives Appendix E. S and Issues Notes: Summaries 249 Appendix F. Case Study Overviews for Eight Countries 287 Appendix G. Summary of Structured Literature Review and Bibliography 352  ummary of Analysis of Country Strategies and Appendix H. S Diagnostics and FSAPs 378 iv Abbreviations ASA advisory services and analytics CPSD Country Private Sector Diagnostic DPO development policy operation EU European Union FY fiscal year GP Global Practice IEG Independent Evaluation Group IFC International Finance Corporation M&E monitoring and evaluation MCPAT Markets and Competition Policy Assessment Tool MFD Maximizing Finance for Development MIGA Multilateral Investment Guarantee Agency PFM public financial management Independent Evaluation Group World Bank Group PPP public-private partnership PRSC Poverty Reduction Support Credit SOE state-owned enterprise SOFI state-owned financial institution All dollar amounts are US dollars unless otherwise indicated. v Acknowledgments Andrew Stone, adviser to the director, led this evaluation for the Indepen- dent Evaluation Group Financial, Private Sector, and Sustainable Develop- ment Department, under the guidance of José Carbajo (director), Marialisa Motta (manager from February 2020 onward), and Stoyan Tenev (senior manager). Contributors include Nadia Asgaraly, Ozlem Onerci, Nadia Ra- mirez, Santiago Rodriguez, and Aurora Siy. Nadia Asgaraly (corporate go- vernance and extractive industries), Thorsten Beck (state-owned financial institutions), John Nellis (privatization, and also adviser to the overall evaluation), and Richard Schlirf Rapti (power sector state-owned enterpri- se reforms) contributed issue notes or deep dives. Amitava Banerjee, José Carbajo, Wasiq Ismail, Riad al Khouri, Ozlem Onerci, Santiago Rodriguez, Andrew Stone, and Izlem Yenice authored the country case studies. Tesfaye Alemu Abebe, Ozlem Onerci, and Aline Weng authored desk-based case studies. Victor Malca, Anqing Shi, and Xuetong Wang provided important inputs. Jacqueline Andrieu, Stefan Apfalter, and Anjali Kumar contributed to the approach paper and early phase of the evaluation. Danni Li conducted a structured literature review with guidance from Ozlem Onerci, who also led authorship of several appendixes. Aurora Medina Siy prepared input on the Multilateral Investment Guarantee Agency. Victoria Alexeeva and Kavi- ta Mathur prepared an Independent Evaluation Group synthesis on utility reform in the power and water sectors. Emelda Cudilla provided adminis- trative support and formatted the report. Helpful comments and guidance State Your Business!  Acknowledgments were received from our three peer reviewers—Robert Cull, Patrick Heller, and William Mako—and from our methods team, including Qihui Chen (econometrics) and Jozef Vaessen. The team thanks key counterparts in the World Bank, International Finance Corporation, and Multilateral Investment Guarantee Agency. We particularly thank Alexandre Arrobio, Pietro Calice, Omar Choudry, Daniel Goldblum, Eva Guttierez, Mona Haddad, Sunita Kikeri, and Martha Licetti, and several other World Bank Group staff for important interactions at multiple stages of this evaluation’s preparation. However, the Bank Group staff bear no responsibility for this evaluation’s content. vi Overview State-Owned Enterprise Challenges and World Bank Group Reforms Purpose and Scope This evaluation assesses the contribution of the World Bank Group’s three main institutions (World Bank, International Finance Corporation [IFC], and Multilateral Investment Guarantee Agency [MIGA]) to enhancing de- velopment outcomes through their support of state-owned enterprise (SOE) reform during fiscal years (FY)08–18. It looks at what works and what does not, the effectiveness of its various approaches, factors that explain success and failure, and the strengths and weaknesses of the Bank Group approach. It parallels efforts in the Bank Group to develop support to SOE reform that is more integrated and to empower staff with new frameworks and tools to address SOE challenges in client countries. To allow for an in-depth analysis, the evaluation’s scope focuses on two key sectors in developing economies where SOEs tend to play a substantial role: the financial sector and the ener- gy sector. Independent Evaluation Group World Bank Group SOEs play a major role in many developing and emerging economies, where governments use them to achieve economic, social, and political objec- tives: to deliver and extend access to services, fill gaps in markets, devel- op key sectors or regions, and provide employment. As of late 2020, SOEs accounted for 71 percent of the Morgan Stanley Capital International (MSCI) Emerging Market Index in utilities, 56 percent in energy, and 39 percent in the financial sector. Although state ownership in commercial banks had declined globally, SOEs often retain a dominant role in banking in emerg- ing markets, such as China and India. For example, India relied primarily on state-owned financial institutions to implement its Jan Dhan Yojana pro- gram, under which more than 300 million basic accounts were opened in less than four years (2014–18). In Kenya, the Kenya Power and Lighting Company vii was the main vehicle for the government’s drive for universal electricity ac- cess, achieving more than 1 million new connections a year. Some SOEs have been run well, and they have made important contributions to economies. Singapore used SOEs successfully to drive development and industrialization after its independence in 1965, with effective efforts in diverse fields, includ- ing shipbuilding, oil refining and petrochemicals, and development finance. As highlighted by the coronavirus crisis, SOEs can be a useful vehicle for govern- ments to channel resources to adversely affected firms and households (for example, through bank loans or power utility payment suspension). However, SOEs’ mixed institutional mandates and their political importance often pose performance and governance challenges. SOEs may reflect the de- sire of the state or political groups to exert political influence over economic outcomes and resource allocation. SOEs may be asked to carry out financially unsustainable functions alongside commercial ones. They may experience political interference or competing mandates, which may reduce their trans- parency and accountability and make their oversight and regulation difficult, World’s Bank: An Evaluation of the World Bank Group’s Global Convening  Overview thus complicating their reform efforts. Although many SOEs are run well, many suffer from low productivity and efficiency, which have a detrimental impact on growth and consumer access to services. Poor financial perfor- mance and management practices can generate substantial public fiscal losses (or contingent liabilities). SOEs can also impose barriers to private participation in sectors where their dominant presence enables anticompeti- tive behavior, often with government protection or subsidy. Where, Why, and How the Bank Group Supports SOE Reforms From FY08 through FY18, the Bank Group implemented 1,008 projects with 2,185 components (interventions) that supported the reform of SOEs in the financial and energy sectors, with an estimated combined value of $71.5 bil- lion in financing (table O.1). This involved financial, technical, analytic, and advisory support for both policy and institutional reforms (upstream) and enterprise-level activities (downstream). Upstream (policy and institu- tional) support was more frequent in upper-middle-income countries, and support for lower-middle-income countries focused more on downstream (enterprise-level) reforms. World Bank lending predominated, constituting viii more than 90 percent of the Bank Group SOE reform portfolio’s value in the energy and finance sectors (table O.1). IFC investment and MIGA guarantees, making up 9 percent of Bank Group commitments, are oriented primarily toward SOEs’ enterprise-level business and operational aspects. IFC advisory engages both upstream and downstream. MIGA is primarily engaged in the power sector. Table O.1. Projects and Commitments by Institution Projects Approved Share of SOE Reform Share of FY08–18 Projects est. Volume Volume Institution (no.) (%) ($, millions) (%) World Bank lending 285 28 64,832 90.6 IFC IS 61 6 3,765 5.3 IFC AS 59 6 51 0.1 MIGA 17 2 2,788 3.9 World Bank ASA 587 58 104 0.1 Total 1,008 100 71,540 100 Source: Independent Evaluation Group portfolio review and analysis. Note: All projects and commitments approved between FY08 and FY18 are projected to the population based on sample size. Advisory services and analytics projects and expenditures are also projected to the population. ASA = advisory services and analytics; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. Independent Evaluation Group World Bank Group The Independent Evaluation Group (IEG) identified relevant SOE adviso- ry services and analytics reform support activities in 142 countries and all other Bank Group SOE reform support in 119 countries. In IEG’s sample, Sub-Saharan Africa was the Region with the highest number of financing projects approved (114), but East Asia and Pacific had a higher average per country (5.3). Bank Group support to reform SOEs has been relatively more focused on lower-middle-income countries (46 percent, excluding advisory services and analytics) and low-income countries (29 percent), followed by upper-middle-income countries (23 percent). This evaluation reviewed these projects, focusing on five major types of SOE reforms (figure O.1): ix » Corporate governance improvements, which aim to enhance the transpar- ency and accountability of SOEs, including by separating SOEs’ ownership and management and improving corporate disclosure (the timely release of accurate financial and business information). » Business and operation reforms at the enterprise level, which aim to improve performance and service delivery, including through enhancement of physical infrastructure, human resource management, product quality, operational efficiency, and organizational structure. » Strengthening competition and regulation in SOE markets, which aims to foster a level playing field among SOEs and private companies and allow private entry under equitable rules. Regulatory reforms compose the majority in this area. » Privatization and other ownership reform, which aims to improve SOE performance by allowing or introducing private ownership of all or some SOE World’s Bank: An Evaluation of the World Bank Group’s Global Convening  Overview activities. It includes privatization and measures to increase the role of the private sector through public-private partnerships (PPPs) and other means. » Macrofiscal, and public financial management reforms, which aim to limit SOEs’ possible negative impact on fiscal soundness or stability through subsi- dy reforms for and improved debt management of SOEs. Over time, emphasis on corporate governance and business and operations has remained relatively high, as has work on strengthening sector regulation and competition, but there has been a low and declining emphasis on pri- vatization. Corporate governance, business and operations, and competition and regulation represent more than 75 percent of the Bank Group portfolio in the evaluation period. Bank Group activity in the macrofiscal, and public financial management areas focusing explicitly on SOEs has remained low (figure O.1). However, the portfolio review is likely to underestimate the lev- el of activity in the macrofiscal, and public financial management areas be- cause much of this work influences SOEs as publicly financed entities with- out necessarily identifying them as an objective. Privatization represents a small and declining share of the portfolio (at less than 6 percent throughout the evaluation period), even though interviews suggest that after a period of client disinterest and political sensitivity, demand for privatization support x has been growing. Even though privatization (and PPPs) is not part of the Equitable Growth, Finance, and Institutions’ new Integrated SOE Framework (intended to provide holistic guidance on SOE reform), the evaluation iden- tified a gap between the high incidence of recommendations on privatization and ownership reform in various Bank Group diagnostic work (for example, Financial Sector Assessment Programs, sector work on energy, and Country Private Sector Diagnostic [CPSDs]) and the low incidence of these topics in the lending portfolio. The Bank Group’s FY18 corporate strategic statement on Maximizing Finance for Development (MFD) and the embedded Cascade approach state a preference for reliance on private finance and private sector solutions and financing, which could provide the framework for future Bank Group work on SOE privatization and ownership reforms.  orld Bank Group SOE Reform Support by Type, FY08–18 Figure O.1. W (no ASA) 40 Number of interventions per year 35 30 25 20 15 10 5 0 Independent Evaluation Group World Bank Group 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Approval fiscal year Business and operations PFM and fiscal policy SOE ownership Competition and regulation Corporate governance Source: Independent Evaluation Group portfolio review and analysis. Note: The figure shows that 217 interventions supported SOEs’ corporate governance reform; 210 sup- ported business and operation reform; 236 sought to improve competition and regulation (of which 126 supported regulation); 167 supported SOE ownership reforms (of which 49 supported privatization); and 40 supported macrofiscal policy, public financial management, and debt. ASA = advisory services and analytics; FY = fiscal year; PFM = public financial management; SOE = state-owned enterprise. xi Effectiveness of Bank Group Support to SOE Reforms Portfolio Performance and Literature On average, the SOE reform portfolio in the financial and energy sectors met the World Bank and IFC corporate targets for project success. Overall, World Bank lending achieved a success rate of 78 percent against a target of 75 percent. Development policy lending achieved a success rate of 85 per- cent versus the investment project finance’s success rate of 67 percent, but the two instruments focused on tackling different SOE reform challenges. Policy lending was more focused upstream, seeking to improve public fi- nances; accountability, transparency, and oversight; or sector competition and productivity. Investment lending was more focused downstream, aiming to strengthen enterprise operational and financial performance as well as World’s Bank: An Evaluation of the World Bank Group’s Global Convening  Overview service delivery and quality. IFC achieved a success rate of 73 percent for investment services and 56 percent for advisory services against an overall target of 65 percent. Two evaluated MIGA guarantees both achieved their outcomes. Across reform types, evaluated privatization and corporate gover- nance reforms showed the highest success rates. In the financial sector, the Bank Group overall engages far more with state-owned commercial banks than with state-owned development banks. The success rate for develop- ment bank SOE reform interventions (77 percent) exceeded that for commer- cial bank interventions (69 percent). In the power sector, SOE reforms in the transmission and distribution subsectors were the most effective, followed by power generation. SOE reform interventions dealing with extractive in- dustries (petroleum, gas, and mining) were successful only half the time. IEG’s in-depth literature reviews yield evidence of SOE reform success in three of the five reform types: privatization, corporate governance reform, and competition. The literature consistently finds superior performance of private and privatized companies over public ones in both the energy and financial sectors and has especially negative findings about state-owned commercial banks. Rigorous national studies also yield evidence of the benefits of corporate governance reform for SOEs. There is strong evidence xii that competition improves SOE performance in both the power and financial sectors and augments the effectiveness of both privatization and regulatory reforms. The World Bank is aware of these benefits through the competitive neutrality framework it applies analytically. IFC also has a policy to focus fi- nancing on SOEs where there is a level playing field for private competition. MIGA’s policy is different. The Bank Group recognizes the importance of competition and competitive neutrality, but analytics on competition have been limited, as has been the application of competitive neutrality in IFC projects. The Markets and Com- petition Policy Assessment Tool, which is the Bank Group’s main diagnostic on competition, has been applied to only nine countries and one subregion during the evaluation period. In addition, only a small number of CPSDs to date have deployed the tool’s framework. IFC policies demand verification that a level playing field for competition exists before engaging with an SOE, but attention to competitive neutrality in project documentation is weak and uneven. MIGA policy emphasizes competitive conditions much less than IFC policy does. Factors of Success and Failure One country characteristic and several project factors are predictive of SOE reform intervention success. Some factors are within the Bank Group’s control, and some are outside of it. Econometric analysis of evaluated proj- ects confirmed the country characteristic and the project factors (originally Independent Evaluation Group World Bank Group emerging from IEG case studies, portfolio review, and the economic litera- ture) as significantly predictive of success. Control of corruption is a country characteristic strongly associated with SOE reform success. Other things being equal, a country with high control of corruption is more than twice as likely to see SOE reform interventions succeed as one with low control of corruption. In conditions of low control of corruption, all five major types of SOE reform are less likely to succeed, and it is more difficult to strengthen the governance, regulation, or performance of public enterprises. Overall, 26 percent of the SOE reform portfolio is in countries with low control of corruption. The marginal effect of weak control of corruption is large, but in practice xiii several factors mitigate its negative influence on SOE reform success, includ- ing selectivity for clients that display commitment, stronger supervision, and good project design and sequencing. In the evaluated portfolio, the success rate for countries with low control of corruption is about 67 percent, but it is 76 percent for those with high control of corruption. The contrast is more striking in low- and lower-middle-income countries—a 67 percent success rate where there is low control of corruption and an 85 percent success rate where control of corruption is high. Five project factors not directly controlled by the Bank Group (though po- tentially influenced by it) are strongly associated with the success of SOE reform interventions: » Client commitment to the reforms and reform activities. This underpinned success in multiple countries, including sustained power sector reforms moti- vated by government commitment to improving electricity supply and access. » Coordination among donors and other stakeholders. This generally contrib- World’s Bank: An Evaluation of the World Bank Group’s Global Convening  Overview utes to effectiveness, but it can be difficult to sustain. It can allow donors to work in complementary support of reform, leveraging one another’s resourc- es and influence. » Client institutional capacity and coordination. High institutional capacity often appeared as a factor of success, but weak coordination among client agencies hindered several reforms. » Political economy. This can work for or against reforms, but vested interests often frustrate them. Political economy factors influencing projects included shifts in commitment arising from political considerations, opposition from vested interests, and a variety of political difficulties caused by electoral cy- cles and regime change. » External shocks. Whether natural or human made, these can create opportu- nity by compelling action, but they can also disrupt reform progress. Four other project-level factors that the Bank Group controls directly are strongly associated with successful SOE reform interventions: » Project design, including appropriate choice of instrument, adaptation to local conditions, and simplicity (versus complexity). xiv » Supervision, including having in-country expertise during project implemen- tation (especially for investment projects). » A strong results framework with active monitoring and evaluation. » Sequencing and complementarity of interventions, including the link of ac- tivities to prior analytic work and internal collaboration. The evaluation found that collaboration among Bank Group institutions, though relatively rare, can provide complementary support that aids SOE reform success through both diagnostics and operations. Sequential analyt- ical and operational engagements built institutional and physical capacity, and the trust of underlying relationships carried reform momentum through difficult periods. Engagements that were more comprehensive involved both sequencing and complementarity of multiple interventions. For example, the CPSDs produced jointly by IFC and the World Bank consistently address SOE reform and feed into Systematic Country Diagnostics and Country Partner- ship Frameworks. Institutional collaboration to mobilize private financing is a key expectation raised in the MFD agenda and the Cascade approach, but neither approach spells out its implications (nor do sector strategies) regard- ing how Bank Group institutions can work together to support SOE reform. Although rare, experiences in several countries show the operational prom- ise of applying a Cascade approach in power generation. However, collabo- ration requires a balancing of benefits and costs. Recent IEG work suggests that such collaboration works best when the roles, division of labor, and Independent Evaluation Group World Bank Group responsibilities among the different Bank Group institutions and respective project teams are clear (World Bank, 2017f). Recommendations to Address Outstanding Challenges in SOE Reform The Bank Group can build on successful features of SOE reform in the financial and energy sectors by enhancing selectivity and mitigation of risk factors and by applying the MFD and its embedded Cascade approach to SOE reform. Enhancing selectivity by addressing corruption and competition. The evaluative evidence indicates that better SOE reform outcomes occur in the xv context of better control of corruption at the country level and competitive conditions at the sector and enterprise level. Both can be incorporated into ap- proaches to selectivity and mitigation of risks when planning for SOE reforms. Although the Bank Group SOE reform portfolio is concentrated in countries with stronger control of corruption, where reform is more likely to succeed, a substantial minority (26 percent) of interventions are in countries with weak control of corruption, where all types of SOE reform support are less likely to succeed. Regarding competitive conditions, the Bank Group recognizes the importance of competition and competitive neutrality principles, but it has not addressed these issues systematically enough and at scale through diagnostic or project work. Recommendation 1: The World Bank Group should apply a selectivity framework for SOE reform support that considers country governance conditions, control of corruption, and sector and enterprise-level competition. First, the Bank Group should adopt a more selective approach World’s Bank: An Evaluation of the World Bank Group’s Global Convening  Overview toward SOE engagement in countries with weak control of corruption, giving full attention to internal and external factors of success. Findings suggest that the Bank Group could ramp up engagement with clients where success is more likely. In conditions of weak control of corruption, one option would be to engage first in addressing overall governance quality before attempting SOE reform. Where disengagement on SOE reform is not possible or desir- able, close attention is needed to the factors that may mitigate corruption’s negative influence on SOE reform success, including selectivity toward clients who display commitment, stronger supervision, good (and simple) project design, and sequencing of activities. Next, the Bank Group should gear up capacity to conduct competition analysis at both the sector and proj- ect levels. The importance of competitive neutrality, especially considering IFC and (to a far lesser extent) MIGA policy requirements, indicates a need to ramp up project-level analysis by carrying out competition assessment systematically and by applying substantial up-front analytic capability to project-specific work on competitive neutrality. This would allow for greater selectivity toward competitive conditions that enhance SOE performance and for establishing up-front mitigating measures if competitive conditions were not conducive to success. xvi Improving Internal Coordination and Support Options by Applying MFD to SOE Reform This evaluation generally finds positive experiences when the Bank Group collaborates internally on SOE reform. Institutional collaboration to mobi- lize private financing and capabilities is a key expectation of the MFD agen- da and the Cascade approach. However, at the corporate level, there is room to spell out the implications of MFD and the Cascade approach for SOE re- form and to ensure that the new Integrated SOE Framework diagnostic treats privatization and PPPs as part of a comprehensive Bank Group approach. Recommendation 2: The World Bank Group should apply the MFD and its embedded Cascade approach for SOE reform. This would enhance internal coordination and mobilize private financing and capacity, especially for ownership reforms. First, the Bank Group should further de- velop and harmonize its diagnostic frameworks applied to SOE reform. This requires developing shared framing tools such as an Integrated SOE Frame- work and CPSD modules treating private sector options, including privatiza- tion and PPPs, for addressing SOE performance challenges. Second, the Bank Group could apply the Cascade approach in offering clients options for SOE reform that mobilize private financing and capacity through privatization and ownership reform. Along with recommendation 1, given appropriate country and sector conditions, there is greater room to apply the Cascade approach through a greater degree of and more routine World Bank, IFC, and Independent Evaluation Group World Bank Group MIGA coordination that builds on their respective comparative advantages. This can be piloted as a sequential process, with upstream interventions fo- cusing on any needed policy and regulatory reforms to create a level playing field for private entry and investment, combined with downstream use of Bank Group instruments to catalyze and mobilize private financing. With careful monitoring and evaluation, such a pilot could inform future efforts to realize the Cascade more fully as a systematic approach to SOE reform. xvii Management Response Management of the World Bank Group institutions would like to thank the In- dependent Evaluation Group (IEG) for its report, State Your Business! An Eval- uation of World Bank Group Support to the Reform of State-Owned Enterprises, FY08–18. Management is pleased to note that IEG’s findings suggest that, on average, the evaluated portfolio met World Bank and International Finance Corporation (IFC) targets for project success, and welcomes the suggestion that efforts continue to be made to improve development effectiveness. World Bank Management Response Management notes with satisfaction that World Bank lending, which accounts for more than 90 percent of the evaluated state-owned enterprise (SOE) reform portfolio volume, achieved a success rate of 78 percent, surpassing the target of 75 percent. Factors explaining the success of SOE reform, including both internal variables and external variables beyond World Bank control, resonate well with management. This is particularly true for the conclusion that “design quality and client commitment are frequently identified as suc- cess factors across all five types of SOE reform support, although the frequen- cy of other factors varies by area” (see page 37). Over the years, successive management self-assessments have recognized the influence of these vari- ables in project success, independent of country type and sector. State Your Business!  Management Response Management agrees with the report’s recommendation to enhance selectivity by more systematically addressing corruption and competition and is already working in that direction. The report shows that only 26 percent of activities are implemented in environments with low control of corruption, suggesting already strong selectivity. Enhanced selectivity should therefore help prior- itize the type of support to SOE reform that is better tailored to a particular intervention in the country context rather than excluding countries. Such support should be accompanied, as suggested by the report, by management’s sustained attention to all factors that mitigate corruption’s negative influ- ence on SOE, particularly those within the World Bank’s immediate influence, xviii such as putting a premium on client commitment, stronger supervision, good project design, and sequencing of activities. These considerations are already part of the Systematic Country Diagnostic and the prioritization that takes place during Country Partnership Framework design. This is proven by the fact that the World Bank has achieved, as stated in the report, a success rate of 67 percent even in countries with low control of corruption. Independent Evaluation Group World Bank Group xix Management Action Record IEG Findings and Conclusions Enhancing selectivity: corruption and competition The World Bank Group has a higher rate of successful outcomes in countries with better control of corruption and competitive conditions at the sector and enterprise level. State-owned enterprises (SOEs) perform better in both focal sectors if competitive conditions prevail at the sector and enterprise level. Both can be incorporated into approaches to selectivity and mitigation of risks when planning SOE reforms. A substantial minority of interventions (26 percent) are in countries with weak control of corruption. In conditions of low control of corruption, all five major types of SOE reforms are less likely to succeed. There is an opportunity for greater Bank Group traction on SOE reform through a more selective approach. Certain internal and external factors can at least mitigate the risk of weak governance to project success. Competition and competitive neutrality at the sector and enterprise level of SOEs can strongly influence performance. Improving competition can enhance the success of other reforms. However, analytics on competition, especially at the project level, are insufficient. International Finance Corporation (IFC) and Multilateral Investment Guarantee Agency (MIGA) policies demand verification that a level playing field for competition exists before engaging with an SOE, but attention to State Your Business!  Management Record Action competitive neutrality in project documentation is uneven and often missing. Realizing SOE reform through a more coordinated and consistent application of Maximizing Finance for Development (MFD) and its embedded Cascade approach This evaluation generally finds positive experiences when the Bank Group collaborates internally on SOE reform. Institutional collaboration to mobilize private financing and capabilities is a key expectation of the Cascade approach. Recent Independent Evaluation Group work suggests that such collaboration works best when the roles, division of labor, and responsibilities among the different Bank Group institutions and respective project teams are clear. Realistically, internal coordination can require additional resources that require balancing of benefits and costs. At the corporate level, there is room to spell out the implications of MFD and xx the Cascade approach for SOE reform. There is a gap between the high incidence of recommendations on privatization and ownership reform in diagnostic work and the low incidence of privatization in the portfolio. Despite priority given to private solutions in the MFD, the Equitable Growth, Finance, and Institutions’ Integrated SOE Framework does not treat privatization (or public-private partnerships) as part of the Bank Group’s approach. However, interviews suggest that after a period of client disinterest and political sensitivity, demand for privatization support has been growing. IEG Recommendations Recommendation 1. The World Bank Group should apply a selectivity framework for SOE reform support that considers country governance conditions, control of corruption, and sector and enterprise- level competition. First, findings suggest that the Bank Group could ramp up engagement with clients where success is more likely. In conditions of weak control of corruption, one option would be to engage first in addressing overall governance quality before attempting SOE reform. Where disengagement on SOE reform is not possible or desirable, close attention is needed to the factors that may mitigate corruption’s negative influence on SOE reform success. Next, the Bank Group should gear up capacity to conduct competition analysis, especially at the project level. The importance of competitive neutrality (the idea that SOEs should be on a level playing field with potential private competitors), especially considering IFC and, to a lesser extent, MIGA policies, indicates a need to ramp up project-level analysis by carrying out competition assessment more systematically and by applying substantial up-front analytic capability to project-specific work on Independent Evaluation Group World Bank Group competitive neutrality. This would allow for greater selectivity toward competitive conditions that would enhance SOE performance and for establishing up-front mitigating measures if competitive conditions were not conducive to success. Recommendation 2. The World Bank Group should apply the MFD and its embedded Cascade approach for SOE reform. This would enhance consistent internal coordination and mobilize private financing and capacity, especially for ownership reforms. First, the Bank Group should further develop and harmonize its diagnostic frameworks applied to SOE reform. This requires developing shared framing tools such as an Integrated SOE Framework and Country Private Sector Diagnostic modules treating private sector options, including privatization and public-private partnerships, for addressing SOE performance challenges. Second, xxi the Bank Group could apply the Cascade approach in offering clients options for SOE reform that mobilize private financing and capacity through privatization and ownership reform. Along with recommendation 1, given appropriate country and sector conditions, there is greater room to apply the MFD and its Cascade approach through a greater degree of and more routine World Bank, IFC, and MIGA coordination that builds on their comparative advantages. This can be piloted as a sequential process, with upstream interventions focusing on any needed policy and regulatory reforms to create a level playing field for private entry and investment, combined with downstream use of Bank Group instruments to catalyze and mobilize private financing. With careful monitoring and evaluation, such a pilot could inform future efforts to realize the Cascade more fully as a systematic approach to SOE reform. State Your Business!  Management Record Action xxii Report to the Board from the Committee on Development Effectiveness The Committee on Development Effectiveness met to consider the report enti- tled State Your Business! An Evaluation of World Bank Group Support to the Reform of State-Owned Enterprises, FY08–18, and the draft management response. The committee welcomed the Independent Evaluation Group’s first system- atic assessment of the Bank Group’s support for the reform of State-Owned Enterprises (SOEs) for 2008–18 and commended it for the quality and learn- ing orientation of the report. Members were pleased to learn that manage- ment broadly agreed with the Independent Evaluation Group’s findings and that it was already working in the direction of the report’s recommendations. Members welcomed the report’s findings that SOEs can play a major role in both developing and emerging economies to achieve economic, social, and political objectives, including promoting growth; delivering and extending access to services; filling market gaps; developing key sectors or regions; creating jobs; and addressing issues of heightened national priority or secu- rity. They noted the use of SOEs by governments and donors, including the Independent Evaluation Group World Bank Group Bank Group, to channel subsidies or benefits such as deferred payment of utility bills or subsidized credit to enterprises in response to the coronavi- rus (COVID–19) crisis. Members highlighted the importance of SOE reforms to help client countries achieve the Sustainable Development Goals and recognized that even in normal times, SOEs’ mixed institutional mandates and political importance often pose performance, financial, and governance challenges. Members acknowledged the evaluation’s findings that reforms have higher rates of successful outcomes in countries with better control of corruption and competitive conditions at the sector and enterprise level and that positive experiences from collaboration across Bank Group institutions are essential for mobilizing private financing and capabilities consistent with the Maximizing Finance for Development and Cascade approach. xxiii Members underscored the importance of corporate governance and corrup- tion, especially in challenging market environments, noting that this should not inadvertently lead to a movement away from these challenging environ- ments, particularly in International Development Association and fragile and conflict-affected situation countries. Members noted that SOE reform is broader than ownership and privatization, and they underscored the impor- tance of regulatory reforms to promote competition and remove potential bottlenecks for private players. Members noted that financial sustainability issues in many SOEs and their impact on general government budget are of- ten a challenge in client countries and encouraged management to continue paying close attention to public financial management reforms. The committee recognized that SOE reforms could be used to help client countries overcome public sector and multilateral development bank limit- ed financing capacity and that by promoting, when relevant, a more robust and commercially organized governance of SOEs, including with the use of de-risking instruments, the Bank Group could set clients on a trajectory from sovereign-guaranteed public sector financing to commercial financing. Members highlighted that mobilizing private financing for SOE reform can have positive externalities, notably in terms of transparency, and under- scored that the sequencing of reforms must be driven by country context, making policy analysis an essential part of the assessment. They recognized that SOEs can impose barriers to private participation in sectors where their dominant presence enables anticompetitive behavior. Members noted that SOEs represent a fertile ground for the implementation of the Cascade approach and for the Bank Group to strengthen its collaboration efforts, State Your Business!  Report to the Board and they encouraged management to continue assisting client countries in creating an enabling environment where private sector participants can also invest and play an important role in providing various services. xxiv 1 | State-Owned Enterprise Challenges and World Bank Group Reforms Highlights This chapter presents why World Bank Group support for state- owned enterprise (SOE) reform matters, the pattern of Bank Group engagement, and a typology of major Bank Group support interven- tions. It assesses the contribution of the World Bank Group’s three main institutions to enhancing development outcomes through their support of SOE reform: what works and what does not, the effective- ness of the Bank Group’s various approaches, factors that explain success and failure, and the strengths and weaknesses of the Bank Group’s approach. To allow for greater depth, it focuses on two lead- ing sectors for SOE support: finance and energy. SOEs play a major role in many developing and emerging econ- omies, where governments use them to achieve economic, social, and political objectives: to deliver and extend access to services, fill gaps in markets, develop key sectors or regions, and provide employment. However, the mixed institutional mandates of SOEs and their politi- cal importance often pose performance and governance challeng- es. This can mute transparency and accountability, making over- sight and regulation difficult. The Bank Group, from fiscal years 2008 through 2018, initiated 1,008 projects with 2,187 components (interventions) that support- ed the reform of SOEs in the financial and energy sectors, with an estimated combined value of $71.7 billion in financing. This 1  involved financial, technical, and advisory support for both policy and institutional reforms (upstream) and enterprise-level activities (downstream). This evaluation focuses on five major types of reforms: corporate governance; business and operations; strengthening competition and regulation in SOE markets; privatization and other ownership reform; and macro, fiscal, and public financial management reforms. 2   Evaluation Purpose and Scope State-owned enterprises (SOEs) have distinctive characteristics, including control by the state, legal and financial autonomy from the state, and par- ticipation in the productive sector (Raballand et al. 2015). Corporate control may be exercised through ownership, administrative and technical manage- ment, interlocking of directorates, and regulatory oversight (Farazi, Feyen, and Rocha 2011). Unlike public agencies, SOEs benefit from a level of auton- omy because of their productive activity. Special laws that are different from private sector laws often govern SOEs (World Bank 2014a).  tate-Owned Enterprises in the Response to Box 1.1. S the Coronavirus Pandemic State-owned enterprises (SOEs) are an inviting channel for government responses to economic crises and natural disasters, given the degree of direct government control. The Independent Evaluation Group found such responses in the portfolio—for exam- ple, responding to the 2008 global financial crisis in part by using state-owned banks (such as the Sri Lanka Small and Medium Enterprise Development Facility, 2011) and responding to floods by restoring power distribution (Serbia Floods Emergency Recov- ery Project, 2015). The International Finance Corporation helped Russian Federation banks (public and private) deal with small and medium enterprises’ nonperforming loans after the financial crisis (the Russian Federation Financial Management Crisis Independent Evaluation Group World Bank Group Management Project), and the World Bank helped Nigeria create an asset manage- ment company to restore financial health to the banking sector. SOEs can provide convenient vehicles to channel resources to adversely affected firms and households (for example, through bank loans or power utility payment abey- ance). Although some crisis response efforts also aimed to reform SOEs, many aimed to use them to reach targeted beneficiaries. Several of the World Bank’s initial private sector support projects in response to the crisis are channeled through state develop- ment banks. The literature provides some explicit precautions: » Temporary subsidies and benefits after crises may be “policy traps” that are politi- cally difficult to reverse. This is true for both utilities and banks. (continued) 3  tate-Owned Enterprises in the Response to Box 1.1. S the Coronavirus Pandemic (continued) » Use of state-owned banks to respond to crises may be costly and inefficient. The World Bank Group’s Global Financial Development Report 2013: Rethinking the Role of the State in Finance finds that although SOE lending may be “less procyclical” than private lending, it “did not always target the most constrained borrowers” and is associated with a “deterioration of the quality of financial intermediation.” Special attention must be paid to governance, including assuring “adequate risk manage- ment processes are in place.” Sources: Bril-Mascarenhas and Post 2012; World Bank 2012a; Independent Evaluation Group state- owned financial institutions deep dive; Independent Evaluation Group portfolio review and analysis. SOEs play a critical role in the energy and financial sectors in many develop- ing and emerging economies. Most countries still depend on SOEs to provide power. SOEs accounted for 71 percent of the Morgan Stanley Capital Interna- tional (MSCI) Emerging Market Index in utilities, 56 percent in energy, and 39 percent in the financial sector in 2020. State control is prevalent in the oil and gas sector, with about 90 percent ownership of reserves and 55 percent of production. Although state ownership in commercial banks declined from 67 percent of total banking assets in 1970 to 22 percent in 2009, SOEs often retain a dominant role in banking (World Bank 2012a). In emerging markets such as China and India, SOEs hold more than half of banking system assets (Bank Group 2012). Singapore successfully used SOEs to drive development and industrialization after its independence in 1965, with successes in di- verse fields including oil refining, petrochemicals, and development finance State Your Business!  Chapter 1 (PwC 2015). Governments use SOEs to pursue economic, social, and political objectives alongside their commercial objectives.1 The mixed objectives demanded of SOEs can include contribution to employment creation, poverty alleviation, fiscal stability, spatial or sectoral development, environmental protection, and sector regulation. For example, India relied primarily on state-owned financial institutions (SOFIs) to implement its Jan Dhan Yojana program, under which about 300 million basic accounts were opened in a short peri- 4 od. In Kenya, the Kenya Power and Lighting Company was the main vehicle for the government’s drive for universal electricity access, achieving more than 1 million new connections a year. During the coronavirus pandemic, many governments have used SOEs to channel resources to adversely affect- ed firms and households (box 1.1). Some SOEs have been run well and have made important contributions to economies. SOEs’ multiple objectives pose several governance and management challenges. Mixed objectives and weak oversight obscure accountability, exacerbate princi- pal-agent challenges, and weaken incentives for performance (box 1.2). SOEs also reflect the desire of the state or political groups to exert political influence over economic outcomes and resource allocation. Although some SOEs are run well, many others suffer from low productivity and efficiency and have a detrimental impact on growth and consumer access to services. Poor financial performance and management practices can generate substantial public fiscal losses, debt, or contingent liabilities.2 SOEs frequently lack adequate governance oversight arrangements, regulation, and levels of transparency and disclosure, which can foster mismanagement, corruption, and underperformance.3 Yet SOEs can also impose barriers to private participation in sectors where their dominant presence enables anticompetitive behavior, often with government protection or subsidy.4  hree Perspectives on State-Owned Enterprises in the Litera- Box 1.2. T ture (Not Mutually Exclusive) Independent Evaluation Group World Bank Group Agency view: There is a discrepancy between the objectives of managers (the agents) and of owners (the principals). Although governments may seek to maximize social welfare, their agents may lack the incentive to maximize the use of resources toward this end. Social view: Governments create state-owned enterprises to address market failures and improve social welfare, mixing profitability goals with social objectives. These mixed objectives create challenges for monitoring outcomes and performance. Political view: State-owned enterprises, in some cases, can be mechanisms for politi- cians to pursue their individual goals, often leading to economic distortion or inequita- ble distribution of resources. Source: Independent Evaluation Group structured literature review. 5 Evaluation Approach This evaluation assesses the contribution of the Bank Group’s three main institutions from fiscal year (FY)08 through FY18 to enhancing development outcomes through their support of SOE reform. In this evaluation, the focus is limited to support intended to improve SOE performance at the national, sectoral, or enterprise level, where it imposes a constraint on development. The evaluation excludes interventions where SOE reform was not the chal- lenge being addressed, but rather SOEs were used as an instrument to ad- dress a development challenge (such as when a line of credit is channeled through a state-owned bank; World Bank 2019f). To allow for greater depth, the evaluation focuses on the two leading sectors for SOE reform support (identified at the approach stage): the financial sector and the energy sector. The evaluation answers four questions: » What is the Bank Group doing to support SOE reform? » How effective are Bank Group SOE reform interventions, and where are these strengths or gaps? » What internal (directly under the Bank Group’s control) and external factors explain the success or failure of Bank Group SOE reform interventions? » Does the Bank Group have a robust approach to achieving development im- pact through SOE reform, considering client priorities and needs and its own goals and principles? To answer these questions, the evaluation employs a theory of change (ap- pendix A) to inform the analysis using mixed methods with quantitative and qualitative evidence: State Your Business!  Chapter 1 » The analysis is multilevel, looking at country, sector, project, engagement area, and intervention mechanisms, covering both upstream SOE reforms (policy, regulatory, and institutional) and downstream SOE reforms (enter- prise level). It excludes Bank Group projects that use SOEs to deliver services without trying to reform them. » The mixed methods include portfolio review and analysis, eight country case studies, subject and sector deep dive studies, a structured literature review, 6 a review of country strategies and diagnostics (including Financial Sector Assessment Programs), and econometric analysis. The financial and energy sectors each have unique features and challenges: » In the financial sector, there are state-owned commercial banks (typically taking deposits and offering credit and other services), development banks (financing public development priorities), and nonbank financial institutions in areas like insurance and pensions. Government ownership of financial institu- tions is understood to have the potential both to overcome market failures in promoting socially and economically desirable investments and to provide a vehicle to channel finance to strategically important sectors or firms. The Bank Group understands national development banks to be key “to help crowd-in the private sector to finance projects with high developmental impact such as infrastructure or projects that can yield a greater public good but which the private sector may not be interested in funding directly” (Pazarbasioglu 2017). Challenges include the performance of such institutions in practice, the diffi- culties of aligning their actions with policy intentions, and the effects of such institutions on financial system competition, efficiency, and stability. » Within the energy sector, the evaluation covers power companies (often util- ities) involved in distribution, generation, or transmission and those engaged in energy extraction. A recent World Bank reconsideration of its approach to power sector reforms recognizes that “among the best-performing power sectors in the developing world are some that fully implemented market-ori- Independent Evaluation Group World Bank Group ented reforms, as well as others that retained a dominant and competent state-owned utility guided by strong policy mandates, combined with a more gradualist and targeted role for the private sector. This reality makes a case for greater pluralism of approaches going forward” (Foster et al. 2020). Recent World Bank work suggests that in less institutionally mature environments, private sector participation is best limited to power generation.5 However, the literature (appendix G) finds governance and operational challenges for many power sector SOEs: financially unsustainable tariffs; mandated cross-subsi- dies; weak or inefficient regulatory environments; poor sectoral planning; high network losses, hidden costs or liabilities; or ambitious government access goals lacking adequate subsidization. Traditionally, the power sector was regarded as a natural monopoly, but technological advances and the 7 identification of huge deadweight losses have changed this; yet few govern- ments or regulators explicitly monitor the adequacy and reliability of energy supply, let alone require their disclosure. Professionalization of SOE staff is also challenging. Five Types of Bank Group Support for SOE Reform From FY08–18, the Bank Group initiated 1,009 projects with 2,187 com- ponents (interventions) that supported the reform of SOEs in the financial and energy sectors, with an estimated combined value of $71.7 billion in financing.6 This involved financial, technical, and advisory support for both policy and institutional reforms (upstream) and enterprise-level activities (downstream). To analyze these projects, the Independent Evaluation Group (IEG) took a representative sample of 88 percent of World Bank, Internation- al Finance Corporation (IFC), and Multilateral Investment Guarantee Agency (MIGA) financing and IFC advisory projects (374) and 20 percent of World Bank advisory services and analytics (ASA) projects (116 sampled). This maps to a total portfolio of 421 financing projects plus 1,184 World Bank ASA projects. Within the financing projects, 893 interventions (components) were within the scope of this evaluation. To expand the pool of evaluated SOE reform projects, IEG also analyzed 132 qualifying projects evaluated between FY08 and FY19 but approved between FY02 and FY07. This evaluation focuses on five major types of SOE reforms supported by Bank Group operations and activities (figure 1.1). These reforms are corpo- rate governance; business and operations; competition and regulation; pri- State Your Business!  Chapter 1 vatization and other ownership; and macrofiscal, and public financial man- agement (PFM). Together, interventions in these areas compose 87 percent of both interventions and financing for SOE reform and are components of 96 percent of projects identified as being in scope. The interventions can be complementary if more than one is used to achieve deeper reform. As shown in figure 1.1, there is considerable year-to-year variation in their use. 8  ank Group SOE Reform Support by Type, FY08–18, and by Figure 1.1. B Institution (no ASA) a. Evolution of SOE reform types 40 35 Interventions per year (no.) 30 25 20 15 10 5 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Approval fiscal year Business and operations PFM and fiscal policy SOE ownership Competition and regulation Corporate governance b. Distribution of SOE reform types by institution MIGA 62 38 IFC AS 28 37 26 10 IFC IS 38 34 24 3 World Bank ASA 14 15 27 37 6 World Bank lending 21 15 28 30 6 Independent Evaluation Group World Bank Group 0 20 40 60 80 100 Distribution (percent) Business and operations PFM and fiscal policy SOE ownership Competition and regulation Corporate governance Source: Independent Evaluation Group portfolio review and analysis. Note: The figures show that 217 interventions supported SOE corporate governance reform; 210 sup- ported business and operation reform; 236 sought to improve competition and regulation (of which 126 supported regulation); 167 supported SOE ownership reforms (of which 49 supported privatization); and 40 supported macrofiscal, policy, public financial management, and debt. Percentages may not add up to 100 due to rounding. ASA = advisory services and analytics; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; PFM = public financial management; SOE = state- 9 owned enterprise. Business and Operations Reforms The largest share of Bank Group–supported reforms aims at improving SOEs’ operations and business practices. Reforms in business and operations aim to improve operational and financial performance and enhance service qual- ity. Projects supporting business and operations reform compose 24 percent of SOE reform interventions but account for 50 percent of commitment val- ue. They are the leading form of energy sector reform support and the sec- ond most popular form in the financial sector. By institution, these reforms are in almost 60 percent of MIGA guarantees, 40 percent of IFC SOE invest- ments, and 20 percent of World Bank lending projects. Ten percent of Bank Group financing support to reform SOEs focused on strengthening their financial management, primarily targeting financial sustainability (including the restructuring or rehabilitation of debts), enhancing revenue collection, and strengthening creditworthiness or expanding financial options.7 About one-third of Bank Group commitments in this category support physical infrastructure improvements (almost entirely for power compa- nies), one-quarter support human resource management, 16 percent support service or product quality improvement, 14 percent support operational or process efficiency improvement, and 11 percent support organizational restructuring. In the World Bank, the Energy and Extractives Global Practice (GP) leads many of these commitments, focusing on improving the business and operations of individual state-owned energy companies. For example, in 2017, to strengthen financing of Bangladesh’s power supply, MIGA provided a guarantee to Standard Chartered Bank for its loan to the SOE North-West Power Generation Company Limited in Bangladesh to build, install, and op- erate a 220 megawatt, dual-fuel, combined cycle power plant. In the finan- State Your Business!  Chapter 1 cial sector, examples include a 2009 World Bank–financed Private Housing Finance Markets Strengthening Project (P112258) in Mexico, which aimed to improve the technical capacity of Sociedad Hipotecaria Federal, a SOFI, to expand access to lower-income groups. The SOFI gained capacity through reengineering, the creation of a new department, simplification of proce- dures, and improved risk monitoring. IFC complemented the World Bank project with advisory and fee-based services. 10 Corporate Governance Reforms Corporate governance reforms are pursued to improve SOE performance where government intends to retain ownership or as a path to privatization. Enterprises, whether public or private, are known to perform better with the right combination of incentives and the institutions to secure those incen- tives.8 Corporate governance arrangements shape internal incentives, bal- ancing a desire for managers to have enough discretion to run the company without unduly interfering with a desire to keep them accountable to the interests and objectives of owners and other stakeholders. Corporate gover- nance reforms composed 24 percent of non-ASA Bank Group interventions in the portfolio and 12 percent of commitments, including 29 percent of World Bank lending, 11 percent of IFC advisory services, and 4 percent of IFC investment services. In addition, corporate governance reform is a subject of 38 percent of World Bank ASA interventions. For financial sector SOE reform, corporate governance reform is the most popular form of interven- tion, ahead of business and operations reform. In the power sector, the World Bank’s flagship study finds that many reform efforts began with the corpora- tization of power utilities (World Bank 2019c). The Bank Group has long been a champion for good corporate governance of SOEs, both as a step toward divestiture and as a self-standing means to strengthen performance. Improved corporate governance is used to achieve other ends—for example, to improve performance, service delivery, finan- Independent Evaluation Group World Bank Group cial sustainability, governance, and access to private capital—but is often described as a project objective on its own. Corporate governance reforms often involve the following elements: » Clarifying SOE objectives; » Improving the legal and regulatory framework for SOE governance; » Strengthening the state’s role as owner or shareholder; » Professionalizing SOE boards and management; » Promoting the financial sustainability of SOEs; and » Enhancing the transparency and accountability of SOEs. 11 Corporate governance reform may accompany corporatization (the estab- lishment of SOEs as corporate entities), but it may also be applied to existing public companies. Bank Group efforts to strengthen the governance of SOEs date at least to the 1980s and were described influentially in a 1995 report (Muir and Saba 1995). More recently, in 2014, the Finance and Markets and the Governance GP teams jointly produced a tool kit on SOE corporate governance rooted in Organisation for Economic Co-operation and Development guidelines. In the SOE reform evaluation portfolio, the level of activity in corporate gover- nance was strong but has fluctuated since FY14 (see figure 1.1). IFC requires corporate governance analysis for every investment transaction as part of its due diligence process and strives for client commitment to good corporate governance practices, including protection of shareholder rights, account- ability to investors and stakeholders, quality of the control environment, and disclosure and transparency practices. Support for corporate governance is often bundled with other reforms because it is seen as a vehicle for enhanced access, efficiency, and quality of services. In the evaluation portfolio, sector reforms tend to focus on the enterprise level, and national-level support often sets standards or builds in- stitutional capacity to regulate or implement policy. The World Bank’s work is most often led by the Macroeconomics, Trade, and Investment; Finance, Competitiveness, and Innovation; and Energy and Extractives GPs. A high percentage but small number of Governance GP interventions were in cor- porate governance. In Kenya, for example, improving the utilities’ corporate governance was a core element of Bank Group engagements in energy sector reform. Capacity-building assistance to improve corporate governance was State Your Business!  Chapter 1 provided to several SOEs in the electricity sector, including the generation company, the distribution company, and the transmission company. For ex- ample, the generation company, KenGen, benefited from a World Bank–sup- ported, comprehensive Corporate Governance Assessment, which informed the KenGen Guarantee Project (P162422), which supports long-term private capital mobilization by the company through a commercial risk guarantee. In another financial sector example, IFC invested in the state-owned Sri Lanka Life Insurance Corporation, taking a seat on the board and working 12 to strengthen corporate governance to improve the SOE’s credibility with investors in preparation for privatization. Privatization and Ownership Reforms Privatization, often recommended in analytic work, has shown a declin- ing trend in financing. Privatization and ownership reforms are pursued to improve SOE performance where a government intends to relinquish all or some portion of its ownership. Beyond privatization, other own- ership reforms include promoting public-private partnerships (PPPs) or other partnership arrangements, opening to private sector investment (including foreign investment in SOEs), and setting up new SOEs while liquidating old ones. Privatization and ownership reforms are employed to obtain many of the objectives of SOE reform, including improvement of competition, enterprise productivity, and innovation (for 24 percent of the Bank Group SOE projects); operational or financial performance (for 22 percent of projects); sectoral efficiency (22 percent); public finances (13 percent); and service quality or delivery (13 percent). Own- ership reform is far more popular in energy (especially in generation) than it is in the financial sector. Privatization is often recommended in analytic work such as Financial Sector Assessment Programs (box 1.3); it has seen ups and downs over several decades (box 1.4) and has shown a declining trend in financing (figure 1.2), with no interventions for 2013 and 2018 and a parallel trend in commitments. This seems at odds with Independent Evaluation Group World Bank Group the intention of the Bank Group’s FY18 corporate strategic statement on Maximizing Finance for Development (MFD) and its embedded Cascade approach, which state a preference for reliance on private finance and private sector solutions. The MFD intends to harness “the power of the private sector and enhance market creation to meet the twin goals and the SDGs [Sustainable Development Goals].” The aim is to “help client countries pursue sustainable private sector solutions [where] they can help achieve development goals, while preserving scarce public resourc- es where they are needed most” (World Bank 2017a). 13  inancial Sector Assessment Programs’ Treatment of State- Box 1.3. F Owned Financial Institution Reform In a stratified sample of 29 joint International Monetary Fund–World Bank Financial Sector Assessment Program (FSAP) reports, the Independent Evaluation Group found that 76 percent substantially discussed state-owned financial institutions and 66 per- cent made recommendations. Two-thirds focused on commercial banks, 24 percent on development banks, and 28 percent on nonbank financial institutions. Upstream policy and institutional recommendations focus on sectoral regulato- ry frameworks (41 percent), governance (38 percent), and ownership (17 percent). Downstream enterprise-level reform recommendations focus on firm-level ownership (51 percent), financial management (17 percent), and corporate governance and busi- ness and operational management (17 percent). Examples include the following: » The Botswana FSAP identifies challenges to state-owned financial institutions in delivering financial services to the poor and rural population. It recommends giving full supervisory authority to the Bank of Botswana for statutory banks and licensing these institutions as a prelude to their privatization. » The Russian Federation FSAP recommends a new legal and regulatory frame- work to increase board effectiveness, the gradual privatization of banks, and revisiting state hybrid and development finance institutions. Source: Independent Evaluation Group Financial Sector Assessment Program review. Overall, for the entire evaluation period, ownership reforms constitut- ed more than one-third of IFC’s investment interventions and more than State Your Business!  Chapter 1 40 percent of its advisory interventions. One-third of MIGA guarantees also supported ownership reform, especially focused on power generation. SOE privatization support, which has waxed and waned over several decades (box 1.4), has been rare recently. For the World Bank, 14 percent of lending interventions and 15 percent of ASA for SOE reform supported ownership reforms of all types. In 2007, for example, IFC supported the privatization of the Energy Development Corporation (25839) in the Philippines, the largest producer of geothermal power. IFC also supported its capital expenditure 14 program and improved corporate governance practices. In 2012, MIGA issued a guarantee (M1367) to support, in Indonesia, a PT Rajamandala Electric Power (an independent power provider) hydropower plant and transmission line investments against the risks of expropriation, transfer restriction, war and civil disturbance, and breach of contract covering the contractual obliga- tions of PT Perusahaan Listrik Negara (the state electricity company) under the power purchase agreement. nterventions Supporting State-Owned Enterprise Privatization Figure 1.2. I 25 700 Commitments per year ($, millions) 600 Interventions per year (no.) 20 500 15 400 10 300 200 5 100 0 - 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Year World Bank ASA (no. interventions) Financing (no. interventions) Commitment ($, millions) Source: Independent Evaluation Group portfolio review and analysis, projected to population of projects. Note: ASA = advisory services and analytics. Independent Evaluation Group World Bank Group  endulum Swings on Privatization in Bank Group State-Owned Box 1.4. P Enterprise Reform World Bank Group state-owned enterprise (SOE) engagements in the 1980s stressed improving operational and financial performance by rationalizing the policy framework, improving governments’ management of their portfolios, reducing SOE fiscal expendi- tures and improving their revenues, and assisting SOEs to improve their financial and operational performance. By the early 1990s, many donors were pushing hard for privatization of SOEs given the disappointing results of past SOE reforms, the altered popular understanding of the government’s role regarding productive enterprises, and assumptions about the (continued) 15  endulum Swings on Privatization in Bank Group State-Owned Box 1.4. P Enterprise Reform (continued) ease of transferring Western systems to dramatically different settings. Between 1988 and 2005, global privatization proceeds were an estimated $2.6 trillion. Aligned with this global trend and guided by concern about government failure, World Bank sup- port for privatization grew rapidly, from 14 percent of World Bank SOE-related loans in the 1980s to 52 percent of operations in the 1990s. Between 1990 and 2003, the Bank Group assisted 120 countries to carry out 7,860 transactions generating nearly $410 billion in proceeds. However, the wave of privatization gave rise to critiques. Questions about attribution arose because factors other than ownership may have accounted for performance improvements, such as the introduction of hard budget constraints and increased competition. Performance comparisons were questioned on the grounds that bet- ter-performing SOEs were “cherry picked” for privatization. Additionally, governments faced difficulties in carrying out privatization, for example in creating, monitoring, and enforcing contracts in infrastructure, along with some failures that led to renational- ization. Then there were the failures in the regulation of privatized industries, the rise of powerful oligarchs in the former Soviet Union, the cases of asset stripping, and the popular opposition to privatization from citizens and workers. Furthermore, in the 2000s, China was seen to offer an alternative model involving substantial state owner- ship and control of enterprises, despite large-scale privatizations of its own. Thus, in the 2000s, there was a step back from privatization. Instead, the emphasis was on corporate governance reform to improve SOE performance through super- vision, transparency, and accountability (along with a focus on other aspects of SOE business and operations) to improve quality, cost recovery, and controls. Tools such as public-private partnerships and risk guarantees aimed to bring some degree of private State Your Business!  Chapter 1 investment and private operation to SOE service delivery. Source: Independent Evaluation Group privatization deep dive. Competition and Regulation in SOE Markets Strengthening competition and regulation in SOE markets can help align 16 the activities of SOEs with development and policy objectives and level the playing field among SOEs and potential private competitors. This area of re- form is supported in more than one-quarter of World Bank lending and ASA and IFC investment and advisory services interventions. It is not a feature of MIGA’s work. Regulation can shape sectoral pricing (29 percent of regulatory interventions) or support the enactment of new laws, regulations, or regulatory institutions (26 percent). Strong sector regulation can set the framework for private partic- ipation and shape incentives for efficient service delivery. Power sector reform often emphasizes creating and empowering an independent regulatory agency, with a strong orientation toward technically driven tariff-setting procedures (Pardina and Schiro 2018). In the energy sector, reforms were more likely to focus on pricing (41 percent) and sector strategies (20 percent). Financial viability of the energy sector through cost recovery pricing is needed to attract private investment, ensure reliable supply, meet universal access targets, and minimize negative macrofiscal impacts (Huenteler et al. 2017). Cross-subsidies built into tariffs add complexity to the policies and politics of pricing reforms. For example, the 2009 development policy operation (DPO) for Burkina Faso (P099011) supported an enhanced regulatory framework with a transparent tariff-setting mechanism for power SOEs, along with establishing a regulator. In the financial sector, reforms focused more on sector laws and regulations and institutions to enforce them for SOEs and other financial institutions. This was the case, for example, for the 2017 World Bank Myanmar Financial Sector Development Project (P154389). Some regulatory reform interventions reset sector policy. A DPO for Senegal Independent Evaluation Group World Bank Group in FY13 (First Governance and Growth Support Project, P128284) sought to improve energy sector efficiency and service through the adoption of a new Energy Sector Development Policy Letter, an action plan, and a financial and operational utility restructuring plan. The letter addressed gaps in the energy sector legal and regulatory framework for both the electricity and hydro- carbon sectors. In Vietnam, a series of Poverty Reduction Support Credits (PRSCs) and DPOs supported sector strategy reform. The PRSCs broadly sup- ported the enactment of an electricity law while working to improve sector strategies for the gas and electricity subsectors and supporting the adoption of market-based pricing mechanisms for electricity. DPOs focused on key 17 sector policy areas: development of a competitive power market, power sector restructuring, electricity tariff reform, and demand-side energy efficiency. Competition work often seeks to remedy weak incentives for SOEs to be- have efficiently and contribute to economic development through improved productivity. In many countries, SOEs enjoy substantial market power, which may extend to both markets for goods and services and input markets. This market power can arise from small market size (or poorly developed markets, weak regulations, and poor oversight of competition) or weak policies (or enforcement) governing the ownership and treatment of SOEs. Enhancing competition is known to improve the performance of enterprises, whether private or public. Leading ways the Bank Group promotes competition are opening entry or actively crowding-in the private sector, promoting PPPs and privatization, and reforming tariffs. Competition objectives are often interwoven with others. Although not all SOE activities are in competitive market segments,9 IEG’s portfolio review found 132 interventions seeking to strengthen compe- tition, innovation, and productivity. Their frequency increased in the mid- 2010s. IEG estimates that more than 125 World Bank ASAs had competition, innovation, or productivity objectives. The 2015 Sustaining Shared Growth development policy loan for Turkey (P146322) exemplifies such support to strengthen competition. It supported competition and transparency in the energy sector through enactment of the electricity market law, which lim- its SOEs’ role in the sector and ensures the development of a competitive environment for electricity markets. The operation benefited from extensive analytical work. In 2005, IFC began assisting the Bank of Beijing through investment and advisory services to prepare it to compete regionally. State Your Business!  Chapter 1 IFC has declared in recent years that it considers principles of competitive neutrality when reviewing SOE financing, but the treatment of competitive neutrality in IFC projects is uneven. Papers submitted to the Bank Group Board of Executive Directors in 2017 and 2019 embraced the competitive neutrality concept, clearly defined by the Organisation for Economic Co-op- eration and Development, which implies that the same rules of market behavior should apply to public and private firms, including the application of regulations and of competition law. This principle seeks to ensure a level 18 playing field, where the SOE has no undue competitive advantage. IFC states that the attributes of competitive neutrality include that the SOE earns a commercial rate of return on goods competing with private businesses; the SOE’s pricing for commercial activities and public services should not “un- duly” distort the playing field through subsidy; the SOE’s access to public contracts and other treatment in public procurement is “open, transparent, and nondiscriminatory”; and the SOE strives toward international standards and practices (IFC 2017).10 IEG reviewed project-related documents for seven recent IFC SOE reform projects and found that the treatment of competitive neutrality remains uneven, with some attention as early as 2012 and some omission as recently as 2017. Only one project package treated all of IFC’s competitive neutrality criteria.11 MIGA’s approach to supporting cross-border investments into SOEs is an- chored on three criteria: government control, public service, and the cred- itworthiness and financial viability of the SOE as a stand-alone entity. SOE investors are eligible for MIGA coverage provided they operate on a “commer- cial basis.” In considering whether an SOE investor operates on a commercial basis, at least with respect to the investment being covered, MIGA assesses several factors, including whether the SOE investor (i) operates on a self-sus- taining basis, (ii) enjoys substantial autonomy from government, and (iii) does not enjoy protection from competition or preferential treatment—factors that closely map to aspects of IFC’s competitive neutrality principle. SOE project enterprises receiving MIGA-insured investments need only be creditworthy Independent Evaluation Group World Bank Group and financially viable, as judged by MIGA’s credit risk assessment.12 In this respect, MIGA’s approach to SOEs is markedly different from IFC’s. IEG’s review of nine recent guarantees against the risk of nonhonoring of financial obligations involving an SOE found only one that addressed whether the SOE enjoyed protection from competition or preferential treatment. In a limited number of countries, the Bank Group (through its Markets and Competition Policy cluster, involving both World Bank and IFC advisory staff) has incorporated competitive neutrality into its analytics, including its Markets and Competition Policy Assessment Tool (MCPAT). The MCPAT analysis examines three areas: antitrust rules and enforcement, procompe- tition market and sector regulation, and competition principles in broader 19 public policies, including SOEs and competitive neutrality. In this context, MCPAT examines SOEs and their behavior, including whether the playing field is level and open; if state aid or other unequal tax, regulatory, debt, or procure- ment treatment inhibits competition; and whether there is a clear separation of commercial and noncommercial activities. MCPAT aims to focus reform on areas that promote competition and crowd-in private sector activity. For exam- ple, the Senegal MCPAT finds that in groundnut processing and fertilizer pro- duction, SOEs are protected by “restrictive government regulations” in value chains that “are traditionally economic activities that can be carried out by the private sector more efficiently than by SOEs” (Pop and Corthay 2018). Since its introduction in 2016, MCPAT has been applied to only a few countries—Argen- tina, Kenya, Mauritania, Mexico, Peru, the Philippines, Senegal, Vietnam, and Ukraine—and the Western Balkans region. The link between this analytic work and World Bank operations is still developing. A competitive market framework is being incorporated into the Integrated SOE Framework that the Equitable Growth, Finance, and Institutions Practice Group’s SOE Working Group is de- veloping as guidance for staff. A module based on MCPAT has been incorporat- ed into several Country Private Sector Diagnostics (CPSDs). Public Fiscal and Financial Management Reforms For decades, the World Bank has addressed public financial issues where SOE finances (including liabilities) threaten fiscal soundness or stability. Thus, SOEs’ macro, fiscal, and public finance aspects become part of a broader policy dialogue between the World Bank and governments on managing public reve- nues, expenditures, debts, and liabilities. SOEs’ fiscal implications are created by the influence that their costs, revenues, and risks have on public revenues, expenditures, debt service obligations, or other liabilities. Thus, it is critical to understand SOEs’ potential direct and indirect impacts on state finances. State Your Business!  Chapter 1 Interventions related to macrofiscal and PFM reform compose only 4 percent of the identified portfolio for the two sectors and appear only in World Bank activities. They make up 6 percent of both World Bank lending and ASA inter- ventions. However, the identified portfolio underrepresents the overall World Bank level of SOE-relevant activity on fiscal soundness and PFM because they are often addressed at the national level, thus affecting all SOEs rather than being tied to a single sector. For example, SOFIs and utilities can both gener- 20 ate public liabilities that destabilize the macroeconomy, as in crises faced by Mozambique and Slovenia.13 Some SOEs (such as oil companies) also provide important revenues to the state. For example, a recent Bank Group analysis of Sri Lanka points to the state-owned business enterprise portfolio representing “significant fiscal costs and fiscal risks undermining the government’s fiscal consolidation efforts” (World Bank 2020a, 4). It recommends that the Min- istry of Finance conduct a “systematic analysis of SOE financial statements, business plans, and investment proposals,” which “could help the government anticipate and mitigate fiscal risks to the budget” (72). The Bank Group tackles SOE macrofiscal issues through analytical work, DPOs, and technical assistance. As Mozambique’s recent SOE debt crisis demonstrates, this may involve engagement at the national level to rational- ize budgeting and public investment, constrain SOEs’ ability to incur debt, implement stronger systems of PFM, and more. In response to the hidden debt revelations in 2016, the World Bank launched a program to strengthen public investment and fiscal management, including debt and SOE fiscal risks. It modified ongoing development policy lending and joined a group of general budget support donors to promote concrete steps toward transpar- ency and accountability for the hidden loans. An FY13 Myanmar develop- ment policy loan, for instance, aimed to reduce the budget deficit partly by legally limiting government subsidies for the raw material requirements of state economic enterprises. A focus on fiscal soundness usually complements other SOE reforms. Of the 21 Independent Evaluation Group World Bank Group countries identified in IEG’s portfolio where the World Bank engaged in fiscal soundness reforms related to SOEs, only one had fiscal soundness as its sole focus. For example, a 2008 Ukraine development policy loan combined in its supported actions emphasis on strengthening public finances and improving SOE corporate governance. Fiscal ASA can also accompany other SOE reform inter- ventions, often focusing on debt management, accounting, and auditing. The Equitable Growth, Finance, and Institutions’ SOE task force recently elab- orated staff guidance on this type of support. The guidance advises assessing the fiscal impacts of SOE reforms, the fiscal sustainability of any subsidies, and the links to the fiscal framework. The Macroeconomics, Trade, and Investment GP focuses on improving public finances, oversight, and transparency, includ- ing lending that supports financial management and macrofiscal policy. 21 Where and How the Bank Group Delivers SOE Reform Support During the evaluation period, for the energy and financial sectors, World Bank lending predominated, constituting more than 90 percent of the value of the Bank Group SOE reform portfolio in those sectors (table 1.1). World Bank lending and ASA for SOE reform constituted about 87 percent of the activity and more than 90 percent of the financing. Within lending projects, DPOs accounted for 516 of the 898 World Bank lending interventions, and invest- ment operations accounted for 382 interventions. IFC delivered $3.8 billion in investment services support in the two sectors through 61 projects and spent $51 million to deliver advisory services through 59 projects. MIGA delivered about $3 billion through four guarantees. Support for the energy sector ac- counted for 57 percent of interventions and support for the financial sector for 30 percent, with the rest treating both sectors more broadly.  orld Bank Group SOE Reform Projects, Commitments by Table 1.1. W Institution, FY08–18 (est.) Share of Share of Projects Projects Interventions Volume Volume Institution (no.) (percent) (no.) ($, millions) (percent) World Bank 285 28 800 64,832 91 lending IFC IS 61 6 93 3,765 5 IFC AS 59 6 91 51 0 MIGA 17 2 19 2,973 4 World Bank 587 58 1,184 104 1 ASA State Your Business!  Chapter 1 Total 1,009 100 2,187 71,724 100 Source: Independent Evaluation Group portfolio review and analysis. Note: Due to rounding, volume shares add to 101 percent. ASA = advisory services and analytics; est. = estimated; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = Interna- tional Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. 22 IEG found relevant SOE ASA reform support activities in 142 countries and all other support in 119 countries. In the sample, the Bank Group financed operations in 34 countries in Sub-Saharan Africa, 21 countries in Europe and Central Asia, 16 in Latin America and the Caribbean, 13 in East Asia and Pacific, 11 in Middle East and North Africa, and 8 in South Asia. Although Sub-Saharan Africa was the Region with the highest number of financing projects, East Asia and Pacific had a higher average per country (5.3). Bank Group financing support to reform SOEs has been focused more on low- er-middle-income countries (46 percent) and low-income countries (29 per- cent), followed by upper-middle-income countries (23 percent). MIGA has the majority of guarantees by value in upper-middle-income countries and by number of projects in lower-middle-income countries. The Bank Group supports SOE reform at both the upstream and downstream levels. Upstream interventions—mostly by World Bank lending and ASA and IFC advisory—focus on regulatory frameworks for SOE activities; governance and accountability; and ownership, including privatization and PPPs. Down- stream interventions (at the enterprise level) focus on SOEs’ business and operations; corporate governance, ownership, and financial management are also substantial areas of engagement. Upstream support was more frequent in upper-middle-income countries, and support for lower-middle-income countries focused more on downstream reforms. IFC investment and MIGA guarantees, which make up 9 percent of Bank Independent Evaluation Group World Bank Group Group commitments, are oriented primarily toward SOEs’ business and oper- ational aspects and SOE ownership (whether through privatization or PPPs). IFC advisory engages both upstream and downstream, most often in business and operations and the upstream and downstream aspects of SOE ownership. MIGA is engaged primarily in the power sector through support of business and operations and ownership reform. Given the range of activities and dimensions of reform supported, chapter 2 casts an analytic light on the question of how effective the Bank Group has been and the factors associated with success. 23 1  “Over the years, the rationale for state ownership of commercial enterprises has varied among countries and industries and has typically comprised a mix of social, economic and strategic interests. Examples include industrial policy, regional development, the supply of public goods and the existence of so called ‘natural’ monopolies” (OECD 2015, 2). 2  For example, in China, state-owned enterprises (SOEs) account for 57 percent of corporate debt (valued at 72 percent of gross domestic product), even though they are responsible for less than 20 percent of output and employment (Lam et al. 2017). For revenue-generating companies (for example, state oil companies), these losses can take the form of foregone revenues to the government. 3  “Compared to other companies, SOEs [state-owned enterprises] have specific corruption risks because of their closeness to governments and public officials and the scale of the assets and services they control. Some of the biggest recent corruption scandals have involved state-owned enterprises, which clearly shows the risks that these companies face. In Brazil, the state oil company Petrobras was the focus of a major corruption scandal involving illegal payments to politicians and bribes that affected the whole country. The Nordic telecoms giant Telia was recently caught bribing for business in Uzbekistan, which resulted in fines of $965 million” (Transparency International 2017). 4  A recent International Monetary Fund study of emerging Europe found that the “profitability and efficiency of resource allocation of SOEs lag those of private firms in most sectors, with substantial cross-country variation. Poor SOE performance raises three main risks: large and risky contingent liabilities could stretch public finances; sizeable state ownership of banks coupled with poor governance could threaten financial stability; and negative productivity spillovers could affect the economy at large” (Böwer 2017, 2). 5  “There seems to be a credible empirical basis for selecting a threshold power system size and per capita income level below which unbundling of the power supply chain is not expected to State Your Business!  Chapter 1 be worthwhile” (Vagliasindi 2012b, 22). 6  Subsequent to the completion of analysis for this evaluation, the International Finance Cor- poration (IFC) provided supplementary portfolio information that the Independent Evaluation Group (IEG) has analyzed. This information suggests that an additional 23 IFC projects could fall into the SOE reform categories, although they were not identified by IEG through consistent application of its methodology. In terms of areas of activity, 37 percent of the projects were in business and operations, 22 percent in SOE ownership, and 26 percent in enterprise-level finan- cial management. Their inclusion was not possible given the late date of receiving this informa- 24 tion and would have only marginally changed the picture of IFC’s pattern of engagement. One additional evaluated project was identified; it was rated “unsatisfactory.” 7  In the financial sector, IEG found examples of IFC supporting the restructuring and rehabilita- tion of state-owned financial institutions’ debts through strengthening asset-liability structure and improving accessibility and pricing of alternative funding sources; the Multilateral Invest- ment Guarantee Agency (MIGA) providing a guarantee for asset-liability management purposes through a US dollar–local currency swap arrangement; and the World Bank supporting a finan- cial restructuring process of SOEs and supporting actions to reduce SOEs’ fiscal liabilities. World Bank projects also supported government acquisition, restructuring of debts, and recapitaliza- tion of state-owned banks. In the energy sector, IFC has advised financial restructuring in the power sector in São Tomé and Príncipe, and the World Bank has supported identifying options for restructuring electric companies, actions to bring new shareholders and investments, debt restructuring processes including tariff reviews, clearance of arrears, evaluation of assets, adop- tion of operational efficiency models, and realignment of roles and responsibilities. 8  External incentives shaping corporate behavior include the business enabling environment, the functioning of financial and labor markets, product and input market competition, and the “market for corporate control.” See Stone, Hurley, and Khemani 1998. 9  “When multiple companies compete head to head for consumers, a market discipline emerg- es, along with pressure to keep costs down to efficient levels and to improve service quality. The large economies of scale in the power sector mean that key activities (for example, trans- mission) are traditionally considered natural monopolies, making it inefficient to have more than one supplier. Even under a natural monopoly, however, it is still possible to have differ- Independent Evaluation Group World Bank Group ent companies compete for the right to supply the market on a monopoly basis for a certain period of time. The liberalization of the power sector therefore often proceeds in incremental stages, beginning with the opening up of generation to independent power producers that compete for the market. Eventually, it may transition to a full single-buyer model where gen- eration is fully divested from the incumbent utility, with the latter acting as the single buyer of generation on behalf of end consumers. The next stage—once the transmission segment has been fully unbundled—is to allow third-party access to the power grid so large customers can purchase power directly from generators on a bilateral negotiated basis. In due course, it may evolve into a wholesale power market, with a centralized price-setting mechanism and a variety of contracts and products being exchanged. In some instances, a final step would unbundle the distribution and retail functions of the utility, allowing the latter to be open to competition for energy supply” (Foster and Rana 2020, 48–9). 25 10  Although competitive neutrality was formalized as a policy in the 2017 IFC Board paper, that paper stated it as an existing principle of IFC investment in SOEs. A 2015 IFC directive, “Investments in State-Owned Enterprises,” clearly establishes the requirement to consider “whether or not the IFC investment avoids (i) displacement of viable private provision of the products or services provided by the SOE and (ii) displacement of private financing to the SOE.” It further requires that an SOE operate in a commercial manner, have operational autonomy from government, and be subject to commercial and corporate laws applicable to private companies (IFC 2015). 11  IFC notes in comments to IEG: “IFC has a robust policy assessment matrix in place to evaluate each SOE investment’s fit with IFC’s private sector mandate and this assessment is required for all SOE projects. Competitive neutrality plays an important but not an over- riding role in this assessment, which includes other variables such as the commercial nature of operations, non-displacement of private alternatives and operational autonomy from the government.” 12  MG−010−FY14. MIGA Guidance: Rating State-Owned Enterprise Risk. June 30, 2014; MG- 014-FY15. MIGA Guidance: Eligibility Guidance Relating to NHFO-SOE Coverage. August 8, 2014; MG-001-FY2016. MIGA Guidance: MIGLC Frequently Asked Questions. August 3, 2015. Attachment 1. Question 49, pages 25 to 26. In a now superseded 2003 Board statement, the Multilateral Investment Guarantee Agency (MIGA) stated: “It should be stressed that the SOEs MIGA is covering: a) operate on a commercial basis; and b) retain the commercial risks for the projects that MIGA is guaranteeing.” 13  “In Slovenia, the state owned not only a sizeable portfolio of non-financial companies but also the three largest domestic banks and holds about 63 percent of the total banking sector’s equity. After the first hit of the global financial crisis, Slovenia experienced another banking crisis in 2012–13, when the mostly state-owned banking system came under pressure and led the sovereign to lose market access. Cross-enterprise ownership structures with SOEs at their State Your Business!  Chapter 1 heart, and pervasive connected lending was believed to have amplified the crises. As a result, bankruptcies were wide-spread, and mounting NPLs [nonperforming loans] ate up bank capi- tal” (Böwer 2017, 15). 26 2 | Effectiveness of Bank Group Support to SOE Reforms Highlights This chapter examines the performance of the World Bank Group’s energy and financial sector portfolio of support for state-owned enterprise (SOE) reform, analyzing factors associated with success. The SOE reform portfolio in these sectors, on average, met the World Bank and International Finance Corporation corporate tar- gets for project success. The Bank Group overall engages far more with state-owned com- mercial banks than with state-owned development banks. The success rate for development bank SOE reform interventions ex- ceeded that for commercial bank interventions. SOE reforms in the transmission and distribution sectors were the most effective, followed by power generation; those dealing with extractive industries (petroleum, gas, and mining) were successful only half the time. Several factors at the country and project levels are predictive of intervention success—some within the Bank Group’s control and some outside of it. Control of corruption at the country level is strongly associated with intervention success. Other things being equal, SOE reform interven- tions in a country with high control of corruption are more than twice 27  as likely to succeed than in a country with low control of corruption. Five project-level factors not directly controlled by the Bank Group (though potentially influenced by it) are strongly associated with successful SOE reform interventions: » Client commitment to the reforms and reform activities; » Coordination among donors and other stakeholders; » Client institutional capacity and coordination; » Political economy and vested interests; and » External shocks (natural or other) posing both obstacles and opportunities. Four internal factors at the project level that are under the Bank Group’s direct control are strongly associated with successful SOE reform interventions: » Project design, including the appropriate choice of instrument, adaptation to local conditions, and simplicity (versus complexity); » Supervision, including having in-country expertise during project imple- mentation (especially for investment projects); » A strong results framework with active monitoring and evaluation; and » Sequencing of interventions, including link to prior analytic work. 28   SOE Reform Performance in the Portfolio and Literature On average, the overall SOE reform portfolio in the financial and energy sectors reviewed met the World Bank and IFC corporate targets for project success (figure 2.1). World Bank lending achieved an overall success rate of 78 percent against a target of 75 percent. Development policy lending (151 evaluated projects) achieved a success rate of 85 percent versus the invest- ment project finance success rate of 67 percent (97 evaluated projects), but the two instruments focused on tackling different SOE reform challenges. Policy lending was more focused upstream, seeking to improve public fi- nances; accountability, transparency, and oversight; or sector competition and productivity. Investment lending was more focused downstream, aim- ing to strengthen enterprise operational and financial performance as well as service delivery and quality. IFC achieved a success rate of 73 percent for investment services (22 evaluated projects) and 56 percent for advisory services against an overall target of 65 percent. Investment services were far more likely than advisory services to support improving service delivery and quality, and advisory services were far more likely to support strengthen- ing financial and operational performance and improving transparency and oversight. Two evaluated MIGA guarantees both achieved their outcomes. Across the five SOE reform types discussed in chapter 1, Bank Group projects Independent Evaluation Group World Bank Group supporting corporate governance, ownership reform, and business and oper- ations showed statistically significantly higher success rates than did reforms in macrofiscal, and PFM and in competition and regulation (figure 2.2). World Bank DPOs were significantly more effective on average when pursuing com- petition and regulatory reforms than they were in the other four areas and significantly more successful at pursuing SOE ownership and corporate gov- ernance reforms than in pursuing business and operations and macrofiscal, and PFM reforms. World Bank investment operations were significantly more successful in pursuing business and operations reforms than they were in the other four areas but were also quite successful in pursuing privatization and corporate governance. They were relatively less successful in supporting reforms in competition and regulation and in macrofiscal and PFM. 29  uccess Rate of State-Owned Enterprise Reform Projects Figure 2.1. S Evaluated, FY08–18 100 75 80 85 65 Share of successful projects 78 60 73 67 66 40 56 (percent) 20 0 World Bank DPO IPF IFC IFC IS IFC AS lending (n=151) (n=97) (n=38) (n=22) (n=16) (n=248) Institution Sources: Independent Evaluation Group portfolio review and analysis; World Bank Corporate Scorecard (updated to October 2017). Note: The figure is based on 286 projects. The analysis excludes six World Bank lending projects for which outcome ratings were not available, rated, or applicable and three Multilateral Investment Guar- antee Agency guarantees, which achieved their intervention outcomes. The orange dots show FY17 corporate satisfactory outcomes targets. IFC updated its scorecard in November 2019 and eliminated its corporate success target of 65 percent. A project is now defined as “above the line,” or successful, if it is achieving or mostly achieving project outcomes. DPO = development policy operation; FY = fiscal year; IFC = International Finance Corporation; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; IPF = investment project financing. State Your Business!  Chapter 2 30  uccess Rate of SOE Reform Interventions by Type and In- Figure 2.2. S strument (Bank Group Evaluated by IEG) 72 71 SOE ownership 68 86 70 76 78 Corporate 68 governance 83 100a 70 SOE reform types 67 Business 74 and operations 56 50 68 68 PFM and 67 fiscal policy 68 74 Competition 63 and regulation 47 83 0 20 40 60 80 100 Share of successful projects (percent) Independent Evaluation Group World Bank Group IFC AS IFC IS IPF DPO Overall Source: Independent Evaluation Group portfolio review analysis. Note: n = 147 SOE ownership evaluated interventions (73 DPO, 47 IPF, 14 IFC IS, 10 IFC AS, and 3 Multi- lateral Investment Guarantee Agency); 137 corporate governance evaluated interventions (88 DPO, 41 IPF, 6 IFC IS, and 2 IFC AS); 136 business and operations evaluated interventions (30 DPO, 89 IPF, 9 IFC IS, and 2 IFC AS); 22 fiscal policy evaluated interventions (19 DPO and 3 IPF); and 244 competition and regulation evaluated interventions (129 DPO, 94 IPF, 15 IFC IS, and 6 IFC AS). DPO = development policy operation; IEG = Independent Evaluation Group; IFC AS = International Finance Corporation advisory ser- vices; IFC IS = International Finance Corporation investment services; IPF = investment project financing; PFM = public financial management; SOE = state-owned enterprise. a. n < 5 interventions. 31 IFC investment operations were significantly more successful at pursuing ownership reforms than they were in other areas but also had a high average success rate for the small number of corporate governance reforms evaluat- ed. Advisory services had a higher average success rate in ownership, corpo- rate governance, and competition and regulation, but with small numbers of evaluated projects for which statistical significance could not be compared. Although overall Bank Group privatization support was highly successful in both the energy and financial sectors, corporate governance reform was highly successful only in the financial sector (85 percent) and had a weaker record in the energy sector (62 percent). The Bank Group’s success rate for development bank SOE reform interven- tions (77 percent) exceeded that for commercial bank SOE reform interven- tions (69 percent). However, the Bank Group overall engages far more with state-owned commercial banks than it does with state-owned development banks. For example, IFC successfully financed the turnaround and further privatization of Pakistan’s largest bank, Habib Bank Limited, through a $50 million loan and a $50 million equity investment. IFC supplemented this with advisory services (training) to strengthen staff and managerial capacity. Bank Group engagement with nonbank financial institutions is limited. World Bank investment projects supporting commercial state- owned bank reform fare poorly (61 percent success rate), but DPOs perform better (74 percent success rate). The reverse pattern holds for state-owned development banks: investment operations fare well (90 percent) and DPOs poorly (50 percent). For both categories of state-owned banks, support for business and operations was the most common, including support for risk management, product service improvement, and human resource man- agement. Of 265 non-ASA financial sector SOFI interventions, 9 supported State Your Business!  Chapter 2 privatization of commercial banks, and 7 more supported privatization of other financial institutions. Eight ASA interventions of the 116 identified supported privatization. In the power sector, SOE reforms were the most effective in the transmis- sion and distribution sectors, followed by power generation; those dealing with extractive industries (petroleum, gas, and mining) were successful only half the time (box 2.1). IEG’s 2019 synthesis of findings on utility reform 32 found that in the power sector, investment projects were more successful than DPOs at improving the overall financial performance of electric power utilities, but DPOs were more effective at influencing tariff adjustments. In power, IFC enjoyed substantial success in both investment services (88 per- cent) and advisory services (78 percent).  ank Group Engagement in Extractive Industries State-Owned Box 2.1. B Enterprise Reform The oil, gas, and mining sector faces a unique set of environmental, social, and eco- nomic challenges and has a wide range of stakeholders (appendix E). State control is prevalent in the oil and gas sector, owning about 90 percent of reserves and 55 per- cent of production. In mining, state-owned enterprises have historically been less influential. The performance of national oil companies varies substantially according to state goals, geology, government interactions with the oil companies, and manage- ment strategy. Managerial and technical capacities are important to value creation. Country case studies highlight interventions in the Arab Republic of Egypt, Kenya, Mo- zambique, Serbia, Ukraine, and Vietnam. Those in Egypt, Kenya, Ukraine, and Vietnam addressed improving the supply, prices, and reliability of natural gas or electricity sup- ply. In Mozambique, reforms supported mining industry compliance with the Extractive Industries Transparency Initiative. Overall, the Independent Evaluation Group’s portfolio analysis identified 65 state- Independent Evaluation Group World Bank Group owned enterprise reform interventions through 34 projects. The majority (94 percent) are upstream World Bank lending interventions in oil and gas. Sub-Saharan Africa has more projects but less success (56 percent) than most other Regions. Success factors include client commitment, project design, and supervision. Negative factors include external shocks, weak monitoring and evaluation, insufficient public sector capacity, design issues (for example, complexity), and lack of client commitment. Sources: NRGI 2019; Wolf 2009; Independent Evaluation Group deep dive on state-owned enterprise reform in extractive industries. IEG’s in-depth reviews and literature review yield evidence of SOE reform success in three of the five reform types: privatization, corporate governance 33 reform, and competition (unfortunately, this literature sheds little system- atic light on the other areas of SOE reform). The literature consistently finds superior performance of private and privatized companies over public companies in both the energy and financial sectors and has especially neg- ative findings about state-owned commercial banks. Multiple national and cross-national studies have shown the benefits of privatization (appendix G). A comprehensive literature review found that the studies focusing on before and after performance of privatized SOEs evidenced “significant improve- ments after companies are divested” (Megginson 2017, 1). It also found that China’s model (a socialist market economy based on a prominent role of public ownership and state-owned enterprises) evidenced “abysmal relative performance of state-controlled versus private firms in key industries—espe- cially petroleum, banking, and technology” (50). On corporate governance, a small number of national studies of reforms have found that there were ben- efits to SOE performance but also that implementation of reforms is often incomplete. On competition, the literature shows that enhanced competition improves SOE performance in both the financial and power sectors on its own and as a complement to other reforms. In the banking sector, research repeatedly finds that state-owned commer- cial banks perform poorly relative to private commercial banks (appendix G). There is “little evidence that government bank ownership provides substan- tial benefits (relative to other types of ownership) to the banking sector, the real economy, or users of banking services, especially in developing coun- tries” (Cull, Pería, and Verrier 2018). Ho, Lin, and Tsai (2016) find that, for 39 countries, privatized banks outperform nonprivatized banks, that this benefit is larger in developing than in developed countries, and that good governance benefits privatization in developing countries. There is some ev- State Your Business!  Chapter 2 idence that well-managed state-owned development banks can direct credit to areas of policy priorities and benefit from a clear yet flexible mandate, adequate regulation and supervision, effective corporate governance and management, financial sustainability, and regular performance assessment (Abraham and Schmukler 2017). State-owned banks played a more positive role in countercyclical credit provision or, to be precise, were less procyclical in some countries than were private banks (Cull, Pería, and Verrier 2018). However, evidence is mostly negative regarding government ownership’s 34 effects on bank competition, efficiency, and the stability of financial systems. There are mixed results on financial access. One study found that “women are more likely to be excluded from the financial sector where... state-owned banks have a bigger share in the banking system” (Morsy 2020). Bank research has found that client country bank performance usually im- proved after privatization (Clarke, Cull, and Shirley 2005). The privatizations of Uganda Commercial Bank and the South African Stanbic Bank improved profitability and financial access (Rabiei and Rezaie 2013). A cross-country study in Southeast Asia and a panel of 22 developing countries found that bank privatization raises bank profitability and efficiency over time, even when the acquirer is a foreign bank (Boubakri et al. 2005; Williams and Nguyen 2005). In the power sector, the World Bank’s research finds that “governance scores tend to be systematically higher for private utilities” (Foster and Rana 2020, 12). The efficiency of privatized utilities is “on par with the top half of per- formers among public utilities.” Only privatized utilities ever achieve full capital cost recovery. However, this research cautions that privatization of distribution utilities is rare and should be pursued only when enabling con- ditions are met, including adequate functioning of the utility and a strong authorizing environment (Foster and Rana 2020, 14). Rigorous evidence of the effectiveness of corporate governance reforms comes only from national studies. For example, Heo (2018) finds a positive relation- Independent Evaluation Group World Bank Group ship between financial performance and board size and transparency and dis- closure for 320 Korean SOEs. For Lithuanian SOEs, Jurkonis, Merkliopas, and Kyga (2016) find that management and board independence relate positively to returns on equity. Rudolph (2009), analyzing four well-performing SOFIs in Canada, Chile, Finland, and South Africa, finds that they share an efficiency and profitability objective that shareholders measure regularly, professional and qualified senior management, proper risk management systems, and inde- pendence from government in their financing. Whatever the theoretical power of corporate governance reforms for SOEs, their realization is often incomplete. The 2019 World Bank power sector re- form flagship report finds evidence that good governance practices are strong- 35 ly associated with improvements in cost recovery and operational efficiency of distribution utilities, but it also finds “a significant governance gap between corporatized public utilities and privatized ones. Also, public utilities practice better governance when they coexist alongside private utilities” (World Bank 2019b). Areas in which public utilities lag include lack of autonomy in deci- sion-making on matters of finance and human resources, considerable inter- ference in the appointment and removal of board members, shortcomings in the rigor of accounting practices, and more lax human resource practices, with less ability to reward good performers and fire bad ones. There is strong evidence that competition improves SOE performance in both the power and financial sectors. An econometric assessment of power sector data for 36 developing and transition countries over 18 years found that economic performance gains arose mainly from the introduction of competition (Zhang, Parker, and Kirkpatrick 2008). Privatization or regulato- ry reforms were less effective without a competitive market. In the financial sector, the negative effects of bank concentration on firms’ access to credit are stronger in countries with higher shares of state bank ownership (Beck, Demirgüç-Kunt, and Maksimovic 2004). The benefits of bank privatization are greater when they take place in more competitive environments (Clarke, Cull, and Shirley 2005). Factors of Success and Failure at the Country and Project Levels Several factors at the country and project levels are predictive of interven- tion success—some within the Bank Group’s control and some outside of it.1 IEG’s review of micro evaluative evidence from 294 projects and 671 inter- State Your Business!  Chapter 2 ventions indicates several internal factors (those under the Bank Group’s control) and several external factors (those beyond its direct control) that are most commonly identified as explaining success or failure (figure 2.3). The leading internal factors include project design and supervision, the monitoring and evaluation (M&E) framework, and sequencing (including the availability of prior analytic work). The most common external factors are client commitment, collaboration with other donors and external actors, political economy, client capacity, agency coordination factors, and shocks. 36 Client commitment and design quality are important across World Bank and IFC instruments, but supervision is a more common factor for IFC invest- ment services, and political economy and agency coordination are more common factors for IFC advisory services. For IFC investments, identification of risks at appraisal is especially important.  actors of Success and Failure for World Bank lending and Figure 2.3. F IFC IS and AS Client commitment 100 Collaboration with 18 16 23 external actors or donors Public sector External 80 9 institutional capacity 9 2 Agency coordination 5 10 and political economy 7 Projects affected (%) 5 3 7 External shocks 60 13 6 14 Design 3 7 17 M&E framework 40 23 9 16 Analytical work Internal 9 3 Supervision 9 20 8 Identification of 23 risks at appraisal 4 18 8 Other 0 World Bank IFC IS IFC AS Lending Institution Independent Evaluation Group World Bank Group Source: Independent Evaluation Group portfolio review and analysis. Note: Based on 857 factors identified for 294 evaluated projects. The projects can have multiple factors of success or failure. Excludes the Multilateral Investment Guarantee Agency. Percentages may not add up to 100 due to rounding. IFC AS = International Finance Corporation advisory services; IFC IS = Interna- tional Finance Corporation investment services; M&E = monitoring and evaluation. Design quality and client commitment are frequently identified as success factors across all five types of SOE reform support, although the frequency of other factors varies by area (figure 2.4). For example, risk at appraisal is more likely to be a factor in PFM and in fiscal (for the World Bank) and corporate governance reforms (for IFC and the World Bank), and supervision is rela- tively more important for privatization and ownership reform. 37  requency of Success and Failure Factors by SOE Figure 2.4. F Reform Type 100 Client commitment 17 17 Collaboration with 18 20 20 external actors or donors Agency coordination External 80 6 9 9 and political economy 6 15 12 Public sector 7 8 institutional capacity 7 5 7 7 External shocks Factors cited (%) 5 8 60 1 8 6 4 3 6 Private sector 1 4 1 2 institutional capacity 18 1 Design 17 16 18 40 15 7 Analytical work 8 Internal 9 10 7 M&E framework 11 8 6 7 4 20 10 Supervision 4 5 8 6 4 4 Identification of 7 4 4 risks at appraisal 8 9 6 11 10 Other 0 an e y ns la on lp / er ss p rn rat ic ce n ca FM ne E hi io gu iti tio ol ow SO op ne ve po rs at re et P d usi go r d p Co an m an B fis Co SOE reform types Source: Independent Evaluation Group portfolio review and analysis. Note: Percentages may not add up to 100 due to rounding. M&E = monitoring and evaluation; PFM = public financial management; SOE = state-owned enterprise. Business and operational reforms are relatively more sensitive to issues of agency coordination and political economy, and it is for these reforms that design and supervision issues are most likely to matter. Client commitment is a more frequent factor for both privatization and corporate governance State Your Business!  Chapter 2 reform than for other reform types. At the country level, control of corruption is strongly associated with SOE reform success. Other things being equal, a country with high control of corruption is more than twice as likely to see SOE reform intervention success as one with low control of corruption. In conditions of weak public governance, it is more difficult to strengthen the governance, regulation, or performance of public enterprises. 38 The marginal effect of weak control of corruption is large, but in practice, several factors mitigate its negative influence on success. In the evaluated portfolio, the success rate for countries with low control of corruption is about 67 percent, but it is 76 percent for those with high control of corrup- tion. Given the Bank Group’s commitment to engage in all client countries, it is not surprising that a significant minority of projects are in countries with characteristics predictive of a lower level of success. Overall, 26 percent of the identified portfolio was in countries with low control of corruption. Corruption powerfully undermines performance. In Ukraine, for example, the Country Partnership Framework FY17–21 review and the case study found a widespread challenge of corruption and state capture impeding SOE reform progress. The Organisation for Economic Co-operation and Development reported that by June 2018, more than 194 of the National Anti-Corruption Bureau’s 793 criminal proceedings dealt with about 50 SOEs and their offi- cials. In Kenya, petty corruption among the field staff responsible for install- ing and reading meters reportedly frustrated efforts to stem power system losses, at least in part. The Vietnam case study found that cross-ownership among banks was a significant problem, opening the door to corruption and conflicts of interest. In Bangladesh, weak governance allowed huge banking scandals that wracked state-owned commercial banks (box 2.2).2 Box 2.2. The Sonali-Hallmark Scandal in Bangladesh Independent Evaluation Group World Bank Group The Sonali-Hallmark scandal was one of several that plagued the state-owned com- mercial bank system after the World Bank–supported drive for strengthened corporate governance, privatization, and better oversight was abandoned in 2009. A single branch of Sonali Bank gave loans valued at about $454 million based on fraudulent documents. Fraudulent letters of credit to fictitious companies, combined with collusion or inaction by the Sonali Bank Board and the Bank of Bangladesh, enabled massive fraud. In 2014, Sonali Bank was reported to have a nonperforming loan ratio of 37 percent. Loans were assessed not according to their business potential, but with an eye toward “the influence or the connections of the person” asking for credit. Observers noted a strong incidence of default for loans approved by party-connected bank directors. Sources: Allchin 2016; Economist 2014. 39 IEG analyzed successful projects in countries with weak control of corrup- tion and found that they shared features that may mitigate adverse country conditions, thus improving their chance of success. These features include the internal factors of simple, selective, and flexible project design; prior analytic work;3 and strong supervision. Externally, they include strong client commitment and collaboration with external actors and donors. These fac- tors combined explain the relatively high success rate (68 percent) of Bank Group projects in countries with low control of corruption. For example, the $150 million Guatemala financial sector adjustment loan (evaluated in FY08) largely succeeded in its sector reform objectives. Rooted in a prior Financial Sector Assessment Program, it was accompanied by technical analysis and support through active supervision. Continuous policy dialogue was key to maintaining government commitment through two administrations. IFC’s Zalkar Bank privatization project (592127) in the Kyrgyz Republic (approved in FY12) achieved its objective to support a bank privatization. IFC identi- fied a buyer capable of implementing a restructuring plan to improve Zalkar Bank’s financial and operational performance. The project benefited from a flexible design that was well adapted to local circumstances as well as strong supervision by a well-composed team that included the local knowledge needed to navigate local regulatory requirements. Collaboration with exter- nal actors (including the International Monetary Fund) enhanced the gov- ernment’s commitment to implement recommendations. External Factors of Success Five factors at the project level not directly controlled by the Bank Group (though potentially influenced by it) are strongly associated with the success State Your Business!  Chapter 2 of SOE reform interventions: » Client commitment to the reforms and reform activities; » Coordination among donors and other stakeholders; » Client institutional capacity and coordination; » Political economy (which can work for or against reforms, whereas vested interests often frustrate them); and 40 » External shocks, whether natural or human made, posing both obstacles and opportunities. Client commitment underpinned success in multiple countries, including sustained periods of power sector reform motivated by strong government commitment to improving power supply and expanding or universalizing ac- cess to electricity. Underlying government commitment helped drive reforms in the Arab Republic of Egypt, Kenya, and Vietnam. In Vietnam, government commitment to rural electrification became the basis for sector reform. In Mozambique, there was strong ownership of the reform program under the sixth, seventh, and eighth PRSCs, and a proactive government stance led to progress in applying the Extractive Industries Transparency Initiative. IFC benefited from client commitment with the 2008 Philippines Olongapo Power project, its first successful PPP transaction with a municipality. The City of Olongapo had demonstrated its support for a PPP by launching a previous tender (though unsuccessful), securing necessary central government approv- als to implement a PPP, and committing $130,000 in fees to IFC to cover staff and travel expenses. Results suffered where client commitment was inconsistent, as seen in the case studies of Bangladesh, Egypt, Kenya, Serbia, and Ukraine (appendix F). For example, in Ukraine, commitment to the implementation of corporate governance reforms in SOEs waned as the 2014 crisis receded. However, with the support of the World Bank and other international financial institutions, Independent Evaluation Group World Bank Group the government prepared a state-owned bank strategy and road map to im- prove state-owned bank governance, though implementation progress was slow and limited. In the energy sector, a supervisory board was established in 2015 to strengthen corporate governance for the SOE giant Naftogaz. Super- visory board members resigned by 2017, citing the government’s attempts to block reform of the company. The government later amended the board’s charter to reduce its power. Bangladesh’s lack of commitment to financial sector reform after 2009 undermined corporate governance reforms intro- duced with World Bank support in the first decade of the 2000s. In Serbia, a changing government agenda challenged the effectiveness of an IFC invest- ment and advisory project. A new government in 2013 wanted to cancel pri- vatization of a key state-owned bank, Komercijalna Banka ad Beograd. Long 41 negotiations managed to restore the agenda with a new timetable, contend- ing with political influence on many levels, and leading to the conclusion of a deal for its sale in early 2020. Strong and durable coordination among donors contributes to effectiveness, allowing donors to work in complementary support of reform and leverage one another’s resources and influence. Ukraine saw strong donor coordina- tion in 2014 in both sectors, but this had eroded in the gas sector by 2019. In Vietnam, donors engaged in a formal consultative group and business forum, and 14 development partners supported the 10-year PRSC series. Vietnam’s power sector has seen remarkable success, including achieving universal access to electricity and growth (and improved efficiency and cost recov- ery) to become the second-largest power system in Southeast Asia, with the expectation that it will soon become the largest. Donor support, led by the Bank Group, also involved major support from the Asian Development Bank and financing from the Japan International Cooperation Agency and the German Bank for Reconstruction. In Bangladesh and Kenya, the World Bank led energy sector donor coordination bodies over key periods. Ukraine and Serbia both sought alignment with the European Union (EU) Energy Package. Serbia’s EU accession drive led to de facto donor coordination. In Ukraine, deposit insurance was part of a broader package involving the International Monetary Fund and the European Bank for Reconstruction and Development to reform the banking sector, wherein donors often conducted missions jointly. Conversely, in Kenya, other donors’ support for a large wind power project impeded least-cost planning and the utilities’ financial viability. The econometric analysis confirms that coordination with other donors and part- ners is a significant component predicting success. State Your Business!  Chapter 2 High public sector institutional capacity aids development effectiveness in SOE reform projects. For example, IEG’s review of the Serbia Country Part- nership Strategy for 2008–11 found that in a period of harmonization with the EU, implementation lagged legislation as capacity was built. “When projects are housed with strong institutions, it can take time to reach ini- tial agreement, but prospects for successful implementation are high. Weak institutions are less likely to implement agreements even if there is a high level of formal ownership” (World Bank 2012b). Portfolio analysis of evalu- 42 ated projects indicates that weak public sector institutional capacity often appeared as a negative factor and was the third most frequently identified external factor in project evaluations. Weak coordination among clients’ agencies could hinder projects. Complex management with multiple government stakeholders yielded coordination challenges and overlaps in authority. In Kenya, IEG’s case study found that while the World Bank was working to build regulatory capacity in the Ministry of Petroleum, new legislation transferred regulatory authority to the Ministry of Energy. In Bangladesh, the Ministry of Finance inserted itself between state- owned commercial banks and the Bank of Bangladesh, weakening the financial sector regulator’s oversight authority. In Ukraine, multiple rival government committees played a role in energy sector reform, which complicated deci- sion-making. For example, the transfer of the state-owned electric transmis- sion company, UkrEnergo, from the Ministry of Energy and Coal Industries to the Ministry of Finance in late 2018 delayed key approvals and payments. Political economy factors influencing projects included shifts in commit- ment arising from political considerations, opposition from vested interests, and a variety of political difficulties caused by electoral cycles and regime change. IEG’s SOFI deep dive (summarized in appendix E) found that coun- tries such as Bangladesh, Egypt, and Indonesia have signaled their intent to privatize state-owned banks but later halted efforts because of internal political constraints. One route of political economy influence is public en- Independent Evaluation Group World Bank Group gagement, which can either broaden ownership of reforms or diffuse opposi- tion. In Egypt, for example, when residents opposed construction of a power plant because of misinformation, rumors, and implementation missteps, the Bank Group reacted swiftly through an extensive public awareness cam- paign, offering jobs in construction projects to the community members and holding several conferences in the Giza North and Cairo areas. This allowed implementation to move forward. However, the overall level of the World Bank’s public outreach on SOE reform was found to be inconsistent, with considerable potential to raise engagement. Vested interests can assert themselves in a wide variety of ways, rang- ing from subtle internal resistance in SOEs to overt legislative action. In Ukraine, for example, the opposition of affected oligarchs frustrated finan- 43 cial sector reforms and the resolution of nonperforming loans. One major impact occurred when they influenced the courts to reverse the nationaliza- tion of PrivatBank—billions of dollars of public resources had been used to nationalize and recapitalize the bank to protect the financial system’s sta- bility. The court ruling would transfer these resources to the private owner, whose actions had necessitated the bailout. In both Bangladesh and Kenya, exceptions at times granted to vested interests disrupted least-cost planning and the competitive award of contracts to independent power producers. In Kenya, near-textbook arrangements for corporatization and corporate governance of the power utilities suggested independence of the boards and the regulator. However, in the 2017 election cycle, the government (mindful of politics) reportedly pressured Kenya Power to continue a high rate of rural connections (more than 1 million)—despite a fiscal shortfall that prevented a promised budgetary allocation—and pressured the regulator to not increase electricity tariffs. The result was a sectorwide solvency crisis with ripple ef- fects from distribution to generation to transmission. IEG’s literature review further documents how electoral cycles can influence SOFI activity.4 External shocks, whether natural or human made, can create opportunity by compelling action, but they can also hinder reforms. In Ukraine, for example, the 2008 global financial crisis, the 2014 conflict with the Russian Federa- tion, and the ensuing civil conflict created windows for reforming financial sector SOEs, although momentum was lost a few years later. In Egypt and Kenya, electricity supply crises drove greater engagement. Conversely, the 2011 Egyptian revolution disrupted the World Bank’s SOE reform engage- ment. In one specification, the econometrics found that external shocks are a negative predictor of success. State Your Business!  Chapter 2 Internal Factors of Success Factors under the Bank Group’s direct control are of strong interest because they are most subject to improvement through Bank Group attention and action. Four such internal factors directly controlled by the Bank Group are strongly associated with successful SOE reform interventions: » Project design, including the appropriate choice of instrument, adaptation to 44 local conditions, and simplicity; » Supervision, including having in-country expertise during project implemen- tation (especially for investment projects); » A strong results framework with active M&E; and » Sequencing and complementarity of interventions, including the link of ac- tivities to prior analytic work and internal collaboration. Choice of instrument was cited in 23 cases as a factor of success or failure. In most cases, it was a positive influence because the instruments chosen responded to country needs, were strategic, and combined financing and technical assistance. IEG’s recent utility reform synthesis report finds that the relatively long implementation periods of investment projects allow more time for hands-on operational support and corrective measures in the process (World Bank 2020b). Conversely, the synthesis found that DPOs, with their financing contingent on policy reform, were more effective than invest- ment projects at influencing tariff adjustments (71 percent versus 55 percent success). Programmatic DPOs achieved better outcomes in utility financial recovery than one-off DPOs, which were found to suffer from complexity, overdesign, and an insufficient time frame for implementation. The econo- metric analysis did not indicate that any one instrument was systematically more successful, but one specification did confirm that the correct choice of instrument was a predictor of success. Flexibility and adaptation of design to capacity were frequently identified Independent Evaluation Group World Bank Group as factors associated with successful development outcomes in SOE reform projects. The Bank Group’s country-driven model is responsive to crises, adapts to differences in client capacity and priorities, and allows the Bank Group to leverage service delivery goals to reform SOEs. Crises pose both a danger and an opportunity regarding SOE reform. Shocks are negatively associated with project success, yet they often provide an opening for reform progress. In banking, over the evaluation period, the World Bank responded twice to crises in Ukraine: after the global financial crisis of 2008 and again after the Russian-Ukrainian conflict beginning in 2014. In between, client demand for reform was weak, but each crisis brought new commitment and some progress. In energy, case studies showed high Bank Group responsive- ness in Bangladesh, Kenya, and Ukraine when they confronted serious power 45 shortages, and in each case, the response included measures advancing SOE reform. In Ukraine, the World Bank provided timely support to the gas sector to address supply uncertainty, diversify sources, and build storage reserves. In Mozambique, when discovery of undisclosed SOE liabilities sparked a fiscal crisis in 2016, the World Bank responded with a program to strengthen public investment and fiscal management, addressing SOE debt and fiscal risks. Adaptability has meant that in several cases where client sectoral capacity or commitment was initially low, the Bank Group has shown an ability to engage over the long term to enhance it. In the power sectors in Bangladesh and Vietnam and in China’s financial sector, capacity and shared under- standings were built over time (box 2.3). Long-term engagement and mobi- lization of multiple complementary and sequential instruments have helped build capacity and Bank Group credibility. In some cases, the Bank Group leverages its efforts through both internal and external coordination. Flexibility in adapting to lower client capacity can yield long-term results. In Mozambique, for example, the World Bank’s support for power sector reform originally aimed at unbundling the state power utility, Electricidade de Moçambique, with a separate transmission company and a newly created private market for distribution and generation. Progress was slow, and when new research signaled that unbundling was not the best course for low-ca- pacity countries with small power sectors, the World Bank dropped unbun- dling and shifted its focus to increasing the role and effectiveness of the nascent national electricity regulator, Conselho Nacional de Electricidade, particularly in monitoring Electricidade de Moçambique. The country economic model allows the Bank Group to adapt to some markedly different client priorities. In Vietnam, for example, the government has main- tained ownership and control of thousands of SOEs, including the four largest State Your Business!  Chapter 2 banks, which make up almost half of total sector assets. Despite the political infeasibility of privatizing large state banks, the Bank Group has remained en- gaged with a large SOE reform program across the financial and energy sectors. With mixed effect, the World Bank has been able to engage (including through major analytic work) on work to separate regulation from ownership, strength- en the legal and institutional foundations of financial markets, and enhance regulation and stability. In energy, the World Bank worked with the government on unbundling in the power sector gradually as part of its efforts to improve 46 sector performance and achieve universal electricity access. However, import- ant differences between the World Bank and the government have remained on the pace of reforms and on pricing mechanisms. In contrast to Vietnam, Serbia was aggressively pursuing a more orthodox set of reforms and EU accession, and the Bank Group mobilized to support its agenda. For example, IFC invested in two state-owned banks (Čačans- ka Banka and Komercijalna Banka ad Beograd). IFC aimed to strengthen Čačanska Banka’s capital base through a capital increase, improve its com- petitive position, enable it to bear likely stresses, and support its lending to the small and medium enterprise segment alongside the European Bank for Reconstruction and Development, all to facilitate privatization. IEG’s case study found that IFC had contributed as a shareholder to improving the cor- porate governance of the two state-owned banks. In 2015, a majority share of Čačanska Banka was sold to a Turkish banking group. One important entry point for the Bank Group on SOE reform is client desire to improve the quantity, quality, and consistency of service delivery. This is an explicit objective of 22 percent of SOE reform interventions that IEG reviewed (the second most common). However, engaging adaptably often achieves further SOE reform. With multiple power utilities, the Bank Group’s support for improved generation eased discussions of reforms to regulation, institutions, and operations reforms and private participation. For exam- ple, Vietnam prioritized rapid expansion of citizen access to electricity, so Independent Evaluation Group World Bank Group the World Bank’s support facilitated trust and broader conversations about utility reform. In the Bangladesh, Kenya, and Ukraine case studies, IEG found that severe deficiencies in power availability and reliability, along with ambi- tious access goals, increased willingness to partner with the Bank Group. If adaptability is often a benefit, complex project designs undermine effec- tiveness, potentially overwhelming both client capacity and World Bank supervisory capacity. This problem appeared only in World Bank projects, which tend to have multiple components. In Kenya, for example, an effort to address the needs of an emerging extractive sector in energy compre- hensively resulted in a highly complex project. The $50 million 2014 Kenya Petroleum Technical Assistance Project originally included 11 project com- ponents or subcomponents and three project implementation units, which 47 had to engage with 21 counterpart executive agencies. At midterm, when it was only 19 percent disbursed, the project had to be restructured to simplify it and reduce the implementation agencies to one project implementation unit. Another example is the $250 million 2004 Enterprise Reform and Bank Modernization Project (P081969) in Bangladesh. This project was faulted in IEG’s Implementation Completion and Results Report Review for com- bining too many elements into a single project. Early attention focused on privatization of manufacturing industries (especially jute mills), with a loss of focus on state-owned commercial bank privatization, which the project also supported. A separate operation might have handled addressing state- owned bank privatization better, and by the project’s end, the window for reform had closed. In Ukraine, IEG found that a World Bank investment loan became too complex when it added substantial energy sector policy reform objectives. “Combining an ambitious sector reform program with signifi- cant investment activities in a fragile political economy carries high risks” (World Bank 2017e). In the Democratic Republic of Congo, a component of the $120 million Private Sector Development and Competitiveness Project (2003) on SOE reform was found to be too ambitious given the limited re- sources allocated, the weak client capacity, and the politically fragile envi- ronment. Supervision is important to success, and having country and regional office experts on site helps. In the power sector in Bangladesh and Kenya and in Ukraine’s financial sector, on-site experts forged trust and partnership over longer tenures, reducing transaction costs and strengthening implementa- tion and oversight. For IFC, evaluation showed that its success in supporting the initial public offering (privatization) of the geothermal energy company Energy Development Corporation in 2007 was enabled by a project team that State Your Business!  Chapter 2 included experienced local business developers with strong client relation- ship skills, and officers with power sector, corporate governance, and SOE knowledge and strong processing capabilities. However, experts not located on site sometimes faced problems, such as in supporting the Extractive In- dustries Transparency Initiative in Ukraine. Positive aspects of team compo- sition include strong local presence, necessary skills (or ready access to tech- nical guidance), and continuity in project teams and supervision. Negative aspects of team composition include a high turnover of task team leaders, 48 lack of adequate expertise in project teams, and delayed establishment of the project implementation unit. For example, the quality of supervision for the Mali Energy Support Project suffered from high turnover of task teams and the fact that most task team leaders were located in Washington, DC, instead of in the field (World Bank 2019b). The econometric analysis supports the predictive power of team composition (a combination of expertise, experi- ence, and stability) as a predictor of success. Projects in which M&E frameworks contributed to success had a clear state- ment of objectives and indicators that captured the achievement of the proj- ect’s development objectives. Such projects had well-specified actions, clear and monitorable outcome indicators, baseline and target values, and sources of information for tracking progress. For example, after restructuring, the modified indicators for the Zambia Increased Access to Electricity Services Project were found to be appropriately linked to the objectives and properly designed to monitor progress toward the project objectives. Furthermore, the M&E framework was useful to monitor progress and aided project refine- ment over the course of implementation (World Bank 2016c). Poor M&E design (including results frameworks) undermined effectiveness. The portfolio review yields many examples of where effectiveness was con- strained by projects failing to do the following: incorporate relevant quanti- tative indicators into the Project Appraisal Document for tracking the prog- ress of projects, have consistent indicators over the life of the project and Independent Evaluation Group World Bank Group across similar projects, generate baseline data, establish a clear relationship between project activities and outcome indicators, report only outcomes attributable to the project, or effectively monitor or recalibrate indicators to reflect project changes. For example, the 2006 China Economic Reform Implementation Project lacked well-defined and measurable outcome indi- cators that could be monitored to facilitate project implementation. It failed to establish targets for its outcome indicators and monitor outcomes at the subproject level. Thus, the lack of relevant monitoring information was found to have impeded a midcourse correction during supervision. For the Ghana 2006 Economic Management Capacity Building Project ($50 million), the lack of a clear relationship between project activities and the outcome indicators made it difficult to see if the project was making adequate prog- 49 ress. In IFC’s 2011 project investing $307 million in VietinBank for small and medium enterprise banking and risk management, evaluation found that the M&E framework did not incorporate relevant indicators to track project results and performance against project objectives for the risk management component. The econometric analysis confirmed the significance of a good M&E framework as a predictor of intervention success (appendix D). Sequential and complementary interventions aid success. Sequential engage- ments involving financing and technical and analytic support built institu- tional and physical capacity, and the trust of underlying relationships carried reform momentum through difficult periods.5 The econometric analysis con- firmed that sequencing can be a significant predictor of intervention success. The case studies exemplify the benefits of sequenced and complementary engagements (box 2.3). In Kenya, for example, sequential engagements in the power sector supported broad reform, including upstream support of sectoral policy and planning; construction and rehabilitation of generation, transmis- sion, and distribution infrastructure; capacity-building assistance to utilities, including improvement of their corporate governance; and strengthening of financing and the ability of state-owned utilities to attract long-term private capital to refinance short-term debt (see Kenya case study, appendix F).  equencing and Complementarity to Build Credibility Box 2.3. S and Capacity The World Bank Group strategies and programs in Bangladesh’s power sector were aligned with successive government five-year plans. From at least 2004, the Bank Group engaged in unbundling and building technical capacity through financing and State Your Business!  Chapter 2 technical assistance. This covered regulation, generation, transmission, and distri- bution. The World Bank also supported the Power Cell, which channeled technical, planning, and coordination support to government while facilitating the role of private power producers. The Power Cell is an acknowledged success, and the client owns it fully. The regulator, Bangladesh Energy Regulatory Commission, has benefited from Bank Group support since its creation. The World Bank, the International Finance Cor- poration (IFC), and the Multilateral Investment Guarantee Agency were all involved in a Cascade-type approach in supporting independent power providers. Over time, sector 50 (continued)  equencing and Complementarity to Build Credibility Box 2.3. S and Capacity (continued) performance improved through reduced losses, reduced arrears, and an elimination of the energy gap. The Bank Group became a trusted partner in energy state-owned enterprise (SOE) reform through its expertise in the field, access to global expertise, long-standing relationships with key government agencies, coordination of donors, and consistent policy view. In Vietnam’s power sector, the Bank Group engaged comprehensively in all aspects of the power sector (rural electrification, generation, transmission, distribution, load dispatch, renewables, development of wholesale and retail power markets, regula- tory aspects, and SOE reform). The credibility and trust generated enabled the Bank Group to support the government in sequencing sectorwide reform. The Bank Group tapped a wide range of instruments to support SOE reform in both the energy and financial sectors, including seven Poverty Reduction Support Credits, three Economic Management and Competitiveness Credits, three power sector development policy operations, an energy sector loan, and four financial sector lending projects, along with significant analytical work. IFC engagement included four advisory services and one investment project in the financial sector. The Multilateral Investment Guarantee Agency provided a guarantee for a hydropower project. China’s financial sector saw complementary support by IFC (piloting state bank privat- izations downstream) and the World Bank (knowledge generation, including flagship policy reports; joint studies; policy dialogue; and technical assistance). When the gov- Independent Evaluation Group World Bank Group ernment chose to partner with IFC, IFC created models meant to have demonstration effects. IFC investments in the first decade of the 2000s supported the privatization or restructuring of three SOEs, and a focus on frontier regions contributed to increasing foreign direct investment flows. IFC worked to attract private investment to diversify state-owned financial institution ownership, also providing advisory services on insol- vency and corporate governance. IFC investments and involvement in investee boards supported good corporate governance practices. Source: Independent Evaluation Group case studies (see appendix F). 51 Analytic work before financing interventions figures prominently in se- quencing and is the second most common factor of success or failure that project evaluations identified. The Bank Group has a plethora of analytic products producing findings relevant to SOE reform (box 2.4). Beyond in- forming operations, the Bangladesh and Ukraine case studies reveal that even where reforms stalled and the World Bank disengaged financially for a period, an ongoing program of analytic work in each case kept the World Bank current and offered a basis for reforms once conditions allowed. In most of the case studies, the Bank Group’s ability to share relevant knowl- edge arising from analytic work was a key source of comparative advantage among donors. In Serbia, a decade of sequential and complementary support built a strong partnership on SOE reform. The Bank Group supported the Serbian govern- ment in reforming commercial SOEs for more than a decade and was one of the government’s few trusted partners. World Bank projects supported SOE restructuring, privatization, and improved corporate governance. IFC adviso- ry services supported corporate governance reform. IFC joined the European Bank for Reconstruction and Development to provide long-term financing to two of IFC’s state-owned bank clients to prepare them for privatization. In 2015, one of the banks was sold, and the other, Komercijalna Banka ad Beo- grad, continued to be reformed and is expected to be sold to a private owner in late 2020. Other World Bank interventions aimed to improve fiscal disci- pline and management, reduce direct and indirect SOE subsidies, implement an electricity tariff adjustment, and support public expenditure reform. The World Bank also supported assistance to workers scheduled to be laid off in SOE restructuring and privatization and supported reforms in M&E, audit- ing, worker safeguards, and environmental standards compliance. Similar State Your Business!  Chapter 2 cases were found in the power sectors in Bangladesh, Ukraine, and Vietnam and in China’s financial sector. IFC has also had sequential SOE reform engagements with key clients with notable effects. In 2005, IFC assisted the Bank of Beijing to strengthen its capital base, introduce international standards and practices, become a com- petitive regional player in the market, improve corporate governance and transparency, and establish an environmental and management system. The 52 project benefited from an effective integration between investment and ad- visory services. IEG’s evaluation found that IFC placed a high regional prior- ity on the bank and had a clear engagement plan, which it updated regularly. In 2007, IFC invested in a risk-sharing facility for the China Utility-Based Energy Efficiency Finance Program, building on the Bank of Beijing’s partic- ipation in IFC’s sustainable finance training. The relationship team actively coordinated introducing the program to the World Bank. An advisory team provided tailored services to the Bank of Beijing headquarters and branches in market development, product design, technical assessment, and relation- ship brokering with energy-efficiency vendors. In FY10, four training courses and three market promotions were conducted, which helped the Bank of Bei- jing establish its capacity in energy-efficiency finance, particularly for small and medium enterprises.  tate-Owned Enterprise Reform in World Bank Group Box 2.4. S Analytic Work The Independent Evaluation Group (IEG) found substantial Bank Group analytic work identifying state-owned enterprise (SOE) reform priorities. Systematic Country Diagnostics (SCDs). In a sample of 46 countries, IEG identified 39 SCDs, and 92 percent identified SOEs as a reform priority. Ninety percent of the SCDs focused on energy sector SOEs, and 26 percent focused on financial sector SOEs. These clearly informed country strategies: for the same sample of countries, Independent Evaluation Group World Bank Group 83 percent of Country Partnership Frameworks foresaw Bank Group work to support SOE reforms, mostly in sector-level regulatory framework reform and enterprise-level business and operational reform. SOE reform in more than one area is common. Country Private Sector Diagnostics (CPSDs). All 10 International Finance Corporation– World Bank Country Private Sector Diagnostics (CPSDs) that IEG reviewed addressed SOE reforms. This relatively new, joint International Finance Corporation–World Bank CPSD tool aims to inform the SCD (and thus the Country Partnership Framework) and operations by assessing constraints to and opportunities for private sector–led growth. The CPSDs identi- fied SOE issues ranging from reducing crowding out within the financial sector to reforming the independent power producer regime and sector planning in the energy sector. Owner- ship reforms were most frequent among the recommendations, but reforms in regulation, (continued) 53  tate-Owned Enterprise Reform in World Bank Group Box 2.4. S Analytic Work (continued) corporate governance, and competition were also common. Two recent CPSDs (Morocco and Rwanda) adopted the Markets and Competition Policy Assessment Tool’s competitive neutrality framework. The Morocco CPSD employs a full competitive neutrality gap analysis focused on SOE advantages, and the Rwanda CPSD makes competitive neutrality and strengthening of competition policy a key focal point. Financial Sector Assessment Programs. Most Financial Sector Assessment Program reports, conducted jointly by the International Monetary Fund and the World Bank, discuss and make recommendations about the reform of state-owned financial institutions. The same sample of countries for SCDs yielded 29 Financial Sector Assessment Programs, three-quarters of which substantially discussed these issues and reforms, most focusing on state-owned commercial banks. Upstream, the greatest focus was on sector regulatory frameworks, governance, and ownership. Downstream, the focus was on firm ownership. Markets and Competition Policy Assessment Tool. This tool was introduced in 2016. By late 2019, IEG found only seven countries and one subregion with comprehensive Bank Group competition analyses. This analytic work systematically addresses both industrial structure and competitive neutrality issues pertaining to SOEs. SOE Corporate Governance Assessments. IEG reviewed nine SOE Corporate Gov- ernance Assessments that followed the 2014 SOE Corporate Governance Tool with some variations. For example, only five of the nine addressed the first Organisation for Economic Co-operation and Development Corporate Governance of SOE Guidelines principle: the rationale for state ownership. Integrated SOE Framework. IEG reviewed two works under the Integrated SOE State Your Business!  Chapter 2 Framework label: one for Niger and one for Sri Lanka. Substantial variations between the two suggest this product is still in development. Source: Independent Evaluation Group reviews of analytic work for this evaluation. Note: Late in the evaluation period, the World Bank introduced a new product, the Infrastructure Sector Assessment Program (InfraSAP), which aims to identify a mix of policy reforms and advisory and investment activities that would maximize commercial and private finance for infrastructure. However, only one relevant country study and one regional study were completed during the evalu- 54 ation period. Collaboration among Bank Group institutions, although relatively rare, can provide complementary support that aids SOE reform success. Among the diagnostic products described in box 2.4, the CPSD, which consistently treats SOE reform, is a strategic tool that IFC and the World Bank produce jointly. CPSDs provide a shared perspective on challenges and opportunities for pri- vate sector development (including concerning SOEs) as an input to System- atic Country Diagnostics and Country Partnership Frameworks. The MCPAT is also a shared World Bank and IFC platform of analysis for regulation and competition. IEG found in several case studies an operationally collaborative approach consistent with Cascade, under which the World Bank, IFC, and MIGA work to their comparative advantages. Institutional collaboration to mobilize pri- vate financing is a key expectation raised in the MFD and Cascade approach- es.6 The MFD and Cascade reinforce the existing World Bank Group Strategy calling for a “One World Bank Group” approach.7 Yet project evaluations identify collaboration as a factor that facilitated or constrained success in only 13 of 294 SOE reform projects. Experiences in several countries show the operational promise of MFD and its embedded Cascade approach in power generation where Bank Group institutions work together to create conditions that attract private invest- ment. For example, in Kenya, the Bank Group helped attract private invest- ment in independent power providers through a risk mitigation package Independent Evaluation Group World Bank Group supported by International Development Association partial risk guarantees, MIGA guarantees, and IFC investment loans. This combination, together with IFC’s leadership in establishing a consortium of other financiers, raised private investors’ comfort level. The initiative stimulated the construction of 302 megawatts of installed power generation capacity, equivalent to about one-quarter of total national power consumption at the time. In Bangladesh, IFC’s joint engagement with MIGA and the Japan International Cooperation Agency enabled a PPP for the Sirajganj 4 generator project. This facilitated construction of one of the country’s largest and most efficient gas turbine combined cycle power stations to address chronic energy shortages and sup- ply instability. 55 As observed in a recent IEG evaluation of joint projects, coordination re- quires an informed perspective on both benefits and costs (World Bank 2017f). For example, joint projects have been especially helpful in high-risk contexts. They have worked best where the Bank Group had a clear compar- ative advantage and where the roles, division of labor, and responsibilities among the different Bank Group institutions and respective project teams were clear. Realistically, the costs involved for internal coordination can require additional resources for administration, preparation, and implemen- tation, regardless of commitment amounts. At a corporate level, there is no clear road map for collaboration to support SOE reform. The implications of MFD and the Cascade and of sector strat- egies for how Bank Group institutions can work together to support SOE reform have not been spelled out. The implications of IFC’s new emphasis on upstream engagement to create, deepen, and expand markets bring new op- portunities and challenges to coordination with the World Bank and MIGA. State Your Business!  Chapter 2 56 1  In this chapter, the Independent Evaluation Group triangulates from multiple evidence sourc- es on World Bank Group effectiveness and factors influencing it. These include the portfolio review, microevaluations, country case studies, deep dives, literature reviews, and econometric analysis. The portfolio analysis draws on 294 evaluated projects with 671 intervention-level ratings. Because World Bank advisory services and analytics has no validated results frame- work, it does not represent a relevant evidence source. The Independent Evaluation Group undertook a rigorous econometric analysis to assess success factors to identify plausible ex- planatory variables associated with the achievement of state-owned enterprise (SOE) reform intervention outcomes, introducing country-level control variables. The analysis was based on a logistic regression model that sought to identify potential predictors of SOE intervention suc- cess. It included individual factors coded in the portfolio review and analysis, composite factors identified through principal component analysis, and country-level variables. The appendixes summarize findings from the portfolio analysis and the country case studies, econometric anal- ysis, and country diagnostics and strategy reviews; however, they are not individually cited for each finding in the chapter. 2  Imam, Jamasb, and Llorca (2019) found that in Sub-Saharan Africa, corruption reduces electricity sector technical efficiency and constrains efforts to increase access to electricity and national income. Chen et al. (2016) found that corruption is associated with underperfor- mance of state banks.  3  Analytical work was an important factor that influenced the positive outcome of 59 evaluat- ed operations. Analysis showed that the Bank Group used analytical work to support project design, implementation, and government and client capacity building, as well as to ensure Independent Evaluation Group World Bank Group continuity in its engagement. For example, in Turkey, the Second Competitiveness and Em- ployment development policy loan (P096840, FY08), which achieved its objective of privat- izing an SOE and selling the state-owned assets, benefited from a large number of analytical works that fed a lasting, stable policy dialogue in the areas supported by the reform. 4  Dinc (2005) provides cross-country evidence that government-owned banks increase their lending in election years relative to private banks; the effect is about 11 percent of a govern- ment-owned banks total loan portfolio. Claessens, Feijen, and Laeven (2008) show that Bra- zilian firms contributing to winning campaigns increase their bank financing relative to a con- trol group after each election, with an economic cost of at least 0.2 percent of gross domestic product. Cole (2009) shows that Indian government–owned banks increase agricultural credit by 5 to 10 percentage points in an election year with no significant impact on agricultural out- put. Bircan and Saka (2018) find for Turkey that state-owned banks systematically adjust their 57 lending in relation to local elections compared with private banks in the same province, based on electoral competition and political alignment of incumbent mayors, with negative effects for firms in opposition-dominated areas. 5  The International Finance Corporation (IFC), in its comments to IEG on the draft evaluation, notes that where privatization is not immediately possible, SOE reform can, over time, help SOEs graduate from sovereign-guaranteed borrowing to borrow from the IFC (directly and through mobilization) and then graduate to commercial-only borrowing, when possible, in- cluding direct access to capital markets. The IFC also notes that SOEs are central to the devel- opment of capital markets in most emerging markets. Underperforming SOEs can potentially limit the development of strong, sustainable capital markets. 6  “The MFD [Maximizing Finance for Development] approach builds on substantial cross– Bank Group experience in working with governments to crowd in the private sector to help meet development goals. MFD seeks to make this systematic. Recent examples of cross–Bank Group collaboration which have crowded in private solutions provide some important les- sons.… These highlight the importance of country ownership and of upstream knowledge and advisory work in helping clients improve investment environments, the complementarity of different Bank Group interventions in transforming the sectors, and the benefits of collabora- tion with other development partners” (World Bank 2017a). 7  In 2013, “One World Bank Group” was enshrined in the World Bank Group Strategy, which stated: “The new Strategy encompasses the concept of acting as One World Bank Group, sig- nificantly increasing collaboration across its agencies.… The One World Bank Group approach will entail joint projects managed more collaboratively than in the past” (World Bank Group, 2013). State Your Business!  Chapter 2 58 3 | Recommendations to Address Outstanding Challenges in SOE Reform Highlights Building on lessons from the two focal sectors, this chapter ex- amines two areas where the World Bank Group can learn from successful features of state-owned enterprise (SOE) reform and, by addressing outstanding challenges, enhance selectivity and improve internal coordination for SOE reforms. The Bank Group faces two key challenges in its work on SOE re- form: selectivity and consistent coordination and implementation of its corporate strategy. The Bank Group should reconsider how it engages in countries where initial conditions for success do not prevail: where there is weak control of corruption and where there is a lack of competitive neutrality for SOEs. Recommendation 1: The World Bank Group should apply a selec- tivity framework for SOE reform support that considers country governance conditions, control of corruption, and sector and enter- prise-level competition. This evaluation generally finds positive experiences when Bank Group institutions collaborate on SOE reform. 59  The spirit of Maximizing Finance for Development is not fulfilled if the menu of options supported does not include the full range of private sector solutions, including ownership reform and privatization. Recommendation 2: The Bank Group should apply Maximizing Finance for Development and its embedded Cascade approach to SOE reform. This would enhance internal coordination and mobilize private financing and capacity, especially for ownership reforms. 60   Enhancing Selectivity: Corruption and Competition Two key findings of the evaluation on the conditions for successful engage- ment to support SOE reforms are that the Bank Group has more successful outcomes in countries with better control of corruption and that SOEs perform better in both focal sectors (and in general) where competitive conditions prevail at the sector and enterprise levels. Both can be incorporated into ap- proaches to selectivity and mitigation of risks when planning for SOE reforms. The Bank Group’s SOE reform portfolio is concentrated in countries where it is more likely to succeed, but a substantial minority (26 percent) of inter- ventions are in countries with weak control of corruption, where all types of reforms are less likely to succeed. This raises questions about how the Bank Group should engage to enhance the likelihood of success of SOE reforms. Competition and competitive neutrality at SOEs’ sector and enterprise levels are vital to SOE performance, yet Bank Group analysis on competition has been insufficient. The Bank Group has an array of valuable strategic and diagnostic instruments on competition but has used them infrequently. The MCPAT, which is the Bank Group’s main tool to diagnose competition issues, has covered only nine countries and one subregion during the evaluation pe- riod. In addition, a small number of CPSDs to date have deployed the MCPAT framework. Although IFC and MIGA policies indicate the need in many cases Independent Evaluation Group World Bank Group to verify that a level playing field for competition exists before engaging with an SOE, attention to competitive neutrality in project documentation is generally weak and uneven. Recommendation 1: The World Bank Group should apply a selectivity framework for SOE reform support that considers country governance conditions, control of corruption, and sector and enterprise-level competition. First, the Bank Group should adopt a more selective approach toward SOE engagement in countries with weak control of corruption, giving full attention to internal and external factors of success. Findings suggest that the Bank Group could ramp up engagement with clients where success is more likely. In conditions of weak control of corruption, one option would 61 be to engage first in addressing overall governance quality before attempting SOE reform. Where disengagement on SOE reform is not possible or desir- able, close attention is needed to the factors that may mitigate corruption’s negative influence on SOE reform success, including selectivity toward cli- ents who display commitment, stronger supervision, good (and simple) proj- ect design, and sequencing of activities. Next, the Bank Group should gear up capacity to conduct competition analysis, especially at the project level. The importance of competitive neutrality (the idea that SOEs should be on a level playing field with potential private competitors), especially consid- ering IFC policy and (to a far lesser extent) MIGA policy, indicates a need to ramp up project-level analysis by carrying out competition assessment more systematically and by applying substantial up-front analytic capability to project-specific work on competitive neutrality. This would allow for greater selectivity toward competitive conditions that enhance SOE performance and for establishing up-front mitigating measures if competitive conditions were not conducive to success. SOE Reform through Coordinated, Consistent Application of MFD and the Embedded Cascade Approach This evaluation generally finds positive experiences when the Bank Group collaborates internally on SOE reform both through joint diagnostic (CPSDs and MCPATs) and joint operational approaches. For example, IEG found in several case studies a collaborative approach consistent with the Cascade approach, where the World Bank, IFC, and MIGA work to their comparative advantages, but these cases are infrequent. At the corporate level, there is State Your Business!  Chapter 3 room to spell out the implications of MFD for how Bank Group institutions can work together to support SOE reform systematically. This is particularly important for privatization and ownership reforms to address governments’ increased requests for support in these areas and considering the importance of mobilizing private financing in them. Although the Bank Group’s SOE diagnostic and strategic work includes a high incidence of recommendations on ownership and privatization, the SOE reform portfolio includes few pri- vatization projects. 62 Recommendation 2: The World Bank Group should apply the MFD and its embedded Cascade approach to SOE reform. This would enhance internal coordination and mobilize private financing and capacity, especially for ownership reforms. First, the Bank Group should fur- ther develop and harmonize its diagnostic frameworks applied to SOE reform. This requires developing shared framing tools such as an Integrated SOE Framework and CPSD modules treating private sector options, including privatization and PPPs, for addressing SOE perfor- mance challenges. Second, the Bank Group could apply the Cascade approach, offering clients options for SOE reform that mobilize private financing and capacity through privatization and ownership reform. Along with recommendation 1, given appropriate country and sector condi- tions, there is greater room to apply the Cascade approach through a greater degree of and more routine coordination by World Bank, IFC, and MIGA that builds on their respective comparative advantages. This can be piloted as a sequential process, with upstream interventions focusing on any needed policy and regulatory reforms to create a level playing field for private entry and investment, combined with downstream use of Bank Group instruments to catalyze and mobilize private financing. With careful M&E, such a pilot could inform future efforts to realize the MFD and its Cascade more fully as a systematic approach to SOE reform. Independent Evaluation Group World Bank Group 63 Bibliography Abraham, F., and S. Schmukler. 2017. “A New Role for Development Banks?” Research and Policy Brief 10, World Bank Group, Washington, DC. Abubakar, Mustafa. 2010. “Ministry of State-Owned Enterprises, Indonesia: Towards World Class Corporations.” PowerPoint presentation. Aftab, S., and S. A. Shaikh. 2013. “Reforming State-Owned Enterprises.” Pakistan Policy Note no. 4, World Bank Group, Washington, DC. http://documents.world- bank.org/curated/en/956641468057347062/Reforming-state-owned-enterprises. Allchin, J. 2016. “Bangladesh’s Other Banking Scam.” New York Times, April 11, 2016. https://www.nytimes.com/2016/04/12/opinion/bangladeshs-other-bank- ing-scam.html. Allen, Richard, and Miguel Alves. 2016. How to Improve the Financial Oversight of Public Corporations. Fiscal Policy How-To Notes. Washington, DC: International Monetary Fund. https://www.imf.org/external/pubs/ft/howtonotes/2016/how- tonote1605.pdf. Andrés, L. A., J. L. Guasch, and S. L. Azumendi. 2011. “Governance in State-Owned Enterprises Revisited: The Cases of Water and Electricity in Latin America and the Caribbean.” Policy Research Working Paper 5747, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/757311468046775569/Gover- nance-in-state-owned-enterprises-revisited-the-cases-of-water-and-electrici- ty-in-Latin-America-and-the-Caribbean. State Your Business!  Bibliography Andrés, L. A., J. Schwartz, and J. L Guasch. 2013. Uncovering the Drivers of Utility Per- formance: Lessons from Latin America and the Caribbean on the Role of the Private Sector, Regulation, and Governance in the Power, Water, and Telecommunication Sectors. Directions in Development Series: Infrastructure. Washington, DC: World Bank. https://doi.org/10.1596/978-0-8213-9660-5. Andrianova, S., P. Demetriades, and A. Shortland. 2012. “Government Ownership of Banks, Institutions, and Financial Development.” Economica 79 (313): 449–69. https://www.jstor.org/stable/23274805. 64 Armstrong, P. 2015. “Corporate Governance and State-Owned Enterprises.” Ethical Boardroom. https://ethicalboardroom.com/corporate-governance-and-soes. Arrobio, A. 2017. EFI Integrated Approach for SOEs. Washington, DC: World Bank Group. Bacchiocchi, Emanuele, Matteo Ferraris, Massimo Florio, and Daniela Vandone. 2019. “State-Owned Banks in the Market for Corporate Control.” Journal of Economic Policy Reform 22 (2): 120–47. Bacon, Robert. 2018. “Taking Stock of the Impact of Power Utility Reform in Devel- oping Countries: A Literature Review.” Rethinking Power Sector Reform. Policy Research Working Paper 8460, World Bank Group, Washington, DC. https://doi. org/10.1596/1813-9450-8460. Balbuena, Sultan S. 2016. “Concerns Related to the Internationalisation of State- Owned Enterprises: Perspectives from Regulators, Government Owners and the Broader Business Community.” OECD Corporate Governance Working Papers no. 19, OECD Publishing, Paris. Balza, Lenin, Raul Jiménez, and Jorge Mercado. 2013. “Privatization, Institutional Reform, and Performance in the Latin American Electricity Sector.” Energy Divi- sion Technical Note IDB-TN-599, Inter-American Development Bank, Washing- ton, DC. https://publications.iadb.org/en/publication/11893/privatization-insti- tutional-reform-and-performance-latin-american-electricity. Independent Evaluation Group World Bank Group Beck, T., A. Demirgüç-Kunt, and V. Maksimovic. 2004. “Bank Competition and Access to Finance: International Evidence.” Journal of Money, Credit and Banking 36 (3): 627–48. Beck, T., S. M. Maimbo, I. Faye, and T. Triki. 2011. Financing Africa: Through the Crisis and Beyond. Washington, DC: World Bank. https://documents.worldbank.org/en/ publication/documents-reports/documentdetail/633671468194645126/financ- ing-africa-through-the-crisis-and-beyond. Berger, Allen, George Clarke, Robert Cull, Leora Klapper, and Gregory Udell. 2005. “Corporate Governance and Bank Performance: A Joint Analysis of the Static, Selection, and Dynamic Effects of Domestic, Foreign, and State Ownership.” 65 Journal of Banking and Finance 29 (8–9): 2179–221. https://doi.org/10.1016/j. jbankfin.2005.03.013. Bertay, A., A. Demirgüç-Kunt, and H. Huizinga. 2015. “Bank Ownership and Credit over the Business Cycle: Is Lending by State Banks Less Procyclical?” Journal of Banking and Finance 50 (January): 326–39. https://doi.org/10.1016/j.jbank- fin.2014.03.012. Bircan, Ç., and O. Saka. 2018. “Political Lending Cycles and Real Outcomes: Evidence from Turkey.” Paper presented at the Royal Economic Society Annual Confer- ence 2018, Sussex, UK, March 26–28. Birdsall, N., and J. Nellis. 2005. Privatization Reality Check: Distributional Effects in Developing Countries. Washington, DC: Center for Global Development. Bonnet, C., P. Dubois, D. Martimort, and S. Straub. 2011. “Empirical Evidence on Satisfaction with Privatization in Latin America.” World Bank Economic Review 26 (1): 1–33. https://doi.org/10.1093/wber/lhr037. Boubakri, Narjess, and Jean-Claude Cosset. 1998. “The Financial and Operating Per- formance of Newly Privatized Firms: Evidence from Developing Countries.” The Journal of Finance 53 (3): 1081–110. https://www.jstor.org/stable/117387. Boubakri, Narjess, Jean-Claude Cosset, Klaus Fischer, and Omrane Guedhami. 2005. “Privatization and Bank Performance in Developing Countries.” Journal of Banking and Finance 29 (8–9): 2015–41. https://doi.org/10.1016/j.jbank- fin.2005.03.003. Boubakri, Narjess, Omrane Guedhami, Dev Mishra, and Walid Saffar. 2012. “Political Connections and the Cost of Equity Capital.” Journal of Corporate Finance 18 (3): State Your Business!  Bibliography 541–59. https://doi.org/10.2139/ssrn.1589688. Bovet, D. 1985. “Financial Aspects of Public Enterprises in Sub-Saharan Africa.” Unpublished paper. Böwer, U. 2017. “State-Owned Enterprises in Emerging Europe: The Good, the Bad, and the Ugly.” IMF Working Paper WP/17/221, International Monetary Fund, Washington, DC. 66 Bril-Mascarenhas, Tomas, and Alison E. Post. 2012. “Broad-Based Consumer Subsi- dies as ‘Policy Traps’: The Case of Utilities Subsidies in Postcrisis Argentina.” Paper presented at the American Political Science Association 2012 Annual Meeting, New Orleans, Louisiana, August 30–September 2. Brissimis, Sophocles, Manthos Delis, and Nikolaos Papanikolaou. 2008. “Exploring the Nexus between Banking Sector Reform and Performance: Evidence from Newly Acceded EU Countries.” Journal of Banking and Finance 32 (11): 2674–83. https://doi.org/10.1016/j.jbankfin.2008.07.002. Bussolo, M., S. J. Commander, and S. Poupakis. 2018. “Political Connections and Firms: Network Dimensions.” Policy Research Working Paper WPS8428, World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-8428. Cameron, Peter D., and Michael C. Stanley. 2017. Oil, Gas, and Mining: A Sourcebook for Understanding the Extractive Industries. Washington, DC: World Bank. http:// hdl.handle.net/10986/26130. Camos, Daniel, Robert Bacon, Antonio Estache, and Mohamad Mahgoub Hamid. 2017. Shedding Light on Electric Utilities in the Middle East and North Africa: Insights from a Performance Diagnostic. Washington, DC: World Bank. https://doi. org/10.1596/978-1-4648-1182-1. Campanaro, Alessandra, and Cuong Duc Dang. 2018. Mobilizing Finance for Local Infrastructure Development in Vietnam: A City Infrastructure Financing Facility. Washington, DC: World Bank. Capobianco, A., and H. Christiansen. 2011. “Competitive Neutrality and State- Independent Evaluation Group World Bank Group Owned Enterprises.” OECD Corporate Governance Working Papers no. 1, OECD Publishing, Paris. https://doi.org/10.1787/5kg9xfgjdhg6-en. Chang, H.-J. 2007. “State-Owned Enterprise Reform.” Policy Notes 4, United Nations, Department of Economics and Social Affairs, New York. Chen, Y.-S., Y. Chen., C.-Y. Lin, and Z. Sharma. 2016. “Is There a Bright Side to Gov- ernment Banks? Evidence from the Global Financial Crisis.” Journal of Financial Stability 26: 128–43. 67 Christiansen, Hans. 2011. “The Size and Composition of the SOE Sector in OECD Countries.” OECD Corporate Governance Working Papers no. 5, OECD Publish- ing, Paris. https://doi.org/10.1787/5kg54cwps0s3-en. Christiansen, Hans. 2013. “Balancing Commercial and Non-Commercial Priorities of State-Owned Enterprises.” OECD Corporate Governance Working Papers no. 6, OECD Publishing, Paris. https://doi.org/10.1787/5k4dkhztkp9r-en. Christiansen, Hans, and Yunhee Kim. 2014. “State-Invested Enterprises in the Global Marketplace: Implications for a Level Playing Field.” OECD Corporate Gover- nance Working Papers no. 14, OECD Publishing, Paris. https://doi.org/10.1787/ 5jz0xvfvl6nw-en. Chunlin, Z., and W. Lihong. 2009. Effective Discipline with Adequate Autonomy: The Direction for Further Reform of China’s SOE Dividend Policy. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/358411468024535236/ Effective-discipline-with-adequate-autonomy-the-direction-for-further-re- form-of-Chinas-SOE-dividend-policy. Cihak, M., and A. Demirgüç-Kunt. 2013. Rethinking the State’s Role in Finance. Pol- icy Research Working Paper 6400, World Bank, Washington, DC. https://doi. org/10.1596/1813-9450-6400. Claessens, S., E. Feijen, and L. Laeven. 2008. “Political Connections and Preferential Access to Finance: The Role of Campaign Contributions.” Journal of Financial Economics 88 (3): 554–80. Claessens, Stijn, and Burcin Yurtoglu. 2012. Focus 10: Corporate Governance and De- velopment—An Update. Washington, DC: International Finance Corporation. State Your Business!  Bibliography Clarke, George R. G., Robert Cull, and Michael Fuchs. 2009. “Bank Privatization in Sub-Saharan Africa: The Case of Uganda Commercial Bank.” World Development 37 (9): 1506–21. https://doi.org/10.1016/j.worlddev.2009.01.007. Clarke, George R. G., Robert Cull, and Mary M. Shirley. 2005. “Bank Privatization in Developing Countries: A Summary of Lessons and Findings.” Journal of Banking and Finance 29 (8–9): 1905–30. https://doi.org/10.1016/j.jbankfin.2005.03.006. Cole, Shawn. 2009. “Fixing Market Failures or Fixing Elections? Agricultural Credit in India.” American Economic Journal: Applied Economics 1 (1): 219–50. 68 Council of European Energy Regulators. 2016. Status Review on the Implementation of Distribution System Operators’ Unbundling Provisions of the 3rd Energy Package. CEER Status Review C15-LTF-43-03. Brussels: CEER. https://www.ceer.eu/docu- ments/104400/-/-/882514d5-c57f-86f8-50a3-90185e270f15. Crane-Charef, Mary. 2015. “Stocktaking of Anti-Corruption and Business Integrity Measures for Southern African SOEs.” OECD Corporate Governance Working Papers no. 18, OECD Publishing, Paris. https://doi.org/10.1787/5jrtd6fghvf1-en. Cull, Robert, Wei Li, Bo Sun, and Lixin Colin Xu. 2013. Government Connections and Financial Constraints: Evidence from a Large Representative Sample of Chinese Firms. Washington, DC: World Bank. https://doi.org/10.1596/1813-9450-6352. Cull, Robert, and María Soledad Martínez Pería. 2008. “Crises as Catalysts for Foreign Bank Activity in Emerging Markets.” In Power and Politics after Financial Crises, edited by Justin Robertson, 52–81. London: Palgrave Macmillan. https://doi. org/10.1057/9780230235366_3. Cull, Robert, María Soledad Martínez Pería, and Jeanne Verrier. 2018. “Bank Own- ership—Trends and Implications.” Policy Research Working Paper 8297, World Bank, Washington, DC. http://documents1.worldbank.org/curated/ en/810621515444012541/pdf/WPS8297.pdf. Cull, Robert, and Connor P. Spreng. 2008. “Pursuing Efficiency while Maintaining Outreach: Bank Privatization in Tanzania.” Policy Research Working Paper WPS4804, World Bank, Washington, DC. http://documents.worldbank.org/curat- Independent Evaluation Group World Bank Group ed/en/2008/12/10140391/pursuing-efficiency-maintaining-outreach-bank-pri- vatization-tanzania. Curi, Claudia, Justas Gedvilas, and Ana Lozano-Vivas. 2016. “Corporate Governance of SOEs and Performance in Transition Countries. Evidence from Lithuania.” Modern Economy 7 (12). De Halleux, Morgane, Antonio Estache, and Tomas Serebrisky. 2019. “Governance Choices and Policy Outcomes in the Latin American and Caribbean Electricity Sector.” Working Papers, ECARES 2019-08, Universite Libre de Bruxelles, Brux- elles, Belgium. 69 de Janvry, A., and E. Sadoulet, eds. 2012. The World Bank Economic Review, vol. 27. Oxford, UK: Oxford University Press. de la Torre, A., S. L. Schmukler, and J. C. Gozzi. 2017. Innovative Experiences in Access to Finance: Market-Friendly Roles for the Visible Hand? Washington, DC: World Bank. De Luna-Martínez, J., and C. Vicente. 2012. “Global Survey of Development Banks.” Policy Research Working Paper 5969, World Bank Group, Washington, DC. http://documents.worldbank.org/curated/en/313731468154461012/Global-sur- vey-of-development-banks. Denizer, Cevdet, Daniel Kaufmann, and Aart Kraay. 2013. “Good Countries or Good Projects? Macro and Micro Correlates of World Bank Project Performance.” Journal of Development Economics 105 (November): 288–302. https://doi. org/10.1016/j.jdeveco.2013.06.003. Dinc, Serdar. 2005. “Politicians and banks: Political influences on govern- ment-owned banks in emerging markets” Journal of Financial Economics 77(2): 453–479. Dinc, Serdar, and Nandini Gupta. 2011. “The Decision to Privatize: Finance and Politics.” The Journal of Finance 66 (1): 241–69. https://doi.org/10.1111/j.1540- 6261.2010.01631.x. Doidge, Craig, G. Andrew Karolyi, and Rene M. Stulz. 2007. “Why Do Countries Mat- ter So Much for Corporate Governance?” Journal of Financial Economics 86 (1): 1–39. https://doi.org/10.1016/j.jfineco.2006.09.002. Dow, W. H., E. C. Norton, and J. T. Donahoe. 2019. “Stata Tip 134: Multiplicative and Marginal Interaction Effects in Nonlinear Models.” The Stata Journal 19 (4): State Your Business!  Bibliography 1015–20. https://doi.org/10.1177/1536867X19893644. Dressen, R., J. Dyer, and Z. Northrip. 2002. “Turning Around State-Owned Banks in Underserved Markets.” Small Enterprise Development 13 (4): 58–67. https://doi. org/10.3362/0957-1329.2002.046. Du, L., A. Harrison, and G. Jefferson. 2011. “Do Institutions Matter for FDI Spillovers? The Implications of China’s Special Characteristics.” Policy Research Working Paper 5757, World Bank Group, Washington, DC. http://documents.worldbank. 70 org/curated/en/542161468220516003/Do-institutions-matter-for-FDI-spill- overs-the-implications-of-Chinas-special-characteristics. Eberhard, Anton, Orvika Rosnes, Maria Shkaratan, and Haakon Vennemo. 2011. Africa’s Power Infrastructure: Investment, Integration, Efficiency. Directions in Development Series: Infrastructure. Washington, DC: World Bank. https://open- knowledge.worldbank.org/handle/10986/2290. Economist. 2012. “The Rise of State Capitalism: The Emerging World’s New Model.” Economist Special Edition, January 21, 2012. Economist. 2014. “State-Owned Banks in Bangladesh: From Cancer to Pimple.” Economist, April 16, 2014. https://www.economist.com/finance-and-econom- ics/2014/04/16/from-cancer-to-pimple. Estache, Antonio. 2020. “Institutions for Infrastructure in Developing Countries: What We Know and the Lot We Still Need to Know.” In The Handbook of Eco- nomic Development and Institutions, edited by J. M. Baland, F. Bourguignon, J. P. Platteau, and T. Verdier, 634–88. Princeton, NJ: Princeton University Press. Estrin, Saul, and Adeline Pelletier. 2018. “Privatization in Developing Countries: What Are the Lessons of Recent Experience?” The World Bank Research Observer 33 (1): 65–102. https://doi.org/10.1093/wbro/lkx007. Farazi, S., E. Feyen, and R. Rocha. 2011. “Bank Ownership and Performance in the Middle East and North Africa Region.” Policy Research Working Paper 5620, Independent Evaluation Group World Bank Group World Bank, Washington, DC. Favaro, E., K. Smits, and M. Bakanova. 2012. “Structural Challenges for SOEs in Belarus: A Case Study of the Machine Building Sector.” Policy Research Work- ing Paper 6010, World Bank, Washington, DC. http://documents.worldbank.org/ curated/en/174931468206680605/Structural-challenges-for-SOEs-in-Belarus-a- case-study-of-the-machine-building-sector. Ferrari, A., D. S. Mare, and I. Skamnelos. 2017. State Ownership of Financial Institu- tions in Europe and Central Asia. Washington, DC: World Bank Group. https:// doi.org/10.1596/1813-9450-8288. 71 Foster, Vivien, and Anshul Rana. 2020. Rethinking Power Sector Reform in the Devel- oping World. Sustainable Infrastructure Series. Washington, DC: World Bank Group. http://hdl.handle.net/10986/32335. Foster, Vivien, Samantha Witte, Sudeshna Ghosh Banerjee, and Alejandro Moreno. 2017. “Charting the Diffusion of Power Sector Reforms across the Developing World.” Rethinking Power Sector Reform, Policy Research Working Paper 8235, World Bank Group, Washington, DC. http://documents.worldbank.org/curated/ en/576801510076208252/Charting-the-diffusion-of-power-sector-reforms- across-the-developing-world. Francisco, M., Y. Mascaró, J. C. Mendoza, and J. Yaron. 2008. “Measuring the Perfor- mance and Achievement of Social Objectives of Development Finance Insti- tutions.” Policy Research Working Paper 4506, World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-4506. Frederick, W. R. 2011. “Enhancing the Role of the Boards of Directors of State- Owned Enterprises.” OECD Corporate Governance Working Papers no. 2, OECD Publishing, Paris. https://doi.org/10.1787/5kg9xfg6n4wj-en. Fu, Xiaoqing, and Shelagh Heffernan. 2009. “The Effects of Reform on China’s Bank Structure and Performance.” Journal of Banking and Finance 33 (1): 35–52. https://doi.org/10.1016/j.jbankfin.2006.11.023. Galal, Ahmed, Leroy Jones, Pankaj Tandon, and Ingo Vogelsang. 1994. Welfare Con- sequences of Selling Public Enterprises: An Empirical Analysis. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/517121468052737074/ Welfare-consequences-of-selling-public-enterprises-an-empirical-analy- sis-a-summary. State Your Business!  Bibliography Gelb, A., S. Tordo, H. Halland, N. Arfaa, and G. Smith. 2014. “Sovereign Wealth Funds and Long-Term Development Finance: Risks and Opportunities.” Policy Re- search Working Paper 6776, World Bank, Washington, DC. http://hdl.handle. net/10986/17313. Ghulam, Yaseen. 2017. “Effect of Privatization and Broader Reforms on Labor Use Adjustments and Efficiency.” Annals of Public and Cooperative Economics 88 (4): 499–518. https://doi.org/10.1111/apce.12164. 72 Girma, S., and Y. Gong. 2008. “FDI, Linkages and the Efficiency of State-Owned Enterprises in China.” The Journal of Development Studies 44 (5): 728–49. https:// doi.org/10.1080/00220380802009233. Gonzalez-Garcia, Jesus, and Francesco Grigoli. 2013. “State-Owned Banks and Fiscal Discipline.” IMF Working Paper 13/206, International Monetary Fund, Washing- ton, DC. https://www.imf.org/external/pubs/ft/wp/2013/wp13206.pdf. Gore, Christopher D., Jennifer N. Brass, Elizabeth Baldwin, and Lauren M. McLean. 2019. “Political Autonomy and Resistance in Electricity Sector Liberalization in Africa.” World Development 120 (August): 193–209. https://doi.org/10.1016/j. worlddev.2018.03.003. Grossi, G., U. Papenfuß, and M.-S. Tremblay. 2015. “Corporate Governance and Ac- countability of State-Owned Enterprises: Relevance for Science and Society and Interdisciplinary Research Perspectives.” International Journal of Public Sector Management 28 (4/5): 274–85. https://doi.org/10.1108/IJPSM-09-2015-0166. Gupta, Nandini., Jhon C. Ham, and Jan Svejnar. 2008. “Priorities and Sequencing in Privatization: Evidence from Czech Firm Panel Data.” European Economic Review 52 (2): 183–208. https://doi.org/10.1016/j.euroecorev.2007.05.004. Habib, Ahsan, Haiyan Jiang, and Donghua Zhou. 2014. “Audit Quality and Market Pricing of Earnings and Earnings Components in China.” Asian Review of Ac- counting 22 (1): 20–34. https://doi.org/10.1108/ARA-05-2013-0034. Independent Evaluation Group World Bank Group Halland, H., M. Lokanc, and A. Nair. 2015. The Extractive Industries Sector: Essentials for Economists, Public Finance Professionals, and Policy Makers. Washington, DC: World Bank. https://doi.org/10.1596/978-1-4648-0492-2. Hallward-Driemeier, M. 2009. “Who Survives? The Impact of Corruption, Competi- tion and Property Rights Across Firms.” World Bank Policy Research Working Paper 5084, World Bank, Washington, DC. Hamilton, Ron, and Alexander Berg. 2008. “Corporate Governance of SOEs in Zam- bia.” Unpublished Manuscript. Washington, DC: World Bank. Havrylyshyn, Oleh, and Donal McGettigan. 2000. “Privatization in Transition Countries.” Post-Soviet Affairs 16 (3): 257–86. https://doi.org/10.1080/106058 6X.2000.10641488. 73 Heo, Kyoungsun. 2018. “Effects of Corporate Governance on the Perfor- mance of State-Owned Enterprises.” Policy Research Working Paper 8555, World Bank, Washington, DC. http://documents.worldbank.org/curated/ en/523421534424982014/Effects-of-corporate-governance-on-the-perfor- mance-of-state-owned-enterprises. Hertog, S. 2010. “Defying the Resource Curse: Explaining Successful State-Owned Enterprises in Rentier States.” World Politics 62 (2): 261–301. https://www.jstor. org/stable/40646201. Ho, Po-Hsin, Chih-Yung Lin, and Wei-Che Tsai. 2016. “Effect of Country Governance on Bank Privatization Performance.” International Review of Economics and Finance 43: 3–18. Huang, Z., L. Li, G. Ma, and L. Xu. 2015. “Hayek, Local Information, and the De- centralization of State-Owned Enterprises in China.” Policy Research Work- ing Paper 7321, World Bank, Washington, DC. http://documents.worldbank. org/curated/en/2015/06/24680786/hayek-local-information-decentraliza- tion-state-owned-enterprises-china. Huenteler, J., I. Dobozi, A. Balabanyan, and S. G. Banerjee. 2017. “Cost Recovery and Financial Viability of the Power Sector in Developing Countries: A Literature Review.” Rethinking Power Sector Reform, Policy Research Working Paper 8287, World Bank Group, Washington, DC. http://documents.worldbank.org/curated/ en/896141513777558023/Cost-recovery-and-financial-viability-of-the-pow- er-sector-in-developing-countries-a-literature-review. Iannotta, G., G. Nocera, and A. Sironi. 2013. “The Impact of Government Ownership on Bank Risk.” Journal of Financial Intermediation 22 (2): 152–76. https://doi. State Your Business!  Bibliography org/10.1016/j.jfi.2012.11.002. IFC (International Finance Corporation). 2014. Corporate Governance Scorecards: Assessing and Promoting the Implementation of Codes of Corporate Governance. Washington, DC: International Finance Corporation. IFC. 2015. Directive: Investments in State-Owned Enterprises. Catalogue IO259, Febru- ary 23, International Finance Corporation, Washington, DC. 74 IFC. 2017. IFC’s Approach to Engaging with State-Owned Enterprises Operating Outside of Home Markets. Washington, DC: International Finance Corporation. Imam, M. I., T. Jamasb, and M. Llorca. 2019. “Sector Reforms and Institutional Corruption: Evidence from Electricity Industry in Sub-Saharan Africa.” Energy Policy 129 (June): 532–45. https://doi.org/10.1016/j.enpol.2019.02.043. Ivanova, Irina A., and M. V. Bikeeva. 2016. “Corporate Social Responsibility: Speci- ficity, Formation Mechanism, Estimation of Management Efficiency.” European Research Studies Journal 19 (3A): 167–84. Jamasb, T., R. Nepal, and G. R. Timilsina. 2015. “A Quarter Century Effort Yet to Come of Age: A Survey of Power Sector Reforms in Developing Countries.” Policy Research Working Paper 7330, World Bank Group, Washington, DC. http://documents.world- bank.org/curated/en/504451467997248704/A-quarter-century-effort-yet-to-come- of-age-a-survey-of-power-sector-reforms-in-developing-countries. Jurkonis, Liudas, S. Merkliopas, and K. Kyga. 2016. “Impact of the Corporate Gover- nance Reform on the Management Effectiveness of State-Owned Enterprises in Lithuania in 2012–2014.” Ekonomika 95 (2): 158–78. Kane, Korin. 2017. “Size and Sectoral Distribution of State-Owned Enterprises: Main Findings of the Latest Review.” Presentation to OECD Steel Committee, Sep- tember 28. https://www.oecd.org/industry/ind/Item_6_3_OECD_Korin_Kane.pdf. Khemani, S. 2010. Political Economy of Infrastructure Spending in India. Washington, Independent Evaluation Group World Bank Group DC: World Bank. https://doi.org/10.1596/1813-9450-5423. Kikeri, Sunita. 2018. Corporate Governance in South African State-Owned Enterprises. Washington, DC: World Bank. https://doi.org/10.1596/30029. Kikeri, Sunita, and Verena Phipps. 2008. “Privatization Trends.” Public Policy for the Private Sector Note no. 317, World Bank, Washington, DC. http://documents. worldbank.org/curated/en/965751468182641229/Privatization-trends. Kowalski, P., M. Büge, M. Sztajerowska, and M. Egeland. 2013. “State-Owned Enter- prises: Trade Effects and Policy Implications.” OECD Trade Policy Papers no. 147, OECD Publishing, Paris. https://doi.org/10.1787/5k4869ckqk7l-en. 75 Kowalski, P., and D. Rabaioli. 2017. “Bringing Together International Trade and In- vestment Perspectives on State Enterprises.” OECD Trade Policy Papers no. 201, OECD Publishing, Paris. https://doi.org/10.1787/18166873. Kufeoglu, S., M. Pollitt, and K. Anaya. 2018. “Electric Power Distribution in the World: Today and Tomorrow.” Working Paper, Faculty of Economics, University of Cambridge, Cambridge, UK. https://doi.org/10.17863/CAM.27667. La Porta, R., and F. Lopez-de-Silanes. 1999. “The Benefits of Privatization: Evidence from Mexico.” The Quarterly Journal of Economics 114 (4): 1193–242. La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. 2002. “Government Ownership of Banks.” The Journal of Finance 57 (1): 265–301. Lam, W. Raphael, Markus Rodlauer, and Alfred Schipke, eds. 2017. Modernizing China: Investing in Soft Infrastructure. Washington, DC: International Monetary Fund. Lam, W. Raphael, A. Schipke, Y. Tan, and Z. Tan. 2017. “Resolving China’s Zombies: Tackling Debt and Raising Productivity.” IMF Working Paper 17/266, Interna- tional Monetary Fund, Washington, DC. Legovini, A., V. Di Maro, and C. Piza. 2015. “Impact Evaluation Helps Deliver Devel- opment Projects.” Policy Research Working Paper 7157, World Bank, Washing- ton, DC. Lehuedé, H. 2013. “Colombian SOEs: A Review against the OECD Guidelines on Corporate Governance of State-Owned Enterprises.” OECD Corporate Governance Working Papers no. 12, OECD Publishing, Paris. https://doi. org/10.1787/5k3v1ts5s4f6-en. State Your Business!  Bibliography Lei, Z., B. L. Liebman, and C. J. Milhaupt. 2016. “SOEs and State Governance: How State-Owned Enterprises Influence China’s Legal System.” In Regulating the Visible Hand? The Institutional Implications of Chinese State Capitalism, edited by B. L. Liebman and C. J. Milhaupt, 203–23. Oxford, UK: Oxford University Press. Lozano-Vivas, A., M. A. Meléndez-Jiménez, and A. J. Morales. 2011. “With Whom to Merge? A Tale of the Spanish Banking Deregulation Process.” SERIEs 2 (June): 159–84. https://doi.org/10.1007/s13209-010-0035-2. 76 McMahon, Gary. 2010. The World Bank’s Evolutionary Approach to Mining Sector Re- form. Extractive Industries and Development Series 19. Washington, DC: World Bank. http://hdl.handle.net/10986/18288. Megginson, William L. 2005. “The Economics of Bank Privatization.” Journal of Banking and Finance 29 (8–9): 1931–80. https://doi.org/10.1016/j.jbank- fin.2005.03.005. Megginson, William L. 2017. “Privatization, State Capitalism and State Ownership of Business in the 21st Century.” Foundations and Trends in Finance 11 (1–2): 1–153. Megginson, William L., Barkat Ullah, and Zuobao Wei. 2014. “State Ownership, Soft-Budget Constraints, and Cash Holdings: Evidence from China’s Privatized Firms.” Journal of Banking and Finance 48 (November): 276–91. https://doi. org/10.1016/j.jbankfin.2014.06.011. Megginson, William L., and Jeffry M. Netter. 2001. “From State to Market: A Survey of Empirical Studies on Privatization.” Journal of Economic Literature 39 (2): 321–89. https://doi.org/10.2307/2698243. Mian, A. 2005. “Foreign, Private Domestic, and Government Banks: New Evidence from Emerging Markets.” Journal of Banking and Finance 31 (7): 1219–410. https://doi.org/10.1111/j.1540-6261.2006.00878.x. Micco, A., U. Panizza, and M. Yañez. 2007. “Bank Ownership and Performance: Does Independent Evaluation Group World Bank Group Politics Matter?” Journal of Banking and Finance 31 (7): 219–41. https://doi. org/10.1016/j.jbankfin.2006.02.007. Milhaupt, C. J., and M. Pargendler. 2017. “Governance Challenges of Listed State- Owned Enterprises around the World: National Experiences and a Framework for Reform.” Cornell International Law Journal 50 (3): 473–542. https://scholar- ship.law.cornell.edu/cgi/viewcontent.cgi?article=1907&context=cilj. Milhaupt, C. J., and W. Zheng. 2016. “Why Mixed-Ownership Reforms Cannot Fix China’s State Sector.” Paulson Policy Memorandum, Paulson Institute, Chica- go, IL. http://www.paulsoninstitute.org/wp-content/uploads/2017/01/PPM_ SOE-Ownership_Milhaupt-and-Zheng_English_R.pdf. 77 Molnar, Margit. 2017. “Revitalizing China’s Economy by Improved Corporate Gover- nance and State-Owned Enterprise Reforms.” Journal of International Commerce, Economics and Policy 8 (3): 1–23. https://doi.org/10.1142/S1793993317500156. Morsy, A. 2014. “The Military Crowds Out Civilian Business in Egypt.” Carnegie Endowment for International Peace, June 24, 2020. https://carnegieendowment. org/2014/06/24/military-crowds-out-civilian-business-in-egypt-pub-55996. Morsy, H. 2020. “Access to Finance–Mind the Gender Gap.” The Quarterly Review of Economics and Finance, February 27. https://doi.org/10.1016/j.qref.2020.02.005. Muir, Russell, and J. P. Saba. 1995. “Improving State Enterprise Performance. The Role of Internal and External Incentives.” World Bank Technical Pa- per 306, World Bank, Washington, DC. http://documents.worldbank.org/ curated/en/489021468186871150/Improving-state-enterprise-perfor- mance-the-role-of-internal-and-external-incentives. Multilateral Investment Guarantee Agency. 2003. MIGA’s Issuance of Coverage to State- owned Enterprises Operating on a Commercial Basis. Washington, DC: MIGA. Munawarah, Muhammad Din, F. Zainuddin, and H. Muharam. 2017. “What Effects Do Privatisation Policies Have on Corporate Governance of State-Owned Enterpris- es?” European Research Studies 20 (4A): 124–32. Musacchio, A., and S. G. Lazzarini. 2014. Reinventing State Capitalism: Leviathan in Business, Brazil and Beyond. Cambridge, MA: Harvard University Press. Nellis, John. 1991. “Reform of Public Enterprises.” In Restructuring Economies in Dis- tress: Policy Reform and the World Bank, edited by Vinod Thomas, Ajay Chhibber, Mansoor Dailami, and Jaime de Melo, 108–30. Washington, DC: International State Your Business!  Bibliography Bank for Reconstruction and Development. http://documents1.worldbank.org/ curated/en/924991468765880583/pdf/multi-page.pdf. Nellis, John. 1994. “Is Privatization Necessary? Viewpoint.” FPD Note 17, World Bank, Washington, DC. Nellis, John. 2006a. “Back to the Future for African Infrastructure? Why State-Own- ership is No More Promising the Second Time Around.” Working Paper 84, Center for Global Development, Washington, DC. https://www.cgdev.org/sites/ default/files/6352_file_WP_84.pdf. 78 Nellis, John. 2006b. “Privatization in Developing Countries: A Summary Assess- ment.” Working Paper 87, Center for Global Development, Washington, DC. https://www.cgdev.org/sites/default/files/6928_file_WP_87.pdf. Nellis, John. 2012. “The International Experience with Privatization: Its Rapid Rise, Partial Fall and Uncertain Future.” SPP Research Paper 12-3. https://doi. org/10.2139/ssrn.1999098. Nellis, John, and Nancy Birdsall, eds. 2005. Reality Check: The Distributional Impact of Pri- vatization in Developing Countries. Washington, DC: Center for Global Development. NRGI (Natural Resource Governance Institute). 2019. The National Oil Company Da- tabase. Washington, DC: NRGI. OECD (Organisation for Economic Co-operation and Development). 2005. OECD Guidelines on Corporate Governance of State-Owned Companies. 2005 ed. Paris: OECD Publishing. OECD. 2010. Corporate Governance: Accountability and Transparency: A Guide for State Ownership. Paris: OECD Publishing. OECD. 2015. OECD Guidelines on Corporate Governance of State-Owned Enterprises. 2015 ed. Paris: OECD Publishing. OECD. 2016a. “Global Knowledge Sharing Network on Corporate Governance of State-Owned Enterprises.” Proceedings from the First Meeting, Mexico City, Mexico, June 7–8. http://www.oecd.org/daf/ca/Global-Knowledge-Sharing-Net- Independent Evaluation Group World Bank Group work-on-SOEs-Proceedings-June-2016.pdf. OECD. 2016b. State-Owned Enterprises as Global Competitors: A Challenge or an Op- portunity? Paris: OECD Publishing. Pardina, Martín Rodríguez, and Schiro, Julieta. 2018. “Taking Stock of Economic Reg- ulation of Power Utilities in the Developing World: A Literature Review.” Policy Research Working Paper 8461, World Bank, Washington, DC. Pazarbasioglu, Ceyla. 2017. “Steps to Increase Cooperation between National De- velopment Banks, the Private Sector and Multilateral Banks.” World Bank Blogs, October 26, 2017. https://blogs.worldbank.org/voices/steps-increase-cooper- 79 ation-between-national-development-banks-private-sector-and-multilater- al-banks. Pearson, M. M. 2015. “State-Owned Business and Party-State Regulation in China’s Modern Political Economy.” In State Capitalism, Institutional Adaptation, and the Chinese Miracle, edited by B. Naughton and K. Tsai, 27–45. Cambridge, UK: Cambridge University Press. Piatkowski, M. 2013. “Poland’s New Golden Age: Shifting from Europe’s Periphery to Its Center.” Policy Research Working Paper 6639, World Bank, Washington, DC. Pop, Georgiana, and Laurent Olivier Corthay. 2018. Senegal—Better Markets for All through Competition Policy. Washington, DC: World Bank Group. http:// documents1.worldbank.org/curated/en/704251545040037438/pdf/Sene- gal-MCPAT-Better-Markets-for-All-Through-Competition-Final-Report.pdf. Popovici, Vlad. 2011. “Power Generation Sector Restructuring in Romania—A Critical Assessment.” Energy Policy 39 (3): 1845–56. https://doi.org/10.1016/j. enpol.2011.01.023. Privatization Barometer. 2017. Privatization Barometer Report 2015/16. Milan: Pri- vatization Barometer. https://assets.kpmg/content/dam/kpmg/it/pdf/2017/01/ ThePBReport2015-2016.pdf. PwC. 2015. State-Owned Enterprises Catalysts for Public Value Creation? London: PwC. https://www.pwc.com/gr/en/publications/assets/state-owned-enterprises-cata- lysts-for-public-value-creation.pdf. Raballand, G., G. Veuillot, L. Habhab, and P. de Meneval. 2015. Middle East and North Africa: Governance Reforms of State: Lessons from Four Case Studies (Egypt, Iraq, State Your Business!  Bibliography Morocco, and Tunisia). Washington, DC: World Bank Group. Rabiei, Mahnez, and Parmida Rezaie. 2013. “Does Privatization Effect Profitability of Banking Sector? (Case Study of Iran).” Actual Problems of Economics 139 (1): 479–85. Raimondo, E. 2016. “What Difference Does Good Monitoring and Evaluation Make to World Bank Project Performance?” Policy Research Working Paper, World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-7726. 80 Rimmer, S., and M. Al-Ani. 2012. “Reform of State Owned Enterprise and Pub- lic-Private Partnerships (PPs) in Iraq.” MENA Knowledge and Learning Quick Notes Series 76: 1–3. https://openknowledge.worldbank.org/bitstream/han- dle/10986/20575/NonAsciiFileName0.pdf. Robinett, D. 2006. Held by the Visible Hand: The Challenge of State-Owned Enter- prise Corporate Governance for Emerging Markets. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/396071468158997475/Held-by- the-visible-hand-the-challenge-of-state-owned-enterprise-corporate-gover- nance-for-emerging-markets. Rodriguez Pardina, M., and J. Schiro. 2018. “Taking Stock of Economic Regulation of Power Utilities in the Developing World: A Literature Review.” Policy Research Working Paper 8461, World Bank, Washington, DC. http://documents.world- bank.org/curated/en/732921527773002514/Taking-stock-of-economic-regula- tion-of-power-utilities-in-the-developing-world-a-literature-review. Rudolph, Heinz P. 2009. “State Financial Institutions: Mandates, Governance, and Beyond.” Policy Research Working Paper 5141, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/114531468159602791/State-finan- cial-institutions-mandates-governance-and-beyond. Schiffbauer, M., A. Sy, S. Hussein, H. Sahnoun, and P. Keefer. 2014. Jobs or Privileges: Unleashing the Employment Potential of the Middle East and North Africa. Wash- ington, DC: World Bank. https://doi.org/10.1596/978-1-4648-0405-2. Independent Evaluation Group World Bank Group Schmitz, J. A., Jr., and A. Teixeira. 2008. “Privatization’s Impact on Private Productiv- ity: The Case of Brazilian Iron Ore.” Review of Economic Dynamics 11 (4): 745–60. https://doi.org/10.1016/j.red.2008.01.001. Sen, Anupama, Rabrindra Nepal, and Tooraj Jamasb. 2016. Reforming Electricity Re- forms? Empirical Evidence from Asian Economies. Oxford, UK: The Oxford Insti- tute for Energy Studies. https://doi.org/10.26889/9781784670504. Sensarma, R. 2006. “Are Foreign Banks Always the Best? Comparison of State- Owned, Private and Foreign Banks in India.” Economic Modelling 23 (4): 717–35. https://doi.org/10.1016/j.econmod.2006.04.002. 81 Sherif, K., M. Borish, and A. Gross. 2003. State-Owned Banks in the Transition: Origins, Evolution, and Policy Responses. Washington, DC: World Bank. https://open- knowledge.worldbank.org/bitstream/handle/10986/14851/258320State1banks- 1Final1022803.pdf. Smith, Gregory, Le Duy Binh, Jim Colvin, and Habib Rab. 2014. Transparency of State Owned Enterprises in Vietnam: Current Status and Ideas for Reform. Wash- ington, DC: World Bank Group. http://documents.worldbank.org/curated/ en/618601468132616527/Transparency-of-state-owned-enterprises-in-Viet- nam-current-status-and-ideas-for-reform. Stewart, Chris, Roman Matousek, and Thao Ngoc Nguyen. 2016. “Efficiency in the Vietnamese Banking System: A DEA Double Bootstrap Approach.” Research in International Business and Finance 36: 96–111. https://doi.org/10.1016/j.rib- af.2015.09.006. Stone, A., K. Hurley, and R. S. Khemani. 1998. The Business Environment and Corpo- rate Governance: Strengthening Incentives for Private Sector Development. Wash- ington, DC: World Bank. Stoyan, T., and C. Zhang. 2002. Corporate Governance and Enterprise Reform in China: Building the Institutions of Modern Markets. Washington, DC: World Bank. Tabachnick, B. G., and L. S. Fidell. 2001. Using Multivariate Statistics. 4th ed. Need- ham Heights, MA: Allyn and Bacon. Tabachnick, B. G., L. S. Fidell, and J. B. Ullman. 2007. Using Multivariate Statistics. 5th ed., 481–98. Boston, MA: Pearson. Tang, H., F. Wang, and Z. Wang. 2014. The Domestic Segment of Global Supply Chains State Your Business!  Bibliography in China Under State Capitalism. Washington, DC: World Bank. https://doi. org/10.1596/1813-9450-6960. Tenev, S., C. Zhang, and L. Brefort. 2002. Corporate Governance and Enterprise Reform in China: Building the Institutions of Modern Markets. Washington, DC: World Bank and International Finance Corporation. Thorne, J., and C. du Toit. 2009. “A Macro-Framework for Successful Devel- opment Banks.” Development Southern Africa 26 (5): 677–94. https://doi. org/10.1080/03768350903303183. 82 Tomasi, M., and M. Brons. 2015. “Firm Ownership and Financial Performance: An Empirical Assessment in the Energy and Rail Sectors.” Paper presented at the State-Owned Enterprises workshop, Brussels, November 24, 2015. https:// ec.europa.eu/economy_finance/events/2015/20151124-workshop/documents/ soes_performance_section_i_ecfin_en.pdf. Tordo, S., B. S. Tracy, and N. Arfaa. 2011. “Natural Oil Companies and Value Cre- ation.” World Bank Working Paper 218, World Bank, Washington, DC. http:// documents.worldbank.org/curated/en/650771468331276655/National-oil-com- panies-and-value-creation. Tran, Ngo My, W. Nonneman, and A. Jorissen. 2015. “Privatization of Vietnamese Firms and Its Effects on Firm Performance.” Asian Economic and Financial Review 5 (2): 202–17. Transparency International. 2017. “Preventing Corruption in State-Owned Enter- prises.” November 28, 2017. https://www.transparency.org/en/news/how-to-pre- vent-corruption-in-state-owned-enterprises#. Transparency International Secretariat. 2017. “Transparency International Launches 10 Anticorruption Principles for State-Owned Enterprises.” Press Release, No- vember 28. https://www.transparency.org/en/press/transparency-internation- al-launches-10-anti-corruption-principles-for-soes. Trimble, C., M. Kojima, I. P. Arroyo, and F. Mohammadzadeh. 2016. “Financial Vi- ability of Electricity Sectors in Sub-Saharan Africa: Quasi-Fiscal Deficits and Independent Evaluation Group World Bank Group Hidden Costs.” Policy Research Working Paper 7788, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/182071470748085038/Finan- cial-viability-of-electricity-sectors-in-Sub-Saharan-Africa-quasi-fiscal-defi- cits-and-hidden-costs. Urpelainen, J., and J. Yang. 2019. “Global Patterns of Power Sector Reform, 1982– 2013.” Energy Strategy Reviews 23 (January): 152–62. https://doi.org/10.1016/j. esr.2018.12.001. Vagliasindi, Maria. 2008a. “Governance Arrangements for State Owned Enterprises.” Policy Research Working Paper 4542, World Bank, Washington, DC. 83 Vagliasindi, Maria. 2008b. “The Effectiveness of Boards of Directors of State-Owned Enterprises in Developing Countries.” Policy Research Working Paper 4579, World Bank, Washington, DC. Vagliasindi, Maria. 2012a. Implementing Energy Subsidy Reforms: Evidence from Devel- oping Countries. Washington, DC: World Bank. http://documents.worldbank.org/ curated/en/566201468177234043/Implementing-energy-subsidy-reforms-evi- dence-from-developing-countries. Vagliasindi, Maria. 2012b. “Power Market Structure and Performance.” Policy Re- search Working Paper 6123, World Bank, Washington, DC. Vagliasindi, Maria, and John Besant-Jones. 2013. Power Market Structure: Revisiting Policy Options. Washington, DC: World Bank. http://documents1.worldbank.org/ curated/en/795791468314701057/pdf/Power-market-structure-revisiting-poli- cy-options.pdf. Vandycke, N., A. Legovini, A. Liaplina, and V. di Maro. 2015. Impact Evaluation to Inform and Transform Investments in Transport and ICT. Washington, DC: World Bank. https://www.worldbank.org/en/topic/transport/brief/impact-evalua- tion-to-inform-and-transform-investments-in-transport-ict. Victor, David G., David R. Hults, and Marl C. Thurber, eds. 2012. Oil and Governance: State-Owned Enterprises and the World Energy Supply. Cambridge, UK: Cam- bridge University Press. Warshaw, C. 2012. “The Political Economy of Expropriation and Privatization in the Oil Sector.” In Oil and Governance: State-Owned Enterprises and the World Energy Supply, edited by David G. Victor, David R. Hults, and Mark C. Thurber, 35–61. State Your Business!  Bibliography Cambridge, UK: Cambridge University Press. Weber, F. 2011. “Verstaatlichung und Privatisierung in Österreich 1946–1986.” Zeitschrift für öffentliche und Gemeinwirtschaftliche Unternehmen: ZögU / Journal for Public and Nonprofit Services 34 (2): 126–47. doi: 10.2307/41318646. Williams, Jonathan, and Nghia Nguyen. 2005. “Financial Liberalisation, Crisis, and Restructuring: A Comparative Study of Bank Performance and Bank Governance in South East Asia.” Journal of Banking and Finance 29 (8–9): 2119–54. 84 Wolf, C. 2009. “Does Ownership Matter? The Performance and Efficiency of State Oil vs. Private Oil (1987–2006).” Energy Policy 37 (7): 2642–52. https://doi. org/10.1016/j.enpol.2009.02.041. Wong, Simon C. Y. 2004. “Improving Corporate Governance in SOEs: An Integrated Approach.” Corporate Governance International 7 (2): 5–15. https://papers.ssrn. com/sol3/papers.cfm?abstract_id=897121. Wong, Simon C. Y. 2018. “The State of Governance at State-Owned Enterprises.” Private Sector Opinion 40, International Finance Corporation, Washington, DC. https://www.ifc.org/wps/wcm/connect/b1adde06-267d-4d79-bfaf-62f17de51f4a/ PSO40.pdf?MOD=AJPERES&CVID=m7T0xLQ. World Bank. 2002. Private Sector Development Strategy: Directions for the World Bank Group. Washington, DC: World Bank. World Bank. 2009. India: Organizational Transformation of State Owned Enterpris- es: Case of West Bengal Power. Outcomes from Knowledge Sharing Workshops. Washington, DC: World Bank. World Bank. 2011. Overview of State Ownership in the Global Minerals Industry. Long Term Trends and Future. Extractive Industries for Development Series 20. Washington, DC: World Bank. http://documents.worldbank.org/curated/ en/339551468340825224/Overview-of-state-ownership-in-the-global-miner- als-industry. Independent Evaluation Group World Bank Group World Bank. 2012a. Global Financial Development Report 2013: Rethinking the Role of the State in Finance. Washington, DC: World Bank. https://openknowledge. worldbank.org/bitstream/handle/10986/11848/GlobalFinancialDevelopmen- tReport2013.pdf?sequence=1. World Bank. 2012b. “Serbia— Review of Country Partnership Strategy, FY08–11.” Country Partnership Strategy Completion Report Review 66003, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2014a. Corporate Governance of State-Owned Enterprises: A Toolkit. Washington, DC: World Bank. https://doi.org/10.1596/978-1-4648-0222-5. World Bank. 2014b. Corporate Governance of State-Owned Enterprises in Latin Amer- ica: Current Trends and Country Cases. Washington, DC: World Bank. http:// 85 documents1.worldbank.org/curated/en/634451468088457556/pdf/894680WP- 00P1260orate0Governance0LAC.pdf. World Bank. 2014c. “Tajikistan: Fiscal Risks from State-Owned Enterprises.” Tajiki- stan Policy Notes on Public Expenditures, Policy Note no. 5, World Bank, Wash- ington, DC. http://documents1.worldbank.org/curated/en/571421468119375495/ pdf/891830NWP0P1475020Box385273B00OUO090.pdf. World Bank. 2014d. Transparency of State-Owned Enterprises in Vietnam: Current Status and Ideas for Reform. Washington, DC: World Bank. https://www.econom- ica.vn/Portals/0/Documents/Transparency%20of%20SOEs%20in%20Vienam%20 WB.pdf. World Bank. 2015a. World Bank Group Support to Electricity Access, FY2000–2014: An Independent Evaluation. Independent Evaluation Group. Washington, DC: World Bank. https://ieg.worldbankgroup.org/sites/default/files/Data/Evaluation/files/ Electricity_Access.pdf. World Bank. 2015b. World Bank Group Support to Public-Private Partnerships: Les- sons from Experience in Client Countries, FY02–12. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/21309. World Bank. 2016a. Republic of Ghana State Owned Enterprises and Governance: Governance of State Owned Enterprises and Financial and Economic Regulators and Trust Synthesis Report. Synthesis Report. Independent Evaluation Group. Washington, DC: World Bank. World Bank. 2016b. “Republic of Paraguay: Challenges of SOE Service Delivery.” Study Note P156523, World Bank Group, Washington, DC. State Your Business!  Bibliography World Bank. 2016c. “Zambia—Increased Access to Electricity Services Project.” Implementation Completion and Results Report ICR3523, World Bank, Wash- ington, DC. World Bank. 2017a. “Maximizing Finance for Development: Leveraging the Private Sector for Growth and Sustainable Development.” Development Committee Report DC2017–0009, World Bank Group, Washington, DC. World Bank. 2017b. Mobile Metropolises: Urban Transport Matters: An IEG Evaluation of the World Bank Group’s Support for Urban Transport. Independent Evaluation 86 Group. Washington, DC: World Bank. https://ieg.worldbankgroup.org/sites/de- fault/files/Data/Evaluation/files/urbantransport.pdf. World Bank. 2017c. Moldova—Support to State Owned Enterprises (SOE) Preliminary Diagnostics and Reform Assessment: Phase 1. Washington, DC: World Bank Group. https://documents.worldbank.org/en/publication/documents-reports/ documentdetail/762831499412473409/moldova-support-to-state-owned-enter- prises-soe-preliminary-diagnostics-and-reform-assessment-phase-1. World Bank. 2017d. A Thirst for Change: The World Bank Group’s Support for Water Supply and Sanitation, with Focus on the Poor. Independent Evaluation Group. Washington, DC: World Bank. https://ieg.worldbankgroup.org/sites/default/ files/Data/Evaluation/files/watersupplyandsanitation.pdf. World Bank. 2017e. “Ukraine—Hydropower Rehabilitation Project.” Implementation Completion and Results Report Review ICRR20606, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2017f. World Bank Group Joint Projects: A Review of Two Decades of Expe- rience. Independent Evaluation Group. Washington, DC: World Bank. https://ieg. worldbankgroup.org/sites/default/files/Data/Evaluation/files/lp_wbgjointproj- ects.pdf. World Bank. 2019a. “The Forward Look and IBRD-IFC Capital Package Implementa- tion.” Report to Governors at 2019 Spring Meetings World Bank Group Update. World Bank, Washington, DC. World Bank. 2019b. “Mali—Energy Support Project.” Implementation Completion Independent Evaluation Group World Bank Group and Results Report ICR4645, World Bank, Washington, DC. World Bank. 2019c. Rethinking the 1990s Orthodoxy on Power Sector Reform: Flagship Report. Washington, DC: World Bank Group. http://documents1.worldbank.org/ curated/en/597911563520308282/pdf/World-Rethinking-the-1990s-Orthodoxy- on-Power-Sector-Reform-Flagship-Report.pdf. World Bank. 2019d. Scaling Private Sector Solutions Is Critical to Achieving the Twin Goals and the SDGs. Washington, DC: World Bank. World Bank. 2019e. “Ukraine—First and Second Programmatic Financial Sector De- velopment Policy Loan.” Project Performance Assessment Report 144601, Inde- 87 pendent Evaluation Group, World Bank, Washington, DC. https://ieg.worldbank- group.org/sites/default/files/Data/reports/ppar_ukraineprogrammaticdpl12.pdf. World Bank. 2019f. World Bank Group Support for the Reform of State-Owned Enter- prises, 2007–2018: An IEG Evaluation. Approach Paper. Independent Evaluation Group. Washington, DC: World Bank. World Bank. 2020a. Integrated State-Owned Enterprises Framework: Sri Lanka Assess- ment. Washington, DC: World Bank Group. World Bank. 2020b. Public Utility Reform: What Lessons Can We Learn from IEG Evalu- ations in the Energy and Water Sectors? Independent Evaluation Group. Washing- ton DC: World Bank. World Bank Group. 2013. A Stronger, Connected, Solutions World Bank Group: An Overview of the World Bank Group Strategy. Washington, DC: World Bank Group. World Bank Group. 2018. 2017 Survey of National Development Banks. Wash- ington, DC: World Bank Group. http://documents.worldbank.org/curated/ en/977821525438071799/2017-Survey-of-National-development-banks. Yusuf, S., K. Nabeshima, and D. H. Perkins. 2006. Under New Ownership: Privatizing China’s State-Owned Enterprises. Washington, DC: World Bank. https://open- knowledge.worldbank.org/bitstream/handle/10986/7399/345170CHA0Unde10 1OFFICIAL0USE0ONLY1.pdf?sequence=1&isAllowed=y. Zagha, Roberto, ed. 2005. “Privatization and Deregulation: A Push Too Far?” In Eco- nomic Growth in the 1990s: Learning from a Decade of Reform, 165–97. Washing- ton, DC: World Bank. http://www1.worldbank.org/prem/lessons1990s/. State Your Business!  Bibliography Zhang, Yin-Fang, David Parker, and Colin Kirkpatrick. 2008. “Electricity Sector Reform in Developing Countries: An Econometric Assessment of the Effects of Privatization, Competition and Regulation.” Journal of Regulatory Economics 33 (2): 159–78. 88 APPENDIXES Independent Evaluation Group State Your Business! Appendix A. Methodology Theory of Change This evaluation employed a theory-driven analysis of the key causal steps identified in the theory of change (figure A.1) and mixed methods drawn on quantitative and qualitative information. The analysis is multilevel and examines country, sector, project, engagement area, and intervention mechanism levels, and looks at both upstream reforms (policy, regulatory, and institutional) and downstream state-owned enterprise (SOE) reforms (enterprise level). It does not examine projects that use SOEs as a vehicle to deliver services without trying to change them (such as a line of credit channeled through a state bank). To allow greater depth, the evaluation focused on the two sectors identified as having the most projects and highest level of commitments pertaining to SOE reform: the energy and financial sectors. Within the energy sector, it covered both power companies delivering energy to households and businesses (often utilities) or engaged in power distribution, generation, or transmission, and those engaged in energy extraction. Within the financial sector, there are state-owned commercial banks (typically taking deposits and offering credit and other services), development banks (financing government development priorities), and nonbank financial institutions in such areas as insurance and pensions. The evaluation assesses the World Bank’s contribution to enhancing development outcomes through its support of SOE reform. 92 Appendix A Methodology Figure A.1. Evaluation Theory of Change Source: World Bank 2019. Note: The scope of this evaluation comprises the green boxes in the energy and power sectors. ASA = Advisory Services and Analytics; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; PFM = public financial management; SOE = state-owned enterprise; WB = World Bank; WBG = World Bank Group. 93 Appendix A Methodology Evaluation Questions This evaluation assesses the contribution of the World Bank Group to enhancing development outcomes through its support for SOE reform. This overarching objective elicited three guiding questions (box A.1). Box A.1. Questions Guiding the Evaluation Question 1—Relevance: Does the Bank Group have a credible approach to achieving development impact through state-owned enterprise (SOE) reforms? What has been the nature of client demands and World Bank Group identified priorities for country, sector, and firm-level SOE reforms? How aligned is Bank Group engagement with SOE reforms with country, sector, and SOE firm-level development priorities and capabilities and the most relevant constraints? To what extent has Bank Group support been aligned with relevant Bank Group strategic objectives? How has the coherence and coordination of the Bank Group’s engagement with SOE reform evolved over time? Question 2—Effectiveness: How effective are the Bank Group’s SOE reform interventions? How effective have the Bank Group’s SOE reform interventions been in helping clients to strengthen strategy and performance of SOEs at the enterprise, sector, or national level? To what extent did Bank Group interventions lead to improved SOE performance at the enterprise, sectoral, or national level? To what extent have Bank Group interventions contributed to improved economic, social, and environmental outcomes at the enterprise, sector, or national level? Question 3—Learning: What factors explain the success or failure of the Bank Group’s SOE reform interventions? What internal factors (for example, design, supervision, team composition, consistency, choice of instrument, monitoring and evaluation framework, sequencing, collaboration, complementarity, and funding) and/or external factors (for example, client commitment and political economy, public sector institutional capacity, private sector capacity and 94 Appendix A Methodology engagement, and activities of other donors and partners) explain observed development outcomes of the Bank Group’s SOE reform interventions? What examples of good practice can be identified from the Bank Group’s experience in SOE reform over the last 10 years? What implications can be drawn from lessons of experience for the future involvement of the Bank Group in SOE reform? Main Methodological Components Three central principles embodied in the evaluation design are multilevel analysis, theory-driven analysis of the key causal steps, and mixed methods drawn on quantitative and qualitative aspects. The evaluation is multilevel, examining countries, projects, interventions, mechanisms, engagement areas, and whether the reform focused on regulatory and institutional environment (upstream reforms) or on addressing firm-level SOE reforms (downstream reforms). The evaluation selected the energy and financial sectors and identified interventions at the national level that supported those sectors. It is theory driven, grounded in and testing the intervention logic of Bank Group SOE reform support elaborated in the approach paper. This model was constructed based on a preliminary review of the literature and portfolio documentation, and multiple interviews with Bank Group staff working on SOE reform and external experts. This analysis is characterized as a multilevel evaluation involving the quantitative aspects (for example, those in the analysis of the portfolio data) and qualitative aspects such as those in the literature review, case studies, and interviews. The evaluation applied a mixed methods approach that combined a range of methods for data collection and analysis and triangulated (especially between quantitative and qualitative evidence) to ensure the robustness of the findings. 95 Appendix A Methodology Design Matrix Table A.1. Evaluation Questions and Methods Applied Portfolio Review Country- Econo- Case and Deep Literature Level FSAP metric Evaluation Questions Studies Analysis Dives Review Reviews Review Analysis 1. Relevance: Does the World Bank Group have a credible approach to achieving development impact through SOE reforms? a) What has been the nature of client demands ✓✓✓ ✓✓ ✓✓ ✓ ✓✓✓ ✓✓ and Bank Group identified priorities for country, sector, and firm-level SOE reforms? b) How aligned is Bank Group engagement with ✓✓✓ ✓✓ ✓✓ ✓✓✓ SOE reforms with country, sector, and SOE firm- level development priorities and capabilities? c) To what extent has Bank Group support been ✓✓✓ ✓✓ ✓✓ ✓✓✓ aligned with relevant Bank Group strategic objectives? d) How has the coherence and coordination of the ✓✓✓ ✓✓ ✓✓✓ ✓✓✓ Bank Group’s engagement with SOE reform evolved over time? 96 Appendix A Methodology Portfolio Review Country- Econo- Case and Deep Literature Level FSAP metric Evaluation Questions Studies Analysis Dives Review Reviews Review Analysis 2. Effectiveness: How effective are the Bank Group’s SOE reform interventions? a. How effective have the Bank Group’s SOE ✓✓✓ ✓✓ ✓✓ ✓✓ reform interventions been in helping clients to strengthen strategy and performance of SOEs at the enterprise, sector, or national level? b. To what extent did Bank Group interventions ✓✓✓ ✓✓ ✓✓ ✓ ✓✓ lead to improved SOE performance at the enterprise, sectoral, or national level? c. To what extent have Bank Group interventions ✓✓✓ ✓✓ ✓✓ ✓ ✓✓✓ contributed to improved economic, social, and environmental outcomes at the enterprise, sector, or national level? 3. Learning: What factors explain the success or failure of the Bank Group’s SOE reform interventions? a. What internal factors (for example, design, ✓✓✓ ✓✓✓ ✓✓ ✓✓ ✓✓ ✓✓ supervision, team composition, consistency, choice of instrument, monitoring and evaluation framework, sequencing, collaboration, 97 Appendix A Methodology Portfolio Review Country- Econo- Case and Deep Literature Level FSAP metric Evaluation Questions Studies Analysis Dives Review Reviews Review Analysis complementarity, and funding) and/or external factors (for example, client commitment and political economy, public sector institutional capacity, private sector capacity and engagement, and activities of other donors and partners) explain observed development outcomes of the Bank Group’s SOE reform interventions? b. What examples of good practice can be ✓✓✓ ✓✓✓ ✓✓✓ ✓ ✓✓ ✓ identified from the Bank Group’s experience in SOE reform over the last 10 years? c. What implications can be drawn from lessons of ✓✓✓ ✓✓ ✓✓✓ ✓ ✓✓ ✓ experience for the future involvement of the Bank Group in SOE reform? Source: Independent Evaluation Group. Note: FSAP = Financial Sector Assessment Program; SOE = state-owned enterprise. 98 Appendix A Methodology Ensuring the Validity of Findings The Independent Evaluation Group (IEG) took several steps to guarantee a consistent approach across the evaluation team members—for example, using a case study template to ensure a common framework and evaluative lens across studies. Among the steps taken for quality control were the following:  The intervention logic and approaches for key methods were reviewed with the methods adviser at the outset.  Portfolio team members peer-reviewed the data entry for the portfolio, and the team held weekly meetings for unifying revision criteria.  The structured literature review was conducted according to IEG’s protocol and quality checked by the team.  Case studies were peer-reviewed within the team.  The task team leader controlled the quality of deep dive studies.  The econometric adviser from the methods advisory team advised and reviewed the econometric analysis. Multiple designs and models were tested, and robustness checks were applied to enhance validity. The team applied triangulation across evaluation components—that is, validating hypotheses or findings based on one source with information from other sources. For example, a series of hypotheses derived from a workshop on the case studies was checked against evidence from deep dives, interviews, portfolio analysis, and the econometric analysis. In addition, IEG’s quality control protocols were followed through both internal and external review processes. Description of Methodologies The principal methods include (i) the intervention logic (theory of change); (ii) the main and focused structured literature reviews, the portfolio review 99 Appendix A Methodology and analysis, and semistructured interviews with Bank Group staff, key informants, and country stakeholders; (iii) case studies of the Bank Group’s role and contribution in supporting SOE reform in eight countries; and (iv) subject matter deep dives on the power sector, state-owned financial institutions (SOFIs), the extractives sector, privatization, and corporate governance, and standard desk review of the portfolio, external relevant databases and indicators, and econometric analysis. The methodologies for portfolio review and econometric analysis are fully elaborated in subsequent appendixes. Beyond those methodologies, further details follow on the methodologies for the case studies, deep dives, country strategies and Systematic Country Diagnostics review, and Financial Sector Assessment Program (FSAP) review. Methodology for Case Studies Eight country case studies involved site visits (the Arab Republic of Egypt, Bangladesh, Kenya, Ukraine, and Vietnam) or desk reviews (China, Mozambique, and Serbia) to capture and assess country-level evidence in SOE reform in the two sectors. Countries were selected purposively based on having significant SOE reform portfolios in one or both sectors and reflecting a diversity of characteristics and types of support. In each study, the team identified priorities; Bank Group strategies and interventions relevant to SOE reform; the complementarity, relevance, and efficacy of interventions and coordination and sequencing of interventions over the evaluation period; the role of key stakeholders (including development partners); the achievement of objectives; and factors explaining success or failure (box A.2). 100 Appendix A Methodology Box A.2. Template of Case Study Questions 1. Country Priorities and Context Role of state-owned enterprises (SOEs) in the economy at the national level and for the energy and financial sectors. Country priorities in SOE reform and the nature of client demands for country, sector, and firm-level SOE reforms. 2. World Bank Group’s Role and Relevance The Bank Group’s stated priorities and objectives on SOE reform in the country strategy documents and projects (at national, sector, and enterprise levels). Bank Group interventions on SOE reform in the country over time (national, sector, and enterprise levels). How aligned is Bank Group engagement with SOE reforms with country, sector, and SOE firm-level development priorities and capabilities and the most relevant constraints? To what extent has Bank Group support been aligned with relevant Bank Group strategic objectives as defined in the Country Assistance Strategy and Country Partnership Strategy? 3. Effectiveness How effective have Bank Group interventions been in implementing the needed sector reforms (at the policy and regulatory framework and at the enterprise and SOE levels, and the sustainability of results)? How effective have the Bank Group’s efforts been in improving the performance of SOEs that its reform efforts targeted (beyond those addressed under previous questions)? To what extent have Bank Group interventions contributed to improved economic, social, and environmental outcomes? 101 Appendix A Methodology 4. Learning about Factors of Success and Failure What are the key internal factors that explain the success or failure (achievement or nonachievement of objectives and development outcomes) of the Bank Group’s SOE reform interventions at the national, sector, and enterprise levels? How are they associated with outcomes? What external factors explain the success or failure (achievement or nonachievement of objectives and development outcomes) of the Bank Group’s SOE reform interventions at the national, sector, and enterprise levels? What examples of good practice can be identified from the Bank Group’s experience on SOE reform over the last 10 years? Methodology for Deep Dives Deep dives consisted of issue briefs, focused literature reviews, and portfolio analysis on SOE reform issues. Each deep dive identified key SOE challenges and reform interventions, the relevance of interventions to achieve reform goals, achievement of goals, and internal and external factors that facilitated or constrained implementation of interventions. The deep dive topics were state-owned financial institutions Thorsten Beck), privatization (John Nellis), power sector (Richard Schlirf Rapti), and corporate governance and extractives industries in energy (Nadia Asgaraly). Most drew from any relevant information from case studies. Each of the deep dives used the combined information sources to identify the following:  SOE challenges and reform interventions  Characteristics of SOE reform and the role of upstream and downstream interventions  Suitability and adaptability of Bank Group instruments for the SOE reform needs  Achievement of SOE reform intervention targets 102 Appendix A Methodology  Internal and external factors that facilitate or constrain SOE reform implementation  Role of the Bank Group in improving economic, social, and environmental outcomes at the enterprise, sector, or national level  Role of other stakeholders (beyond the Bank Group) at the country and global levels Methodology for Structured Literature Review A structured literature review of the academic and professional literature used a rigorous search strategy (with established keywords and phrases and inclusion criteria), following IEG’s protocol, to better understand the typology of needs and priorities for developing countries in reforming SOEs, the typology of interventions supporting SOE reform, evidence on the effectiveness of intervention types, and contextual factors and characteristics important for the effectiveness of reforms. The identification of relevant literature followed four main domains: key publications of international organizations on economic development, such as the Organisation for Economic Co-operation and Development and the International Monetary Fund; Google Scholar; the EconLit publications database; and the World Bank Open Knowledge Repository. It also included a few other references, including those cited in the approach paper and publications found through a back-referencing strategy. Overall, the search sources generated a list of 1,981 publications. Removing duplicates and using only the latest version of a paper eliminated 171 publications. Filters: A second step involved scanning papers’ introductions and/or data sections to determine if their definition of SOE reform conformed to IEG’s definition for this evaluation and if the article included evidence-based findings whose empirical data covered the year 2008 and onward. An extra 103 Appendix A Methodology quality filter was applied to references identified through Google Scholar: the journals of the selected publications were checked on Ulrichsweb (http://www.ulrichsweb.com) to select those that are refereed. Additional filters were applied to articles identified in the EconLit database to control for relevance by scanning the introductions and data sections to ensure they discussed the energy or financial sectors and included coverage of developing countries or regions. To limit the excessive representation of China, any publications published before 2017 that study only China with fewer than five references were eliminated. After applying all of the filters, the list was further narrowed down to 369 publications that formed the basis for the literature review, summarized in appendix G. Methodology for FSAP Review An FSAP review examined a sample of FSAP reports to better understand the SOFI reform challenges and recommendations to address these challenges. Based on the stratified random sample for the country strategy review, IEG reviewed the most recent FSAP reports within the evaluation period for 29 countries where there was also an FSAP to better understand the SOFI reform challenges in the financial sector. The review identified the coverage of issues, challenges, and recommended actions on SOFIs, particularly on state-owned commercial banks and development banks. The team also identified the upstream and downstream engagement areas and mechanisms. Country Strategy and Systematic Country Diagnostics Reviews A review of country strategies (Country Partnership Frameworks and Country Assistance Strategies) and Systematic Country Diagnostics involved a structured review of a stratified random sample of country strategy 104 Appendix A Methodology documents and Systematic Country Diagnostics to understand the level of alignment and coherence of Bank Group country-level strategies and diagnostics and SOE reform priorities. Of 114 countries with an IEG- reviewed Country Partnership Strategy, the team drew a random sample of 46 stratified by region and income level (40 percent). For countries substantially treating SOE reform, the review aimed to identify the instrument, intervention levels, engagement areas, mechanisms, sectors, and subsectors for SOE reform interventions proposed in the strategy documents. Achievement of objectives for these interventions were checked by reviewing the Country Assistance Strategy Review documents. It identified patterns based on country characteristics (for example, income level and fragility, conflict, and violence status). Limitations of Methodologies Notwithstanding these steps, the evaluation methodologies had limitations related to the choices about scope and focus and inherent to data availability, resource constraints, and specific methodologies.  Evaluation scope and focus limit the generalizability and audience. The team made a necessary trade-off between breadth and depth of analysis. The overall scope of the evaluation excluded important activities related to SOEs: The evaluation focused on reform intervention in two sectors (energy and finance); interventions that used SOEs for development purposes but did not try to reform the SOEs were not covered. Furthermore, deep dives could only cover a limited number of areas within the broader range of Bank Group areas of activity.  Country case studies were selected purposively, based partly on the richness of the portfolio. This biased the team toward countries where the Bank Group had adopted multiple interventions and instruments. In seeking depth versus breadth, the team recognized that additional cases would have represented more variations in context and intervention 105 Appendix A Methodology patterns but, given limited time and resources, at the price of depth of understanding.  The portfolio analysis was limited in part by the extent to which projects were closed and had been validated by IEG. Additionally, the lack of a consistent and validated framework for advisory services and analytics evaluation means that the performance of a vast amount of work is largely unexamined except through case studies. The 20 percent sample of potential SOE advisory services and analytics, necessitated by the size of the advisory services and analytics portfolio and resource constraints, also limited the analysis.  The econometric analysis brought with it perils limiting the ability to draw valid inferences from the results, including data limitations, potential biases from small numbers of observations, and potential omitted variable biases. For example, some of the external indicators of country characteristics could not be used because there were insufficient rated countries during the evaluation period for meaningful results to emerge. Among the mitigation strategies were a grounding in hypotheses generated from other evidence, the use of a predictive rather than causal model, the use of principal components analysis, and numerous robustness checks.  Interviews were chosen based on expertise or position. They were opportunistic or “snowball” and do not constitute a representative sample or survey of the population of experts and stakeholders in the field of SOEs. This potentially limits the generalizability of findings. Reference World Bank. 2019. World Bank Group Support for the Reform of State-Owned Enterprises, 2007–2018: An IEG Evaluation. Approach Paper. Independent Evaluation Group. Washington, DC: World Bank. 106 Appendix B. Portfolio Review Framework and Identification Methodology The evaluation’s portfolio review framework and identification methodology benefited from informative interactions with stakeholders and subject matter experts and from a review of available literature and project-level documentation. During the early phases of the review, the Independent Evaluation Group (IEG) interacted with World Bank Group staff working on state-owned enterprise (SOE) reform generally, and in the energy, financial, and transport sectors. These interactions, together with a review of relevant internal and external literature and project-level documentation, informed the evaluation approach by highlighting important concepts and frameworks and revealing industry coding, system flags, and keywords that would facilitate the identification of the portfolio and its initial classification. Portfolio Review Framework IEG’s portfolio review framework was designed to reflect the main interventions and areas of engagement used by each of the Bank Group institutions to engage with or support SOE reform in client countries. Although the framework that IEG developed reflected underlying patterns identified in the portfolio, the framework also relied on consultations with IEG stakeholders and Bank Group stakeholders to test the internal validity of the instrument. The framework was applied to all energy and financial sector projects that provided support for SOE reform (many of which also engaged with SOEs in other ways) to arrive at a unified portfolio view of the Bank Group’s support in this regard. The portfolio review framework was also used to understand the effectiveness of these interventions. The evaluation framework accounted for the fact that SOE support may be one of many elements addressed by a 107 Appendix B Portfolio Review Framework and dentification Methodology project. Therefore, to understand the effectiveness of the SOE intervention in this context, IEG designed an effectiveness framework parallel to the intervention framework, which relied on evaluative information available in Implementation Completion and Results Reports, IEG’s Implementation Completion and Results Report Reviews (ICRRs), Expanded Project Supervision Reports (XPSRs), IEG evaluation notes, Project Completion Reports, Project Evaluation Reports, and validation notes. It used indicators and their results and qualitative information on the achievement of their targets. Interventions Description A typology of intervention dimensions was developed to capture the breadth of SOE reform interventions undertaken by the Bank Group in client countries. This review framework was used to better understand the characteristics and their effectiveness in reaching outcomes. The framework acknowledges that these interventions may co-occur within a project. The evaluation focuses on the two SOE reform intervention types defined in table B.1. The evaluation does not focus on related and complementary areas such as broad enabling sectors and conditions. Bank Group SOE interventions were classified into four categories: two pursuing SOE reforms and two using SOEs for development goals. The first two were the focus of this evaluation. Table B.1. State-Owned Enterprise Reform Intervention Types Reform or Intervention Support Type Description Type SOE reform Upstream reform Interventions focused on the reform of the enabling, World to improve the regulatory, and institutional environment for SOEs to Bank enabling, policy, enhance their performance. Includes interventions 108 Appendix B Portfolio Review Framework and Identification Methodology Reform or Intervention Support Type Description Type and regulatory that seek to change the “rules of the game” for SOEs— Group DPO environment for for example, by introducing market discipline and and IPF SOEs competitive neutrality in SOE markets and sectors; World rationalizing tariffs or SOE subsidy pricing; assessing Bank ASA or reforming market dynamics in pursuit of an optimal IFC mix of public and private ownership; and promoting, advisory designing, or implementing public financial management systems to assess and report on SOE liabilities and to deal with SOEs’ fiscal effects. Downstream May include support to improve SOE governance, Bank reform addressing transparency, and accountability by strengthening the Group DPO firm-level SOE state’s ownership and oversight function over them and IPF concerns through and/or SOEs’ financial accountability, controls, and IFC policy advice, transparency; to improve SOEs’ business and advisory, technical operational performance through company investment assistance, and restructuring, market assessments, product mix and s direct investment process efficiency, performance management MIGA systems, restructuring debts and assets, and rehabilitating assets and infrastructure; and to improve environmental and social aspects. SOE engagement Indirect support Includes interventions aimed to advance SOEs’ World for SOEs through delivery of development goals without reforming the Bank IPF upstream and SOEs. Includes interventions that affect SOEs capacity IFC enabling or position without reforming SOEs themselves or any advisory, engagements that assets that they own or operate. These interventions investment generate external do not set rules for SOEs versus upstream s benefits engagement. Examples include the modernization of MIGA a national payment system that can be used by state- owned financial institutions, or construction of a regional natural gas pipeline that reduces distribution costs for natural gas utilities in a specific country. 109 Appendix B Portfolio Review Framework and dentification Methodology Reform or Intervention Support Type Description Type Direct support for Operations that directly benefit SOEs with the purpose World SOEs through of achieving country and sector development Bank IPF downstream objectives without reforming the SOEs themselves. IFC finance and Support may include expanding, sharpening, investment technical focusing, or mandating an SOE’s role in underserved s assistance to segments of the market; supporting state-owned MIGA leverage their role financial institutions to advance financial inclusion in in pursuit of rural or extreme poverty areas and their use of no- development frills, basic saving accounts; and country-level support objectives from development banks, including their role in development agendas, partnerships with other institutions, and disbursements through apex banks. Note: ASA = advisory services and analytics; DPO = development policy operation; IFC = International Finance Corporation; IPF = investment project financing; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. Identification of SOE/State-Owned Financial Institutions Portfolio Methodology (All Sectors) IEG’s identification methodology of potential SOE reform projects used the Bank Group’s internal project coding framework and targeted keyword searches in text-based data sets to systematically capture and categorize the portfolio subsets relevant to SOE reform and engagement. IEG employed the following steps to identify the evaluation’s portfolio of projects: IEG retrieved projects identified using the Bank Group’s systems and system codes (for example, sector, thematic, and industry codes).  For projects that did not contain at least one of the relevant system codes, IEG performed a targeted keyword search in text-based data sets (for example, project-level abstracts, objectives, and descriptions), 1 and 110 Appendix B Portfolio Review Framework and Identification Methodology for World Bank lending and advisory services and analytics (ASA) also in the institution’s operations portal. After identifying energy and financial sector projects with substantial shares of potential SOE reform and engagement projects compared with other sectors, IEG performed a manual portfolio review of these two sectors’ projects. The manual review was made for 91 percent of Bank Group financing projects, 2 and for 20 percent of World Bank ASA projects identified through the above-mentioned search methods to identify false positives and systematically categorize relevant projects with the goal of developing a unified picture of the features underpinning the SOE reform portfolio in the energy and financial sectors. For the World Bank, IEG identified several Operations Policy and Country Services sector and theme codes relevant to the SOE evaluation. Given that projects may contain one or more sector and theme codes, IEG included in the SOE portfolio any project that contained at least one relevant code. In addition, for development policy operations, IEG searched inside the prior actions database for operations that contained at least one prior action classified under a relevant sector or theme code (table B.2). IEG also ran a targeted keyword search in project titles (both financing and ASA), in a text- based data set that contains project abstracts and other memo fields (lending only), and in the operations portal. This resulted in a list of 1,265 World Bank lending projects and additional finance operations and 1,426 World Bank ASA activities (totaling 2,691 projects), accounting for 22 percent and 24 percent of the World Bank’s overall portfolio, respectively. For International Finance Corporation (IFC), there were no sector or industry codes that could facilitate identification, yet the targeted keyword search proved fruitful. Applying a targeted keyword search strategy to project descriptions resulted in the identification of 194 IFC investment operations (10 percent of their total portfolio) and 142 IFC advisory activities (15 percent of their total portfolio). 111 Appendix B Portfolio Review Framework and dentification Methodology Table B.2. World Bank Group System Codes and Keyword Search Strategy to Identify SOE Portfolio IFC Sector and World Bank Industry Lending and ASA MIGA Sector Codes: Theme code: Codes: None available State-owned enterprise reform and None available Keyword searches privatization (436) in Search carried out in project and prior Keyword searches in project title, IFC action level for World Bank IPF, DPF, and MIGA portal project AS memos, and World Bank ASA (ESW/technical briefs IFC IS SPI memos assistance) Additionally, MIGA Keyword searches in SOE includes list of project title, project abstracts, operations nonhonoring projects. portal, and prior actions Source: Independent Evaluation Group review and interviews with World Bank Group subject matter experts and management. Note: Stemmed keywords used include state_own, public_own, publicly_own, government_own, state_enter, privatiz, government_bus, government_enter, crown_corp, commercial_government, public_sector_under parastatal, nationalized, municipalized, SOE, SOB, and SOFI. ASA = advisory services and analytics; DPF = development policy financing; ESW = economic and sector work; IFC = International Finance Corporation; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; IPF = investment project financing; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise; SOFI = state-owned financial institutions; SPI = Summary of Proposed Investment. Similarly, for MIGA, IEG used a targeted keyword search plus the list of nonhonoring projects to identify and review projects that support SOE reform in client countries. This targeted keyword search strategy resulted in the identification of 38 guarantees that account for 22 percent of the institution’s portfolio in number of projects. In addition, IEG used existing system codes to identify relevant sector-based portfolio segments. To achieve this, IEG retrieved projects using relevant Bank Group system codes (for example, sector, thematic, and industry 112 Appendix B Portfolio Review Framework and Identification Methodology codes). For more on the system codes used to identify these portfolio segments, see table B.3. Table B.3. World Bank Group System Codes to Identify Sector-Based Portfolio Segments IFC Sector and World Bank Lending Industry and ASA MIGA Sector Energy: Energy: Energy: Primary sector: electric Sector codes: LA, LB, LD, LE, Sectors: power power LH, LI, LN, LT, LU, LW Theme codes: 86 Finance: Finance: Finance: Primary sector: finance and Sector codes: FX (financial Sectors: banking, capital insurance sector) markets, financial markets, Theme codes: 30 (finance) financial services, leasing ICT: ICT: ICT: Primary sector: information Sector codes: CX (ICT) Sectors: telecommunications Theme codes: 26 (ICT), 261 (ICT solutions), 262 (ICT policies) Transport: Transport: Transport: Primary sector: Sector codes: TX Sectors: transportation transportation and (transportation) warehousing Theme codes: 713 113 Appendix B Portfolio Review Framework and dentification Methodology IFC Sector and World Bank Lending Industry and ASA MIGA Sector Water: Water: Water: Secondary sector: Sector codes: WX (water and Sectors: water, water and waste treatment and sanitation) wastewater management Theme codes: 716 (urban water, wastewater, and water and sanitation) district heating and cooling Source: Independent Evaluation Group Review and interviews with World Bank Group subject matter experts and management. Note: ASA = advisory services and analytics; ICT = information and communication technology; IFC = International Finance Corporation; MIGA = Multilateral Investment Guarantee Agency. Manual Review Strategy for Potential SOE Reform Energy and Financial Sector Projects IEG’s manual review focused on potential SOE reform projects in the energy and financial sectors. It sought to identify and classify Bank Group SOE reform activities at the project and subproject levels and divided the analysis into four dimensions: relevance, effectiveness, lessons learned, and environmental and social objectives. IEG collected SOE reform project and subproject level information at three levels: projects, interventions, and mechanisms. First, an intervention was defined as a component, subcomponent, activity, or other element within a project that sought to achieve SOE reform objectives and was linked to either qualitative or quantitative outcome indicators. Additionally, an intervention was always connected to a single engagement area of support in SOE reform. These engagement areas are defined as SOE ownership, corporate governance, business and operations, regulatory frameworks and competition, financial management, environmental and social aspects, and macroeconomic and public financial management issues (including fiscal 114 Appendix B Portfolio Review Framework and Identification Methodology policy and debt). Interventions (that is, engagement areas) were thus meant to identify the nature of an SOE reform engagement (tables B.1 and B.4). Second, within each intervention, different mechanisms of engagement were identified. Mechanisms of engagement are defined as the channels through which SOE reform objectives were pursued; they were meant to identify how an intervention achieved its objective (see mechanism definitions in table B.4.). Finally, intervention objectives sought to capture what a specific intervention sought to do; they identified the end to which SOE reform was pursued through an engagement area and its specific mechanism or mechanisms. With both this framework and the theory of change in mind (described in appendix A), IEG developed a data collection tool to review energy and financial sector projects identified as potentially having interventions on SOE engagement and SOE reform. The tool’s structure followed the evaluation questions. It consisted of sections on relevance, effectiveness, and learning, plus a section on environmental and social objectives. Descriptions of each of these sections follow: Relevance: This section sought to identify SOE reform interventions and their level (upstream or downstream), engagement areas, mechanisms, objectives, sectors, and subsectors. In addition, to measure the commitment amounts allocated to SOE reform interventions, the team estimated an average of SOE reform commitments of evaluated projects by engagement area and sector (energy and financial) and subsequently extrapolated these proportions to the entire portfolio. Effectiveness: This section sought to capture ratings of individual SOE reform interventions of IEG-evaluated projects in terms of their development effectiveness, focusing on intermediate-level outcomes. It rated interventions using a four-point scale: achieved, mostly achieved, mostly not achieved, and not achieved. These levels of achievement were applied in line with the World Bank’s outcome ratings of satisfactory, 115 Appendix B Portfolio Review Framework and dentification Methodology moderately satisfactory, moderately unsatisfactory, and unsatisfactory; IFC’s ratings of successful, mostly successful, mostly unsuccessful, and unsuccessful; and MIGA’s ratings of excellent, satisfactory, partly unsatisfactory, and unsatisfactory, though at the intervention level, not the project level. Rating of interventions was based on ICRR information on achievement of objectives that applied to SOE reform, information from evaluation notes of XPSRs on development outcome and project business success that applied to SOE reform, and Project Evaluation Report information on development outcome—including business performance and contribution to private sector development—that applied to SOE reform. The effectiveness section also identified or estimated SOE reform intervention commitment amounts depending on whether they were the explicit amount shown in the project appraisal documents or not. If not, the amount was estimated based on the overall project original commitment amount divided by the number of components, subcomponents, and activities or—for development policy operations—prior actions. 3 The section identified donors and partners that participated or contributed to the SOE reform interventions identified. Learning: This section sought to identify factors of success and failure behind SOE reform development outcomes. These factors were identified at the project level and included only factors that applied to SOE reform interventions previously identified for the project. As in the effectiveness section, factors of success were identified only for projects that IEG evaluated on or before January 31, 2019. This section also identified factor direction—whether the factor was a positive or negative for SOE reform success—and classified factors in two categories: external factors (those exogenous to the Bank Group’s influence), 4 and internal factors (those endogenous to Bank Group actions), such as project design, supervision, team composition, choice of instrument, monitoring and evaluation framework, collaboration among Bank Group institutions, analytical work, and sequencing of operations, among others. Information on factors of 116 Appendix B Portfolio Review Framework and Identification Methodology success was extracted from the lessons sections of the ICRRs, evaluation notes for XPSRs, and the emerging lessons section of the Project Evaluation Reports. Environmental and social aspects: This section sought to identify explicit environmental or social objectives at the project level, whether connected or not connected to SOE reform interventions, and their level of achievement. It was important that the benefits of these objectives must have clearly extended beyond SOEs to a wider population. For example, a downstream SOE reform that sought to improve the business and operations of an energy sector SOE by constructing a more efficient power plant may have an explicit objective to reduce CO2 emissions. Because the benefits of CO2 emission reduction extend beyond the SOE, an environmental objective would be identified for this project. The section would classify social objectives in terms of topic areas, including addressing the environment, poor and excluded populations, quality of service delivery, employment and labor issues, health-related issues, corruption issues, and security and safety issues. It would also classify objectives in terms of how much attention they were given in a project based on the length of the reference: brief (one line or paragraph), substantive (two or more paragraphs or a separate section on the issue), and evidenced (two or more paragraphs or a separate section on the issue plus references to academic literature or Bank Group analytic work). Information was drawn from ICRRs, evaluation notes for XPSRs, and Project Evaluation Reports. This section was only applied to IEG-evaluated projects. Overall, most World Bank Group SOE reform interventions fall into nine engagement areas aiming either to improve the policy and institutional framework (upstream reforms) or to strengthen them at the enterprise level (downstream reforms) and are associated to mechanisms. The four upstream areas of engagement are regulatory framework; governance, transparency, and accountability; SOE ownership; and macro-fiscal policy, public financial management, or debt. The five downstream areas of engagement are 117 Appendix B Portfolio Review Framework and dentification Methodology business and operations, corporate governance, SOE ownership, financial management, and environmental and social aspects (table B.4). After the data analysis, IEG decided to examine five types of Bank Group support for SOE reform in-depth:  Business and operations: This is the engagement area with the biggest share of interventions. The analysis captures support to improve the quality of services provided by SOEs, improve operational efficiency, rehabilitate infrastructure that affects SOEs, and restructure SOEs’ organization and build human capacity.  Corporate governance: This is the second most important engagement area, in which the Bank Group supports both at the upstream and downstream levels. It includes reforms aimed to strengthen SOEs’ corporate governance practices, including those related to transparency, accountability and oversight, risk management, minority shareholder rights, financial reporting, compliance with international standards, board independence, and improved management performance.  Privatization: This was a frequent intervention mechanism through which the Bank Group supported SOE reform objectives between fiscal years 2008 and 2018. Macro-fiscal and public financial management: This is an engagement area that supports SOE reform at the upstream level through the introduction of macroeconomic policy, fiscal policy, or public financial management systems and practices to assess and report on SOE liabilities and/or that enables dealing with SOEs’ macroeconomic or fiscal effects. Regulation and competition: This area captures support to align SOEs’ behavior with development and policy objectives and level the playing field to allow private entry under equitable rules where SOEs have enjoyed competitive advantages. It includes the interventions that aimed to 118 Appendix B Portfolio Review Framework and Identification Methodology strengthen regulatory frameworks and those that have the objective to improve competition, productivity, or innovation in SOEs. Table B.4. Definition of Engagement Areas and Mechanisms. Definition and Mechanisms (mechanism codes Engagement Area included in parentheses) Corporate governance Strengthening SOEs’ corporate governance practices, including (upstream and those related to financial reporting and disclosure requirement downstream reforms) (financial_reporting), compliance with international standards (compliance_international_stds), independent external auditing and oversight (oversight), professionalizing or improving independence of boards of directors (board_ directors), improving risk management or control systems (risk_mgt), anticorruption measures on SOE executives and employees (anticorruption_disclosure), modernizing SOE performance management (performance_mgt), enhancing protection of minority shareholder rights (minority shareholder rights), improving procurement and contracting practices (procurement), transparency (transparency), and enhancing policies that promote competition and level playing field (policies_for_competition). Business and operations World Bank Group support to efforts to improve the quality of (downstream reforms) services of products provided by an SOE (product_service_improvement), improve SOEs’ operational and process efficiency (process_efficiency), build or rehabilitate infrastructure that affects SOEs (infrastructure), undertake organizational restructure (for example, creating or eliminating departments or functions; organizational_restructuring), and enhance human resource management, which captures Bank Group support to training and other capacity-building activities within specific SOEs and/or that benefit specific SOEs (human_resource_mgt). 119 Appendix B Portfolio Review Framework and dentification Methodology Definition and Mechanisms (mechanism codes Engagement Area included in parentheses) SOE ownership Bank Group support for SOE to fully or partially privatize SOEs (upstream and through divestment or equitization of shares (privatization), downstream reforms) promotion of public-private partnerships and other partnership arrangements that involve SOEs (ppp_&other_partnerships), corporatization of government entities into SOEs (corporatization), other private sector crowding-in and opening to entry mechanisms (for example, allowing foreign ownership of SOEs, private_sec_crowding_in), setting up new SOEs and liquidation of existing (soe_setup_or_liquidation), introducing competition for the market to allow private participation, and national SOE ownership policies and institutions (for example, policies for openness to private sector participation and setup of privatization agencies). Regulatory framework Bank Group support to SOE reform at the upstream level (upstream) through national or sectoral laws, regulations or legal and regulatory institutions that affect SOEs’ enabling environment (laws_regulations), set up of regulatory agencies (regulatory_agency), preparation or strengthening of sector strategies (strategies), adoption of pricing or tariff structures (tariffs_or_pricing), and payment systems. Financial management Bank Group support to improve the financial sustainability of (downstream) SOEs, including through restructuring and rehabilitating their debts/assets (restructuring), strengthening their creditworthiness or expanding their financing options (strengthen creditworthiness), improving expenditure/investment management (expenditure_investment_mgt), improving budgeting practices (budget_mgt), or enhancing revenue collection (revenue_mgt). Environmental and Bank Group support to review and/or reform environmental social aspects aspects (e&s_environment), poverty and inclusiveness aspects (downstream) (e&s_poverty), public health and safety aspects (e&s_health&safety), gender aspects (e&s_gender), and labor aspects (e&s_labor). 120 Appendix B Portfolio Review Framework and Identification Methodology Definition and Mechanisms (mechanism codes Engagement Area included in parentheses) Macro-fiscal, policy, Bank Group support of subsidy reduction, reform or control for PFM, and debt SOEs (subsidies), improving debt management of SOEs (upstream) (debt_mgt), improving expenditure management (expenditure_investment_mgt), improving revenue collection and revenue management, improving planning processes (improve_planning_process) and budget management practices (budget_mgt), and improving accounting/auditing (accounting_or_auditing). Note: PFM = public financial management; SOE = state-owned enterprise. Notes 1 Stemmed keywords used: state_own, public_own, publicly_own, government_own, state_enter, privatiz, government_bus, government_enter, crown_corp, commercial_government, public_sector_under parastatal, nationalized, municipalized, SOE, SOB, and SOFI. 2 Financing projects refers to World Bank lending, International Finance Corporation investment, International Finance Corporation advisory, and Multilateral Investment Guarantee Agency projects. 3 For investment projects, commitments were identified through either state-owned enterprise (SOE) reform allocated amounts as specified in project appraisal documents’ components and subcomponents, when available, or by dividing the project’s full commitment amount by the number of components (or subcomponents if available) and multiplying the result by the number of SOE reform interventions in a specific engagement area within a specific sector. For development policy loans, commitments were identified by dividing the loan’s commitment amount by the number of prior actions and multiplying the result by the number of SOE reform prior actions found. 4 External factors included counterpart commitment, public and private sector institutional capacity, collaboration with external partners and donors, coordination, and political economy external shocks. 121 Appendix C. State-Owned Enterprise Reform Support: Financial and Energy Sectors Results of Manual Review of Potential State- Owned Enterprise Reform Projects The initial review identified 2,242 potential state-owned enterprise (SOE) reform projects (952 financing projects 5 and 1,290 World Bank advisory services and analytics [ASA]) from the financial and energy sectors (including oil, gas, and mining) that had been approved between fiscal year (FY)08 and FY18 or evaluated between FY08 and FY18 but approved on or after FY02. 6 From the 2,242 projects, the Independent Evaluation Group (IEG) reviewed 1,123 projects (50 percent), composed of 868 World Bank Group financing projects (91 percent) and 255 ASA projects (20 percent sample). An additional 26 Bank Group financing projects were identified and reviewed during the evaluation through country case studies and desk reviews, for a total of 1,149 reviewed projects: 894 financing projects and 255 ASA projects. 7 From the identified financing projects, 507 were evaluated and were all reviewed, and 473 nonevaluated, of which 387 (87 percent) were reviewed. IEG assigned the original portfolio to its team of reviewers for manual revision based on a random sample, stratified by institution to allow for the possibility of reviewing only a portion of the population at the end of the evaluation. 8 Similarly, the 20 percent random sample drawn from the population of ASA projects was stratified by income level, which is expected to be representative of its population. The success rate for SOE reform project identification strategy for the energy and financial sector was 54 percent. From the 1,149 projects, 623 (507 financing and 116 ASA) were classified as having SOE reform interventions relating to the financial sector, energy sector, or both, or at the national level. 9 For financing projects, identification success rate was 57 percent (507/894), which 374 were approved during the FY08–18 period; 10 the remaining 133 122 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors projects were approved before this period, except for 1 project which was approved for the Arab Republic of Egypt in FY19 and was included as part of that country’s case study. Identification success rates between sectors were different. Using the World Bank’s sector classification, less than half (46 percent) of the projects reviewed in the financial sector contain an SOE reform component, while 78 percent of energy projects were correctly identified. Also, there were differences between World Bank sector classification and the portfolio review analysis’s interventions sector classification. So, in the end, SOE reform interventions were most frequently found in the energy sector, followed by the financial sector and lastly the national-level reforms. For ASA projects, identification success rate was lower (45 percent: 166/255). The financial sector was confirmed as having the largest number of projects with SOE reform interventions, followed by national-level interventions and the energy sector. Portfolio Overview The majority of financing to SOEs in the financial and energy sectors have been delivered through the World Bank. Over the 10-year period between FY08 and FY18, IEG estimated that the Bank Group has approved and delivered 421 SOE reform financing operations, which accounted for $71.4 billion in SOE reform commitments. Much of this support is delivered by the World Bank: Its lending operations accounted for 68 percent (285 projects) and 91 percent ($64,832 million) of volume when accounting for total project volume. International Finance Corporation (IFC) support accounts for 14 percent of projects, delivered through investment services (IS) ($3.8 billion, with 61 projects) and advisory services (AS; $51 million, with 59 projects). The Multilateral Investment Guarantee Agency (MIGA), delivered 17 guarantees over the evaluation period, accounting for $2.8 billion. In addition, an estimated 587 of pieces of analytical work have been delivered through World Bank’s ASA, accounting for $104 million. 123 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.1. SOE Reform Projects, Interventions and Commitments by Institution Approved FY02–07 and Evaluated FY08–18 Approved FY08–18 Total Projects Interventions Volume Volume Total Total Proj. Intrv. Vol. Share Share Proj. Intrv. ($, Institution (no.) (no.) ($, millions) (no.) (%) (no.) (%) ($, millions) (%) (no.) (no.) millions) Reviewed sample World Bank 117 375 16,593 253 68 710 80 57,569 90.8 370 1,085 74,162 lending IFC IS 13 33 769 54 14 83 9 3,343 5.3 67 116 4,112 IFC AS 1 2 3 52 14 81 9 45 0.1 53 83 48 MIGA 1 1 4 15 4 15 2 2,476 3.9 16 16 2,480 Subtotal 132 411 17,369 374 100 889 100 63,433 100 506 1,300 80,802 World Bank 0 0 116 100 234 100 20 100 116 234 20 ASA Total 132 17,369 490 100 1,123 100 63,453 100 622 1,123 80,822 124 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Approved FY02–07 and Evaluated FY08–18 Approved FY08–18 Total Projects Interventions Volume Volume Total Total Proj. Intrv. Vol. Share Share Proj. Intrv. ($, Institution (no.) (no.) ($, millions) (no.) (%) (no.) (%) ($, millions) (%) (no.) (no.) millions) b. Population projection World Bank 117 375 16,593 285 68 800 80 64,832 90.8 402 1,175 81,425 lending IFC IS 13 33 769 61 14 93 9 3,765 5.3 74 126 4,534 IFC AS 1 2 3 59 14 91 9 51 0.1 60 93 54 MIGA 1 1 4 17 4 17 2 2,788 3.9 18 18 2,792 Subtotal 132 411 17,369 421 100 1,001 100 71,436 100 553 1,412 88,805 World Bank 0 0 0 587 100 1,184 100 104 100 587 1,184 104 ASA Total 132 17,369 1,008 100 2,185 100 71,540 100 1,140 2,185 88,909 Source: Independent Evaluation Group portfolio review and analysis. Note: SOE reform financing project and commitment projections in section b are the result of multiplying sample projects and commitments in panel a by a factor of 723/642, based on population and sample sizes from the operations approved between FY08 and FY18. Similarly, SOE reform ASA project and commitment projections are the result of multiplying sample projects and commitments by a factor of 1290/255. ASA = Advisory Services and Analytics; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; Intrv. = interventions; MIGA = Multilateral Investment Guarantee Agency; proj. = projects; SOE = state-owned enterprise. 125 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors SOE Reform Engagement: Portfolio Approved between FY08 and FY18 The Bank Group delivered its support for the reform of SOEs in the financial and energy sectors mainly through three Global Practices (GPs) and two IFC industry groups. 11 For the World Bank’s lending portfolio from FY08 to FY18, three GPs (Macroeconomics, Trade, and Investment; Energy and Extractives; and Finance, Competitiveness, and Innovation) accounted for most of its support to reform SOEs in the financial and energy sectors (95 percent and $62.7 billion for World Bank lending), while World Bank ASA was delivered it mainly through the Finance and Market GP (72 percent and $50 million). For IFC advisory, most of its support is concentrated in the Cross-Cutting Advisory Solutions, Public-Private Partnerships Transaction Advisory, and Financial Institutions Group (65 percent and $27 million). For IFC investment, 93 percent of investments were delivered by the Infrastructure industry group and Financial Markets. Most of the MIGA guarantees (87 percent and $2.6 billion) were delivered by the power sector. 126 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.2. SOE Reform Portfolio Summary by Operational Unit, FY08–18 Projects Estimated Amount Institution World Bank Group Operational Unit (no.) (%) ($, millions) (%) World Bank lending Macroeconomics, Trade, and Investment 123 43 16,742 26 Energy and Extractives 77 27 27,670 43 Finance, Competitiveness, and Innovation 71 25 18,322 28 Other 15 5 2,099 3 Total, World Bank lending 285 100 64,832 100 IFC IS Infrastructure 38 63 2,944 78 Financial Markets 18 30 578 15 Other 5 7 243 6 Total, IFC IS 61 100 3,765 100 IFC AS Cross-Cutting Advisory solutions 17 29 14 27 PPP Transaction Advisory 11 19 3 6 Financial Institutions Group 10 17 10 20 Other 20 35 27 47 127 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Projects Estimated Amount Institution World Bank Group Operational Unit (no.) (%) ($, millions) (%) Total, IFC AS 59 100 51 100 MIGA Power 15 87 2,606 90 Other 2 13 182 10 Total, MIGA 17 100 2,788 100 Subtotal 421 71,436 World Bank ASA Finance and Markets 425 72 50 48 Energy and Extractives 86 15 35 34 Governance 40 7 3 3 Other 35 6 16 16 Total World Bank ASA 587 100 104 100 Total 1,008 71,540 Source: Independent Evaluation Group portfolio review and analysis. Note: Population projection. SOE reform financing project and commitment projections are the result of multiplying sample projects and commitments by a factor of 723/642, based on population and sample sizes. SOE reform ASA project and commitment projections are the result of multiplying sample projects and commitments by a factor of 1290/255. ASA = Advisory Services and Analytics; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. 128 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Support for the reform of SOEs has remained stable. The World Bank approved 37 financing operations per year during FY08 to FY18, except for FY10, the year in which was approved 53 projects. World Bank lending approvals were on average 25 operations per year (excluding FY10, when there was a corporate push to respond to the global financial crisis). IFC has remained at around 6 investment and 5 advisory operations, and MIGA has increased its activities since 2011 (from 1 to 2 operation per year between FY08–12 and FY13–18). For its part, World Bank ASA increased its approvals from 54 to 64 operations per year in the same period. Support to reform SOEs in the financial sector declined over time, while support in the energy sector remained stable, but commitments in both sectors increased. The highest level of support to SOE reform in the financial sector came in FY09, FY10, and FY11, with 17, 26 and 17 financing projects approved, respectively, in those years. However, after FY11, the pace was reduced, with an average of 12 financing projects approved annually between FY12 and FY18. In the energy sector, approvals remained on average at 23 financing SOE reform projects per year between FY08 and FY18. Overall commitments to reform SOEs have increased from $4,875 million per year to $6,193 million per year between FY08–12 and FY13–18 (except from FY10, in which commitments were $14,781 million). For its part, ASA support to reform SOEs in both sectors has increased in number of approved projects and amount committed. 129 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.1. Evolution of World Bank Group Support for the Reform of SOEs by Sector, FY08–18 (Population projection) a. Financing support (no. of projects) b. Financing support (SOE reform commitments) c. World Bank ASA support (no. of projects) d. World Bank ASA support (SOE reform commitments) Source: Independent Evaluation Group portfolio review and analysis. Note: Projects can be double counted across sectors if they have SOE reform interventions in more than one sector. The unique value of projects in panels a and b is 421 and in panels c and d is 587. ASA = Advisory Services and Analytics; FY = fiscal year; Mov avg = moving average; SOE = state-owned enterprise. 130 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Description of Revised Sample: Projects Approved FY08–18 Based on the revised sample, Bank Group support to reform SOEs has been relatively more focused on lower-middle (46 percent, not considering World Bank ASA) and low-income countries (29 percent), followed by upper-middle countries (23 percent) and only seven projects on high-income countries (figure C.2). The share of SOE reform projects to the total World Bank lending portfolio increases as income level decreases (but is more focused on lower-middle countries), while more than 50 percent of IFC and MIGA portfolios were concentrated in lower-middle-income countries and upper- middle countries (+23 percent). World Bank ASA has also been more focused on lower-middle-income countries (38 percent), followed by upper-middle countries (29 percent) and low-income countries (26 percent). Using the World Governance Indicator Control of Corruption (table C.3), which measures the extent to which public power is exercised for private gain and capture of the state by elites and private interest, 30 percent of SOE reform interventions were in countries with low control of corruption. 131 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.2. Distribution of World Bank Group Reform Projects by Income and Institutions, FY08–18 (Sample) a. By country income level and institution, no b. Distribution ASA Source: Independent Evaluation Group portfolio review and analysis. Note: Figure is based on 367 projects; excludes 7 regional or income-unclassified country/territory projects. ASA = Advisory Services and Analytics; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency. Table C.3. State-Owned Enterprise Reform by Country Corruption Level, FY08–18 Projects Interventions Control of Corruption Level (no.) (%) (no.) (%) Low 90 27 239 30 Lower middle 85 25 191 24 Upper middle 93 27 224 28 High 71 21 154 19 Total 339 100 808 100 Source: World Bank World Governance Control of Corruption Indicator; Independent Evaluation Group portfolio review and analysis. Note: Excludes interventions in regional projects (16) and in countries with no Worldwide Governance Indicators 65). The control of corruption indicator is measured from –1.9 to 2.5, where higher values correspond to better outcomes. Cutoffs are based on quartiles of this indicator. Sample with no Advisory Services and Analytics. 132 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors SOE Reform by Region During the evaluation period, the Bank Group provided financing support for reforming of SOEs in 103 countries and ASA in 59 countries (table C.4). Bank Group financing operations were active in 34 countries in Sub-Saharan Africa (approving 114 projects in this region, which accounted for 30 percent of the financing projects), 21 countries in Europe and Central Asia, 16 in Latin American and the Caribbean, 13 in East Asia and Pacific, 11 in the Middle East and North Africa, and 8 in South Asia. For its part, World Bank ASA supported 15 countries in Europe and Central Asia (where most operations were approved), 15 in Sub-Saharan Africa, 12 in Latin America and the Caribbean, 8 in East Asia and Pacific, 5 in Middle East and North Africa, and 4 in South Asia. Most of the supported countries had more than three interventions (80 percent) and more than 3 projects (50 percent); the rest of the countries had 1 or 2 projects, almost equally distributed. In the energy sector, most countries had more than three interventions (76 percent), but projects were more distributed among countries, with 1, 2, or 3 approved projects. In the financial sector, interventions and projects were more clustered, having more countries with one or two interventions (52 percent) and countries with 1 approved project (63 percent). 133 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.4. Countries, Projects, and Interventions by Region, FY08–18 (number) Financing Projects World Bank ASA Inter- Inter- Region Countries Projects ventions Countries Projects ventions SSA 34 114 292 15 25 60 ECA 21 78 193 15 29 47 LAC 16 34 65 12 17 26 EAP 13 69 142 8 14 31 MNA 11 33 91 5 6 15 SAR 8 39 90 4 12 26 RGN 7 16 13 29 Source: Independent Evaluation Group portfolio review and analysis. Note: ASA = Advisory Services and Analytics; EAP= East Asia and Pacific; ECA= Europe and Central Asia; FY = fiscal year; LAC= Latin America and the Caribbean; MNA= Middle East and North Africa; RGN = Regional; SAR= South Asia; SSA= Sub-Saharan Africa. Nature of SOE Reform: SOE Reform Engagement Areas, Mechanisms and Objectives Bank Group financing support by reform type was focused on strengthening them at the enterprise level (53 percent of financing interventions), while World Bank ASA focus more on upstream reforms (66 percent; figure C.4). 12 Downstream financing support put more emphasis on improving SOEs’ business and operations (24 percent), corporate governance (12 percent), ownership (10 percent) and competition and regulation (7 percent); Upstream support centered on strengthening competition and regulation in SOE markets (20 percent) and SOEs’ governance transparency and accountability rules (13 percent). World Bank ASA interventions focused on improving SOE’s governance (26 percent), competition and regulatory framework (22 percent) at the upstream level, and improving SOEs business and operations practices (14 percent). However, approach differed by 134 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors institution. Most interventions undertaken by IFC and MIGA sought to improve SOEs’ business and operations and/or their ownership. By contrast, World Bank lending and ASA more often focused on strengthening SOEs’ regulatory framework and competition and governance rules, in addition to improving business operation. This suggests that World Bank and, to a limited extent, IFC Advisory, focus more on upstream reforms. Upstream support was relatively more frequent in upper-middle-income countries while support to lower-middle-income countries focused more on downstream reforms. Financing interventions in low-income countries focused on reforming SOEs’ corporate governance, business and operations, and competition and regulatory framework (+24 percent) through rehabilitation of infrastructure, improving human resource and performance management, modernizing SOE performance management, financial reporting, preparing sector strategies and supporting regulatory agencies and laws and regulations. For lower-middle income countries, support focused on business and operations (28 percent through rehabilitation of infrastructure) and on strengthening competition and regulation (25 percent through adopting pricing or tariff structures and preparing laws and regulation). Financing interventions in upper-middle- and high-income countries concentrated on corporate governance (+28 percent), with a focus on performance and risk management and oversight. Relative to the other income groups, support to high income countries focused less on SOE’s ownership. World Bank ASA support in low and lower-middle-income countries most commonly targeted corporate governances, while support to upper-middle- and high-income countries focused relatively more on SOE Ownership. 135 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.3. SOE Reform Interventions by Reform Types, FY08–18 (Sample) A. Distribution of the interventions by reform type for the World Bank Group financing projects B. Distribution of the interventions by reform type for the World Bank ASA projects 136 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors C. Distribution of interventions by reform types and institution Source: Independent Evaluation Group portfolio review and analysis. Note: Figures are based on 870 financing and 238 World Bank advisory services interventions identified for the SOE reform types. Interventions can be double counted if they supported more than one reform type. 137 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Variation by sector. Financing interventions in the energy sector were commonly targeted SOE’s business and operations and competition and regulation (28 percent and 29 percent, respectively) and focused relatively more on SOE’s Ownership than the other sectors. For the financial sector the focus was on Corporate Governance and Business and Operations (31 percent and 28 percent) while almost half of the interventions at the national level focused on Corporate Governance reforms (45 percent). World Bank ASA interventions in the energy sector focused on improving competition and regulation for SOEs and their Business and Operations (37 percent and 25 percent, respectively), while those in the financial sector and at the national level were concentrated in Corporate Governance. Financing interventions in ECA and MNA supported SOEs’ policies and institutional frameworks, while in SAR, LCR, AFR, and EAP interventions focused on improving SOE at the enterprise level. Business and Operations and Competition and Regulation were more common in LAR, SAR, and SSA, but with differences among mechanisms. The most frequent mechanism was product service improvement in LAC, human resource management and product service improvement in SAR, and infrastructure investments in SSA. A focus on SOE Ownership was more common in MNA (especially in private sector crowding in), ECA (privatization) and EAP (support to PPPs and other partnerships). For its part, World Bank ASA focused on improving SOEs’ Corporate Governance in all regions (+31 percent). 138 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.4. Distribution of SOE reform types, FY08–18 (Sample) a. Financing interventions by income level b. World Bank ASA interventions by income level c. Financing interventions by sector d. World Bank ASA interventions by sector Source: Independent Evaluation Group portfolio review and analysis. Note: Figure A is based on 854 interventions (excludes 16 part of regional or income-unclassified countries/territories), figure B on 210 interventions (excludes 28 also part of regional or income unclassified countries), figure C on 870 and figure D on 238 interventions. Interventions can be double counted if they supported more than one reform type. Source 139 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Mechanisms of Engagement: How Objectives Were Pursued The most frequent intervention mechanisms were concentrated in two engagement areas (business and operations and SOE ownership) (figure C.5). Bank Group support to SOE business and operations reform was provided mainly through the two of the most common mechanisms -- construction and rehabilitation of infrastructure (8 percent) and capacity-building activities (6 percent). For its part, support to SOE ownership was provided through PPP and other partnership arrangements (7 percent) and privatization (6 percent). The World Bank lending intervention mechanisms of SOE reform were more evenly distributed, with a slight focus on human resource management (6 percent of interventions), rehabilitation of infrastructure (6 percent) and oversight (5 percent). For its part, IFC investment concentrated almost half of its mechanisms on rehabilitation of infrastructure, PPPs that involve SOEs, and privatization, while half of IFC advisory interventions supported PPP, product service improvement, and human resource management. MIGA guarantees clearly focused on rehabilitation of infrastructure and PPPs (80 percent or interventions) as a mechanism of SOE reform. 140 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.5. Intervention Mechanisms in SOE Reform, FY08–18 (Sample) Independent Evaluation Group portfolio review and analysis. Note: The figure excludes mechanisms with fewer than 20 actions. FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; PPP = public-private partnership; SOE = state-owned enterprise. 141 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.6. Objectives of Support, No ASA, FY08–18 (Sample) Source: Independent Evaluation Group portfolio review and analysis. Note: The figure is based on 889 interventions of 374 projects; excludes World Bank ASA projects. Interventions may have more than one objective. ASA = Advisory Services and Analytics; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; SOE = state- owned enterprise. Intervention Objectives Overall support to SOE reform most frequently sought to improve sectoral efficiency or strengthen financial or operational performance, with some difference by Bank Group institution (figures C.6, C.7). The focus of the support is reflected in the types of objectives undertaken by the institution. The World Bank Lending operations place more emphasis on sectoral efficiency interventions (24 percent), strengthening SOEs financial or operational performance (17 percent), and improving SOEs’ transparency, accountability, or oversight (17 percent). In contrast, more than half (57 percent) of all interventions undertaken by IFC IS sought to improve SOEs’ service delivery and quality or to improve competition, productivity, 142 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors or innovation in SOEs. IFC AS focused their support on strengthening SOEs’ financial or operational performance (25 percent), and MIGA’s centers on the improvement of service delivery, service quality, or sectoral efficiency (94 percent). Figure C.7. Distribution of SOE Reform Interventions by Objectives and Engagement Area Source: Independent Evaluation Group portfolio review and analysis. Note: Interventions may have more than one objective. N = 889 interventions. E&S = environmental and social; Mgt. = management; PFM = public financial management; SOE = state-owned enterprise 143 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors SOE Reform by Sector Financing support to SOE reform was more frequent in the energy sector than in the financial sector, while ASA focused on the financial sector. Over half of SOE reform financing interventions supported the energy sector (57 percent, of which 82 percent went to low or lower-middle-income countries), followed by interventions in the financial sector (30 percent, 76 percent of which went to lower-middle or upper-middle-income countries) and interventions supporting SOE reform at the national level (13 percent, 42 percent of which went to upper-middle countries). 13 Sub- Saharan Africa was the Region with most interventions in the energy sector (213), Europe and Central Asia and South Asia were the Regions that had most interventions in the financial sector (60 and 54, respectively), and Europe and Central Asia and Sub-Saharan Africa for reforms at national level (43 and 40, respectively). Development policy operations (DPOs) were more frequent at national level than investment project financings (IPFs) (20 percent of DPO interventions were at the national level versus 9 percent for IPF), while IPFs were more frequent than DPOs in the energy and finance sectors (59 percent and 32 percent, respectively versus 55 percent and 25 percent). For its part, World Bank ASA’s interventions focused on reforming SOEs in the financial sector (50 percent), followed by support on reforming SOEs at the national level (27 percent) and in the energy sector (24 percent). 144 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.8. SOE Reform Interventions by Sector, Income Level and Region, no World Bank ASA, FY08–18 (Sample) a. Interventions by income level and sector b. Interventions by region and sector Source: Independent Evaluation Group portfolio review and analysis. Note: Panel a is based on 873 interventions; 16 interventions from regional projects or unclassified territories were excluded from the analysis. Panel b is based on 889 interventions. ASA = Advisory Services and Analytics; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; RGN = regional; SAR = South Asia; SOE = state- owned enterprise; SSA = Sub-Saharan Africa. 145 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors SOE Reform in the Energy Sector The energy sector was the largest in terms of both number of approvals and estimated SOE reform commitment volume. From the revised portfolio, the team found 224 projects in FY08–18 worth $39 billion. World Bank lending accounted for 145 projects worth $33.8 billion in SOE reform commitments: 68 of these projects were mapped to the Macroeconomics, Trade, and Investment GP, 14 and 67 to the Energy and Extractives GP, worth $32.6 billion or over 84 percent of energy sector estimated commitment value. 15 For its part, IFC investment accounted for 37 energy sector projects worth $2.8 billion in SOE reform commitments: 34 of these projects were mapped to the Infrastructure industry group, worth $2.6 billion. 16 IFC advisory accounted for 29 energy sector projects worth $21 million in SOE reform commitments, of which 14 were Cross-Cutting Advisory solutions worth $12 million. For example, IFC approved a project for Indonesia in FY08 (26215) for advising PLN (state-owned power utility), in designing/preparing a bidding process for private sector participation to build, own, and operate a greenfield coal-fired central java power. The team also found 13 MIGA guarantees in the energy sector worth $2.3 billion in SOE commitments, which were delivered by the power sector. For example, a guarantee to a K-Water (M1162) in Pakistan approved in FY12 for the construction and operation of a hydropower plant, which will sell its electricity to an SOE under a power purchase agreement (SOE ownership engagement area). 146 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.5. SOE Reform Portfolio FY08–18 Summary (Sample) Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total World Macroeconomics, 10 10 50 37 26 4 16 153 Bank Trade, and 1,139.4 859.7 1,644.6 798.7 447.6 1,913.8 1,540.7 8,344.4 lending Investment Energy and 82 38 31 36 25 16 8 236 Extractives 14,792.5 4,823.5 849.8 541.6 472.5 2,169.2 615.5 24,264.5 Other 5 1 3 2 2 — — 13 1,025.0 65.1 80.8 4.6 4.2 — — 1,179.7 Total interventions 97 49 84 75 53 20 24 402 (no.) Total estimated 16,956.9 5,748.2 2,575.1 1,344.9 924.3 4,083.0 2,156.2 33,788.6 amount ($, millions) IFC IS Infrastructure 25 19 — — 1 7 — 52 1,469.6 376.9 — — 9.4 758.5 — 2,614.4 Other 1 3 — — — — — 4 120.0 95.2 — — — — — 215.2 147 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total Total interventions 26 22 — — 1 7 — 56 (no.) Total estimated 1,589.58 472.05 — — 9.40 758.51 — 2,829.5 amount ($, millions) IFC AS Cross-Cutting 1 14 1 — — — — 16 Advisory solutions 1.9 9.7 0.3 — — — — 12.0 Public-Private — 8 — — 1 — — 9 Partnerships — 2.22 — — 0.01 — — 2.2 Transaction Advisory Infrastructure and — 4 1 2 2 — — 9 Natural Resources — 4.3 0.4 0.3 0.5 — — 5.6 Other 1 1 — — — 1 — 3 0.2 0.6 — — — 0.2 — 1.0 Total interventions 2.0 27.0 2.0 2.0 3.0 1.0 — 37.0 (no.) 148 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total Total estimated 2.1 16.9 0.7 0.3 0.5 0.2 — 20.8 amount ($, millions) MIGA Power 8 5 — — — — — 13 2,161.1 152.6 — — — — — 2,382.6 Total interventions 8 5 — — — — — 13 (no.) Total estimated 2,161.1 152.6 — — — — — 2,382.6 amount ($, millions) Grand total financing 133 103 86 77 57 28 24 508 interventions (no.) Grand total estimated 20,709.7 6,389.8 2,575.8 1,345.2 934.2 4,841.8 2,156.2 38,952.7 amount ($, millions) World Energy and 11.0 5.0 13.0 9.0 5.0 2.0 1.0 46.0 Bank Extractives 4.6 0.5 0.3 0.2 0.2 1.0 0.1 6.8 ASA Other 2 2 1 2 2 — — 9 0.4 0.1 0.02 0.02 0.02 — — 0.6 149 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total Total interventions 13 7 14 11 7 2 1 55 (no.) Total estimated 4.9 0.7 0.3 0.2 0.2 1.0 0.1 7.4 amount ($, millions) Source: Independent Evaluation Group portfolio review and analysis. Note: ASA = Advisory Services and Analytics; E&S = environmental and social; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; mgt = management; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. (no.) of interventions and SOE reform estimated commitments. 150 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Energy sector support focused more on nonrenewable electricity than on renewables. A total of 61 interventions addressed upstream reform for the nonrenewable electricity sector and 130 interventions supported nonrenewable electricity utilities, followed by the renewable energy subsector (107) and interventions for both renewable and nonrenewable energy (75). Renewable electricity projects were more recent; on average the Bank Group approved five renewable energy projects per year in FY08–12, while in FY13–18 it approved an average of seven projects per year. Nonrenewable electricity projects have remained stable approximately nine operations per year between FY08 and FY18. Support to SOEs in the nonrenewable electricity subsector was mostly for infrastructure, PPP and other partneships, and tariff or pricing policies, and support for renewable electricity was for PPP and other partnerships and for infrastructure. Figure C.9. Support to Reform SOEs Power Subsector by Source Energy, No World Bank ASA, FY08–18 (Reviewed Sample) a. By institution (share of interventions) 151 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors b. Evolution of Bank Group energy subsector support (no. of projects) Source: Independent Evaluation Group portfolio review and analysis. Note: Panel a is based on 373 power subsector interventions. Panel b shows projects with SOE reform interventions identified in the electricity subsectors. N = 195. Excludes World Bank ASA. ASA = Advisory Services and Analytics; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; Mov avg = moving average; SOE = state-owned enterprise. The Bank Group’s electricity support for SOE reform in lower and upper- middle-income countries focused more on generation through renewable sources. One-third of lower-middle and upper-middle-income interventions supported generation, transmission, or distribution through renewable resources. Support to low-income countries, however, had lower emphasis on renewables (19 percent) and greater emphasis on nonrenewable sources of electricity (57 percent). Support for SOE reform in the context of nonrenewable sources of energy was provided to countries in Sub-Saharan Africa (58 percent). Although renewable energy interventions were more spread out across regions, each region allocated approximately one-third of its interventions for the generation, transmission, or distribution of renewable energy, except for Sub-Saharan Africa, where renewable accounted for approximately 15 percent of its interventions. 152 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.10. World Bank Group support to reform SOE electricity subsector no World Bank ASA, FY08–18 a. Overlap with income level b. Overlap with region Independent Evaluation Group portfolio review and analysis. Note: Sample. Panel a is based on 364 interventions in electricity; excludes 9 interventions from regional or unclassified territories. Panel b is based on 373 interventions. ASA = Advisory Services and Analytics; EAP = East Asia and Pacific; ECA = Europe and Central Asia; FY = fiscal year; H = high income; L = low income; LAC = Latin America and the Caribbean; LM = lower-middle income; MNA = Middle East and North Africa; RGN = regional; SAR = South Asia; SOE = state-owned enterprise; SSA = Sub-Saharan Africa; UM = upper-middle income. 153 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors SOE Reform in the Financial Sector During the evaluation period, the team found 139 projects that support SOE reform in the financial sector, worth $21.4 billion in estimated SOE reform commitments. This support was led by the World Bank’s Finance, Competitiveness, and Innovation GP, which approved 56 projects summing $15.7 billion in estimated commitments or 74 percent of all the financial sector SOE reform volume. 17 Another GP that supported SOE reform for the financial sector was Macroeconomics, Trade, and Investment, which approved 38 projects accounting $4.4 billion in estimated SOE reform commitments. 18 For its part, IFC investment supported SOE reforms in the financial sectors mainly through the Financial Market industry group. 19 The team also found 21 IFC advisory services projects that supported SOE reforms in the financial sector, mainly through the financial institutions group. One example of business and operations support provided by this group is the advisory approved for Oman in FY13 (596907), which assisted Bank Muscat (a state-owned financial institution) to strengthen its small and medium enterprise banking operations, including reengineering the credit underwriting and risk process and enhancing capacity to offer nonfinancial services to small and medium enterprise clients. Finally, there were two MIGA guarantees that supported SOE reform in this sector. 154 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.6. Financial Sector SOE Reform Portfolio, FY08–18 Summary (Reviewed Sample) Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total World Finance, 37 13 21 54 12 2 2 141 Bank Competitiveness, 8,694.4 710.8 1,296.8 3,240.7 268.0 1,551.7 10.9 15,773.4 lending and Innovation Macroeconomics, 4 12 8 18 8 2 52 Trade, and 1,400.0 709.7 444.1 1,369.8 430.6 54.5 4,408.8 Investment Other 3 1 1 5 415.0 31.6 25.4 472.1 Total interventions 44 26 29 73 20 2 4 198 (no.) Total estimated 10,509.4 1,452.2 1,740.9 4,635.9 698.7 1,551.7 65.4 20,654.2 amount ($, millions) 155 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total IFC IS Financial Markets 7 7 2 7 3 26 190.6 146.2 12.1 62.1 101.9 512.8 Other 1 1 0.5 0.5 Total interventions 7 7 — 3 7 3 — 27 (no.) Total estimated 190.6 146.2 12.6 62.1 101.9 513.4 amount ($, millions) IFC AS Financial Institutions 9 3 1 3 16 Group 5.5 0.6 0.1 3.0 9.2 Access to Finance 5 2 1 2 1 11 2.1 0.2 0.01 0.2 0.05 2.6 Equitable Growth, 6 1 7 Finance, and 10.0 0.2 10.2 Institutions 156 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total Other 3 2 2 1 8 0.8 0.3 0.1 0.1 1.2 Total interventions 23 5 3 6 2 3 42.0 (no.) Total estimated 18.4 0.6 0.1 1.0 0.1 3.0 23.2 amount ($, millions) MIGA Financial Markets 1 2 22.1 117.1 Banking 1 1 139.7 139.7 Total interventions 2 2 (no.) Total estimated 161.8 161.8 amount ($, millions) Total financing interventions (no.) 74 38 32 82 31 8 4 269 157 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Business and SOE Regulatory Corporate Financial Fiscal Global Practice Operations Ownership Frameworks Governance Mgt E&S Policy Total Total estimated amount ($, 10,718.4 1,599.0 1,741.0 4,649.5 922.6 1,656.6 65.4 21,352.6 millions) World Finance & Markets 18 13 19 44 8 1 7 110 Bank ASA 3.7 0.8 0.7 2.1 0.3 0.2 0.2 8.0 Other 1 1 2 1 1 6 0.1 0.01 0.03 0.02 0.01 0.2 Total interventions 19 14 19 46 9 1 8 116 (no.) Total estimated 3.8 0.8 0.7 2.1 0.4 0.2 0.2 8.2 amount ($, millions) Source: Independent Evaluation Group portfolio review and analysis. Note: SOE reform estimated commitments were either identified or calculated for each subproject intervention and then aggregated by engagement area. ASA = Advisory Services and Analytics; E&S = environmental and social; FY = fiscal year; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; mgt = management; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. (no.) of interventions and commitment levels in millions of US dollars. 158 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors World Bank Group support to SOE reform in the financial sector was mostly directed at state-owned commercial banks (excluding World Bank ASA) (figure C.11). Based on the revised sample, the World Bank lending and IFC investment and advisory focused on the commercial banking subsector (48 percent, 78 percent, and 52 percent of the interventions between FY08 and FY18, respectively), while all of MIGA’s were for development banks. Also, 21 percent of the World Bank lending interventions in the financial sector were for development banks, and 32 percent of IFC’s were for other nonbank financial institutions. The Bank Group provided support for the reform of state-owned banks and development banks through similar channels. Business and operations and corporate governance were the most common engagement areas in interventions for reforming both state-owned commercial banks and development banks. Support to risk management (18), product service improvement (16), human resources management (11), and law regulation (10) were the most common mechanisms for commercial banks, while assistance for improving product service (7), human resource management (6), and risk management (6) were most common for development banks. Support for improving ownership was more common for commercial banks than for development banks. Nine interventions involved privatization of commercial banks while only one intervention sought to privatize a development bank. Geographically, support for reform of state- owned commercial banks was spread out across income levels yet focused on Europe and Central Asia, East Asia and Pacific, and Sub-Saharan Africa countries. On the other hand, support for state-owned development banks was more focused on upper-middle-income countries, especially in especially in Latin America and the Caribbean. 159 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.11. Share of Financial Subsector Interventions, FY08–18 (Sample) Note: The figure is based on 265 interventions; excluded 5 for which these subsectors did not apply. ASA = Advisory Services and Analytics; FY = fiscal year; IFC AS Source: Independent Evaluation Group portfolio review and analysis. = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; NBFI = nonbank financial institution. Sample with no Advisory Services and Analytics. 160 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.12. Financing Projects by Financial Subsector FY08–18 (Sample) a. By income level b. By region Source: Independent Evaluation Group portfolio review and analysis. Note: Panel a is based on 258 interventions, excluding 5 for which these subsectors did not apply and 6 for regional projects. Panel b is based on 264 interventions, excluding 5 for which these subsectors did not apply. EAP = East Asia and Pacific; ECA = Europe and Central Asia; FY = fiscal year; H = high income; L = low income; LAC = Latin America and the Caribbean; LM = lower-middle income; MNA = Middle East and North Africa; NBFI = nonbank financial institution; RGN = regional; SAR = South Asia; SSA = Sub- Saharan Africa; UM = upper-middle income. 161 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors SOE Reform through National Level Interventions Support for SOE reform at national level involved 59 projects and $3.1 billion in SOE reform estimated commitments. It was led by the World Bank’s Macroeconomics, Trade, and Investment GP with 38 projects and an estimated SOE reform commitment value of $2.1 billion. For example, a project approved by this GP (a development policy loan for Ukraine in FY08, P096389) sought to improve governance in SOEs to create fiscal space for growth through strengthened public finances and public sector reform. The development policy loan required the enacted law on the management of SOEs to set an appropriate framework for dealing with management and governance concerns. Another GP that supported SOE reform at the national level was Finance, Competitiveness, and Innovation, which in the revised sample had approved 10 projects accounting for $489 million in estimated SOE reform commitments. For example, a development policy loan approved for Serbia in FY09 (Second Programmatic Private Financial Sector Development, P096711) sought to strengthen financial discipline through the reform of SOEs. The program included actions for selling socially owned enterprises (SOE ownership reform). Finally, the most frequent mechanism used at national-level interventions was privatization (29 interventions), followed by finance reporting (22), oversight of SOEs (14), and laws and regulations (10). 162 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.7. National Level SOE Reform Portfolio, FY08–18 Summary (Reviewed Sample) Business & SOE Regulatory Corporate Financial Fiscal Grand Practice Name Operations Ownership Frameworks Governance Mgt E&S Policy Total World Macroeconomics, Trade, 3 7 2 39 1 9 61 Bank and Investment 80.0 256.8 36.2 1,500.9 0.7 238.3 2,113.0 lending Finance, 10 4 11 2 1 28 Competitiveness, and 231.6 10.3 37.0 200.0 10.2 489.1 Innovation Other 8 2 7 2 2 21 104.2 17.3 64.8 200.0 137.9 524.3 Total interventions (no.) 3 25 8 57 1 4 12 110 Total estimated amount 80.0 592.7 63.9 1,602.7 0.7 400.0 386.4 3,126.4 ($, millions) IFC AS Cross-Cutting Advisory 1 1 solutions 0.2 0.2 Environment, Social, and 1 1 Governance 0.8 0.8 Total # of interventions 1.0 1.0 2.0 163 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Business & SOE Regulatory Corporate Financial Fiscal Grand Practice Name Operations Ownership Frameworks Governance Mgt E&S Policy Total Total estimated amount, 1.0 $, millions Total financing interventions (no.) 3 26 8 58 1 4 12 112 Total estimated amount ($, 80.0 592.7 63.9 1,602.7 0.7 400.0 386.4 3,126.4 millions) World Finance & Markets 12 4 19 2 3 40 Bank 0.4 0.1 0.7 0.01 0.1 1.4 ASA Other 2 2 3 13 3 23 0.4 0.06 2.49 0.50 0.11 3.6 Total interventions (no.) 2 14 7 32 2 6 63 Total estimated amount 0.4 0.5 2.6 1.2 0.01 0.2 4.9 ($, millions) Source: Independent Evaluation Group portfolio review and analysis. Note: SOE reform estimated commitments were either identified or calculated for each subproject intervention and then aggregated by engagement area. ASA = Advisory Services and Analytics; E&S = environmental and social; FY = fiscal year; IFC AS = International Finance Corporation advisory services; mgt = management; SOE = state-owned enterprise. (no.) of interventions and commitment levels in millions of US dollars. 164 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Effectiveness This section analyzes the achievement of Bank Group SOE reform interventions. To yield more information, in addition of the projects evaluated and approved between FY08 and FY18, the team also analyzed projects evaluated in this period but approved on or after FY02 and identified as having a potential SOE reform engagement in the financial and energy sectors. The team reviewed 507 evaluated financing projects, of which 294 had a SOE reform intervention (58 percent) and assessed the achievement of project objectives (with IEG-validated ratings and data available at the individual intervention level 20). The final portfolio included 294 evaluated projects (245 World Bank lending projects, 22 IFC investments, 16 IFC advisory services projects, and 2 MIGA guarantees), which provided 691 interventions. However, for 9 interventions no relevant data were provided on their effectiveness; hence, the denominator for the calculations reflected in the figures below is 682. 165 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.8. Distribution of Evaluated SOE Reform Projects and Interventions by Institution SOE Reform Interventions (no.) SOE Reform Projects Total Total Energy Finance National (no.) (%) (no.) (%) World 300 217 104 621 90.0 254 86 Bank lending IFC IS 12 31 43 6.2 22 7 IFC AS 10 14 24 3.5 16 5 MIGA 3 1 3 0.3 2 1 Total (no.) 324 263 104 691 100.0 294 100 (%) 47 38 15 Source: Independent Evaluation Group portfolio review and analysis. Note: Nine World Bank lending interventions (six for the energy sector and three for the financial sector) have no data about their effectiveness. IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. By Institution Most Bank Group financing projects supporting SOE reforms in the financial and energy sectors achieved their development objectives. Within the World Bank, 78 percent of projects (194 out of 248) successfully achieved their outcomes. 21 This is above the 75 percent FY17 World Bank corporate scorecard target. Effectiveness was different across lending instrument type. Policy support operations were more successful, with 85 percent achieving development outcomes (above the scorecard target), compared with 67 percent successes for Investment lending operations (below the scorecard target). The share of successful IFC SOE reform investment at 73 percent was above the FY17 target of 65 percent, but advisory work projects were below the target at 56 percent. The three MIGA guarantees for which there are 166 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors information achieved their development outcomes. At the intervention level, the Bank Group SOE reform financing interventions had a success rate of 72 percent (that is, they achieved or mostly achieved their intervention outcomes). Within the World Bank, 72 percent of interventions (439 out of 612) successfully achieved their outcomes, suggesting they performed relatively less well than the projects they were part of. Yet effectiveness was different across lending instrument type. Policy support interventions were more successful, with 74 percent achieving outcomes, as compared with 68 percent success for investment lending operations. The share of successful IFC SOE investment and advisory work interventions was 69 percent, and the three evaluated MIGA guarantees achieved their intervention outcomes. Figure C.13. Success Rate by Institution (percent above the line) a. Success rate of SOE reform projects 167 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors b. Success rate of SOE reform interventions Source: Independent Evaluation Group portfolio review and analysis; World Bank Corporate Scorecard (updated to October 2017). Note: Panel a, N= 286 projects, excludes 6 projects (World Bank lending) for which outcome ratings were not available, and three MIGA guarantees that achieved their intervention outcomes. The orange dots show FY17 corporate satisfactory outcomes targets. IFC updated its scorecard in November 2019 and eliminated its corporate success target of 65 percent. The project was defined “above the line” if achieving or mostly achieving project outcomes. Panel b, N = 679 interventions; excludes three MIGA downstream guarantees that achieved their intervention outcomes. DPO = development policy operation; FY = fiscal year; IEG = Independent Evaluation Group; IFC = International Finance Corporation; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; IPF = investment project financing; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. Bank Group evaluated by Independent Evaluation Group. By Country Income Level and Region The intervention success rate was higher for lower-middle-income countries (78 percent), followed by low (68 percent), upper-middle (67 percent), and high-income countries (50 percent) (figure C.14). For lower-middle-income countries downstream reforms were more successful than upstream reforms (80 percent versus 75 percent), while for low, upper-middle, and high- income countries upstream reforms were more successful. Across regions, SOE reform interventions in Latin America and the Caribbean, Europe and Central Asia, and East Asia and Pacific appear to be the most successful with 81 percent, 75 percent and 74 percent of success rate, respectively. Downstream SOE reform interventions were more successful in Latin 168 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors America and the Caribbean, East Asia and Pacific, and South Asia, while upstream reforms were most successful in Europe and Central Asia and in Middle East and North Africa. Figure C.14. Success Rate of SOE Reform Interventions by Income Level and Region (percent above the line) a. By income level b. By Region Source: Independent Evaluation Group portfolio review and analysis. Note: Panel a is based on 663 interventions (excludes 19 interventions part of regional projects or unclassified countries/territories), 299 upstream interventions (excludes 4), and 365 downstream interventions (excludes 15). Panel b is based on 676 interventions (excludes 6 interventions part of regional projects), 303 upstream interventions, and 374 downstream interventions. One intervention was classified as upstream and downstream reform. EAP = East Asia and Pacific; ECA = Europe and Central Asia; IEG = Independent Evaluation Group; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; SAR = South Asia; SOE = state-owned enterprise; SSA = Sub-Saharan Africa. 169 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors By Engagement Area Across engagement areas, environmental and social SOE reform interventions appear to be relatively less successful. Upstream corporate governance reform interventions were more successful than the other upstream engagement areas, while financial management was the most successful engagement area across the downstream reform areas. Across institutions, effectiveness was slightly different across engagement areas. World Bank lending SOE reform interventions achieved a better intervention outcome rating than the average in financial management (44 out of 58 intervention achieved or mostly its intervention outcome) and corporate governance (97 out of 129). IFC performed relatively better on corporate governance and SOE ownership (88 percent and 79 percent, respectively) than World Bank lending interventions (75 percent and 70 percent, respectively). But IFC had a lower average success rate for its intervention outcomes in business and operations (9 of 17) and financial management (4 of 7). 22 In general, projects that had interventions in more than one engagement area tended to have a greater achievement rate. For example, operations that had interventions in more than two engagement areas were more successful, with more than 85 percent achieving intervention outcomes, compared with 75 percent success for operations with interventions in one engagement area. 170 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.9. Success Rate by Engagement Area and Institution (percent above the line) World Bank Lending IFC IS IFC AS MIGA Success Success Success Success (no.) (%) (no.) (%) (no.) (%) (no.) (%) Business 119 72 9 56 8 50 and operations Corporate 129 75 6 83 2 100 governanc e SOE 120 70 14 86 10 70 3 100% ownership Financial 58 76 6 50 1 100 mgt E&S 21 67 8 63 1 0 aspects Regulatory 143 69 2 100 framework s PFM and 22 68 fiscal policy 612 72 43 70 24 67% 3 100% Source: Independent Evaluation Group portfolio review and analysis. Note: E&S = environmental and social; IEG = Independent Evaluation Group; IFS AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; mgt = management; MIGA = Multilateral Investment Guarantee Agency; PFM = public financial management; SOE = state-owned enterprise. Evaluated by the Independent Evaluation Group. The most significant factors 23 that positively affected World Bank lending projects that include interventions for improving SOEs financial management were client commitment (mentioned 19 times), 24 collaboration with external 171 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors actors or donors (18), positive project design (15), 25 and analytical work that accompanied project design or implementation (14). Although for interventions that sought to improve SOE corporate governance, the factors were client commitment (38 times), analytical work (28), collaboration with external actors or donors (26), and project design (18). In the case of IFC IS: i. Positive factors for the successful implementation of projects with SOE ownership interventions were client commitment (three times), choice of instrument, supervision, collaboration between World Bank institutions and with external actors/donors (two times each), and for corporate governance interventions were collaboration between IFC investment and advisory to add value to the client and collaboration with external actors (from the eight positive factors listed for those projects, each one was mentioned two times). ii. Factors that negatively affected the implementation of business and operations interventions were lack of an adequate project design (24135, 22418, and 26027) and external shocks (24316 and 24135).26 Absence of full client commitment (3) and inadequate project design (2) were factors that negatively affected the implementation of projects with financial management interventions. For example, the Banco de Finanzas (26027) loan had over-optimistic growth targets and the portfolio review did not identify completed vulnerabilities (EVNote, 9). In the case of IFC advisory, projects with SOE ownership interventions were positively influenced by project designs that adapted to the reality in the field and clearly defined objectives and deliverables (2 times identified as a positive factor) and by team composition (2). Advisory services projects to business and operations were affected mainly by lack of client commitment (3) and project design (2) with weak causal chains that make it difficult to measure the changes or that did not take into consideration the time needed to perform the activities. 172 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors By SOE Reform Type Figure C.15. Success Rate of SOE Reform Interventions by Type and Instrument Sourrce: Independent Evaluation Group portfolio review and analysis. Note: N = 147 SOE ownership evaluated interventions (73 DPO, 47 IPF, 14 IFC IS, 10 IFC AS, and 3 MIGA); 137 corporate governance evaluated interventions (88 DPO, 41 IPF, 6 IFC IS, and 2 IFC AS); 136 business and operations evaluated interventions (30 DPO, 89 IPF, 9 IFC IS, and 2 IFC AS); 22 fiscal policy evaluated interventions (19 DPO and 3 IPF); and 244 competition and regulation evaluated interventions (129 DPO, 94 IPF, 15 IFC IS, and 6 IFC AS). * denotes n < 5 interventions. DPO = development policy operation; IEG = Independent Evaluation Group; IFC AS = International Finance Corporation advisory servies; IFC IS = International Finance Corporation investment services; IPF = investment project financing; MIGA = Multilateral Investment Guarantee Agency; PFM = public financial management; SOE = state-owned enterprise. Evaluated by the Independent Evaluation Group. Across lending instruments, effectiveness was different across intervention types. DPOs achieved a better intervention outcome rating in SOE reforms of corporate governance (78 percent) and competition and regulation (74 percent), while IPF interventions were more successful in business and operations reforms (74 percent). Also, IFC investment and advisory performed well in corporate governance and SOE ownership but faced difficulties achieving success in interventions that sought to reform SOEs’ business and operations. In the energy sector, interventions that sought to support fiscal policies were more successful (78 percent) but low in number, while those that sought to reform SOEs’ corporate governance in the financial sector were more successful (85 percent). 173 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Table C.10. Reform Types Success Rate by Institution and Sector (percent above the line) Overall DPO IPF IFC IS IFC AS Energy Financial Suc- Suc- Suc- Suc- Suc- Suc- Suc- cess cess cess cess cess cess cess (no.) (%) (no.) (%) (no.) (%) (no.) (%) (no.) (%) (no.) (%) (no.) (%) SOE 147 72 73 71 47 68 14 86 10 70 57 68 61 75 ownership Corporate 137 76 88 78 41 68 6 83 2 100 42 62 60 85 governance Business and 136 70 30 67 89 74 9 56 8 50 79 71 51 71 operations PFM and 22 68 19 68 3 67 9 78 5 0 fiscal policy Competition 244 68 129 74 94 63 15 47 6 83 113 65 92 72 and regulation Source: Independent Evaluation Group portfolio review and analysis. Note: DPO = development policy operation; IEG = Independent Evaluation Group; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; IPF = investment project financing; PFM = public financial management; SOE = state-owned enterprise. Evaluated by the Independent Evaluation Group. 174 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors By Sector Most evaluated SOE reform interventions are in the energy sector (47 percent) as opposed to the financial sector (38 percent) or at the national level (15 percent). Interventions for reforming SOEs’ at the national level and financial sector were slightly more successful than interventions in the energy sector (+73 percent versus 69 percent). In the financial sector, upstream and downstream interventions are equally successful with 73 percent rated as achieved or mostly achieved. In the energy sector, upstream interventions were more successful than downstream interventions (73 percent versus 67 percent), while at the national level downstream reforms were more successful (76 percent versus 74 percent). In the financial sector, the most successful interventions were in lower and upper-middle-income countries, while in the energy sector they were in high (but the number of evaluated interventions is relatively small) and lower- middle-income countries. SOE reform interventions for the energy sector in upper-middle and for the financial sector in high-income countries underperformed due to the lack of client commitment or inadequate project design. For example, most of the electricity sector report supported by the Lebanon Reform Implementation Development Policy Loan (P094288) have not been initiated or achieved due to the lack of government agreement with the project objectives and implementation plan. Also project design did not take into consideration how it might agitate vested interests and sets ambitious targets (World Bank 2010, 17). Overall, countries that had more than one intervention had on average a greater achievement rate (+71 versus 67 percent), but this is not true for countries with two interventions for the energy sector. 175 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.16. Success Rate by Sectors (percent above the line) a. By sector b. By income level and sector Source: Independent Evaluation Group portfolio review and analysis. Note: Panel a is based on 682 interventions (318 for the energy sector, 260 for the financial sector, and 104 at national level), 303 upstream interventions (128, 91, and 84), and 380 downstream interventions (190, 169, and 21). Panel b is based on 315 interventions for the energy sector (excludes 3 interventions as part of regional projects or unclassified countries or territories), 248 for the financial sector (excludes 12 interventions), and 100 interventions at national level (excludes 4 interventions). One intervention was classified as upstream and downstream reform. * denotes n <5 interventions. IEG = Independent Evaluation Group; SOE = state-owned enterprise. Evaluated by the Independent Evaluation Group. 176 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Energy Subsectors SOE reform energy sector interventions were most successful in macro-fiscal policies (78 percent) and financial management (77 percent) and least successful in the areas of environmental and social aspects (63 percent), corporate governance (62 percent), and ownership (68 percent). SOE reforms dealing with the subsector of transmission and distribution were the most effective (82 percent), followed by those dealing with the extractives industries (72 percent) and with the power subsector (62 percent). The factors that most affected results in power sector interventions were project designs, external shocks, and public sector institutional capacity, while projects with interventions for extractives industries were mostly affected by external factors such as external shocks, public sector institutional capacity, and client commitment. Within the power sector, effectiveness was different across source of energy. Reforms dealing with renewables were more successful than nonrenewables (67 percent versus 58 percent). Also, for this subsector, IFC has the highest success rate of the institutions (88 percent of the IFC IS interventions achieved their outcomes, and 78 percent of the IFC AS interventions had similar outcome ratings), while the World Bank has the lowest success rate, being interventions for DPOs more successful than those for investment projects (63 percent versus 57 percent). Across engagement areas, interventions involving power sector financial management reforms achieved their targets 77 percent of the time, while less than half of those involving corporate governance had similar results. 177 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.17. Success Rate of Energy Sector Interventions a. Energy sector by engagement area b. By energy subsector Source: Independent Evaluation Group portfolio review and analysis. Note: Panel a is based on 318 interventions, and panel b on 311 (excludes 6 interventions for which these subsectors did not apply). In panel b, 4 interventions are double counted as they address both power and extractive subsectors. E&S = environmental and social; IEG = Independent Evaluation Group; mgt = management; PFM = public financial management; SOE = state-owned enterprise. Evaluated by the Independent Evaluation Group. 178 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.18. Success Rate of Power Subsector Interventions a. By kind of resource b. By instrument 179 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors c. By engagement area Source: Independent Evaluation Group portfolio review and analysis. Note: N = 180 interventions; 4 interventions address both power and extractive subsectors. * denotes n < 5 interventions. DPO = development policy operation; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; IPF = investment project financing; Mgt = management; PFM = public financial management; SOE = state-owned enterprise. World Bank Group interventions evaluated by the Independent Evaluation Group. Financial Subsectors In the financial sector, the commercial bank SOE reform interventions success rate was below the average sector success rate (69 percent versus 73 percent), while for development banking it was above the average (77 percent versus 73 percent). But, in the case of development banking, downstream interventions are most successful with 83 percent rated as achieved or mostly achieved, compared with upstream interventions (55 percent). In general, projects with interventions for commercial banks fell short on project design and external factors as shocks and client commitments also affected their success. While client commitment and analytical work were the most important positive factors that influenced the success of projects with interventions for development banks. Among nonbanking financial institutions, the Bank Group was more successful in supporting sovereign wealth fund (all evaluated interventions were successful, but the number is relatively small) and insurance reforms (92 percent) and was less successful in supporting pensions funds reforms 180 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors (60 percent). With regard to commercial banks, DPOs were the most successful instrument (74 percent), followed by IPF (61 percent) and IFC IS (62 percent), while IFC advisory operations have the lowest success rate (57 percent). Also, the Bank Group commercial banks operations’ strongest success rate was related to corporate governance reforms (83 percent) and ownership reforms (74 percent) and the lowest was with regard to environmental and social aspects (33 percent). Regarding development bank SOE reforms, IFC just had two advisory interventions, which were successful. Withing the World Bank, IPF was the most successful instrument, with 90 percent of interventions achiveing their outcomes, but DPOs underperformed: just half of them achieved its outcomes. In terms of engagement areas, corporate governance (83 percent) and business and operation (83 percent) reforms were more successful than ownership (50 percent) and regulatory framework (57 percent). Figure C.19. Successful Financial Sector Interventions Source: Independent Evaluation Group portfolio review and analysis. Note: Interventions are double counted when they address more than one subsector. Figure excludes 4 interventions for which these subsectors did not apply. * denotes n < 5 interventions. NBFI = nonbank financial institution. World Bank Group interventions evaluated by the Independent Evaluation Group. 181 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Learning about Factors Affecting Success or Failure IEG identified 857 factors that affected the success or failure of SOE reform development outcomes. 27 These factors could be identified at project level (not at intervention level). Additionally, they were tagged with the direction of their effect, whether it was positive (417), negative (337), or mixed (102), 28 and one factor without information about direction. The most common factors that facilitate or constrain the achievement of development outcomes were client commitment (18 percent of factors) and project design (17 percent of factors). Other important contributing external factors included collaboration with external actors, agency coordination or political economy, public sector institutional capacity, and external shocks. Beyond design, other internal factors (under Bank Group control) include the accompanying analytic work, a good monitoring and evaluation (M&E) framework, and good supervision. Among institutions, for IFC IS another important factor was supervision and for IFC AS agency coordination and collaboration with external actors were important. Looking at the five reform types, certain factors are important for all of them, such as design quality. Other factors vary substantially. For example, collaboration with external actors is more likely to be important in public financial management and corporate governance reforms, while agency coordination and political economy is relatively more important for privatization and ownership reform. Business and operational reforms are relatively more sensitive to issues of supervision. Client commitment is a more frequent factor in both public financial management and corporate governance reform than for other reform types. 182 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.20. External and Internal Factors Affecting Outcomes in Evaluated Projects, FY08–18 . Source: Independent Evaluation Group portfolio review and analysis. Note: Distribution of factors (vertical bars denote 95 percent confidence intervals)The figure shows the most frequent factors (800) identified for the 294 evaluated projects;, excludes team composition (which was identified 21 times), sequenced operations (14), collaboration among World Bank institutions (13), privte sector institutional capacity (7), and funding (2) factors. Projects may be double counted since they can have multiple factors that affected their outcomes. FY = fiscal year; Inst. = institutional; M&E = monitoring and evaluation; Pub. = public. 183 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.21. Factors of Success/Failure in SOE Reform Project Evaluations by Five Reform Types Source: Independent Evaluation Group portfolio review and analysis. Note: The figure shows the distrubution of factors identified for the five types of SOE reform evaluated projects, business and operations (336 factors), competition and regulation (480), SOE ownership (359), corporate governance (319), and PFM/fiscal policy (71). Projects may be double counted since they can have multiple factors that affected their outcomes. Inst. = institutional; M&E = monitoring and evaluation; PFM = public financial management; Priv. = private; Pub. = public; SOE = state-owned enterprise. There are some variations to observe regarding the direction in which the factors affect project outcomes in the sectors analyzed.  Client commitment is key for supporting the implementation of reforms regardless of the sector. Collaboration with external actors and analytical work were also important factors that positively influenced the success of SOE reform projects, especially for reforms at national level. For the energy sector, analytical work was the second most important factor, while for the financial sector it was collaboration with external actors. 184 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors  Design, external shocks, public sector institutional capacity, and M&E framework negatively affected the success rate. External shocks and public sector institutional capacity especially affected SOE reform projects in the energy sector and at national level.  Project design is an important factor that could affect implementation success of SOE reform projects either positively or negatively. Among regions, project design was the second most important positive factor in South Asia and in Middle East and North Africa, and the third most important in East Asia and Pacific and in Latin America and the Carribean. Coordination complexities among agencies and political economy were more pronounced in Middle East and North Africa, Europe and Central Asia, and East Asia and Pacific Regions, while evaluation (M&E) framework was a prominent negative factor in Latin America and the Caribbean and in Europe and Central Asia. Project outcomes in Latin America and the Caribbean were most affected by external shocks. In general, among engagement areas, the factors that positively influenced SOE reform projects’ outcomes were client’s commitment and collaboration with external actors, except for projects that sought to review environmental and social aspects and to improve SOEs’ business and operations. For operations on those engagement areas, supervision and project design were important factors that positively influenced their outcomes. In the case of operations that aimed to improve SOEs’ corporate governance, regulatory framework, and macro-fiscal situation, another positive factor was the analytical work. On the other hand, in addition to the project design, operations outcomes on regulatory frameworks, business and operations, SOE ownership, and corporate governance were affected by the lack of an adequate M&E framework. The capacity of public sector institutions was an important negative factor for operations that aim to improve SOEs’ financial management. 185 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.22. Direction of the Factors That Affect Outcomes in Evaluated Projects a. Distribution of positive factors by sector b. Distribution of negative factors by sector Source: Independent Evaluation Group portfolio review analysis. Note: Panel a is based on 417 factors that positively affect success of SOE reform projects, and panel b on 337 factors that negatively affect success of SOE reform projects. Factors may be double counted if projects have interventions in more than one sector. Pub. = public; Inst. = institutional; M&E = monitoring and evaluation. 186 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors These findings are exemplified in portfolio analysis and the evaluation country cases of studies. Internal Factors  Project design o (+, mentioned 54 times) Projects that identified design as a positive factor had simple, selective, flexible designs or a design that suits the country’s situation. Designs were also based on project-related studies and analytical work, which took into consideration previous findings and recommendations and interactions with stakeholders. For example, in Vietnam, the project’s design of the Payment System and Bank Modernization 2 considered the World Bank’s experience in other financial infrastructure development projects and followed a flexible approach in the overall solution, a design that contributed to the success of the project. In Kenya, the World Bank sequentially and to some extent simultaneously addressed the major institutions in the electricity sector, building capacity in all of them. Electricity SOE reform projects also included stakeholder consultation of industrial (but not of household consumers) in project preparation. o (–, 67) On the contrary, projects for which this factor affected their implementation, had a complex design with ambitious objectives, addressing multisectoral needs, multiple components and/or several implementation agencies. Also, design that did not take into consideration the government’s institutional capacity or the resources allocated. For example, in Ukraine, one important factor that negatively affected the success of development policy loan II (FY12) and its SOE reforms was its overly flexible and complex design. Given the lack of sustained commitment, it may have been preferable to focus on specific reforms with a different instrument design to have greater impact and success. In Kenya, the KEPTAP 187 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors investment project (FY15) was overly complex, basing implementation on 21 executing agencies, three project implementation units, and too many components, resulting in only a 19 percent disbursement rate by the time of the midterm review.  Analytical work o (+, 59) Operations that highlighted analytical work as a positive factor benefited from it either during design or implementation, through the provision of technical assistance or analytical and advisory activities to support government counterparts or to build client capacity. The Bank Group also used analytical work for ensuring continuity in its engagement. For example, in Bangladesh after 2009, the Bank Group dealt with lost government buy-in and commitment to reform by focusing its engagement on analytical work to diagnose state-owned commercial banks and financial sector stability. In Ukraine, due to lack of progress in adopting a sustainable macroeconomic framework and structural reforms over FY10–14, the World Bank opted for technical assistance that identified and tracked progress on utility sector and public sector governance reform, including the privatization of state-owned banks. In Kenya, the World Bank not only supported the government’s program to increase access to electricity by investing in geothermal generation and transmission infrastructure but also provided continuous capacity-building assistance to sector utilities to support their financial sustainability.  M&E framework o (+, 16) Project aspects highlighted on projects with good M&E frameworks were a clear statement of objectives and indicators that captured their achievement. Projects with well-specified policy actions and clear monitorable outcome indicators, baseline and target values, and sources of information for tracking progress. For 188 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors example, in Zambia the indicators for increased access to electricity services project (P077452, FY08) were appropriately linked to the objectives and properly designed to monitor progress toward the project development objectives. The M&E framework was also used to inform project progress and aided project refinement during implementation (World Bank 2016b). o (–, 39) On the other hand, projects with weak M&E frameworks did not incorporate all relevant indicators for tracking progress, had inconsistent indicators, lack of baseline data or a clear relationship between project activities and outcome indicators, and outcomes that are not attributable to the project. For example, in China, the Economic Reform Implementation project (P085124, FY06) lacked well-defined and measurable outcome indicators that could facilitate the assessment of the project. Each of the subprojects had their own result framework with mostly output indicators, which made it difficult to assess overall achievement of the project.  Team composition o (+, 12) Positive aspects of team composition were strong local presence, team with the necessary skills or that benefited from technical guidance on complex process, and continuity in project teams and supervision. For example, in Ethiopia the first Poverty Reduction Support Operation (P074014) had an early and continuous engagement with the borrower by a multidisciplinary project team. The project team included experts from different thematic groups, engaged early, and had close interaction with their counterparts, which allowed prompt follow up of issues (ICR, 42). o (–, 7) In contrast, projects that identified this as a negative factor had a high turnover of task team leaders, lack of adequate expertise in project teams and a late establishment of the project implementation unit. For example, in Mali, the quality of the 189 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors supervision for the Growth Support Project (P080935, FY05) was affected by the turnover of task teams and that three of the four task team leaders were located in Washington instead of being in the field (ICRR, 7).  Choice of instrument o (+, 13) Projects that were positively influenced had instruments that responded to the country’s needs, a strategic choice of the instrument, and combined financing and technical assistance. For example, the First Programmatic Reform Implementation (P083927) in Uruguay used a development policy loan to deepen the relationship with the borrower. Also, aligned the priorities with the government agenda and tailored the program to suit local conditions (ICR, 30). o (–, 6) On the contrary, an inappropriate choice of instrument, and the use of a country-specific instrument in a project with regional implications (in the case of P131027), negatively affected the success of SOE reform projects. In addition, one of the projects mentioned that the World Bank did not have a well-defined range of instruments for crisis lending (P116020).  Collaboration between the Bank Group institutions in SOEs reform have not been fully seized, despite the strong potential. 29 o (+,9) Positive features include collaboration between IFC and the World Bank or between investment and advisory services or the use of analytical reports to add value to the client. The country case studies found some but limited levels of collaboration between Bank Group institutions (only four of eight had some collaboration between institutions). Although it played an important positive factor of reform where it occurred. For example, an important feature of the Kenya portfolio was the Bank Group’s Risk Mitigation 190 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Package to attract independent power producers, supported by International Development Association partial risk guarantees, MIGA guarantees, and IFC investment loans. 30 The combination of instruments and IFC’s leadership in establishing a consortium of other financiers raised the comfort level of investors and attracted independent power producers. o (-, 4) Projects that identified this as a negative factor had inadequate coordination between their components or to manage the transition within the World Bank teams or inadequate information sharing between institutions. For example, the Implementation Completion and Results Report Review of the Growth Support Project in Mali (P080935) reported that the collaboration between the World Bank and IFC staff was at time hampered by inadequate information sharing (7). External Factors  Client or government commitment o (+, 93) Client commitment was the most important factor behind the success of SOE reform operations. For example, for the energy sector the government in Kenya showed substantial willingness to continue the unbundling process and to build stronger institutions, more independent SOEs, and well-qualified leadership, policies closely following Bank Group advice. Government reliably came in to reassure private investors and/or restore solvency to SOEs (IEG SOE Kenya Country Case Study). Also, in Vietnam the government’s commitment to expand rural electrification was one of the most significant positive factors to the success of interventions. The demonstrated ownership of the government of Mauritius in addressing emerging issues confronted by the Private Sector Competitiveness Development Policy Loan (P126903) is another 191 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors example. Government constantly engages with stakeholders and was transparent in carrying out its fiduciary functions. o (–, 34) But lack of commitment was also an important factor that constrains the effectiveness of operations. For example, in Ukraine, government willingness for reform was more motivated by difficult conditions created by financial crises. Only after the 2014 crisis was the new recapitalization framework (supported previously by the World Bank) implemented. In Bangladesh, state-owned bank performance was seen as an essential factor in maintaining good macroeconomic and fiscal management until 2009, but was subsequently not seen as urgent, in part due to a healthier fiscal situation.  Donor collaboration o (+, 61) Positive aspects of collaboration among donors were a strong coordination in ensuring reforms were implemented, building consensus about the reform agenda, and ensuring synergies and effective support. Also, other projects benefited from complementary analytic and advisory activities and technical assistance from other donors, which helped build consensus. For example, in Ukraine, the 2014 crisis triggered the importance of donor collaboration and assistance in SOE reform. At this time, donors started working together toward a unified set of policies. Donors and international financial institutions met regularly and prepared joint letters reflecting a cohesive position on policy reform. In the energy sector, IFIs are regarded as the “ambassadors of corporate governance” by sector representatives. However, most recently the position of the IFIs on the unbundling of the gas sector became more fragmented. 192 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors  External shocks o (–, 46) Projects that highlighted this as an important factor mostly were affected by economic downturn, crisis, challenging political situation, negative external environment, or price volatility and changes in the government priorities. For example, in Egypt external shocks hindered the effectiveness of the World Bank interventions. The 2011 revolution halted engagement and the Bank Group had to establish itself as a trusted development partner with the new government and with a civil society that played a key role.  Political economy can also hinder the effectiveness of interventions. o (–, 16) Projects influenced negatively by this factor faced difficult political situations, changes of agenda, delays in the congressional approval process, changes in government, sharp inflation, vested interests, and confrontation between major political parties. Also, some of those projects had complicated management with multiple government stakeholders, coordination issues, overlaps among government ministries or agencies, and institutional conflicts. For example, the outcome for the IFC KBB SubDebt (31515, FY12) was affected by change of government agenda due to the unwillingness to proceed with the privatization agenda stabilized by the project. Political influences were constant, and the government was bureaucratic and politically sensitive and a difficult partner for day- to-day communication and management of the privatization process (Ev. Note, 13). Another example was the Financial Sector Technical Assistance Credit (P040177 FY03) for Honduras, whose implementation was delayed due to the congressional approval process taking one year; there were four changes in government and multiple changes in the leadership of the project, resulting in an uneven commitment to the project (ICR, 20). 193 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Factors That Influenced Outcomes Positively Client commitment and supervision were the most frequent factors that facilitated the achievement of development outcomes in countries with low- level corruption controls. Supervision was especially important; the frequency of this factor was 16 percent for projects with successful interventions in countries with low control of corruption versus 8 percent for projects with successful SOE reform interventions in countries with medium and high corruption control, and this difference is statistically significant at 5 percent. Client commitment was also a key positive factor that facilitated the implementation of SOE reforms, especially for projects in countries with a low level of corruption control (this factor was frequent at approximately 26 percent versus 22 percent), but the difference is not significant. Another important contributing factor was collaboration with external actors and donors, with its accompanying analytic work. Evidence suggests that analytical work, client commitment, and collaboration with external donors tend to influence the success of SOE reform projects more in upper-middle and high-income countries. Project design and supervision were more frequent success factors for low-income countries, although these differences by income are not statistically significant at the 5 percent level. 194 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Figure C.23. Positive Factors That Influenced Successful Outcomes (evaluated projects) a. Distribution by control of corruption level b. Distribution by country income level Source: Independent Evaluation Group portfolio review and analysis. Note: The figure shows the distribution of positive factors identified for projects with successful state- owned enterprise reform interventions by control of corruption and country income levels. Projects may be double counted since they can have multiple factors of success. Inst. = institutional; M&E = monitoring and evaluation; Pub. = public. 195 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors References World Bank. 2007. “Ethiopia - First and Second Poverty Reduction Support Operations Project.” Implementation Completion and Results Report, World Bank, Washington, DC. World Bank. 2010a. “Lebanon—Reform Implementation Development Policy Loan.” Implementation Completion and Results Report ICR00001237, World Bank, Washington, DC. World Bank. 2010b. “Uruguay—First and Second Programmatic Reform.” Implementation Completion and Results Report ICR1585, World Bank, Washington, DC. World Bank. 2013. “Honduras—Financial Sector Technical Assistance Credit.” Implementation Completion and Results Report ICR2224, World Bank, Washington, DC. World Bank. 2014. “Mali—Growth Support Project.” Implementation Completion and Results Report Review ICRR14443, World Bank, Washington, DC. World Bank. 2016a. XPSR Evaluative Note 2016 - KBB SubDebt. Independent Evaluation Group. Washington, DC: World Bank. World Bank. 2016b. Zambia—Increased Efficiency and Access to Elec SIL (FY08). Independent Evaluation Group. Washington, DC: World Bank. 196 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors Notes 5 These projects were the total of 664 World Bank lending, 160 International Finance Corporation (IFC) investment services, 103 IFC advisory services, and 25 Multilateral Investment Guarantee Agency (MIGA) projects. 6 Projects evaluated during the period but approved from fiscal year (FY)02 to FY07 were included to improve the data available for analyzing effectiveness and were used only for the analysis of achievements and factors. 7 Of the 26 projects, 8 are from Vietnam, 7 from Serbia, 7 from Ukraine including one Recipient Executed Activities (REA) (P151927), and 4 from Egypt including one project approved in FY19 (P168630).8 The team reviewed 87 percent of nonevaluated projects because the review process was time consuming due to the large size of the instrument designed to review the projects, so a decision was made to stop the process after a certain point. Additionally, no documents were available or accessible for a subset of IFC projects. 9 Part of the manual review involved classifying interventions by sector, superseding the original World Bank sector classification. National-level interventions may capture several sectors but were identified only when involving the energy or financial sectors. 10 Compared with the higher success rates reported in the approach paper, in the present case the rates reflect success in identifying state-owned enterprise (SOE) reform interventions only, instead of SOE engagement and SOE reform interventions. 11 Numbers in this section were calculated based on the projection of the reviewed sample. 12 Reform interventions fall into 9 EAs aiming either to improve SOE’s policy and institutional framework (upstream) or to strengthen them at the enterprise level (downstream). The 4 upstream areas are i) the regulatory framework, (ii) governance and accountability, (iii) ownership, (iv) macro fiscal policy. The 5 downstream areas are (i) business and operations, (ii) corporate governance, (iii) ownership, (iv) financial management, and (v) E&S aspects. 13 National level captures interventions that support several sectors and were identified only when involving energy or financial sectors. 14 An example of an operations approved by the Macroeconomics, Trade, and Investment Global Practice is a development policy operation for Burkina Faso in fiscal year (FY)09 197 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors (P099011) for, among other things, enhancing the regulatory framework for SOEs on electricity. The mechanisms were the establishment of a regulator responsible for maintaining a balance between actors and facilitating access to information, clarifying mandates of key sector institutions, and maintaining a clearer segmentation of producers, transporters, and distributors, including establishing a legislative framework for an affermage (PPP) of SONABEL, the power utility, and adoption of a more transparent tariff setting mechanism. 15 A project approved by the Energy and Extractives GP included an investment project financing in FY08 for Zambia (P077452) for the reinforcement of ZESCO’s (state-owned power utility) distribution, assisted in implementing management measures to mitigate the current power shortage, and provided capacity building of key sector institutions. 16 For example, a project approved to Umeme (a parastatal company) in Uganda in FY09 (25788) for improving its business and operations through the rehabilitation of the company’s distribution networks, and a project approved for the Philippines in FY08 to support the privatization of Magat Hydro Electric Power Plant (SOE ownership) (26041). 17 This Global Practice approved in India a project for improving the operations of the Infrastructure Finance Company (IIFCL, P102771—FY10) through the implementation of a market-oriented human resource strategy and providing training and support for the development of new products. Also, a project for improving the Development Bank of Central African States’ corporate governance (P099833—FY09) through establishing risk assessment and internal control practices, guidelines for auditors in line with international practices and financial reporting. 18 For example, this Global Practice approved a third Programmatic Fiscal Management & Competitiveness in Peru (P106720—FY10) for improving, among other things, FONIPREL’s (a sovereign fund for subnational investment) corporate governance through the evaluation of the implementation of the fund. Also, a development policy loan (Third Governance, Opportunities, and Jobs, P150950—FY16) in Tunisia for improving state-owned bank’s business and operations through organizational restructuring of its board of directors and a restructuring strategy for BNA (another public bank). 19 For example, this practice approved a project for Maldives (26089, FY09) for supporting the privatization of a government-owned housing finance institution and transforming it into a commercially viable, private sector–led company and playing a key role in providing housing 198 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors finance to low and middle-income households. Also, there was another project (26027) in Nicaragua to strengthen the financial management of Banco de Finanzas (an SOE) through helping it to strengthen its asset liability structure, improve its access to and the cost of alternative funding sources, and help it to implement best practices in micro, small, and medium enterprises. 20 Only intermediate-level outcomes were assessed. A four-point rating system was used to evaluate achievement. A score of “achieved,” “mostly achieved,” “mostly not achieved,” or “not achieved” was given to each intervention. This scoring system allowed mapping to the Bank Group’s rating systems, which allowed comparisons across the rating systems of different institutions and product lines. 21 The analysis excludes seven projects for which development outcomes are not available. Successful achievement means a development cutcome rating of moderately satisfactory or above. Overall project outcome ratings are based on the overall performance of the whole project, while the intervention level analyzes the achievement of Bank Group SOE reform interventions. 22 In the case of International Finance Corporation (IFC) investment servies, most of the SOE ownership and corporate governance interventions achieved their outcomes but fell short on business and operations and financial management interventions. IFC’s advisory had an average success rate for interventions outcomes in SOE ownership of 70 percent, while the ones that sought to improve business and operations were successful half of the time. 23 Factors are reported at project level and cannot be directly associated with specific interventions. 24 Projects may be double counted if one factor affected the project several times. 25 Designs that were appropriate, aligned with county strategies and priorities, included interaction with stakeholders, influenced by previous findings and recommendations, and adapted to the countries’ political situation and capacities. 26 For example, the lack of progress on the structure of Guangzhou Development Group Incorporated (GDIH, 22418), justification of IFC’s support, was due to unfavorable Shanghai stock market conditions and that this objective was not mentioned in the loan documents or 199 Appendix C State-Owned Enterprise Reform Support: Financial and Energy Sectors an expression of interest. The latter may indicate that this objective may not have been realistic at the time of appraisal (EvNote, 3–4). 27 As in the effectiveness section, factors were identified only for projects evaluated by the Independent Evaluation Group on or before January 31, 2019. Sources of information to identify factors were the same as for assessing effectiveness. 28 Means that the factor supported implementation in some aspects of SOE reform but constrained it in others. 29 Collaboration among Bank Group institutions was rarely mentioned as a factor that facilitated or constrained the achievement of project development outcomes. The team also found limited collaboration between the World Bank Advisory Services and Analytics and lending windows (only 2 percent of the revised World Bank Advisory Services and Analytics operations have a parental World Bank lending project). 30 The Bank Group’s Risk Mitigation Package included the following projects: (1) P122671 “Kenya Private Sector Power Generation Support Project”—FY12; (2) 29801 IFC investment services “Thika IPP”—FY11; (3) 29418 IFC investment services “Gulf Power Ltd.”—FY12; (4) 9722 Multilateral Investment Guarantee Agency (MIGA) “Thika Power Ltd.”—FY12; (5) 9993 MIGA “Triumph Power Generating Company Limited”—FY13; and (6) 10646 MIGA “Gulf Power Limited”—FY14. 200 Appendix D. Econometric Analysis Introduction Country conditions are key predictors of the odds of state-owned enterprise (SOE) reform success. Specifically, a high level of control of corruption dramatically increases the likelihood of success over a low level of control. Deeper private credit markets are associated with a lower chance of success. Being in a low-income country is a significant but less robust predictor of the likelihood of success. The odds of SOE reform success are not significantly associated with the SOE reform sector or engagement area; however, they are positively associated with the overall volume of country SOE reform commitments. The joint movement of several project-level explanatory factors is strongly associated with the likelihood of SOE reform success. Principal components analysis indicates that the odds of SOE reform success are positively associated with: Strength of project implementation tools, where sound monitoring and evaluation (M&E) frameworks, adequate team composition, and close project supervision are components Elements for sustainable SOE reform engagement, where collaboration with external actors and partners, sequencing of projects, and relevant analytical work are components Analysis of individual factors finds a positive and robust association between strong M&E frameworks and the likelihood of reform success. However, external shocks (for example, crises) are negatively and robustly associated with the odds of success. There is additional, but less robust, evidence of a significant positive relationship between other factors and intervention odds of success, including collaboration with external partners, a programmatic 201 Appendix D Econometric Analysis nature of operations, and appropriate choice of the type of financing instrument. Approach The Independent Evaluation Group (IEG) sought to apply econometric methods to relate SOE reform intervention outcomes (of closed projects) to possible predictors of SOE reform success, including factors of success and country-level characteristics. The purpose was to reveal statistical relationships and triangulate them with findings from other methods conducted during the evaluation. Econometric methods were not used to evidence causal relationships between variables. Instead, the main evaluation combines findings from econometric analyses with those from country case studies, literature review, portfolio review and analysis, and sector and topic-specific deep dives to reach overarching conclusions. This appendix describes this econometric analysis and its main findings in terms of SOE reform success and its predictors. Their interpretation and the integration of these findings with findings from different methodological approaches are left for the main body of the evaluation. The main source of data for the econometric analysis was a database constructed for the portfolio review and analysis (based on a uniform template developed and implemented by the IEG SOE team) aligned with IEG-wide standards and guidance. This template was designed to collect information both at the project level and at the SOE reform or intervention level, depending on the dimension being analyzed. 31 All IEG-evaluated projects with World Bank–assigned codes for energy and financial sectors and tagged by the IEG SOE team as having SOE reform interventions were included in the econometric analysis. Based on hypotheses generated from case studies, deep dives, and descriptive statistics, IEG also gathered potential explanatory external variables from World Bank Group borrowing countries, including the World Bank’s Worldwide Governance Indicators (WGIs) for 2002–17, World Development Indicators (WDI) for 2002–18, 202 Appendix D Econometric Analysis and the International Monetary Fund’s World Economic Outlook (April 2019). After testing several models, the team selected WGI’s control of corruption and WDI’s domestic credit as a share of gross domestic product (GDP) as country-level predictors because they reflected the best model fit compared with other specifications. 32 On this basis, the main objective of the econometric analysis was to identify explanatory variables (or predictors, depending on the model specification) significantly (and robustly) associated with the success of SOE reform intervention outcomes. The main unit of analysis was the SOE reform intervention. The main outcome of interest was captured with a dummy variable signaling whether the SOE reform intervention was successful (above the line) or not (below the line). Note that to assign success or failure to project components, the portfolio analysis team assigned intervention- specific ratings. (A separate specification using a four-point rating of intervention success did not substantively change the findings.) Box D.1. Hypotheses Generated from IEG SOE Evaluation Case Study Workshop Internal Factors Continuous World Bank Group engagement with clients in both the energy and financial sectors is a positive factor of project success. Strong team composition in the form of competent task team leader, technical expertise, and strong rapport is a positive factor of project success. Team presence in the field (“field presence”)— with local language ability, local knowledge, and technical expertise – is a positive factor. Joint interventions by Bank Group institutions are a positive factor. Use of a combination of instruments and a mix of technical and financial assistance is a positive internal factor. Prior diagnosis/analysis through analytic work is a positive internal factor. 203 Appendix D Econometric Analysis Positive design features include flexible approach, sequential approach, and capacity building are positive factors. Negative design features include overly flexible approach, overly broad focus, and excess complexity. Strength of the project implementation unit (PIU) can be a positive factor, while weakness of the PIU is a negative factor. External Factors Vested interests and corruption undermine reform effectiveness and sustainability. Donor coordination is another key external factor for project success. External shocks plays a role in pushing reform forward or hindering it. Weak rule of law is a negative external factor. Lack of independence of the sector regulator is a negative external factor. Note: IEG = Independent Evaluation Group; SOE = state-owned enterprise. Based on hypotheses generated from case studies (box D.1), deep dives, and descriptive statistics, the data set produced by the team used several combinations of explanatory variables at the intervention level, including the type of SOE reform intervention (upstream or downstream), SOE reform engagement areas (ownership, business and operations, cooperate governance, environmental and social aspects, financial management, macro-fiscal policy, and regulatory frameworks), mechanisms through which such reforms were pursued (for example, SOE privatization), estimated commitment amounts per SOE reform intervention, and sector. 33 Explanatory variables at the project level included factors of implementation success, financing instrument, approval year dummies, the previously mentioned country-level controls from WGI and WDI, World Bank region, 34 and country-year income levels. Most predictors included in the regressions were dummy variables. 204 Appendix D Econometric Analysis Methodology Basic Model The resulting data set for analysis was a pooled cross-section of SOE reform project interventions. Because the dependent variable of interest took the form of a dichotomous “successful” or “unsuccessful” SOE reform intervention outcome (that is, y = 1 or 0), the estimation technique chosen was a logistic model, where the variable coefficients are estimated by maximum likelihood. 35 The functional form of the basic model is the following: ( = ) ( ) = � � 1 − ( = ) = 0 + ′ + ′ + ′ + ′ + (1) where yisjt is the SOE reform intervention outcome variable, taking the values of 1 if “successful” or 0 if “unsuccessful” for intervention i in project s in country j at approval year t; β0 is an intercept; is a vector of independent variables of interest that includes the type of SOE reform intervention (for example, upstream = 1, downstream = 0), individual engagement areas (for example, ownership reforms = 1, base engagement area = 0), estimated SOE reform intervention commitments (continuous variable), sector (for example, finance = 1, base sector = 0) and other intervention-level variables; is a vector of project-level variables, including individual factors of implementation success or failure and composite factors identified through principal component analysis (PCA), 36 financing type (for example, World Bank adjustment loan = 1, base instrument type = 0), and number of interventions per project; is a vector of country-level variables, including WGI’s control of corruption, WDI’s domestic credit to the private sector as a share of GDP, Region, income-level dummies, and projects per country-year; and is a vector of approval year groups capturing time-related external shocks. 205 Appendix D Econometric Analysis In the logistic model, logit denotes the natural logarithm of the odds, where the odds are defined as odds = p / (1 − p), where p = probability of the SOE reform intervention being successful. Thus logit = ln[p / (1 − p)]. As such, if both sides of equation (1) are elevated to the power e = 2.7183, the equation coefficients can be interpreted and reported as odds ratios. For example, one could test the hypothesis that the odds of success of SOE reform corporate governance interventions are higher than those of other engagement areas. 37 It is important to emphasize that the interpretations of the estimated odd ratios (odds of SOE reform success) across this exercise are applicable only to this sample and model specification. 38 Descriptive statistics of the SOE reform success dummy variable and its predictors suggest that there is enough variation in most variables to justify including them in the regression analysis: almost 80 percent of the variables show coefficients of variation above 1 (equal to the standard deviation divided by the mean.) More than 70 percent of SOE reform interventions were successful, and most of the individual factors occurred for somewhere between 11 percent and 23 percent of projects. Lower-middle-income countries accounted for 42 percent of country-years, followed closely by low-income countries with 37 percent. Additionally, 91 percent of SOE reform interventions came from World Bank financing projects (see table D.1 for variable definitions). Defining Individual Factors of Success The team individually identified factors associated with SOE reform outcomes for projects evaluated by IEG on or before January 31, 2019. Information on factors of success for World Bank lending and International Finance Corporation projects was extracted from the lessons sections of the Implementation Completion and Results Report Reviews, Expanded Project Supervision Report Evaluation Notes, and—if they were of good quality according to IEG ratings—Implementation Completion and Results Reports and Expanded Project Supervision Reports. IEG categorized 16 possible 206 Appendix D Econometric Analysis factors behind SOE reform success (table D.1). The definitions of these individual factors are listed in the table and were coded only when judged to affect the project’s SOE reform(s) success. 39 Table D.1. Variables Definitions—Categories Used by IEG in SOE Project Coding Variable Definition Coding Individual factors—internal: factors under the World Bank Group’s direct control Project design Project design was coded whenever lessons dummy (0-1) addressed issues of simplicity, selectivity, or flexibility of project design, including clarity of objectives, parsimony of components, number of implementation agencies, and/or if the design was a good fit in relation to the countries’ situation. The importance of project design could have also been identified through project-related studies and analytical work which took into consideration previous findings, recommendations, and/or interactions with stakeholders that informed project design. Supervision Supervision was coded for projects in which dummy (0-1) lessons addressed the quality of World Bank supervision of the project and the quality of supervision missions, in terms of whether it (they) was (were) well equipped for implementation support. Team Team composition was coded whenever dummy (0-1) composition lessons addressed the adequacy of expertise or experience in project teams and highlighted them as influential in project implementation success. Choice of Choice of instrument was coded when dummy (0-1) instrument lessons addressed the adequacy of the lending instrument used, for example, a DPL, to set appropriate conditions for project 207 Appendix D Econometric Analysis Variable Definition Coding implementation or otherwise deliver the outcomes sought by the project. M&E M&E framework was coded when lessons dummy (0-1) framework addressed the clarity and/or monitorability of objectives and outcome indicators of a project; for example, the adequacy of baseline and target values, or the quality of sources of information for tracking indicator progress; and when these elements were emphasized as clear influencers of project implementation success. Collaboration Collaboration among Bank Group institutions dummy (0-1) among Bank was coded when lessons highlighted the Group degree of collaboration between World Bank, institutions IFC, and MIGA and its influence on project implementation success. Collaboration could have been in the form of information sharing between institutions, cross-fertilization of ideas, or fostering partnerships in the interest of a higher quality product. Sequenced Sequenced operations were coded when dummy (0-1) operations lessons highlighted the importance of a series of complementary or “stepwise” activities or a programmatic approach to operations in influencing project implementation success. This could have taken place when sequencing was important for the implementation of structural reforms or when they were part of the Bank Group’s longer- term support and commitment in a sector. Evidence of its importance could also have been identified in appraisal documents of either the project or related ex ante or ex post projects. Analytical work Analytical work was coded when lessons dummy (0-1) highlighted its importance for the design or 208 Appendix D Econometric Analysis Variable Definition Coding implementation of projects, including informing its design and setting up adequate conditions for project implementation such as providing training and/or supporting government’s/client’s capacity building. Identification of Risks at appraisal was coded when micro dummy (0-1) risks at evaluative evidence highlighted the appraisal adequacy of risk assessments at project appraisal and its influence in project implementation success. Funding Funding was coded when lessons highlighted dummy (0-1) the influence of the allocation of funds on project implementation success. For example, as a result of one or several restructurings that diverted funds from one component to another. Individual factors—external: factors outside of Bank Group’s direct control Client Client commitment was coded when lessons dummy (0-1) commitment highlighted the government’s support/commitment/willingness for reform and its influence on project implementation success. External External shocks was coded when lessons dummy (0-1) shocks showed that project implementation was affected (for example, interrupted) by exogenous events such as economic crises, political unrest, or climate-related incidents like droughts. Public sector Public sector institutional capacity was coded dummy (0-1) institutional when lessons highlighted the adequacy of capacity government or relevant government implementing agencies’ capacity to implement the reform program in a timely manner. 209 Appendix D Econometric Analysis Variable Definition Coding Private sector Private sector institutional capacity was coded dummy (0-1) institutional when lessons highlighted the private sector’s capacity (private enterprises and institutions) capacity to aid in implementing projects; for example, when its participation was an important part of carrying out SOE reform interventions. Collaboration Collaboration with external dummy (0-1) w/ external actors/donors/partners was coded when actors/donors lessons highlighted the extent of cooperation and/or coordination between the Bank Group and external development partners and its influence on project implementation success. Agency Agency coordination & political economy was dummy (0-1) coordination & coded when lessons highlighted the extent to political which projects were influenced by political economy situations, government agenda, the congressional approval process, changes in government, and/or coordination among government ministries or agencies. Aggregated factor variables Factor 1. The added version is the sum of M&E Value between 0 and 3 Strength of framework, supervision, and team project composition. implementation The standardized version of the factor has A standardized variable tools mean = 0 and standard deviation = 1. with mean of 0 and standard deviation of 1 Factor 2. The sum of analytical work, collaboration with Value between 0 and 3 Elements for external actors, and sequencing of operations. sustainable The standardized version of the factor has A standardized variable engagement mean = 0 and standard deviation = 1. with mean of 0 and standard deviation of 1 Factor 3. The sum of agency coordination/political Value between 0 and 3 Internalization economy, project design, and risks at appraisal. 210 Appendix D Econometric Analysis Variable Definition Coding of local The standardized version of the factor has A standardized variable conditions mean = 0 and standard deviation = 1 with mean of 0 and standard deviation of 1 Intervention-level variables Success A dummy variable signaling whether the SOE dummy (0-1) reform intervention was successful (1) or not (0). Projects were defined as “above the line”/successful if achieving or mostly achieving project outcomes. SOE reform Estimated amount of Bank Group resources US$ millions commitments for the intervention committed to SOE reform in US$ millions. Estimates for all interventions were made by sector and engagement area, based on the average share of project commitment amounts in each sector- engagement area accounted for by manually coded intervention amounts (intervention amounts were manually coded only for a subset of interventions whenever it was judged possible to do so based on project information). The estimates were made in the following way: each manually coded intervention amount was divided by the corresponding project’s commitment amount and the resulting shares were averaged within each sector-engagement area. These average shares were then used to calculate the estimated portion of project commitment amounts corresponding to SOE reform in all interventions Log SOE Log of SOE reform commitments log reform commitments Upstream An intervention was assigned a 1 if it was dummy (0-1) reform geared toward policy and institutional reforms 211 Appendix D Econometric Analysis Variable Definition Coding (upstream), and a 0 if it was geared toward enterprise-level activities (downstream). Operational-level variables World Bank A dummy variable indicating whether the dummy (0-1) adjustment operation was World Bank adjustment lending. IFC–AS A dummy variable indicating whether the dummy (0-1) operation was IFC advisory services. IFC– IS A dummy variable indicating whether the dummy (0-1) operation was IFC investment services. World Bank A dummy variable indicating whether the dummy (0-1) investment operation was World Bank investment lending. 2002–2008 A dummy variable for projects approved from dummy (0-1) FY02 to FY08. 2009–2010 A dummy variable for projects approved from dummy (0-1) FY09 to FY10. 2011–2016 A dummy variable for projects approved from dummy (0-1) FY11 to FY16. Country-level variables Low income Income levels were defined according to the dummy (0-1) country's level at the time of project approval—a dummy variable indicating whether the operation was approved in a low- income country. For example, for the current 2021 fiscal year, low-income economies are defined as those with a GNI per capita, calculated using the World Bank Atlas method, of $1,035 or less in 2019. Lower-middle A dummy variable indicating if the operation dummy (0-1) income was approved in a lower-middle-income country. 212 Appendix D Econometric Analysis Variable Definition Coding Upper middle A dummy variable indicating if the operation dummy (0-1) and High was approved in either an upper-middle or a income high-income country. Low level of This variable uses the World Governance Quartiles control of Indicator Control of Corruption, which corruption measures the extent to which public power is exercised for private gain and capture of the state by elites and private interest. The control of corruption indicator is measured from −1.9 to 2.5, where higher values correspond to better outcomes. Cutoffs are based on quartiles of this indicator. Domestic credit This indicator is the ratio of domestic credit to Ratio to private the country's gross domestic product. sector Domestic credit to private sector refers to financial resources provided to the private sector, such as through loans, purchases of nonequity securities, and trade credits and other accounts receivable, that establish a claim for repayment. For some countries these claims include credit to public enterprises. Region Sub-Saharan A dummy variable for countries classified as dummy (0-1) Africa Sub-Saharan Africa: Angola Ethiopia Niger Benin Gabon Nigeria Botswana Gambia, The Rwanda Burkina Faso Ghana São Tomé and Príncipe Burundi Guinea Senegal Cabo Verde Guinea-Bissau Seychelles Cameroon Kenya Sierra Leone Central African Republic Lesotho Somalia Chad Liberia South Africa Comoros Madagascar South Sudan Congo, Dem. Rep. Malawi Sudan Congo, Rep Mali Tanzania Côte d'Ivoire Mauritania Togo Equatorial Guinea Mauritius Uganda Eritrea 213 Appendix D Econometric Analysis Variable Definition Coding Mozambique Zambia Eswatini Namibia Zimbabwe. East Asia and A dummy variable for countries classified as dummy (0-1) Pacific East Asia and Pacific: American Samoa Korea, Rep. Philippines Australia Lao PDR Samoa Brunei Darussalam Macao SAR, China Singapore Cambodia Malaysia Solomon Islands China Marshall Islands Taiwan, China Fiji Micronesia, Fed. Sts. Thailand French Polynesia Mongolia Timor-Leste Guam Myanmar Papua New Guinea Hong Kong SAR, China Nauru Tonga Indonesia New Caledonia Tuvalu Japan New Zealand Vanuatu Kiribati Northern Mariana Islands Vietnam Korea, Dem. People's Rep. Palau Europe and A dummy variable for countries classified as dummy (0-1) Central Asia Europe and Central Asia: Albania Gibraltar Norway Andorra Greece Poland Armenia Greenland Portugal Austria Hungary Romania Azerbaijan Iceland Russian Federation Belarus Ireland San Marino Belgium Isle of Man Serbia Bosnia and Herzegovina Italy Slovak Republic Bulgaria Kazakhstan Slovenia Channel Islands Kosovo Spain Croatia Kyrgyz Republic Sweden Cyprus Latvia Switzerland Czech Republic Liechtenstein Tajikistan Denmark Lithuania Turkey Estonia Luxembourg Turkmenistan Faroe Islands Moldova Ukraine Finland Monaco United Kingdom France Montenegro Uzbekistan Georgia the Netherlands Germany North Macedonia Latin America A dummy variable for countries classified as dummy (0-1) and the Latin America and the Caribbean: Antigua and Caribbean Barbuda Curacao Paraguay Argentina Dominica Peru Aruba Dominican Republic Puerto Rico Bahamas, The Ecuador Sint 214 Appendix D Econometric Analysis Variable Definition Coding Maarten (Dutch part) Barbados El Salvador St. Kitts and Nevis Belize Grenada St. Lucia Bolivia Guatemala St. Martin (French part) Brazil Guyana St. Vincent and the Grenadines British Virgin Islands Haiti Suriname Cayman Islands Honduras Trinidad and Tobago Chile Jamaica Turks and Caicos Islands Colombia Mexico Uruguay Costa Rica Nicaragua Venezuela, RB Cuba Panama Virgin Islands (US) Middle East A dummy variable for countries classified as dummy (0-1) and North Middle East and North Africa: Algeria Jordan Africa Qatar Bahrain Kuwait Saudi Arabia Djibouti Lebanon Syrian Arab Republic Egypt, Arab Rep. Libya Tunisia Iran, Islamic Rep. Malta United Arab Emirates Iraq Morocco West Bank and Gaza Israel Oman Yemen, Rep. South Asia A dummy variable for countries classified as dummy (0-1) South Asia: Afghanistan India Pakistan Bangladesh Maldives Sri Lanka Bhutan Nepal Source: Independent Evaluation Group portfolio review and analysis and econometric analysis. Note: Each success or failure factor from evaluations was coded if (i) it was highlighted in the lessons of a project’s evaluative document(s) (or in additional documents for specific factors), and (ii) the factor was judged as relevant for SOE reform interventions. DPL = development policy loan; FY = fiscal year; GNI = gross national income; IEG = Independent Evaluation Group; IFC = International Finance Corporation; M&E = monitoring and evaluation; MIGA = Multilateral Investment Guarantee Agency; SOE = state-owned enterprise. Principal Component Analysis After identifying individual factors associated with SOE reform interventions outcomes, IEG applied PCA to these individual factors to identify latent composite factors that both simplified the analysis and enabled a more holistic understanding of SOE reform success. 40 To do so, several steps are required. First, new variables are constructed that are combinations of the 215 Appendix D Econometric Analysis individual success factors. These new combination variables are ordered based on how much they predict the variance of the individual factors, thus retaining the most important features of those variables in constructing a predictive model. Results from this exercise could potentially reveal how different individual factors may jointly affect SOE reform success. Risks with Adopted Methodology The methodology presents risk of omitted variable bias to the extent that relevant covariates of SOE reform success that may also be correlated with some of the independent variables cannot be observed. For example, because SOE reform success can be part of the success of a larger project, and the portfolio review identified only factors associated with SOE reform components, some variables associated with overall project success may be omitted. Other variables may be omitted for lack of data, such as identity of project manager and managerial turnover during the project cycle, which have been documented as being strongly associated with project outcomes (for example, Geli, Kraay, and Nobakht 2014; Legovini, Di Mario, and Piza 2015) and that may be correlated with at least the M&E framework, project design, supervision, or team composition factors. The analysis does not include this in the model and potentially other variables related with project-level outcomes because the portfolio review exercise focused on identifying potential predictors of SOE reform success as opposed to broader project success. But despite this potential source of bias, the omitted variables are not strongly negatively associated with the predictors, so the omission of these variables would not cause sign changes in estimation results, and hence the findings are still valid. In addition, the main results are consistent with those documented in the literature of project outcome determinants. For example, the model identifies significant relationships between the joint movement of several project-level factors and SOE reform success, in line with the evidence found by Denizer, Kaufmann, and Kraay (2013), according to which 80 percent of the variation in World Bank project outcomes can be explained by within-countries and across-projects variations rather than by 216 Appendix D Econometric Analysis country characteristics. Similarly, the M&E framework variable shows a significant and positive relationship with SOE reform success, aligned with Raimondo (2016), which found that the quality of M&E is significantly and positively associated with project outcome as institutionally measured by the World Bank. That said, the methodology sought to uncover potential associations between intervention-level variables, project-level variables, and SOE reform success, not project success, and sought to uncover possible associations between these variables, not causal relationships. The results should thus be interpreted with caution and in the context of SOE reform success, not overall project success. A second risk is the potential for perfect collinearity between variables because of their dichotomous nature and potential few observations. This may have prevented the inclusion of controls for which few interventions have information and more generally prevented the testing of relevant hypotheses that involve interacting dummy variables. Perfect collinearity problems in logistic regression models are common in small samples where all or most variables analyzed are dichotomous. This risk was sought to be mitigated by minimizing the number of relevant projects that were not coded because of coder uncertainty and/or lack of information that could somehow be recovered. 41 A third risk related to sample representativeness of the underlying population of SOE reform projects approved in fiscal years (FY)02–18, which would determine if inference was possible. The sample is based on all IEG- evaluated SOE reform projects identified through sector code and/or keyword searches and the ensuing manual revision done through a random (stratified by institution) assignment of projects to reviewers. 42 The sample accounts for more than half (57 percent) of the projected population of SOE reform financing projects. 43 This said, however, the sample is not random and thus there is a potential risk of it not being representative and invalidating any inference. To the extent this is true, the results reported here must be interpreted with caution. Another key assumption is that the 217 Appendix D Econometric Analysis choice of sector codes and keywords to identify the initial SOE reform portfolio did in fact capture the population of SOE reform projects approved during FY02–18. Results IEG sought a stepwise approach when performing the analysis. First, it performed the PCA and with it identified possible composite factors behind SOE reform success. Second, it estimated equation (1) with the use of these composite factors and other intervention-level, project-level, and country- level variables. Third, it again estimated equation (1) but this time with the original individual factors of success identified through the manual review. PCA Results Applying PCA to the individual factors yielded six principal components with eigenvalues greater or equal to 1, which is the main criterion for selecting principal components. 44 The six components explained 54 percent of their cumulative variance (table D.2, panel a). After applying component factoring to these principal components, rotating the resulting factor loadings matrix to ease interpretation, and choosing items with factor loadings above 0.4 in line with standard cutoff values when applying this simple assessment criterion, IEG identified three composite factors (table D.2, panel b). The first composite factor is based on three individual factors: M&E framework, quality of supervision, and quality of team composition. IEG called this composite factor “strength of project implementation tools.” The second composite factor is based on another three individual factors: analytical work, collaboration with external actors/partners, and sequencing of projects or operations. IEG named this composite factor “elements for sustainable engagement.” The third composite factor was again made up of three individual factors: political economy/agency coordination, project design, and identification of project risks at appraisal. IEG named this composite factor “internalization of local 218 Appendix D Econometric Analysis conditions/context in project preparation.” As a final step, IEG used two methods to predict factor scores for regression analysis. First, it simply added the individual underlying factors of each of the three composite factors; this is intuitively appealing because a larger sum of the factor dummies corresponds to a larger number of factors present in any given project; second, it used the standardized scoring coefficients (shown in table D.2, panel c) from the three composite factors to build the factor scores. 45 Estimation results of equation (1) are reported with the first version of these factor scores, and the second (standardized) version is used for robustness checks. Table D.2. Principal Component Analysis Results a. Principal components 219 Appendix D Econometric Analysis b. Factor analysis Principal components Eigenvalue Difference Proportion Cumulative variance Component 1 1.708 0.212 0.114 0.114 Component 2 1.496 0.173 0.100 0.214 Component 3 1.324 0.029 0.088 0.302 Component 4 1.295 0.147 0.086 0.388 Component 5 1.147 0.018 0.077 0.465 Component 6 1.130 0.146 0.075 0.540 Component 7 0.984 0.058 0.066 0.606 Component 8 0.926 0.028 0.062 0.667 Component 9 0.898 0.072 0.060 0.727 Component 10 0.825 0.063 0.055 0.782 Component 11 0.762 0.053 0.051 0.833 Component 12 0.709 0.063 0.047 0.880 Component 13 0.646 0.060 0.043 0.923 Component 14 0.586 0.023 0.039 0.962 Component 15 0.563 0.038 1.000 220 Appendix D Econometric Analysis c. Factor scores Standard scoring coefficients Variables Factor 1 Factor 2 Factor 3 Political Economy/Agency Coordination 0.001 0.110 -0.400 Analytical work -0.047 0.383 0.105 Choice of instrument 0.010 -0.159 -0.253 Client commitment 0.096 0.196 0.215 Collaboration w/external actors -0.014 0.412 -0.075 Project design 0.064 0.063 0.424 Identification of project risks at appraisal -0.178 -0.059 0.369 M&E Framework 0.424 -0.024 0.095 Public sector capacity 0.173 -0.130 -0.046 Sequenced projects/operations -0.004 0.373 -0.086 Quality of supervision 0.442 -0.012 0.030 Quality of team composition 0.435 0.173 -0.169 External shocks -0.018 -0.236 -0.006 Collaboration between WBG institutions 0.023 0.031 -0.225 Private sector capacity -0.032 -0.143 0.003 Source: Independent Evaluation Group. Note: In panel a, only principal components 1–6 have eigenvalues greater than 1. Principal components having eigenvalues below 1 explain less than the equivalent of one variable’s variance, which makes them unhelpful for data reduction (Tabachnick and Fidell 2012). In panel b, Varimax orthogonal rotation is applied, producing uncorrelated factors. Only items with a factor loading above the cutoff value of 0.4 are included when rounded to the nearest integer. The communality is each variable’s proportion of variability that is explained by the factors. The closer the communality is to 1, the better the variable is explained by the factors. Communality values do not change across unrotated and rotated factor loadings. In panel c, scoring coefficients are estimated by regression. M&E = monitoring and evaluation; WBG = World Bank Group. Composite Factor Regression Results IEG initially sought to understand the drivers of SOE reform success holistically based on the composite factors identified through PCA. The objective was to test whether certain factors were jointly associated with the odds of success of an SOE reform intervention. Other regressors included in the estimations were relevant intervention and project- and country-level variables. Three specifications of equation (1) were estimated, where each subsequent specification improved on the previous one based on model fit 221 Appendix D Econometric Analysis statistics (for example, the likelihood ratio test statistic). Although all three composite factors were initially tested as regressors, the third factor, “internalization of local conditions/context in project preparation,” was never statistically significant and did not change the model fit nor the significance or magnitude of other variable coefficients, so it was dropped from the reported estimations. 46 Table D.3 shows logistic regression results of SOE reform success on its predictors, including the first two composite factors: “strength of project implementation tools” (factor 1) and “elements for sustainable engagement” (factor 2). The model fit improves substantially from the first to second specifications when the two composite factors are included in the model. 47 Results suggest that the estimated odds of success for an SOE reform intervention in which the strength of project implementation tools (through improved quality of the M&E framework, supervision, and team composition) increases by one standard deviation (0.76), are 1.36 times greater than the odds of success for an SOE reform intervention with lesser strength in these implementation tools, all other controls kept constant. In the preferred specification in model 3, 48 where country income group, control of corruption level, and domestic credit to the private sector as a share of GDP are included as additional regressors, the coefficient for factor 1 remains statistically significant at the 5 percent level, and the odds of SOE reform success associated with it remain robust at an estimated odds ratio of 1.31. That said, the coefficient of this composite factor loses its statistical significance at standard levels when inferences are based on robust standard errors (the associated p value increases to 0.19). In the case of factor 2, the estimated odds of success for an SOE reform intervention in which the elements for sustainable engagement increase by one standard deviation (through analytical work, collaboration with external actors/partners and sequencing of projects), are 1.39 times greater than the odds of success for a SOE reform intervention where these elements of sustainable engagement are not in place, all other controls kept constant. 222 Appendix D Econometric Analysis The estimated odds ratio for factor 2 increases to 1.44 in model 3, and its associated coefficient remains statistically significant at the 5 percent level (table D.3, panel a). In contrast to the strength of project implementation tools factor, after adjusting the standard errors, the significance of the coefficient for the elements for sustainable engagement factor holds at the 10 percent significance level. Overall, the analysis yields evidence suggestive of a positive association between both composite factors and the odds of SOE reform success, but it is more robust for factor 2, elements of sustainable engagement. There are additional important results from the logistic regressions reported in table D.3. The first result is that high levels of control of corruption, as measured by the WGIs, are associated with substantially better estimated odds of success for SOE reform interventions compared with low levels. These results remain robust after adjusting the standard errors to address the potential heteroscedasticity issue, with the coefficient remaining significant at the 5 percent level. Lower-middle and upper-middle levels of control of corruption, however, were not statistically significant compared with low levels. Results suggest that in countries with high levels of control of corruption, the estimated odds of SOE reform success are 2.7 times higher than in countries with low levels, all other controls in the model kept constant. The second result is that, holding other controls in the model constant, a 1 percent increase in domestic credit as a share of GDP is associated with 1.2 percent (1 – 0.988 = 0.012) lower estimated odds of SOE reform success. This means, for example, that a one standard deviation increases in the domestic credit ratio (+27.21 percent) is associated with a reduction in the odds of success of 32.6 percent. This result holds after adjusting the standard errors to address the potential heteroscedasticity issue. The third result is that in upper-middle- and high-income countries the estimated odds of SOE reform success appear to be lower than in low-income countries, specifically by a factor of 0.48, or equivalently, by 52 percent, all 223 Appendix D Econometric Analysis other controls in the model kept constant. However, this result did not hold when standard errors were adjusted. Fourth, higher SOE reform commitment amounts are associated with higher estimated odds of success, all other controls in the model constant, with the associated coefficient being significant at the 5 percent level. This result is robust to the inclusion of both types of standard errors. In addition, some variables of interest, such as SOE reform sectors, engagement areas, and number of projects and interventions per country- year, show statistically insignificant coefficients at standard levels. This indicates that the estimated odds of SOE reform success are not associated with these variables (table D.3, panel b). For example, whether a project engages on corporate governance reform or regulatory reform is not significantly associated with the odds of success. As shown in panel c of table D.3, the stated results are robust to using standardized factor scores instead of the added individual factors. The positive relationships between project implementation tools, elements for sustainable SOE reform engagement, and SOE reform success hold when country-level variables are included (model 2 versus model 3). Country-level variables’ coefficients also retain their original sign and statistical significance so that high levels of control of corruption, income group, and domestic credit to the private sector as a share of GDP remain significantly associated with SOE reform success. Additionally, the statistical insignificance of the coefficients for SOE reform sectors, engagement areas, and volume of projects and interventions are again observed. The changes described for the results in panels a and b of table D.3, resulting from the inclusion of robust standard errors, apply in the same way to the results in table D.3, panel c. 49 224 Appendix D Econometric Analysis Table D.3. Logistic Regressions of State-Owned Enterprise Reform Success with Composite Factors a. Logistic regressions models with added individual factors. Dependent variable is SOE reform intervention success dummy (1 = above the line, 0 = below the line). Model 1 Model 2 Model 3 Predictor Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Intercept 0.548 0.327 0.522 Ln (intervention amount) 0.104* 1.109 0.087 0.140** 1.150 Intervention type: upstreama 0.077 0.096 0.095 Factor 1: Strength of proj. impl. toolsb — 0.306** 1.358 0.272** 1.313 Factor 2: Elements for sust. eng. c — 0.327** 1.386 0.366*** 1.443 Financing loan typed World Bank adjustment loan 0.207 0.172 0.091 IFC AS 0.279 0.426 0.813 IFC IS 0.233 0.377 0.542 SOE reform sector dummiese Energy −0.112 −0.021 −0.060 National 0.210 0.242 0.421 225 Appendix D Econometric Analysis Model 1 Model 2 Model 3 Predictor Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Year dummiesf 2002−2008 −0.187 −0.199 −0.496* 0.609 2011−2016 −0.245 −0.290 −0.344 Income groupg Low-middle income — — 0.346 Upper-middle and high income — — −0.739* 0.478 Level of control of corruptionh Low-middle level — — 0.214 Upper-middle level — — 0.314 High level — — 0.978*** 2.660 Domestic credit to private as % GDP — — −0.012*** 0.988 Region dummiesi Sub-Saharan Africa −0.089 −0.170 −0.464 East Asia and Pacific 0.249 0.063 0.176 Europe and Central Asia 0.310 0.149 0.088 226 Appendix D Econometric Analysis Model 1 Model 2 Model 3 Predictor Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Latin America and the Caribbean 0.558 0.525 0.488 Middle East and North Africa −0.045 −0.052 0.064 −2 Log L 781.766 770.540 747.713 Pseudo R square 0.0244 0.0407 0.0730 Df used 14 16 22 N 666 666 666 Number of countries 88 88 88 Source: Independent Evaluation Group, World Bank Development Indicators, and Worldwide Governance Indicators. Note: Statistical inferences of the three models are based on default standard errors. Coeff. = coefficient; Df = degrees of freedom; GDP = gross domestic product; IFC AS International Finance Corporation advisory servies; IFC IS = International Finance Corporation investment services; M&E = monitoring and evaluation; SOE = state- owned enterprise. a. Binary coding (1 = yes upstream, 0 = downstream). e. Reference group is financial sector. b. Factor 1 (strength of project implementation tools) is the sum of M&E f. Reference group is year 2009−10. framework, supervision, and team composition individual factors. g. Reference group is low-income countries. c. Factor 2 (elements for sustainable engagement) is the sum of analytical h. Reference group is low level of control of corruption. work, collaboration with external actors, and sequenced operations individual i. Reference group is South Asia. factors. *p <.10 **p <.05 ***p <.01. d. Reference group is World Bank investment project financing. 227 Appendix D Econometric Analysis b. Logistic regressions: Role of engagement area, number of interventions, and number of projects. Dependent variable is SOE reform intervention success dummy (1 = above the line, 0 = below the line). Role of Engagement Role of Number of Area Intervention’ and projects Predictor Coeff Odds ratio Coeff Odds ratio Intercept 0.135 0.461 Ln (intervention amount) 0.196** 1.216 0.124* 1.132 Intervention type: upstreama 0.245 0.102 Factor 1: Strength of proj. impl. tools b 0.294** 1.342 0.264** 1.302 Factor 2: Elements for sust. eng. c 0.369** 1.446 0.398*** 1.488 (no.) of interventions per country per year — −0.051 (no.) of projects per country per year — 0.286 Financing loan typed World Bank adjustment loan −0.009 0.046 IFC AS 1.129* 3.091 0.627 IFC IS 0.581 0.494 SOE reform sector dummiese 228 Appendix D Econometric Analysis Role of Engagement Role of Number of Area Intervention’ and projects Predictor Coeff Odds ratio Coeff Odds ratio Energy −0.020 −0.061 National 0.410 0.416 Engagement areaf SOE ownership 0.190 — Business and operations −0.004 — Corporate governance 0.434 — E&S aspects −0.276 — Financial management 0.518 — Macro-fiscal policy −0.236 — Year dummiesg 2002−2008 −0.435 −0.464 2011−2016 −0.317 −0.326 Income grouph Low-middle income 0.368 0.328 229 Appendix D Econometric Analysis Role of Engagement Role of Number of Area Intervention’ and projects Predictor Coeff Odds ratio Coeff Odds ratio Upper-middle and high income −0.702* 0.495 −0.689* 0.480 Level of control of corruptioni Low-middle level 0.178 0.122 Upper-middle level 0.320 0.279 High level 0.950*** 2.586 0.928*** 2.530 Domestic credit to private as % GDP −0.012*** 0.988 −0.012*** 0.988 Region dummiesj Sub-Saharan Africa −0.428 −0.445 East Asia and Pacific 0.201 0.103 Europe and Central Asia 0.058 0.086 Latin America and the Caribbean 0.486 0.418 Middle East and North Africa 0.008 0.010 −2 Log L 742.824 745.117 Pseudo R square 0.0798 0.0766 230 Appendix D Econometric Analysis Role of Engagement Role of Number of Area Intervention’ and projects Predictor Coeff Odds ratio Coeff Odds ratio Df used 28 24 N 666 666 Number of countries 88 88 Sourrce: World Bank Independent Evaluation Group, World Bank Development Indicators, and Worldwide Governance Indicators. Note: Statistical inferences of the two models are based on default standard errors. Coeff. = coefficient; Df = degrees of freedom; GDP = gross domestic product; E&S = environmental and social; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; Interv’ns = interventions; M&E = monitoring and evaluation; SOE = state-owned enterprise. a. Binary coding (1 = yes upstream reform, 0 = downstream reform). e. Reference group is financial sector. b. Factor 1 (strength of project implementation tools) is the sum of M&E f. Reference group is regulatory framework. framework, supervision, and team composition individual factors. g. Reference group is year 2009−10. c. Factor 2 (elements for sustainable engagement) is the sum of analytical h. Reference group is low-income countries. work, collaboration with external actors, and sequencing of operations i. Reference group is low level of control of corruption. individual factors. j. Reference group is South Asia. d. Reference group is World Bank investment project finance. *p <.10 **p <.05 ***p <.01. 231 Appendix D Econometric Analysis c. Models with standardized factor scores. Dependent variable is SOE reform intervention success dummy (1 = above the line, 0 = below the line). Model 1 Model 2 Model 3 Model 4 Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Intercept 0.023 0.064 0.114 0.091 Ln (intervention amount) 0.148** 1.160 0.146** 1.157 0.219** 1.245 0.204*** 1.227 Intervention type: upstreama 0.140 0.206 0.209 0.222 Factor 1: Strength of proj. impl.b — 0.244** 1.276 0.170* 1.185 0.164 Factor 2: Elements for sust. eng.c — 0.248** 1.281 0.287*** 1.332 0.305*** 1.356 Factor 3: Inter. of local cond.d — 0.044 0.101 0.113 Number of interventions — — — −0.060 Number of projects — — — 0.281 Lending loan typee World Bank adjustment loan 0.069 0.085 0.027 −0.014 IFC AS 0.556 0.692 1.167* 1.005 IFC IS −0.022 0.266 0.725 0.681 SOE reform sectorf Energy −0.083 0.032 −0.003 0.003 232 Appendix D Econometric Analysis Model 1 Model 2 Model 3 Model 4 Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio National 0.197 0.246 0.453 0.449 Engagement areag SOE ownership 0.508 0.557 0.504 0.492 Business and operation 0.337 0.326 0.281 0.302 Corporate governance 0.716 0.693 0.730 0.759 E&S 0.409 0.349 0.347 0.347 Finance 0.831 0.841 0.840 0.836 Macro-fiscal policy 0.207 0.176 0.093 0.123 Year dummiesh 2002−2008 −0.246 −0.245 −0.461 −0.422 2011−2016 −0.303 −0.322 −0.319 −0.313 Income groupi Low-middle income — — 0.358 0.343 Upper-middle income — — −0.746* 0.474 −0.702* 0.495 233 Appendix D Econometric Analysis Model 1 Model 2 Model 3 Model 4 Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Level of control of corruptioni Low-middle level — — 0.176 0.086 Upper-middle level — — 0.362 0.333 High level — — 1.035*** 2.816 1.004* 2.728 Domestic credit to private as % GDP — — −0.013*** 0.988 −0.013*** 0.987 Region dummiesk Sub-Saharan Africa −0.051 −0.148 −0.414 −0.401 East Asia and Pacific 0.300 0.072 0.184 0.111 Europe and Central Asia 0.243 0.071 0.071 0.069 Latin America and the Caribbean 0.600 0.548 0.502 0.409 Middle East and North Africa −0.119 −0.208 −0.067 −0.112 234 Appendix D Econometric Analysis Model 1 Model 2 Model 3 Model 4 Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio Coeff Odds ratio −2 Log L 792.93 779.496 742.493 739.788 Df used 20 23 29 31 Pseudo R square 0.028 0.047 0.080 0.084 N 679 679 666 666 Source: World Bank Independent Evaluation Group and World Bank Development Indicators. Note: Statistical inferences of the four models are based on default standard errors. Coeff. = coefficient; Df = degrees of freedom; E&S = environmental and social; GDP = gross domestic product; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; M&E = monitoring and evaluation; SOE = state-owned enterprise. a. Binary coding (1 = yes upstream, 0 = no). f. Reference group is finance sector. b. Factor 1 (strength of project implementation tools) is the standardized g. Reference group is regulatory framework. factor score of three variables: M&E framework, supervision, and team h. Reference group is year 2009−10. composition. i. Reference group is low-income countries. c. Factor 2 (elements for sustainable engagement) is the standardized factor j. Reference group is low level of control of corruption. score of three variables: analytical work, collaboration with external actors, k. Reference group is South Asia. and sequencing of operations. *p <.10 **p <.05 ***p <.01. d. Factor 3 (internalization of local conditions/context in project preparation) is the standardized factor score of three variables: agency coordination and political economy, design, and risks appraisal. e. Reference group is World Bank investment project finance. 235 Appendix D Econometric Analysis Individual Factor Regression Results In addition to the logistic regressions with composite factors, IEG also estimated equation (1) by using the original individual factors of success identified during the portfolio manual review. When all individual factors are introduced one by one as regressors, results suggest that preparation of sound M&E frameworks, Bank Group collaboration with external actors and partners, sequenced operations, and an adequate choice of financing instrument may be important individual factors positively associated with SOE reform success. Coefficients for external shocks are also significant but are negatively associated with the odds of success. 50 However, only the coefficients for M&E frameworks and external shocks retain their statistical significance after adjusting the standard errors in the models to address the potential heteroscedasticity issue. Nonetheless, IEG used the original five statistically significant individual factors to perform one last exercise. Of these five individual factors, three of them underpin composite factors (M&E framework, sequenced operations, and collaboration with external actors), and two of them are unrelated to the composite factors (choice of instrument and external shocks). Table D.4, panel b, shows results for two final model specifications in which these five factors are introduced simultaneously in equation (1) alongside usual controls. Model 2 in the table reports these results with the team’s preferred specification (per the Hosmer-Lemeshow Goodness of Fit Test), showing that all five factor coefficients remain statistically significant and with their original signs. These five factors are robust to the inclusion of several control variables, including control of corruption, income group, and domestic credit to the private sector as a share of GDP. Results suggest, for instance, that an SOE reform intervention for which collaborating with external actors and partners was important during project implementation, is associated with 1.56 times higher odds of success than an intervention where such collaboration was not salient, all other controls in the model kept constant. Similarly, projects for which sequencing was referenced 236 Appendix D Econometric Analysis as influencing implementation is associated with 2.56 times higher odds of SOE reform success than projects for which sequencing was not considered relevant for SOE reform success, all other controls kept constant. Reference to and relevance for SOE reform success of M&E frameworks and choice of instrument were also positively associated with a project’s odds of SOE reform success by 1.76 and 2.27 times, respectively. As in the composite factor estimations, the coefficients for control of corruption and domestic credit to the private sector remain significant (and with the same sign), and the coefficient for upper- middle- and high-income status does not. Table D.4. Logistic Regressions with Individual Factors a. Logistic regressions: SOE reform success, individual factors, and other correlates. Dependent variable is SOE reform intervention success dummy (1 = above the line, 0 = below the line). Coefficient F1 Variable (entered one by without Country Coefficient with one) Vars Country Vars1 Analytical work 0.304 0.337 Choice of instrument 0.708** 0.766** Collab. World Bank Group institutions 0.588 0.714 Project design 0.315* 0.307 Identification of project risks at –0.284 –0.113 appraisal Monitoring and evaluation framework 0.618*** 0.459* Sequenced operations 0.800 0.833* Supervision 0.444** 0.356 Team composition 0.151 –0.049 Political economy and agency –0.317 –0.311 coordination Client commitment 0.046 0.093 Collab. with external actors 0.413* 0.432* External shocks –0.630*** –0.704*** 237 Appendix D Econometric Analysis Coefficient F1 Variable (entered one by without Country Coefficient with one) Vars Country Vars1 Private sector capacity –0.398 -–0.608 Public sector capacity –0.147 –0.308 Source: World Bank Independent Evaluation Group, World Bank Development Indicators, and Worldwide Governance Indicators. Note: For models without country variables, the predictors include intervention amounts in log, intervention type—upstream, lending type dummies, sector dummies, year dummies, and regional dummies. For models with country variables, three country-level variables are added, which are income dummies, control of corruption, and domestic credit to private sector as percentage of gross domestic product. Statistical inferences are based on default standard errors. Collab. = collaboration; SOE = state- owned enterprise. *p <.10 **p <.05 ***p <.01. b. Logistic regressions: SOE reform success, selected individual factors, and other correlates. Variable Model 2 with Number of Interventions and Model 1 projects Predictor Coeff Odds ratio Coeff Odds ratio Intercept 0.501 0.497 Ln (intervention amount) 0.158** 1.171 0.148** 1.160 Intervention type: upstreama 0.034 0.050 F1 M&E framework 0.581** 1.788 0.569** 1.767 F1 collaboration with 0.395* 1.484 0.445* 1.560 external actor F1 sequenced operations 0.842a 0.941* 2.562 F1 external shocks −0.556** 0.573 −0.517** 0.596 F1 choice of financial 0.779** 2.178 0.821** 2.272 instrument Financing loan typeb World Bank adjustment loan 0.184 0.123 IFC AS 0.951 0.802 238 Appendix D Econometric Analysis Model 2 with Number of Interventions and Model 1 projects Predictor Coeff Odds ratio Coeff Odds ratio IFC IS 0.579 0.527 SOE reform sectorc Energy 0.011 0.020 National 0.502 0.495 Number of interventions per −−− −0.062 country per year Number of projects per −−− 0.246 country per year Year dummiesd 2002−2008 −0.398 −0.352 2011−2016 −0.269 −0.248 Income groupe Lower-middle income 0.320 0.311 Upper-middle and high −0.794* 0.452 −0.735* 0.480 income Level of control of corruptionf Low-middle level 0.139 0.059 Upper-middle level 0.258 0.213 High level 0.705** 2.024 0.649* 1.914 Domestic credit to private as −0.012* 0.988 −0.013*** 0.988 % GDP Region dummiesg Sub-Saharan Africa −0.339 −0.317 East Asia and Pacific 0.211 0.178 Europe and Central Asia 0.172 0.161 239 Appendix D Econometric Analysis Model 2 with Number of Interventions and Model 1 projects Predictor Coeff Odds ratio Coeff Odds ratio Latin America and the 0.654 0.561 Caribbean Middle East and North Africa 0.005 −0.021 −2 Log L 736.81 734.55 Df used 25 27 Pseudo R square 0.088 0.091 Max rescaled R square 0.126 0.131 Hosmer-Lemeshow 8.22 9.463 goodness of fit test Chi-square Df used 8 8 Pr > chi-square 0.412 0.305 N 666 666 Source: World Bank Independent Evaluation Group, World Bank Development Indicators, and Worldwide Governance Indicators Note: Statistical inferences are based on default standard errors. Marginal p =.106. Coeff. = coefficient; Df= degrees of freedom; GDP = gross domestic product; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; Intv. = intervention; Proj. = project; Pr= probability; M&E = monitoring and evaluation; SOE = state-owned enterprise. 240 Appendix D Econometric Analysis a. Binary coding (1 = yes upstream, 0 = f. Reference group is low level of control of downstream). corruption. b. Reference group is World Bank investment g. Reference group is South Asia. project finance. *p <.10. c. Reference group is finance sector. **p <.05. d. Reference group is year 2009−10. ***p <.01. e. Reference group is low-income countries. Final Tests of Robustness The PCA and logistic regression results suggest that the odds of SOE reform success are improved when (i) the composite factor reflecting the Bank Group’s strength of project implementation tools, composed of M&E frameworks, qualified teams, and close supervision, is enhanced as a whole;and (ii) when the composite factor reflecting elements for sustainable SOE reform engagements, made up of collaboration with external actors and partners, thoughtful sequencing of projects, and continuous analytical work, is enhanced. However, only the factor capturing the inclusion of elements for sustainable reform engagements is robust to all of the tested changes in specification. Similarly, initial evidence that collaboration with external actors and partners, sequenced operations, and choice of financing instrument are positively associated with the odds of SOE reform success was not robust to all tested specifications, although the variable sound M&E frameworks proved robust. In addition, when these factors are included jointly with country-level variables and adjusted standard errors, none of the results hold. External shocks, however, are negatively associated with the odds of SOE reform success, and this result is robust to all changes in specification attempted. Certain country conditions seem to be positively associated with odds of SOE reform success, namely high levels of control of corruption, lower levels of income, and shallower private credit markets. However, only high levels of control of corruption and depth of private credit markets are robust to all changes in specification attempted. 241 Appendix D Econometric Analysis Descriptive Statistics Table D.5. Descriptive Statistics of SOE Reform Success Outcome and Predictors Variable Mean SD Min. Max. Success dummy 0.71 0.45 0 1.00 SOE reform commitments ($, 53.29 116.95 0.01 1,195.00 millions) Log SOE reform commitments 2.68 1.79 (4.37) 7.09 Upstream reform 0.45 0.50 0 1.00 Individual factors F1: M&E framework 0.23 0.42 0 1.00 F1: Collaboration with external 0.27 0.44 0 1.00 actors F1: Sequence adopted 0.06 0.23 0 1.00 F1: External shocks 0.17 0.38 0 1.00 F1: Choice of instrument 0.11 0.31 0 1.00 PCA Factor 1. Strength of project 0.52 0.76 0 3.00 implementation tools a Factor 2. Elements for 0.58 0.76 0 3.00 sustainable engagement b 0.86 0.67 0 3.00 Factor 3. Internalization of local conditionsc Interventions per country-year (no.) 4.20 2.32 1.00 10.00 Projects per country-year (no.) 1.36 0.71 1.00 5.00 Sector Energy 0.47 0.50 0 1.00 Finance 0.37 0.48 0 1.00 National 0.16 0.36 0 1.00 242 Appendix D Econometric Analysis Variable Mean SD Min. Max. Type of financing instrument World Bank adjustment 0.51 0.50 0 1.00 IFC AS 0.03 0.18 0 1.00 IFC IS 0.06 0.23 0 1.00 World Bank investment 0.40 0.49 0 1.00 Approval fiscal year 2002–08 0.61 0.49 0 1.00 2009–10 0.19 0.40 0 1.00 2011–16 0.20 0.40 0 1.00 Income level Low income 0.37 0.480 0 1.00 Lower-middle income 0.42 0.49 0 1.00 Upper-middle and high income 0.21 0.41 0 1.00 Country-level variables Low level of control of 0.25 0.43 0 1.00 corruption (1–25 percentile) Low-middle-level control of 0.26 0.44 0 1.00 corruption (26–50 percentile) Upper-middle-level control of 0.24 0.43 0 1.00 corruption (51–75 percentile) High level control of corruption 0.25 0.43 0 1.00 (76–100 percentile) Domestic credit to private sector 33.49 27.21 0.49 126.58 (% of GDP) Region Sub-Saharan Africa 0.30 0.46 0 1.00 East Asia and Pacific 0.20 0.40 0 1.00 Europe and Central Asia 0.22 0.42 0 1.00 243 Appendix D Econometric Analysis Variable Mean SD Min. Max. Latin America and the Caribbean 0.08 0.27 0 1.00 Middle East and North Africa 0.06 0.23 0 1.00 South Asia 0.14 0.35 0 1.00 Source: Independent Evaluation Group. Note: N = 666 interventions. GDP = gross domestic product; IFC AS = International Finance Corporation advisory services; IFC IS = International Finance Corporation investment services; M&E = monitoring and evaluation; Max. = maximum; Min. = minimum; PCA = principal component analysis; SD = standard deviation. SOE = state-owned enterprise. a. Factor 1 is the sum of M&E framework, supervision, and team composition. The standardized version of the factor has mean = 0 and standard deviation = 1. b. Factor 2 is the sum of analytical work, collaboration with external actors, and sequencing of operations. The standardized version of the factor has mean = 0 and standard deviation = 1. c. Factor 3 is the sum of agency cooperation or political economy, project design, and risks at appraisal. The standardized version of the factor has mean = 0 and standard deviation = 1. References Denizer, Cevdet, Daniel Kaufmann, and Aart Kraay. 2013. “Good Countries or Good Projects? Macro and Micro Correlates of World Bank Project Performance.” Journal of Development Economics 105 (November): 288–302. https://doi.org/10.1016/j.jdeveco.2013.06.003. Geli, Patricia, Aart Kraay, and Hoveida Nobakht. 2014. “Predicting World Bank Project Outcome Ratings.” Policy Research Working Paper 7001, World Bank, Washington, DC. Hamilton, Lawrence C. 2013. Statistics with Stata: Version 12. Boston, MA: Cengage Learning. Kleinbaum, David G., Lawrence L. Kupper, Keith E. Muller, and Azhar Nizam. 1997. Applied Regression Analysis and Other Multivariable Methods. 3rd ed. Pacific Grove, CA: Duxbury Press. Legovini, A., V. Di Maro, and C. Piza. 2015. “Impact Evaluation Helps Deliver Development Projects.” Policy Research Working Paper 7157, World Bank, Washington, DC. 244 Appendix D Econometric Analysis Norton, E., and B. Dowd. 2018. “Log Odds and the Interpretation of Logit Models.” Health Services Research 53 (2): 859–78. doi:10.1111/1475-6773.12712. Raimondo, E. 2016. “What Difference Does Good Monitoring and Evaluation Make to World Bank Project Performance?” Policy Research Working Paper, World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-7726. Tabachnick, B. G., and L.S. Fidell. 2001. Using Multivariate Statistics. 4th ed. Needham Heights, MA: Allyn and Bacon. Notes 31 The template had four dimensions: relevance, effectiveness, factors of implementation success, and environmental and social aspects, of which the first three were used in the econometric analysis. Relevance and effectiveness were analyzed at the state-owned enterprise (SOE) reform or intervention level, and factors for implementation success and environmental and social aspects were analyzed at the more aggregate project level. The reason behind this difference is that elements of these latter two dimensions are recorded at the project level in micro evaluative documents (for example, Implementation Completion and Results Reports, and Implementation Completion and Results Report Reviews [ICRRs]). In the factors dimension, for example, only those factors judged to be relevant for the achievement of SOE reform or intervention objectives were recorded. However, for effectiveness, although achievement of specific SOE reform or interventions is not typically recorded, tracking it in micro evaluative documents is possible. For example, whether the outcome of a development policy loan’s specific SOE reform prior action was achieved can sometimes be identified in micro evaluative content. 32 Results are available on request. 33 Sectors include financial, energy, and national level. National level captures several sector interventions, in which at least one of them is financial or energy. 34 Sub-Saharan Africa, East Asia and Pacific, Europe and Central Asia, Latin America and the Caribbean, Middle East and North Africa, and South Asia. 35 In the logistic regression analysis, the regression coefficients are calculated by using the maximum likelihood method, that is, a method that by an iterative calculation routine identifies the regression coefficients that maximize the probability of the observed data (Kleinbaum et al. 1997, 639–55). 245 Appendix D Econometric Analysis 36 See footnote 10 for an explanation of how principal component analysis (PCA) was used to identify composite factors for the analysis. 37 For example, if the probability of success of an SOE reform corporate governance intervention (p (success|corporate governance = 1)) is 60 percent, and the probability of success of an SOE reform intervention in any of the other engagement areas ((p (success|other engagement areas = 1)) is 40 percent, then the odds ratio ([0.6/0.4] / [0.4/0.6] = [1.5/0.67]) will be 2.24. If this reported coefficient is statistically significant at standard levels (such as 5 percent), one would interpret the result as “the estimated odds of success for an SOE reform intervention that engages in corporate governance is 2.24 times greater than the odds of success for an SOE reform intervention that engages in other areas, conditional on the controls included in the model.” 38 Norton and Dowd (2018, 865) remind readers that an odds ratio estimated from a multivariate logit model is conditional on the sample and on the model specification, and state that a correct, precise interpretation might be the following (using an example): “The estimated odds ratio is 1.5, conditional on age, gender, race, and income, but a different odds ratio would be found if the model included a different set of explanatory variables.” 39 The funding factor was excluded from the analysis because it was found only twice. 40 PCA finds the linear combination that explains the maximum amount of variance among the observed variables, called the “first composite factor.” It also finds another, orthogonal (uncorrelated) linear combination that explains the maximum amount of remaining variance (“second composite factor”), and so on until all variance is explained. From k variables, k principal components are extracted, which between them explain all the variance. PCA serves as a data reduction technique because fewer than k components will often explain most of the observed variance. If further work concentrates on those components, the analysis can be simplified (Hamilton 2013, 315–16). PCA also helps avoid multicollinearity issues because it exacts the common variation among individual factors. For example, if this common part is big, then the individual factors would be highly correlated, but PCA pools this common variation into a single factor, which avoids including many highly correlated factors while keeping their common variation. 41 Weekly meetings were carried out with the coding team during the implementation of the template so that all coders understood what to code and where to find relevant information, and to troubleshoot coding issues that increased the number of coded projects. This process 246 Appendix D Econometric Analysis sought the maximum number of projects and interventions coded to maximize the number of observations that entered the regressions, thereby reducing the risk of perfect collinearity. 42 Evaluated through ICRRs or Expanded Project Supervision Report evaluation notes. The econometric analysis is constrained to World Bank lending, International Finance Corporation investment services (IFC IS), and International Finance Corporation advisory services (IFC AS) projects. 43 From 894 manually reviewed financing projects, 507 were evaluated through ICRRs, Expanded Project Supervision Report Evaluation Notes, or Project Evaluation Reports. In turn, from these 507 projects, 294 World Bank lending, IFC IS, IFC AS, and Multilateral Investment Guarantee Agency projects were positives in containing SOE reform interventions (254 World Bank lending, 22 IFC IS, and 16 IFC AS, and 2 Multilateral Investment Guarantee Agency). The 292 World Bank and IFC projects account for more than half of the projected population of 553 SOE reform financing projects for the period fiscal years 2002–18 based on portfolio sample and projections described in appendix B. 44 Principal components having eigenvalues below 1 were not chosen because they explain less than the equivalent of one variable’s variance, which makes them unhelpful for data reduction (Tabachnick and Fidell 2001). 45 Factor scores are linear composites, formed by standardizing each variable to zero mean and unit variance and then weighting with factor score coefficients and summing for each factor. 46 The statistical insignificance of the third composite factor holds when the standardized version of the factor scores is used (table D.3). Additional results are available on request. 47 The likelihood ratio test statistic shows that factors 1 and 2 explain an additional 11.226 units of –2LogL (781.766 – 770.540 = 11.226), relative to use-up (change) of 2 df (16 – 14 = 2). The critical value in the chi-square statistical table for 2 df, associated with significance at p < .05 level of significance, is 5.991, and thus the team concludes that the model with the additional two factor variables substantially improves on the model without the two factors. 48 Model 3 adds the country-level predictors (income dummies, a series of dummies representing the different levels of control of corruption, and domestic credit to private sector as percentage of gross domestic product). This model captures 747.713 unit of –2LogL, which indicates that three country-level variables capture an additional 22.827 units of –2LogL (770.540 – 747.713 = 22.827), relative to the change of 6 df, which is highly significant at p < 247 Appendix D Econometric Analysis .05 (the critical value for 6 df associated with p < 0.05 level of significance is only 11.070). In other words, model three significantly improves on model 2. 49 The level of significance results of table D.3, panels a–c, are robust to changing the dependent variable to the original four-point scale measure of SOE reform achievement and estimating the models through ordinary least squares. 50 Additional controls were control of corruption, income group, and domestic credit to the private sector as a share of gross domestic product (table D.4, panel a). 248 Appendix E. State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Issues Note and Literature Review on State- Owned Financial Institutions by Thorsten Beck The question of state ownership in the financial system has been a controversial issue in academic and policy debates alike. 51 State-owned financial institutions (SOFIs) are seen as necessary to help overcome market failures and to “go” where privately owned financial institutions do not dare to go. However, SOFIs are associated with holding back the development of efficient and thriving market-based financial systems. This note surveys the theoretical and empirical literature to assess the evidence for these different hypotheses. Government involvement in the financial sector comes at different levels and through a multitude of tools and institutions, including the regulation and supervision of financial institutions, intervention in the same institutions during times of crisis, credit guarantee and other support programs, and lending requirements and prohibition of certain activities. Ownership of financial institutions is thus only one form (though a rather strong form) of government intervention into the financial system. Given poor experience with and failure of many SOFIs, governments in advanced and developing countries alike embarked on privatization—often part of more general financial liberalization—programs in the 1980s and 90s. This has resulted in improvement in performance of these banks but not necessarily in deepening of financial systems. Market failures continued to loom large. Where not liquidated or privatized, development finance 249 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries institutions were often given new mandates, including as second-tier, wholesale financial institutions. More recently, a rethinking of the role of the government as owner of financial institutions has taken place. The 2008 crisis has led to an appreciation of the role of state-owned banks in maintaining the flow of credit to the real economy at times when privately owned financial institutions withdraw. However, there has been a renewed appreciation of development banks, including as first-tier lending institutions. State ownership is especially common in less developed countries and in countries with poorly protected property rights, heavy government intervention in the economy, and underdeveloped financial systems (La Porta, Lopez‐de‐Silanes, and Shleifer 2002). Although the two schools of thought (one that focuses on market failures and one that focuses on government failures) have contrasting predictions on the role of SOFIs in the financial and economic development process, their hypotheses also have implications for the relationship between state ownership of financial institutions and financial stability. Market failures in private financial systems suggest that SOFIs should be more stable, not subject to bank runs (given explicit government guarantees) and less subject to speculative credit bubbles given developmental objectives for banks. However, agency problems and political interference in SOFIs cannot lead only to misallocation but also high losses that will ultimately result in insolvency. Although bank runs might indeed be less of a problem for SOFIs, fragility—hidden through regulatory forbearance but ultimately addressed through government recapitalization—can be prominent in SOFIs. In summary, theory suggests contrasting hypotheses on the role of SOFIs in the financial and economic development process. Over the past 20 years, a significant body of empirical evidence has been collected that tests these different hypotheses. 250 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries There is substantial evidence on performance differences between state- owned and private financial institutions. However, there is also evidence that a higher share of government-owned banks does not support financial and economic development and might even hinder it. There is some evidence that state-owned banks can have an important role in supporting access to external finance during recessions, though it is less clear what the misallocation costs are of such lending. There is evidence of higher losses at government-owned banks, often related to political interference, even though there is no clear-cut evidence on the relationship between government ownership and the incidence of financial crises. There has been a general decline in state ownership across the globe. Some of the failing banks were closed, but many were privatized to domestic buyers or to foreigners. One important driver of privatization is the need for macroeconomic stabilization. More generally, banking and government debt crises can be an important trigger for privatization of SOFIs (Cull and Martínez Pería 2008). Privatization is often associated with foreign bank entry in countries where domestic resources and banking skills are limited. However, although privatization of SOFIs is often part of larger financial liberalization and stabilization programs, often supported by multilateral lenders such as the International Monetary Fund and the World Bank, political resistance has often delayed privatization programs. Countries as diverse as the Arab Republic of Egypt, Bangladesh, and Indonesia have signaled their intent to privatize state-owned banks but have halted these attempts later on, often because of internal political constraints. Megginson (2005) concludes that bank privatization yields significant performance improvements in advanced economies. As discussed by Clarke, Cull, and Shirley (2005), although the experience with bank privatization has been, on average, positive, there has also been a lot of variation across countries. The positive effects are greater when the government fully relinquishes control, when banks are privatized to strategic investors (rather than through voucher privatization), when foreign banks are allowed to 251 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries participate in the privatization process, and when privatization takes place in competitive environments. Boubakri et al. (2005) finds that several, but not all, performance measures improved after privatization. Specifically, profitability, on average, increases, but risk also increases while their capital buffers decrease if the privatized banks are controlled by local industrial groups (thus serving as part of larger conglomerates). One important success factor in bank privatizations seems to be the subsequent ownership structure. Voucher privatization, as undertaken in the Czech Republic in the early 1990s, did not result in any performance improvement because dispersed owners were unable and unwilling to exercise any control over management. Privatization to a strategic owner, however, is more likely to lead to performance improvements, especially in the case of foreign owners. An alternative approach to establishing privately owned banking systems was to allow free entry of new privately owned banks. The Russian Federation, other former Soviet countries, and Central European countries took this approach. There was rapid growth in the number of banks. The experience, however, has been mostly negative. Many of these new banks lacked the necessary skills and scale to be sustainable. Regional or sectoral specialization resulted in fragility. Perhaps the most important impact of foreign bank entry in the former transition economies was on cutting entrenched relationships between politically connected enterprises and the banking system. Experiences in Uganda and Zambia have shown that careful privatization of these institutions can increase efficiency and stability without reducing outreach. As Megginson (2005) noted, privatization alone does not transform the efficiency of divested banks. Although it generally leads to performance improvements, these are far less than is typically observed in studies of nonfinancial industries. One of the main reasons is the continued bailout expectations and possibly government interference in newly privatized banks. In summary, even though state-owned banks will not be 252 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries able to overcome market failures in their current form, privatization is not a panacea in itself to address financial underdevelopment. Several reforms have been suggested as alternatives to privatization of SOFIs. However, the literature review found no rigorous empirical research to assess and compare the effectiveness of these different reforms. Beyond privatization or management contracts, a more systematic policy approach toward SOFIs can be helpful, as discussed by Beck et al. (2011). One model, referred to as corporatization, involves corporatizing a state institution so that it is operated as an autonomous joint-stock company while the state remains the majority stockholder and the institution is run by state entities separate from the central government administration (Tenev, Zhang, and Brefort 2002). Such corporatization also involves upgrades in corporate governance, including professional and independent board members (in addition to representatives of the government), accountability of the chief executive officer in relation to the board, an effective risk management system, and external audits. A formal ownership policy can be helpful, which includes clearly defined mandates for different SOFIs and making funding subsidies transparent. However, these are necessary and not sufficient conditions for performance improvements of SOFIs; the relationship between governments and SOFIs is not only determined by formal rules and policies but also informal norms and customs. Development banks have also played an important role over the past 50 years in developing and developed countries alike. 52 Market failures justify development finance institutions (DFIs) to help achieve social and developmental objectives. However, agency problems between owners (that is, the government) and managers of DFIs might be even larger than for commercial SOFIs, given the more limited regulation and supervision that DFIs are subject to, and even lower market and depositor discipline, given DFIs’ funding structure. 253 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Unlike for commercial state-owned banks, there is little quantitative cross- country research, given the absence of reliable data, though there is an abundance of case studies and policy papers on DFIs. The most recent survey has been undertaken by Xu, Ren, and Wu (2019), covering 499 national or subnational DFIs across 147 countries. Analysis of these DFIs shows that half of all DFIs have general development as their mission, and 40 percent have a single-sector focus and the rest a multisector focus. There seems to be an inverted U-shape in terms of the number of DFIs across different income levels—on average, middle-income economies have more DFIs than high- and low-income economies. However, the sectoral focus varies across different income groups; DFIs in high-income economies focus on the promotion of national trade, and DFIs in middle-income economies focus on infrastructure and agriculture. One important question for DFIs is on the mandate and whether it should be narrow or broad. The Business Development Bank of Canada study (Gutierrez et al. 2011) finds that the six most common target markets for development banks are microenterprises and start-ups, small and medium enterprises, international trade and globalization, housing, infrastructure, and rural and agricultural sector. Even where DFIs have a specific mandate, De Luna-Martínez and Vicente (2012) and World Bank Group (2012) report that many DFIs have a very broad mandate. A second important decision is on products. Although the focus has been traditionally on lending, other support structures, including equity stakes and guarantees, are also important products of DFIs. Guarantees might be better to target credit- constrained firms, and subsidized credit helps also non-credit-constrained firms. However, subsidized credit might be more appropriate for addressing externalities. The governance structure of DFIs is critical in determining their success. Rudolph (2009) concludes that the presence of an independent and qualified board of directors, professional management, and the South African Treasury as an active shareholder have contributed to banks’ strong 254 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries corporate governance practices. Thorne and du Toit (2009) identify six dimensions of success for DFIs, namely: (i) an enabling environment, (ii) a clear but flexible mandate, (iii) adequate regulation and supervision, (iv) effective governance and management, (v) financial sustainability without repeated recapitalization rounds, and (vi) performance assessment on a regular basis against an agreed set of financial and social or developmental objectives. Theory provides contrasting views on the role of SOFIs for financial development, financial inclusion, and financial stability. The evidence is more consistent with the government failure than the market-failure view in terms of SOFI performance and their effect for financial deepening and economic growth. Both inefficiencies and political interference drive this negative effect. There is some tentative evidence that SOFIs can help address the procyclicality of private sector lending, even though the allocative consequences are still to be assessed. Privatization of SOFIs can improve their performance; the effect of such privatization on financial sector development and stability very much depends on the broader macroeconomic, regulatory, and institutional framework in the country. What remains without doubt is the existence of market failures in both developed and (even more so) developing countries and the positive role that governments can play in helping overcome these market failures. The challenge is to identify the best instrument to do so. The past decade has seen an array of government initiatives to foster financial deepening, ranging from seed funding for M-Pesa in Kenya (provided by the UK Department for International Development) to Financiera de Desarrollo Nacional in Colombia taking a lead role in private-public partnerships for infrastructure finance. Equity funds can play a critical role, especially in low- income countries with shallow or no public capital markets. However, high risk and the fixed costs element might prevent them from entering small developing economies, which calls for support through guarantees or cost subsidies. 255 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries In summary, rather than starting the conversation with the institutional ownership debate (state-owned versus private), a more useful starting point would be to identify the market failures and the best way to overcome them, and then consider different institutional options. Issues Note on Power Sector State-Owned Enterprise Reform The various surveys of the literature reveal a reasonable consensus on the main drivers of the decision to reform the power sector. For developing countries, the main reform drivers usually listed are (i) state-owned enterprises’ (SOEs’) poor performance in relation to service quality (commercial and technical) and service coverage, (ii) generation capacity shortages, (iii) lack of financial resources to finance capital expenditures and efficient maintenance, and (iv) global institutional and regulatory efficiency (for example, governance and tariffs). When the Washington Consensus was defined, many believed that privatization and other reforms as such would solve all the issues of the public sector. This belief is no longer widespread, at least in the academic world. All recent stocktaking exercises point to the fact that the one-size- fits-all approach, the standard textbook reform model derived from the Washington Consensus, has not been as successful as initially expected. One important aspect is that institutions matter, whether the providers are public or private. Estache (2020) surveys the conceptual and empirical evidence produced by the economic literature of the essential role of institutional weaknesses as a driver of poor effectiveness of infrastructure reforms in general. Furthermore, implementation differences and the slowness of reform processes in developing countries and the incoherence of reform packages explain underperformance in many ways. There is evidence that often countries are picking up the reform components that suit them mainly because they are less politically conflictual. However, they ignore the 256 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries complementarity of the various components and that partial reforms may be counterproductive. The third recurring topic of the literature focuses on the privatization design failures, probably the most sensitive component. In the academic literature, different statistical approaches are used to review SOE reform concerns in the power sector. Bacon (2018) emphasizes that the literature he reviewed is concentrated heavily on statistical evaluations based on large panels of countries over a long time and considering a partial set of reform components. Foster et al. (2017) documents different components and factors of reform in 88 countries over 25 years (1995–2015). To get an overall picture of the extent of reform status in each country, Foster et al. (2017) also created an aggregated power sector reform index based on a scoring methodology, and the scores are based on giving equal weights to each step on each dimension of power sector reform. The authors focus on the usual four dimensions of reform (regulation, restructuring, competition, and private sector participation). Not only does it analyze a large sample of countries over a long period, but it also considers many country characteristics and offers more details and data on stages of restructuring and competition reform, degree of unbundling, and private sector participation (allowing for analysis of more combinations of factors) than most of the previous large studies. Trimble et al. (2016) focuses on determinations of the financial viability of the power utilities. The study takes the dominant SOE in the power sector of each country and estimates the quasi-fiscal deficit under two cost scenarios: current and improved operational efficiency. Camos et al. (2017) focuses specifically on the performance of the public power utilities in the Middle East and North Africa. Bacon puts in evidence that the literature of the last decade has concentrated on the following reform components as the key steps for analysis: unbundling, private sector participation, regulation, and competition. Unbundling is a key component that will also affect private sector participation and competition. Unbundling can be partial or can lead 257 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries to the complete separation of all main functions (for example, Vagliasindi 2012). In the European Union, according to the European Commission directives, the definition of unbundling means only and specifically the separation of distribution from the other activities. Unbundling is a perfect example of a decision with a big impact on the sector organization but that could face resistance from the incumbent SOEs. The financial characteristics may matter. For instance, it is important to assess the extent that this will end cross-subsidies across production stages that will render one of the stages unsustainable without subsidies. Coordination matters more generally. Potential loss of coordination among the main functions have to be compensated by adequate system operations while also considering the technological evolution. Otherwise, it could have negative impacts on the system performance, whether the operator is public or private. Private sector participation can take various forms and involve different functions of the power market (generation, transmission, distribution, and retail). It does not require the unbundling of these functions, but in that case, the public monopoly on all functions will be replaced by a private monopoly. Regarding regulation, Bacon (2018) points to the existence of a set of four indicators typically used by different academics to build a regulatory index: existence of specific electricity law, autonomous regulator, own financing for regulator, and pay scales. Regarding competition, competition in the market and for the market have to be distinguished. Competition in the market requires unbundling and privatization. Competition for the market does not require unbundling because it could be achieved through the privatization of a vertically integrated power SOE though an efficient auction. Bacon (2018) concludes that the results on the effectiveness of changes of ownership and of reforms are quite mixed, even if some findings are shared by a majority of the studies reviewed. Jamasb et al. (2015) reached the same 258 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries conclusions and gave various reasons for that situation related to the inherent difficulty in capturing reform factors through statistical methods, including the availability of data samples that capture a mix of reform experiences, and the difficulty in isolating the effects of particular reform steps. Regarding private sector participation, many (but not all) studies suggest that privatization is often associated with improvements in sector performance with emphasis on labor performance and operational efficiency. Another important factor linked to private sector participation relates to the different types of privatization and their different effects. The heterogeneity of the contract types has to be considered. Another relevant result is that the effects of regulation and competition are more ambiguous. Foster et al. (2017) find that 40 percent of the developing countries remain largely unreformed, and these countries are mostly fragile states, low- income countries, countries with weak rule of law, or countries with small systems. The analysis also shows that 34 percent of the 88 developing countries reviewed have some private participation (most probably in generation) without unbundling. The analysis confirms the importance of institutional strength and the existence of a problem of pick-and-choose. Some components of reform have indeed been more popular than others, such as the creation of regulatory entities (the most implemented) and some forms of private sector participation mainly in generation with the introduction of independent power producers. Politically, these components are not that challenging to implement, even in the case of the regulatory agency that theoretically takes powers away from the ministry. This removal of powers can be easily mitigated by limiting the powers and prerogatives of the agency and by controlling its budget and the selection of its managerial staff. A last observation relates to the sequence of implementation of reform components: power sector reforms end up being packaged in ways unrelated to the original logic. Trimble et al. (2016) finds that combined network and collection losses on average represent a larger hidden cost and are less politically sensitive to address than underpricing, so it could be a smart area 259 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries for policy focus to reduce quasi-fiscal deficits. Underpricing remains an issue to address over the medium term, as service quality improves. Huenteler et al. (2017) finds that for the majority of the population in many countries, the main political barrier to tariff reform is not affordability or inflationary concerns but the political economy of electricity subsidies. Camos et al. (2017) concludes that the median quasi-fiscal deficit is much larger than in Sub-Saharan Africa (median value of 3.9 percent), and this higher value is almost entirely due to the much greater degree of underpricing (six times higher in percent of gross domestic product than in Sub-Saharan Africa). Their findings on ownership of the utilities (public or private) is a small part of the results, but the study finds private ownership of distribution companies to be significantly positively correlated with cost efficiency, labor efficiency, and the return on equity indicators. However, it was not correlated with system and operational efficiency, losses efficiency, cost structure, cost recovery, and balance sheet indicators. Bacon (2018) points to a few important gaps in the studies. An important one is that reform policies come in packages. This means that looking at partial reforms can be misleading. Another important omission in the studies reviewed concerns the relevance of the degree of reform, the required implementation of all or some of the usual components (unbundling, private sector participation, regulation, and competition), and of the logical sequence of introduction of each of these components. Another important gap is the fact that the literature studying the macro links of the reforms seems scarce (Jamasb et al. 2015). A few recent papers not covered in Bacon’s review also discuss issues that need to be considered for the next wave of reforms to improve SOEs’ performance. Gore et al. (2019) focuses on Sub-Saharan Africa but brings additional global perspectives and hints related to the political factors. The analysis focuses on the timing, pace, and extent of reforms. The main conclusions are the following: (i) The timing of reform was largely contingent on economic factors, primarily the need for financing to improve 260 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries sector efficiency and access rate; (ii) international political considerations— namely, differences in the countries’ reliance on aid from key donors— largely shaped the pace of reform; and (iii) internal national political factors primarily determined the extent of reform. Gore et al. (2019) emphasize the conditionality of power sector reform to obtain loans and grants from multilateral development banks such as the World Bank. The authors also show that there is no obvious direct relationship between the level of reliance on donors’ aid and the volume and pace of reforms. Imam, Jamasb, and Llorca (2018) focus on the impact of reforms on the corruption that characterizes the sector in Sub-Saharan Africa. In terms of the impact of the two reform components used (independent regulator and private sector participation), the authors find mixed effects on a range of variables of interest. On the positive side, both components increased global electricity consumption, although they reduced technical efficiency compared with a context in which SOEs regulated by the ministry are the main providers. On the less positive side, they find that the privatization policies adopted by the Region had no statistically significant impact on access rates in the Region. In other words, SOEs and private providers deliver access at the same rate. The creation of independent regulators was more effective on this front because it reduced the adverse association between corruption and access rates, though at the cost of reinforcing the negative association between corruption and technical efficiency. Private sector participation also did not do any better than SOEs on the association between corruption and access rates and technical efficiency. Sen, Nepal, and Jamasb (2016) confirm insights learned from the assessment of other regions and add other insights learned from electricity reforms in Asia. The paper shows that the most popular reform components are allowing independent power producers (16 of the 17 countries covered) and setting up regulators (15). The less popular ones are allowing third-party access (5 countries) and privatization of distribution (4 countries). The authors observe that in adopting independent 261 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries power producers, the investment risk was transferred to utilities and in some cases ultimately consumers (through higher tariffs) through the take-or-pay clauses prevalent in many contracts. De Halleux, Estache, and Serebrisky (2019) focus on the impact on various standard policy indicators (efficiency, equity, and accountability) of the adoption of governance reforms. The authors find that reforms could be statistically significantly associated with higher technical quality but not social or service improvements. Kufeoglu, Pollitt, and Anaya (2018) document the status of power distribution in the world and discuss some options for the future of this system function. The authors state that as of 2015, only 189 distribution system operators of about 2,400 were legally unbundled in Europe. In Europe, full-ownership unbundling is required by law only in the Netherlands (Council of European Energy Regulators 2016). Given the technological evolutions, Kufeoglu, Pollitt, and Anaya (2018) note that coordination of the transmission system operators and the distribution system operators and the allocation of activities and responsibilities are among the top priorities of the electricity industry. Overall, this paper emphasizes that the reform decisions in the sector also need to account for the technological and service evolutions. Urpelainen and Yang (2019) track the global patterns of reforms in the sector that cover long periods during which countries switched ownership type and implemented various types of reforms. The authors find that allowing independent power producers and setting up regulators are the most popular reforms before privatization and introducing competition. The implementation gap between these reform components is wider for relatively poor and authoritarian countries with low institutional capacity. 262 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Table E.1. Main Issues to Consider in State-Owned Enterprise Reform in Power Sector Issue Discussion Data Multilateral development banks currently cannot rely on large power utilities’ databases to be regularly updated with organizational, regulatory, financial, and operational data to better inform their reforms’ analysis and proposals. Tariff system All the three pillars of tariff systems (cost recovery, tariff structure, and cost indexation and pass-through mechanism) should be better considered—especially the tariff structure— and not only the average cost recovery level. Economic regulation More analysis should be dedicated to the economic regulation of power SOEs —such regulation is not only relevant for private operators. Professionalization Multilateral development banks should consider reenergizing and increasing the offer of large, wide-spectrum professionalization packages for power SOEs because the business and market environment has drastically changed in the last 20 years. Internal and external More information, research, and support are required on SOEs governance internal (management) limitations and on external governance processes and tools to inform the SOEs’ performance diagnostic and decisions for reform. Unbundling Dogmatic decisions should be avoided regarding unbundling, especially for the separation of transmission and distribution. Not only sector size but also potential losses of coordination and economics of scale, transaction costs, technological evolutions, and so on, should be considered. Cost of capital More research is required on power SOEs’ and cost of capital and the impact of private sector participation on the service’s cost of capital and eventually tariffs. Macro links Macro factors need to be better considered in power sector reform, such as the government’s access to concessional loans and its budget trade-offs, labor, and educational factors. Note: SOE = state-owned enterprise. 263 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Issues Note on Privatization By John Nellis Starting with the World Bank’s first operations in developing countries, 53 its projects usually contained institutional and/or managerial measures to strengthen the capacities of the infrastructure public enterprises and project management units involved. 54 Initial public enterprise reform actions stressed improving operational and financial performance through means other than ownership change. By the mid-1980s, the recognition grew among many World Bank staff and management, and in client countries, that World Bank–assisted public enterprise reforms were not producing the needed results. A more drastic approach was required. Poor results of past operations were an important reason for the shift, but there were other contributing factors, including the change in the previously prevailing social democratic tone of political discourse, the collapse of the European–Central Asian communist economies (and that of the USSR in particular), and wider discussion of the term and process of “globalization.” Reasoning Supporting Privatization The arguments for increased private involvement were based on more than the failures of public ownership reform, the political context, and exhortation. Private ownership would, supposedly, improve public enterprise performance because it creates a market for managers, an area of noted deficiency in public enterprises. Furthermore, capital markets subject privately owned firms to greater financial scrutiny and discipline than governments do their public enterprises. Another reason is the fact that public officials interfere less in the workings of private firms than they do public enterprises. Finally, private firms are supervised by self-interested board members and shareholders rather than by (theoretically) disinterested bureaucrats (Nellis 1994). 264 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Privatization Takes Center Stage The heyday of enthusiasm for privatization, inside and outside the World Bank, was about the period of 1990–2005. In the 90s, the scope and pace of divestiture was to grow greatly in the Organisation for Economic Co- operation and Development (OECD) 30 except for Luxembourg, Norway, Switzerland, and the United States. This process was to continue in these countries, at least up to the time of the financial crisis in 2008–09. This partly was in response to the European Union’s limitations placed on direct state aid by member states to enterprises operating in competitive markets. The World Bank’s concern was public enterprise reform and divestiture in its client countries. After 1988, divestiture components of World Bank operations became more numerous, more expansive and more demanding. Many more loans included measures to transform public enterprises by means of management contracts, joint ventures, and concessions to private operators; corporatization followed by offerings of shares; and the sale of ownership, partial or full. Fourteen percent of public enterprise–related loans in the 80s contained a divestiture component, and the incidence rose to 52 percent of operations in the 1990s (World Bank 2005). By the mid-1990s, some 30 unleashed or new states had arisen from the communist ashes, and most of them embarked on privatization programs of one sort or another, the vast majority with assistance from the World Bank. Outside the OECD countries and a few outlying countries, the World Bank was deeply implicated in the privatization process as promoter, instigator, financier, implementer, and evaluator. Positive First Results In the mid-1990s, detailed assessments of the first wave of privatizations in non-OECD settings began to appear (Boubakri and Cosset 1998; Galal et al. 1994; Havrylyshyn and McGettigan 2005; La Porta and Lopez-de-Silanes 1999; Megginson and Netter 2001). All were positive, showing widespread 265 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries and significant performance improvement in the studied privatized firms as measured by productivity gains, profitability, return on sales, and other indicators. Despite the glowing academic reviews, privatization never achieved the status of a panacea for troubled public enterprises, though critics insisted that the World Bank regarded it as such. Many World Bank analysts had long acknowledged the problems that could arise from privatizing firms— especially infrastructure firms in weak legal, regulatory, and institutional settings—and they continued to argue their viewpoint. But the need for improved performance was usually judged as sufficiently urgent to outweigh their concerns and justify an emphasis on speedy transformation. A Shift in Perspective Well before the end of the 1990s, criticisms of privatization and the World Bank’s involvement in the process emerged. These concerns were of three main types: issues of evaluation of results, issues of implementation, and issue of unintended consequences. Concerning evaluations, critics acknowledged that the early positive findings were real, but the improvements may not have been due to ownership change or ownership change alone but might be attributable to the concomitant liberalizing shifts. Furthermore, the good results might have arisen through “selection bias.” Perhaps it was not that privatization made bad firms into good ones, but rather it was the good firms that had been chosen to be privatized. Regarding implementation, the main argument was that much more positive and sustainable outcomes could have resulted from privatization, especially in infrastructure firms, had market liberalization, regulatory, legal, and institutional reforms preceded ownership change. It was problems of consequences that most affected the World Bank’s efforts. Private involvement in infrastructure quickly encountered several problems: institutionally weak governments had very great difficulty creating, monitoring, and enforcing the detailed contracts that guided lessees, 266 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries concessionaires, management contractors, and independent power project operators. The large Russian mass privatization program that seemed well launched in 1994 slowed and became much less transparent and much more politicized. The infamous oligarchs rose to visible prominence after 1995. They manipulated the system to become the majority owners of the great mass of privatized assets, usually paying very little for them. By the early years of the 21st century, the stylized facts on privatization were these: A shift to private ownership of a previously state-owned and operated firm usually led to improved financial and operational performance. Improvements were more likely to occur and endure where private enterprises were divested into competitive or potentially competitive markets. These findings were strongest in middle- to higher-income countries possessing an adequate or at least modicum institutional framework. Privatization less often lived up to the expectations in infrastructure divestitures, particularly in low-income, institutionally weak settings. Privatization began to lose the popularity and acceptance it might have once had because of miserable past public enterprise performance. By 2005, this state of affairs was evident to all, including decision makers at the World Bank. Taking note of the difficulties of infrastructure privatization in low-income countries, the well-publicized lurid tales of corrupt and ineffective divestitures, and the extent and intensity of anti-privatization pushback among borrower government officials and populations, the World Bank shifted its tone on public enterprise reform away from privatization as a first-best option and back to a more agnostic position regarding the importance of the ownership question. Megginson concluded, “Through the early 21st century, there was an unambiguous global trend toward reducing government ownership of business enterprise, but this trend has since at least been slowed, and perhaps even reversed” (Megginson 2017, 1). One important catalyst for this 267 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries shift in approach was the dramatic rise of the distinctly different, evolutionary Chinese road to privatization. China Privatizes China started to reform its gigantic public enterprise sector gradually in the early 1980s. In the early 1990s, somewhat more concrete steps were taken. China introduced a stock market; the aim was to sell minority shares to Chinese citizens in majority government-owned and operated firms. Government control was not—or only partially—ceded, but it became apparent that private initiative was being tolerated and encouraged in a variety of sectors because of the need for increased efficiency and production. At the same time, government continued to shield core public enterprises from competition and cost-cutting measures to maintain employment, social stability, and political control. This ambiguous ownership policy was associated with an excellent and sustained rate of growth without (after 1989) major challenges. The sale of minority of shares in “corporatized” Chinese public enterprises, with decision-making on major issues remaining in the hands of state- appointed officials, became the hallmark of the Chinese approach. After 2005, China made all divested shares tradeable and the insider shares sellable. Both measures boosted the privatization process considerably. A recent survey of 80 mega-privatization transactions—that is, all those raising more than $5 billion per transaction—in the decade 2005–15 reveals that Chinese divestitures accounted for 16 of the sales, with a total value of $148 billion. Divestiture has generally produced improved performance at the level of the firm. The larger the percentage of equity divested, the greater the performance improvement. The more private the new owners and the less direct government intrusion, the more performance improves. 55 China has continued to privatize public enterprises mainly by allowing them “to raise capital by selling newly issued primary shares to investors, thus 268 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries diluting state ownership only indirectly, rather than having the state sell its existing shareholdings directly” (Megginson 2017, 9). The Chinese state exercises its authority through retention of a large public enterprise sector and control of the banking and financing systems for all firms: public enterprises, private, and partially private firms. State officials still serve on the boards and controlling bodies of the large and important partially privatized entities. The state plays active policy and lender roles. The Chinese success with this mixed approach suggests to both World Bank client countries and the World Bank itself that there is a viable alternative policy path to economic progress that involves neither rigid austerity nor complete surrender of state control over enterprise direction. Moreover, since 2013, China has been promoting its approach through a rapidly increasing aid and investment program in developing countries. Total Chinese aid and external investment currently averages about $40 billion per year, much of it going to infrastructure finance. A Swing Away from Privatization Between 2000 and 2005, concern grew in client countries and the institution over the World Bank’s comparatively unnuanced stress on privatization. In response, more attention was paid to public-private partnerships and policy and project risk guarantees. The World Bank also rethought and redoubled its efforts to improve corporate governance inside the public enterprises, aiming to improve government policy toward and supervision of public enterprises and the quantity and quality of public enterprise products and services, aid cost recovery and control inside the enterprise, foster expansion of networks to better serve the previously excluded, and improve firm-level management. Conclusion The conclusions of Megginson’s (2017) most recent and extensive survey of research on the effects of privatization of SOEs are as follows: (i) There has 269 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries been an explosion of rigorous research on the issue since 2005 (Megginson reviewed more than 100 articles published after that date); (ii) all 17 surveyed studies focusing on before and after performance “document significant improvements after companies are divested . . . all the [other] empirical studies surveyed show that private ownership is much more efficient than state ownership, sometimes massively so” (137–8); (iii) government decisions regarding what and when to divest are always intensely politicized; (iv) “most forms of pre-divestment corporate restructurings reduce net prices received” (138); (v) all but one of the studies of bank privatization “show that state ownership distorts banking decisions, capital allocation efficiency, and/or the arms-length provision of credit to firms with the most promising investment prospects” (139); (vi) “political connections…are beneficial for the companies involved, but these private benefits are usually associated with significant costs for the overall economy and financial system” (139); (vii) “…state ownership has a generally distortive effect on corporate financial policies, most importantly capital investment spending” (139); (viii) “government guarantees of private financial transactions and bailouts of failing private firms are inherently distortive”; and (ix) “state capitalism is an inherently failed model, at least for all but the most underdeveloped economies... the economic rise of China…made this seem a plausible model for development, but the abysmal relative performance of state-controlled versus private firms in key industries clearly shows the model’s inherent weakness” (140). One sees at once that the situations in which privatization is more difficult to launch, more difficult to implement, and less likely to yield the anticipated benefits are precisely the situations and circumstances in which many World Bank client countries are found. Many borrower states still lack that degree of institutional, legal, and managerial capacity required to make a success of full-scale privatization, especially in water, energy, banking, and transport. The World Bank and its clients search for public enterprise reform methods that are less contentious, and more palatable socially and 270 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries politically, than divestiture. Yet performance problems persist in many of the large remaining public enterprise sectors. Public enterprises continue draining government budgets, infecting the financial and banking systems, and straining client countries’ scarce administrative resources. Although the recent focus on increased institutional capacity may turn out to be the ultimate solution to the question of development, the least one can say is precisely how a society moves from a state of institutional weakness to one of strength remains something of a mystery, and what is known is that the process of institutional growth is more evolutionary than revolutionary, meaning it will take a considerable amount of time for increased institutional capacity to incubate and take root, even when the formative actions are the “right” ones. To date, persuasive evidence of the superiority of the non-ownership– related reform program seems lacking. Thus, until it is solidly established by rigorous research that the present approach to public enterprise reform is producing improvements of the needed size, in a reasonable time frame, it would be incorrect of the World Bank to dismiss privatization from its inventory of possible actions. Back to the Future? The World Bank has not totally turned away from recommending and supporting privatization of public enterprises, at least in those cases when and where the client is on board and the circumstances appear to justify the action. Indeed, support for divestiture may soon return to prominence as the World Bank policy pendulum has quite recently swung back toward greater enthusiasm for more direct private involvement in development operations in general. The needs for investment capital in developing countries still greatly exceed current projected demand. This is especially true for infrastructure creation and renewal. The Maximizing Finance for Development (MFD) initiative, 271 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries launched by the Development Committee of the Boards of the World Bank and the International Monetary Fund in 2017, aims to “reserve scarce public financing for those areas where private sector engagement is not optimal or available. This means . . . testing—and advising clients on—whether a project is best delivered through sustainable private sector solutions” (World Bank 2017). The World Bank envisions its role as “supporting governments to crowd in the private sector to help meet development goals.” MFD indicates a substantial shift in approach and emphasis from that put forward in 2005. The clear implication is that in most situations, the World Bank will look first at the efficacy of private involvement as creditor, manager, or partial or full owner. Despite the opposition that will arise to MFD, the studied ambiguity of the initial MFD documents seems a reasonably judicious way forward for an institution that needs to harness private sector dynamism and resources but tame its rougher edges to fit its clients and the watchful public’s needs. Deep Dive on Corporate Governance This deep dive covers the topic of corporate governance in SOES. The methodology includes the following elements: (i) focused literature review, (ii) portfolio review and data analysis, (iii) synthesis of draft case studies and issues notes, and (iv) synthesis of the role of stakeholders. The summary of the key findings follows. For Claessens and Yurtoglu (2012), the relationship between corporate governance and development is based on the following channels: (i) increased access to external financing, (ii) lowered cost of capital, (iii) improved operational performance, (iv) reduced risk of financial crises, and (v) improved stakeholder relationships. The significance of these channels has grown as a result of the increasing role of market-based investment processes in global economies. 272 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Muir and Saba (1995) state that there are internal organization incentives and external structural arrangements that determine the performance of all corporations. These are highlighted through common elements identified in successful SOE reform in five case countries. Through an empirical study, Heo (2018) illustrates the relationship between governance and performance of SOEs using data from 320 SOEs in the Republic of Korea. Results suggest that there is a positive relationship between (i) board size and indicators of financial performance, (ii) transparency and disclosure and indicators of financial performance, and (iii) corporatization and customer satisfaction and the firm’s debt ratio. Although this demonstrates some significance in the relationship between SOE governance and performance, endogeneity is an issue. Focusing on the SOE management and the composition and independence of boards, Jurkonis, Merkliopas, and Kyga (2016) find that there is a significant relationship with returns on equity. They also demonstrate management practices—which are isolated from policy formation and instead pursue goals related to effectiveness—have an important impact of quantitative returns. These results emphasize the importance of reforms pushing for the full independence of SOEs and their boards. In the financial sector, Berger et al. (2005) assess changes in performance resulting from changes in ownership in Argentina. State-owned banks tend to have poorer performance in the long term than domestic and foreign private banks, and privatization appears to improve performance through a decline in nonperforming loans. However, it is unclear if these improvements are sustainable because trends demonstrate a deterioration in performance. The restructuring of state-owned banks also improves performance in the same way but to a lesser extent. Still relevant to the financial sector, Rudolph (2009) highlights that a significant gap exists in the relevant literature, specifically in the 273 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries measurement of public policy performance of SOFIs. More research is needed to assess the achievement of SOEs’ policy objectives. In the power sector, the World Bank (2019) finds based on evidence that good governance practices are strongly associated with improvements in cost recovery and operational efficiency of distribution utilities. Key findings on restructuring and governance improvement of power utilities from 15 countries are summarized. As discussed in the World Bank Group corporate governance tool kit, the key difference in governance challenges faced by the private and the state- owned sector lies in the multiplicity of stakeholders and objectives in SOEs (World Bank 2014). Wong (2004) finds that SOE-specific challenges in corporate governance revolve around their operating model, including (i) the need to balance commercial and policy objectives, (ii) risk that political stakeholders will abuse their authority for self-interested reasons, and (iii) fewer tools available to incentivize and discipline employees. Improvements in corporate governance reforms are often informed through anecdotal evidence and at times through empirical information. According to the World Bank toolkit, benefits for SOEs include improved performance, increased access to alternative sources of financing, using alternative models in financing of infrastructure, reduced fiscal burden, and decrease in corruption and transparency. A few frameworks exist to provide guidance for SOEs to implement such corporate governance reforms. These include the World Bank Group Corporate Governance Toolkit (2014), which builds on broader OECD guidelines. Milhaupt and Pargendler (2017) are critics of such prescriptions because of (i) the lack of practical guidance on institutional practices needed to achieve objectives listed, (ii) the assumption that centralization results in improved governance, and (iii) the idea that recommendations emphasizing “active 274 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries ownership” and recommendations condemning political interference are contradictory for the state. The following country case studies were identified as relevant to corporate governance reforms: China, Kenya, Mozambique, Serbia, Ukraine, and Vietnam. In China and Vietnam, changes in ownership and corporatization were used as a tool to improve corporate governance and firm performance in the early 2000s. In Mozambique, reforms were focused on improving the transparency of state operations in the mining sector. Bangladesh and Ukraine provide good examples of the importance of good governance in the state-owned financial sector, with a specific focus on the independence of boards. In Serbia, corporate governance reforms were put in place to precede privatization. The overall corporate governance portfolio is composed of 237 projects and 282 unique interventions. Fifty percent, or 140 of the total of 282 interventions, are evaluated for their effectiveness, and 319 factors are identified as relevant to projects involving some form of corporate governance reform. Overall, the vast majority of interventions are part of World Bank lending projects as opposed to International Finance Corporation (IFC) investment services and advisory services. For all IFC projects, all but two advisory services interventions targeted downstream reforms. However, given that some projects addressed both, World Bank lending interventions are relatively evenly divided between upstream (55 percent) and downstream (55 percent).  Comparing the nature of interventions for the SOE sectors selected, the team finds that at the national level, a vast majority (88 percent) of interventions are upstream, and financial and energy sector interventions tend to target downstream reforms, with 60 percent for each sector. The most common mechanism in the overall portfolio is risk management (n = 54), and this is also true at the downstream level (n = 275 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries 32), but at the upstream level, the most common mechanism is financial reporting (n = 33; figure E.1).  Overall, 74 interventions (58 percent) were rated as achieved, 25 (19 percent) were rated as mostly achieved, 19 (15 percent) were rated as mostly not achieved, and 9 (7 percent) were rated as not achieved. The number of evaluated interventions for IFC are relatively small, with 2 interventions for IFC advisory services and 6 for IFC investment services. Seven of the 8 IFC interventions (87.5 percent) are considered to have achieved their objectives, and only 1 was considered mostly not achieved. Figure E.1. Top Seven Corporate Governance Mechanisms in World Bank Group SOE Reform Portfolio Source: Independent Evaluation Group review. Note: cg = corporate governance; SOE = state-owned enterprise; stds = standards; mgt = management. 276 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Deep Dive on Extractives Sector This deep dive covers the state-owned extractives sector, in accordance with the structure established by the Independent Evaluation Group for deep dives for its evaluation of Bank Group support for reform of SOEs. The methodology will include the following elements: (i) focused literature review, (ii) portfolio review and data analysis, (iii) synthesis of draft case studies and issues notes, and (iv) synthesis of the role of stakeholders. The summary of the key findings follows. According to Halland, Lokanc, and Nair (2015), the oil, gas, and mining sector faces a unique set of environmental, social, and economic challenges while also involving a wide range of stakeholders. For this reason, the efficient development of this sector requires strong legal and regulatory foundations, regardless of ownership. According to Cameron and Stanley (2017), common features of the extractives industry contribute to the sector’s complexity. However, key differences within the extractives industry—between oil and gas and mining—also contribute to the difficulty of identifying clear, cross-cutting solutions to improve the sector’s performance. State control is most relevant to the oil and gas sector, with approximately 90 percent ownership over reserves and 75 percent ownership over production. In mining, SOEs have historically been less influential. State ownership in the sector is very much motivated by political reasons instead of commercial ones. In Victor et al. (2012), Warshaw highlights that political economy and institutions play an important role in defining ownership in the sector. Wolf (2009) assesses the role of ownership on performance of oil and gas firms. Through a panel data regression, he finds that national oil companies (NOCs) underperform significantly in terms of profitability and output efficiency when compared with privately owned international oil companies. 277 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Wolf also highlights limitations on the quality of data and financial information disclosed by key Organization of the Petroleum Exporting Countries member countries compared with other state-owned extractive companies. Contrary to Wolf (2009), Victor et al. (2012) find that the performance of NOCs varies. Four key factors explain the difference in performance: (i) state goals, (ii) geology, (iii) state-NOC interactions, and (iv) management strategy, listed in order of importance. For these authors, the role of NOCs is unclear—much of what could be nationalized already is, and the most significant trend that will affect future performance is the increasing complexity of natural resource extraction. Tordo, Tracy, and Arfaa (2011) highlight that internal governance mechanisms relating to managerial and technical capacity are important to value creation, and contrary to general belief, the pursuit of national objectives is not necessarily a constraint for this—it can instead enhance value creation when clearly defined and captured correctly. Given the multiplicity of objectives and the strong potential for adverse environmental and social impacts, a number of organizations and resources are available for the extractives sectors as a whole, with additional guidance provided for SOEs. Some of these include the Extractive Industries Transparency Initiative Standard and the Natural Resource Governance Institute Resource Governance Index and its Guide to Extractive Sector State-Owned Enterprise Disclosures. Country case studies from this evaluation highlight a significant number of relevant interventions in Egypt, Kenya, Ukraine, Vietnam, Serbia, and Mozambique. The first four countries listed are relevant to the oil and gas sectors, with the objectives to address unreliable supply of natural gas and support the industry to improve electricity supply, prices, and/or reliability. In Serbia, reforms were very much targeted at improving the performance of the state-owned JP Srbijagas to support the company’s financial 278 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries stabilization. In Mozambique, reforms supported mining industries, with Bank Group support to Extractive Industries Transparency Initiative Standard compliance. The portfolio review analysis conducted by the Independent Evaluation Group identified 65 reform interventions through 34 unique projects. A majority of these (94 percent) are World Bank lending interventions in oil and gas with upstream interventions. The Region with the highest number of interventions is Sub-Saharan Africa, with a total 38 interventions approved and evaluated between 2008 and 2018 and 56 percent of targeted reforms achieved successfully. All other Regions except for Middle East and North Africa perform better than Sub-Saharan Africa. The most relevant positive factors in the subsector are (i) client commitment, (ii) design, and (iii) supervision. The most significant negative factors are (i) external shocks, (ii) monitoring and evaluation framework, (iii) public sector capacity, (iv) project design, and (v) client commitment—identified as relevant five times each. References Bacon, Robert. 2018. “Taking Stock of the Impact of Power Utility Reform in Developing Countries: A Literature Review.” Policy Research Working Paper 8460, Rethinking Power Sector Reform, World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-8460. Beck, Thorsten, Samuel Maimbo, Issa Faye, and Thouraya Triki. 2011. Financing Africa: Through the Crisis and Beyond. Washington, DC: World Bank. https://documents.worldbank.org/en/publication/documents- reports/documentdetail/633671468194645126/financing-africa-through- the-crisis-and-beyond. Berger, Allen, George Clarke, Robert Cull, Leora Klapper, and Gregory Udell. 2005. “Corporate Governance and Bank Performance: A Joint Analysis of the Static, Selection, and Dynamic Effects of Domestic, Foreign, and State Ownership.” Journal of Banking and Finance 29 (8–9): 2179–221. https://doi.org/10.1016/j.jbankfin.2005.03.013. 279 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Boubakri, Narjess, and Jean-Claude Cosset. 1998. “The Financial and Operating Performance of Newly Privatized Firms: Evidence from Developing Countries.” The Journal of Finance 53 (3): 1081–110. https://www.jstor.org/stable/117387. Boubakri, Narjess, Jean-Claude Cosset, Klaus Fischer, and Omrane Guedhami. 2005. “Privatization and Bank Performance in Developing Countries.” Journal of Banking and Finance 29 (8–9): 2015–41. https://doi.org/10.1016/j.jbankfin.2005.03.003. Cameron, Peter D., and Michael C. Stanley. 2017. Oil, Gas, and Mining: A Sourcebook for Understanding the Extractive Industries. Washington, DC: World Bank. http://hdl.handle.net/10986/26130. Camos, Daniel, Robert Bacon, Antonio Estache, and Mohamad Mahgoub Hamid. 2017. Shedding Light on Electric Utilities in the Middle East and North Africa: Insights from a Performance Diagnostic. Washington, DC: World Bank. https://doi.org/10.1596/978-1-4648-1182-1. Claessens, Stijn, and Burcin Yurtoglu. 2012. Focus 10: Corporate Governance and Development—An Update. Washington, DC: International Finance Corporation. Clarke, George R. G., Robert Cull, and Mary M. Shirley. 2005. “Bank Privatization in Developing Countries: A Summary of Lessons and Findings.” Journal of Banking and Finance 29 (8–9): 1905–30. https://doi.org/10.1016/j.jbankfin.2005.03.006. Council of European Energy Regulators. 2016. Status Review on the Implementation of Distribution System Operators’ Unbundling Provisions of the 3rd Energy Package. CEER Status Review C15-LTF-43-03. Brussels: CEER. https://www.ceer.eu/documents/104400/-/-/882514d5-c57f-86f8-50a3- 90185e270f15. Cull, Robert, and María Soledad Martínez Pería. 2008. “Crises as Catalysts for Foreign Bank Activity in Emerging Markets.” In Power and Politics after Financial Crises, edited by Justin Robertson, 52–81. London: Palgrave Macmillan. https://doi.org/10.1057/9780230235366_3. De Halleux, Morgane, Antonio Estache, and Tomas Serebrisky. 2019. “Governance Choices and Policy Outcomes in the Latin American and Caribbean 280 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Electricity Sector.” Working Papers, ECARES 2019-08, Universite Libre de Bruxelles, Bruxelles, Belgium. De Luna-Martínez, José, and Carlos Leonardo Vicente. 2012. “Global Survey of Development Banks.” Policy Research Working Paper 5969, World Bank, Washington, DC. World Bank Group. 2018. 2017 Survey of National Development Banks. Washington, DC: World Bank Group. http://documents.worldbank.org/curated/en/977821525438071799/2017- Survey-of-National-development-banks. Estache, Antonio. 2020. “Institutions for Infrastructure in Developing Countries: What We Know and the Lot We Still Need to Know.” In The Handbook of Economic Development and Institutions, edited by J. M. Baland, F. Bourguignon, J. P. Platteau, and T. Verdier, 634–88. Princeton, NJ: Princeton University Press. Foster, Vivien, Samantha Witte, Sudeshna Ghosh Banerjee, and Alejandro Moreno. 2017. “Charting the Diffusion of Power Sector Reforms across the Developing World.” Policy Research Working Paper 8235, Rethinking Power Sector Reform, World Bank Group, Washington, DC. http://documents.worldbank.org/curated/en/576801510076208252/Charting -the-diffusion-of-power-sector-reforms-across-the-developing-world. Galal, Ahmed, Leroy Jones, Pankaj Tandon, and Ingo Vogelsang. 1994. Welfare Consequences of Selling Public Enterprises: An Empirical Analysis. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/517121468052737074/Welfare- consequences-of-selling-public-enterprises-an-empirical-analysis-a- summary. Gore, Christopher D., Jennifer N. Brass, Elizabeth Baldwin, and Lauren M. McLean. 2019. “Political Autonomy and Resistance in Electricity Sector Liberalization in Africa.” World Development 120 (August): 193–209. https://doi.org/10.1016/j.worlddev.2018.03.003. Gutierrez, Eva, Heinz P. Rudolph, Theodore Homa, and Enrique Blanco Beneit. 2011. “Development Banks: Role and Mechanisms to Increase Their Efficiency.” Policy Research Working Paper 5729, World Bank, Washington, DC. 281 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Halland, H., M. Lokanc, and A. Nair.. 2015. The Extractive Industries Sector: Essentials for Economists, Public Finance Professionals, and Policy Makers. Washington, DC: World Bank. https://doi.org/10.1596/978-1-4648-0492-2. Havrylyshyn, Oleh, and Donal McGettigan. 2005. “Privatization in Transition Countries.” Post-Soviet Affairs 16 (3): 257–86. https://doi.org/10.1080/1060586X.2000.10641488. Heo, Kyoungsun. 2018. “Effects of Corporate Governance on the Performance of State-Owned Enterprises.” Policy Research Working Paper 8555, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/523421534424982014/Effects- of-corporate-governance-on-the-performance-of-state-owned-enterprises. Huenteler, Joern, Istvan Dobozi, Ani Balabanyan, and Sudeshna Ghosh Banerjee. 2017. “Cost Recovery and Financial Viability of the Power Sector in Developing Countries: A Literature Review.” Policy Research Working Paper 8287, Rethinking Power Sector Reform, World Bank, Washington, DC. Imam, Mahmoud I., Tooraj Jamasb, and Manuel Llorca. 2018. “Sector Reforms and Institutional Corruption: Evidence from Electricity Industry in Sub-Saharan Africa.” MIT CEEPR Working Paper Series 2018-006, Massachusetts Institute of Technology Center for Energy and Environmental Policy Research, Cambridge, MA. Jamasb, Tooraj, Rabindra Nepal, and Govinda R. Timilsina. 2015. “A Quarter Century Effort Yet to Come of Age: A Survey of Power Sector Reforms in Developing Countries.” Policy Research Working Paper 7330, World Bank Group, Washington DC. http://documents.worldbank.org/curated/en/504451467997248704/A- quarter-century-effort-yet-to-come-of-age-a-survey-of-power-sector- reforms-in-developing-countries. Jurkonis, Liudas, Šarūnas Merkliopas, and Karolis Kyga. 2016. “Impact of the Corporate Governance Reform on the Management Effectiveness of State- Owned Enterprises in Lithuania in 2012–2014.” Ekonomika 95 (2): 158–78. Kufeoglu, S., M. Pollitt, and K. Anaya. 2018. “Electric Power Distribution in the World: Today and Tomorrow.” Working Paper, Faculty of Economics, University of Cambridge, Cambridge, U.K. https://doi.org/10.17863/CAM.27667. 282 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries La Porta, Raphael and Florencio Lopez-de-Silanes. 1999. “The Benefits of Privatization: Evidence from Mexico.” The Quarterly Journal of Economics 114 (4): 1193–242. La Porta, Raphael, Florencio Lopez‐de‐Silanes, and Andrei Shleifer. 2002. “Government Ownership of Banks.” The Journal of Finance 57 (1): 265–301. Megginson, William L. 2005. “The Economics of Bank Privatization.” Journal of Banking and Finance 29 (8–9): 1931–80. https://doi.org/10.1016/j.jbankfin.2005.03.005. Megginson, William L. 2017. “Privatization, State Capitalism, and State Ownership of Business in the 21st Century.” Foundations and Trends in Finance 11 (1– 2): 1–153. Megginson, William L. and Jeffry Netter. 2001. “From State to Market: A Survey of Empirical Studies on Privatization.” Journal of Economic Literature 39 (2): 321–89. Milhaupt, C. J., and M. Pargendler. 2017. “Governance Challenges of Listed State- Owned Enterprises around the World: National Experiences and a Framework for Reform.” Cornell International Law Journal 50 (3): 473–542. https://scholarship.law.cornell.edu/cgi/viewcontent.cgi?article=1907&conte xt=cilj. Muir, Russell, and Joseph P. Saba. 1995. “Improving State Enterprise Performance. The Role of Internal and External Incentives.” World Bank Technical Paper 306, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/489021468186871150/Improvi ng-state-enterprise-performance-the-role-of-internal-and-external- incentives. Nellis, John. 1994. “Is Privatization Necessary?” FPD Note 7, World Bank, Washington, DC. Rudolph, Heinz P. 2009. “State Financial Institutions: Mandates, Governance, and Beyond.” Policy Research Working Paper 5141, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/114531468159602791/State- financial-institutions-mandates-governance-and-beyond. 283 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Sen, Anupama, Rabrindra Nepal, and Tooraj Jamasb. 2016. Reforming Electricity Reforms? Empirical Evidence from Asian Economies. Oxford U.K.: The Oxford Institute for Energy Studies. https://doi.org/10.26889/9781784670504. Tenev, Stoyan, Chunlin Zhang, and Loup Brefort. 2002. Corporate Governance and Enterprise Reform in China: Building the Institutions of Modern Markets. Washington, DC: World Bank and International Finance Corporation. Thorne, Janine, and Charlotte du Toit. 2009. “A Macro-Framework for Successful Development Banks.” Development Southern Africa 26 (5): 677–94. Tordo, S., B. S. Tracy, and N. Arfaa. 2011. “Natural Oil Companies and Value Creation.” World Bank Working Paper 218, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/650771468331276655/National -oil-companies-and-value-creation. Trimble, Chris, Masimi Kojima, Inez Perez Arroyo, and Farah Mohammadzadeh. 2016. “Financial Viability of Electricity Sectors in Sub-Saharan Africa: Quasi-Fiscal Deficits and Hidden Costs.” Policy Research Working Paper 7788, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/182071470748085038/Financia l-viability-of-electricity-sectors-in-Sub-Saharan-Africa-quasi-fiscal- deficits-and-hidden-costs. Urpelainen, Johannes, and Joonseok Yang. 2019. “Global Patterns of Power Sector Reform, 1982–2013.” Energy Strategy Reviews 23 (January): 152–62. https://doi.org/10.1016/j.esr.2018.12.001. Vagliasindi, Maria. 2012. “Power Market Structure and Performance.” Policy Research Working Paper 6123, World Bank, Washington, DC. Victor, David G., David R. Hults, and Marl C. Thurber, eds. 2012. Oil and Governance: State-Owned Enterprises and the World Energy Supply. Cambridge, U.K.: Cambridge University Press. Wolf, Christian. 2009. “Does Ownership Matter? The Performance and Efficiency of State Oil vs. Private Oil (1987–2006).” Energy Policy 37 (7): 2642–52. https://doi.org/10.1016/j.enpol.2009.02.041. 284 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries Wong, Simon C. Y. 2004. “Improving Corporate Governance in SOEs: An Integrated Approach.” Corporate Governance International 7 (2): 5–15. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=897121. World Bank. 2005. “Privatization and Deregulation: A Push Too Far?” In Economic Growth in the 1990s: Learning from a Decade of Reform, 163–194–97. Washington, DC: World Bank. http://www1.worldbank.org/prem/lessons1990s/. World Bank. 2014. Corporate Governance of State-Owned Enterprises: A Toolkit. Washington, DC: World Bank. DOI: 10.1596/978-1-4648-0222-5. License: Creative Commons Attribution CC BY 3.0 IGO. World Bank. 2017. “Maximizing Finance for Development: Leveraging the Private Sector for Growth and Sustainable Development.” Development Committee Report DC2017–0009, World Bank Group, Washington, DC. World Bank. 2019. Rethinking the 1990s Orthodoxy on Power Sector Reform: Flagship Report. Washington, DC: World Bank Group. http://documents1.worldbank.org/curated/en/597911563520308282/pdf/Wo rld-Rethinking-the-1990s-Orthodoxy-on-Power-Sector-Reform-Flagship- Report.pdf. World Bank Group. 2018. 2017 Survey of National Development Banks. Washington, DC: World Bank Group. http://documents.worldbank.org/curated/en/977821525438071799/2017- Survey-of-National-development-banks. Xu, Jiajun, Xiaomeng Ren, and Xinyue Wu. 2019. Mapping Development Finance Institutions Worldwide: Definitions, Rationales, and Varieties. NSE Development Financing Research Report 1. Beijing: Peking University. Notes 51 Thorsten Beck wrote this issues note. The note reflects the author’s views and does not reflect the views of the Independent Evaluation Group or the World Bank Group. 52 In the note, the term “development finance institution (DFI)” is used interchangeably with development bank. 285 Appendix E State-Owned Enterprise Reform Deep Dives and Issues Notes: Summaries 53 John Nellis wrote this issues note. The note reflects the author’s views and does not reflect the views of the Independent Evaluation Group or the World Bank Group. 54 The name public enterprise is given to a range of government-controlled bodies that fulfill a variety of functions: commercial, research, educational, health, and so on. This note deals mainly with government-owned, semi-autonomous commercial, or semi-commercial entities required to raise a substantial part of their operating revenues from the sale of goods and/or services. 55 In China, shares of public entities on the market are often acquired by other state-owned entities. 286 Appendix F. Case Study Overviews for Eight Countries Summaries of Findings of Workshop on State-Owned Enterprise Reform Country Case Studies The World Bank Group has a credible approach to achieving development impact through state-owned enterprise (SOE) reforms to the extent that these reforms are not undermined by government interference and excessive reliance on conventional wisdom where adaptation is required. The Bank Group’s credible approach is evidenced by its continuous engagement with case study countries in both the energy and financial sectors, in most cases since before the evaluation period, reflecting its continuity and commitment. Continuity has allowed the Bank Group to become a leading and trusted partner in several of the country sector cases and to innovate in select cases with successful Maximizing Finance for Development (MFD)–type joint interventions by the World Bank, the International Finance Corporation (IFC), and the Multilateral Investment Guarantee Agency (MIGA). However, its approach is weakened by  Reforms on paper that are not exercised in practice, as in some corporate governance arrangements and “independent regulators” (Bangladesh, Kenya energy); and  An overconcentration of its efforts on unbundling and privatizing without considering whether underlying country conditions allow the benefits of these reforms to materialize in the first place. Examples include Ukraine, where the absence of the rule of law undermines the relevance of the ownership structure of companies to improve their 287 Appendix F Case Study Overviews for Eight Countries performance, or countries where less than fully unbundled structures yield relatively good SOE and sector performance (Bangladesh energy, Vietnam energy,). The Bank Group was somewhat effective in achieving the objectives of its SOE reform interventions, but it was constrained by the degree of government commitment to reforms and the appropriateness of reforms supported to country contexts. In the energy sector, the Bank Group was successful in financing the construction and rehabilitation of generation, transmission, and distribution infrastructure to increase energy access but less consistently so in measures to reduce technical and commercial losses. This was usually the result of competition between social and political objectives with commercial objectives. In such cases, the former usually dominated. In these cases, SOEs’ financial sustainability was likely to be compromised, more so when any previously established governance or regulatory or corporate governance reforms were not “irreversible,” hence unable to guarantee SOE independence in pursuing commercial viability. In both sectors, commercial operation of SOEs seems dependent on the level of government commitment over time to allow reforms to be implemented and the degree to which vested interests in the country undermine the sustainability of reforms. Privatization of state-owned banks or, in the case of Ukraine, nationalization to improve sector stability, sometimes suffered from lack (or change) of government commitment to implement reforms and/or the power of vested interests to overturn it. Considering this, there are at least three crucial factors of success behind SOE reform interventions by the Bank Group: (i) continuity of Bank Group engagement (including capability in the field), (ii) government commitment to support and uphold the integrity of the reforms, and (iii) the underlying political economy context that either enhances or threatens the sustainability of reforms. Other important factors stem from these three. For example, improvements in project design, strong team composition in the form of competent task team leaders that stakeholders trust, donor 288 Appendix F Case Study Overviews for Eight Countries coordination to avoid overlap of interventions and maximize impact and joint interventions by Bank Group institutions are all characteristic of continuous engagement. Bangladesh (Field Based) SOEs are represented in all major sectors of the Bangladesh economy, especially the services, utility, and industry sectors. SOEs’ operational revenues represent more than 10 percent of gross domestic product (GDP) in Bangladesh. Currently, there are 45 SOEs and about 200 subsidiary enterprises engaged in diverse economic activities. SOE employment accounted for 7 percent of total employment. Given the lack of a centralized authority for ownership, as many as 19 Bangladeshi ministries own SOEs, according to the Bangladesh Enterprise Institute. The energy sector faces many challenges. Over the next decade, gas production in Bangladesh is expected to decline, and some of the existing gas-fired plants will have to shift their source of gas from the country’s gas fields to imported liquefied natural gas, which will require investments in the necessary infrastructure to handle the imports. In addition, to improve its energy security by diversifying its fuel sources for electricity, the country also needs to add between 1 and 2 gigawatts of sustainable and cost- competitive capacity expansion annually to meet the rising demand. The change in energy mix away from gas will increase the cost of supply of electricity. With a political reluctance to raise the cost of electricity so that the rates can cover operating and capital costs, it is difficult to attract the private sector to invest more in electricity. The state-owned electricity utility has large financial deficits, and private investment is limited because of the populist pricing policies. Reducing financial and operational inefficiencies in SOEs in the energy sector is critical. The country government holds a key strategic role in the energy sector through the regulatory agency Bangladesh Energy Regulatory Commission and SOEs like Bangladesh Oil, Gas, and Mineral Corporation 289 Appendix F Case Study Overviews for Eight Countries (Petrobangla), Bangladesh Power Development Board, and Bangladesh Petroleum Corporation. In 2017, electricity subsidies from government transfer, in the forms of subsidized cash loans and direct budgetary support, reached $489 million (Bangladesh Ministry of Finance 2018). The government of Bangladesh has adopted a complex energy development strategy, committed by its ambitious goal of ensuring all citizens’ access to affordable and reliable electricity. The Power System Master Plan (PSMP), adopted in 2010, focuses on building imported coal-powered and oil-fired power stations, creating liquefied natural gas facilities, importing electricity generated by hydropower from neighboring countries or joint development, introducing high-efficiency power supply and low carbon dioxide–emission technology, and improving thermal efficiency. However, energy development is not on track compared with the PSMP 2010 plan in Bangladesh. Currently, many power plants in Bangladesh cannot generate electricity as specified in terms of power, thermal efficiency, and so on, for each unit. Daily shortage of power does not allow facilities to stop and undertake periodical maintenance regularly. The legal framework does not stipulate preventive maintenance works as an obligation for plant owners. Low financial soundness of public generating companies because of low electricity tariffs does not permit purchasing necessary spare parts in advance. To further formulate an extensive energy and power development plan, the PSMP 2016, covering energy balance, power balance, and tariff strategies, was drafted with the help of the Japan International Cooperation Agency. The regulator, Bangladesh Energy Regulatory Commission was established in 2003 with World Bank support as a regulatory agency for regulating gas, electricity and petroleum products in Bangladesh. Later, under the Power Sector Development Technical Assistance Project, the World Bank aimed to create effective capacity in the government to develop power sector policies, industry structures, and a gas supply strategy for the balanced development of Bangladesh’s power sector, and to create capacity within the Energy 290 Appendix F Case Study Overviews for Eight Countries Regulatory Commission to regulate the sector effectively. Yet the Bangladesh Energy Regulatory Commission’s success was limited due to its multiplicity of roles, shared oversight and lack of functional independence. Although intended to foster fair competition in the power sector, its key role was at times undermined by the government, which directed the sale of power to the end-users below cost. Although formally independent, its commissioners acknowledge challenges in rate setting and licensing, because it is not the sole agency with the oversight of the energy sector. There is additional oversight from the government through the Ministry of Energy and Resources. Banks dominate the financial system, and they accounted for 60 percent of total assets as of the end of 2015. There are six state-owned commercial banks (SOCBs) and two state-owned specialized developmental banks, accounting for more than 27 percent of the financial system assets. SOCBs contribute significantly to the ease of access to finance in Bangladesh because the SOCBs own about 56 percent of total bank branches and have the deepest branch penetration across Bangladesh, including remote areas and villages, but they are inefficient (World Bank 2016). Furthermore, poor governance has resulted in weak balance sheets of SOCBs, often as a consequence of a series of financial scams and resultant loan defaults in the SOCBs. SOCBs are subject to political influence and interference; they suffer from overstaffing, lower quality of services, lack of innovation, and inefficiency. This results in additional costs to the taxpayers in the form of direct costs from recapitalization of these banks and indirect costs through the misallocation of resources, provision of credit to inefficient borrowers, and incompetent but politically connected borrowers who are often unable to repay their loans (IMF 2018). Under successive bank recapitalization programs, the government has provided additional capital to shore up these banks and will continue to do so in the medium term to enable them to meet the international bank capitalization norms under the second of the Basel Accords. 291 Appendix F Case Study Overviews for Eight Countries Compared with private banks and foreign-owned banks, the SOCBs have been underperforming and in a much weaker financial position. The SOCBs’ nonperforming loan (NPL) ratio was almost five times as high as those for the private banks in 2017, and their capital adequacy ratio is much lower than private banks and foreign-owned banks, despite the several rounds of recapitalization implemented by the government. The SOCBs show weak performance for the standard indicators of profitability, return on assets, and return on equity. In 2017, the SOCBs losses accelerated, with return on equity reaching a negative 20 percent (IMF 2018). SOCB modernization is prioritized in the Bangladesh government’s Seventh Five-Year Plan (2016–20). The government has taken some initiatives in the past few years to improve governance and financial position of the SOCBs, but its past efforts do not provide great confidence for a major reform program to improve the performance of the SOCBs. After a series of earlier reforms in the early 2000s led by the World Bank, in 2007, the Bank of Bangladesh (the country’s central bank) was given the regulatory authority to supervise all banks in the country, including SOCBs. However, after an elected government came to power in 2009, some of the central bank’s authority over SOCBs was diluted. Currently, SOCBs report to the Banking and Financial Institutions Department, a department in the Ministry of Finance. The Ministry of Finance also controls the appointment of board of directors, the managing directors, and the deputy managing directors for SOCBs, and the operating budgets of the SOCBs. The government also mandates the SOCBs to support its development programs, sometimes at subsidized rates, and requires them to provide some services for government free of charge. A comprehensive risk management policy is lacking, and early fraud detection is still a challenge. The SOCBs have a history of weak governance that has led to outright fraud and large losses. Independent, qualified, and reputable professionals are not always present on their boards. 292 Appendix F Case Study Overviews for Eight Countries In both the energy sector and the financial sector, the Bank Group strategies and programs were aligned with the stated objectives of successive government five-year plans (the fifth, sixth, and seventh five-year plans). For example, the government’s sixth five-year plan clearly stated the need for increasing energy availability, reforming the banking sector, and improving governance of key institutions, but as was noted previously, there is marked gap between the government’s stated intentions and its actions in the field. It transpired that after an initial period of support, especially under the caretaker government (run by technocrats) between 2007 and 2009, the subsequent elected government’s appetite for continuing with the reforms waned over time. Because the elected government of 2009 faced an energy crisis with demand for electric power far outstripping supply, reforms in the energy sector were carried out more fully and with greater urgency. The reforms of the financial sector, including the privatization of the state- owned banks, were generally not followed after 2009, and some progress regarding oversight and corporate governance were actually reversed. Although the World Bank found traction and effectiveness in promoting banking sector reform until 2009, it has been ineffective since then and largely excluded from reforms affecting state-owned banks. As noted above, as of 2017, SOCB performance was poor and deteriorating. Although private banks now account for the majority of assets, the share of assets of SOCBs, after declining to about 27 percent in 2009, has held about steady since then. As of 2019, the Independent Evaluation Group (IEG) found a state-owned banking sector that was performing badly, with high NPLs, inadequate capitalization, negative profitability, and inadequate oversight. Although the World Bank had been able to support the capacity development of Bank of Bangladesh, after 2009, the bank’s oversight authority of state-owned banks was sharply constrained and mediated by the direct involvement of the Ministry of Finance. The World Bank was able to engage only in distinct niches such as insurance and long-term capital. IFC engaged with private 293 Appendix F Case Study Overviews for Eight Countries banks in such areas as small and medium enterprises and trade finance but not with sector reform. The World Bank had been able to muster a Financial Sector Assessment Program (FSAP) mission (which coincided with IEG’s mission), which undoubtedly will raise alarms on the state of the SOCBs and the systemic risk they pose to the financial system. The major factors affecting the outcome of interventions in the energy sector are alignment and political will; multiple, continuous, and complementary support; building institutional capacity; insufficient transmission capacity; politicization of tariffs and licensing; incomplete yet effective unbundling; and donor coordination. The early progress from Bank Group engagement from 2004 on unbundling and building technical capacity appear to have had important payoffs over time, even as the sector grew dramatically in capacity in recent years. In general, sector performance improved in terms of reduced losses, reduced arrears, and an elimination of the energy gap. Although unbundling seems to have improved incentives for good performance, the causes of efficiency improvements are not fully documented, though engagement in different segments (oversight, generation, transmission, and distribution) seems overall positive. Under the 2007 Development Support Credit IV (P074801), the government reduced Bangladesh Power Development Board’s payment arrears to independent power producers and increased efforts to recover power utilities’ accounts receivables. It also reinforced its commitment to close the pricing gap. As an interim measure, it provided explicit support to Bangladesh Power Development Board to put the power sector on a financial recovery path. A World Bank technical assistance project (P078707) to the Power Cell was designed to help the government scale up sector reforms by preparing a restructuring plan to shape government policies and regulatory arrangements. It provides initiatives for power sector reform, facilitating the role of private power producers, and coordination between the different power sector agencies. The Power Cell is an acknowledged success, providing 294 Appendix F Case Study Overviews for Eight Countries technical support and monitoring and through contracting, policy support for the sector, and the major organizations within it. The regulator, Bangladesh Energy Regulatory Commission, created with World Bank support, was established in 2003 as a regulatory agency for regulating gas, electricity, and petroleum products in Bangladesh. The spurt in generation capacity has helped meet the power shortage in the country, but it has also led to much higher subsidies from the government to keep power costs under control for the end user. Furthermore, among the challenges in the power sector in Bangladesh is the need to address the country’s transmission system bottlenecks. Other donors (for example Asian Development Bank, the Asian Infrastructure Investment Bank, and Japan International Cooperation Agency) are coming into the power sector with financing on favorable terms, often with fewer attached requirements and less bureaucracy. However, the government acknowledges that the Bank Group’s expertise in the field in Bangladesh, its long-standing relationships with key government agencies, and its consistent policy view are distinguishing features that make it a trusted partner in the SOE reform process in the energy sector. The main factors that impact the achievement of objectives in financial sector are complementary and continuous engagement through fiscal year (FY)09, ownership and political will, and finding opportunities. The World Bank was able to support substantial progress through a rich complement of projects and advisory services and analytics (ASA) up to 2009. Under the caretaker regime, “fit and proper” guidance for executives and board observed was followed more consistently. Once political commitment to sector reform dried up in 2009, the World Bank’s program was limited to the periphery of the financial system, and no major projects were approved involving SOCB reform. The World Bank’s strong emphasis on improved oversight and corporate governance in the financial sector generated pushback after 2009 and an extended “quiet” period in sector reform support. The cancellation of a recent state-owned financial institution 295 Appendix F Case Study Overviews for Eight Countries modernization project indicates continuing differences between priorities of counterparts and the World Bank’s view of sector reform. Nonetheless, the World Bank is finding distinct niches where it can have impact, including insurance, housing finance, and long-term finance. There is much capacity building needed in domains yet to be fully developed and where entrenched political interests have not yet formed. China (Desk Based) 56 SOEs played an essential role in Chinese economic growth and are still a significant and growing component in the economy. However, SOEs in China are facing challenges in terms of low efficiency, and the policy of favoring SOEs is having negative impacts on the macroeconomy. To address these problems, the government has been reforming SOEs since the 1970s. Under the pressure of changing the growth model, a new round of SOE reform was initiated in 2013, aiming to shift state control from the management of the company to the management of capital. In 2015, a major SOE reform policy document, the Guiding Opinions of the Central Committee and State Council on Deepening the Reform of SOEs, classified SOEs as “public class” and “commercial class”. Different types of SOEs were to have different supervision mechanisms, mixed ownership reform plans, and corporate governance mechanisms, and so on. Recent SOE reforms focus on corporate governance with the institutionalization of Party leadership within SOEs. The financial system, dominated by the banking sector and featuring state ownership, is facing resource allocation distortion in China. Since 1999, the banking sector experienced recapitalization and ownership diversification intending to improve corporate governance and efficiency. Despite the effort and the progress of the financial sector reform, the formal lending sectors that are led by SOCBs are exposed to higher risk with the increasing number of NPLs. The energy sector in China has experienced reforms corresponding with the economic reforms. The reform concentrated on separating regulation and 296 Appendix F Case Study Overviews for Eight Countries operation in the SOEs, with priorities shifted from alleviating energy shortage to improving market efficiency, targeting regulation, the pricing system, and separation of ownership from the operation. As the only multilateral development bank that maintained a partnership with China on SOE reform, the Bank Group’s extant relevance can be attributed to its willingness to adapt to China’s reform strategy. Instead of recommending to China the widespread privatization that it offered to other countries in transition, the Bank Group has been helping the government to implement its reform where state ownership is maintained for the key enterprises. The China Country Assistance Strategy Completion Report for FY03–05 found that additional Bank engagement on SOE reform was likely to be remain limited, and that the most effective way for the Bank Group to support SOE reform would be through analysis and advice on social safety nets and inter-government financing (World Bank Group 2006). The China Country Partnership Strategy (CPS) for FY07–12 aims to improve public and market institutions under pillar 5, including reforming SOEs and attracting private investment. the International Bank for Reconstruction and Development supported this objective with a study on SOE portfolio and SOE dividend policy in FY06. In 2011, the State-Owned Assets Supervision and Administration Commission issued guidelines on corporate social responsibility. However, the CPS Completion Report Review stated that there was no information on whether the listed SOEs and other large private corporations have adopted international best practice in these areas. The CPS Progress Report noted that only modest progress had been made in simplifying cumbersome or anticompetitive business regulations, while many institutional and market obstacles remained, including the lack of progress in creating more competitive markets for energy and water. The 2017 FSAP for China prepared in 2017 stated that the slow pace of SOE reform and limited exit of weak firms had resulted in efficiency losses and reinforced the perception of implicit guarantees. It recommended that China 297 Appendix F Case Study Overviews for Eight Countries continue to diversify the ownership structure of state-owned financial institutions and SOEs. Furthermore, it advised to precisely define the mandates of the policy banks by targeting specific market gaps where commercial banks and capital markets are unable or unwilling to fill. Improving the corporate governance arrangements for China Development Bank and the policy banks to ensure the operational autonomy of the institutions, and the competencies and objectivity of the boards was also listed as a high priority action. In China, although there was a small number of Bank Group interventions, those targeting SOE ownership, corporate governance, and financial management reform achieved their objectives. For example, IFC supported the Industrial Bank China in becoming more private sector oriented through a 2008 project (21114) that reinforced the independence of board directors and the establishment of subcommittees to ensure checks and balances at different levels. IFC supported the Bank of Beijing (23943, FY10) in being listed on the Shanghai Stock Exchange in August 2007, attracting funding of ¥1.9 trillion. The bank improved its corporate governance and management practices through the presence of foreign directors on its board and the establishment of several specialized committees (including an internal audit committee recommended by IFC). However, an effort to support the Shenzhen VTB (29386, FY10) to expand financial inclusion for small and medium enterprises and improve loan growth, profitability and asset quality fell short of expectations. The Guangzhou Development Industry Holdings Project (22418, FY06) failed to meet its goal of increasing private sector shareholding in the SOE, due to unexpected market conditions and a lack of realism. Client commitment played a clear role in the performance of Bank Group operations in China. According to the review of the SOE portfolio, client commitment issues made up 25 percent of the factors cited as influencing project outcome, mostly positively. In a context where capital was not as scarce as before, IFC's strategy to migrate from being a capital investor to a 298 Appendix F Case Study Overviews for Eight Countries knowledge investor was positive. An essential part of this role was IFC’s technical assistance program. Flagship reports on China 2030 and Urban China provided a solid foundation for policy dialogue and program design, underpinning both knowledge services and lending. However, there was a divergence in the alignment between the government priorities, the Bank Group strategy, and the Bank Group engagement during the evaluation period. The Bank Group and the government SOE reform agendas were well aligned from 2006 to 2012, but they have diverged since 2013. Despite the new round of SOE reform in China initiated in 2013, the CPS for 2013 to 2016 recognized the necessity of SOE reform, but it did not propose any specific measures to support the reform. Concerning the sector focus, the Bank Group SOE reform portfolio in China from 2008 to 2018 is closely aligned with the government’s ambition on the sector reform. However, in terms of the energy sector, where energy price mechanism reform was addressed in both the Bank Group’s and the government’s strategies, limited SOE reform projects were delivered. IFC and the World Bank underpinned the SOE reform in China through distinctive mechanisms. The World Bank’s endeavor was on upstream knowledge generation and sharing through joint studies, and meanwhile, it provided pilot projects for specific problems solution. Ten of the 11 projects delivered by the World Bank were through advisory services. IFC did the heavy lifting on investment projects and had been actively working with SOEs at the downstream level of the reform, specifically on attracting private investment to diversify the state-owned financial institutions’ ownership. The progress of SOE reform in China is modest. There is progress in leveraging SOEs’ role in expanding access to finance and in simplifying cumbersome or anticompetitive business regulations. Nevertheless, many institutional and market obstacles remain. Furthermore, as the the government increased support to SOEs to counter the global economic downturn of 2009 and beyond, the private sector faced challenges where 299 Appendix F Case Study Overviews for Eight Countries Government continued to restrict competition in protected sectors while merged centrally owned firms increased their market share and power. Given the Bank Group’s resource constraints, a more focused program will likely produce better results. The Bank Group contributed to several reform key issues in China. However, under inconsistent strategic emphasis, the overall approach is less holistic compared with other development partners. The Bank Group activities—both lending and advisory services—could adopt a programmatic approach rather than one-off interventions to ensure consistent support and that initiatives are supportive rather than prescriptive to ensure government ownership. The Arab Republic of Egypt (Field Based) Overall, on SOEs and other areas, Egyptian conventional wisdom in the street (and among elites but less so) is that the public sector is “good” and the private sector is “bad.” This dichotomy explains in part why SOE reform efforts in general (including privatization), while noticeable, have not been rapid. There is strong popular suspicion about the integrity of some reforms, including privatization transactions, which are suspected of corruption. After the 2011 revolution, third parties brought cases in court for reversal of privatization deals, and in several of those, Egyptian courts ruled to reverse the sale of former public companies to the private sector. 57 External shocks also played a role in pushing SOE reform forward, but sometimes they also hindered it. Positively, the electricity crisis that erupted in summer 2014 pushed the government to look more seriously at energy sector reform. Negatively, the country’s 2011 revolution halted engagement with the Bank Group, which then had to prove itself again as a trusted development partner to the new government and a civil society that played a key role in the big political changes that took place. As such, the political economy of Egypt has been important in determining the extent and success of SOE reform. However, the Bank Group has limited 300 Appendix F Case Study Overviews for Eight Countries tools to manage or engage with the political economy, and in any case, these are nontraditional. An example of this is in cross-fertilization between government counterparts and World Bank staff, with a noticeable movement in both directions. At the same time, Bank Group institutions worked together in complementary ways on SOE reform in a modest fashion. In this regard, limitations existed in stakeholder engagement, complementarity of projects (within Bank Group and across donors), and in knowledge sharing. For the energy sector in general, alignment and reform ownership did exist but less so in the financial field. In the latter, the Bank Group reacted through a shift in focus away from privatizing or otherwise engaging with large state-owned banks toward such areas as housing finance, and micro, small, and medium enterprise funding. IEG’s validation of the Completion and Learning Review (CLR) for Egypt covering the period of FY06–14 reported that the objective of reducing the fiscal deficit by the end of the Interim Strategy Note period (June 2012– December 2013) was only partially achieved and energy subsidies did not decrease as a share of the budget. The budget deficit increased from 10.5 percent of GDP in 2011/12 to 13.6 percent of GDP in 2013/14, as the government did not introduce a coherent energy subsidy reduction plan until July 2014. Slow progress on energy subsidies was considered as one of the key barriers to an International Monetary Fund (IMF) program on which the World Bank development policy loan (DPL) in this area was contingent. In July 2014, the government launched a strategy to gradually reduce energy subsidies. The World Bank provided support through Energy Efficiency Strategy (FY11 and FY12) and Energy/Social Safety Nets Sector Reform (FY14). Between 2015 and 2017, World Bank development policy financing (DPF) of $3.15 billion delivered financial support and technical assistance to transform the energy sector and improve financial viability in line with the national energy strategy. Changes included the enactment of a new renewable energy law, adjusting electricity tariffs to allow recovery of 301 Appendix F Case Study Overviews for Eight Countries operational costs, gradually phasing out fuel subsidies, revising the feed‐in tariff policy, and establishing a regulatory framework for competitive bidding for independent power producers. As part its comprehensive program of economic reforms, the government of Egypt introduced a solar feed‐in tariff program to leverage private sector capital and expertise to support Egypt’s energy goals (Performance and Learning Review of the Country Partnership Framework for FY15–19). The World Bank also supported the energy pricing and subsidy reform through technical assistance. First, the World Bank delivered support via two related projects: Energy Pricing and Subsidy Phase I (FY15) and Support on Energy Subsidy Reforms Phase II (FY16). These two technical assistance projects delivered analytical work on energy pricing, capacity building on modelling of price reforms, and a communication strategy. Second, the World Bank provided support through the Energy/Social Safety Nets Sector Reforms Technical Assistance. In the financial sector, the World Bank aimed to improve the sector competitiveness and efficiency. The indicators on developing a regulator for nonbank financial institutions and improving the soundness of the Egyptian Financial Supervisory Authority were met, and Egypt’s central bank had strengthened corporate governance in the banking system by issuing corporate governance regulations in 2011. Furthermore, the central bank improved its own governance through amendments to the central bank, banking system, and money law. With IFC and World Bank support for financial sector reform and capacity building, the Egyptian government was able to develop a collateral registry and implement several reforms targeting areas like capital issuance, directors’ conflict of interest, calling of extraordinary board meetings by minority investors, disclosure index, shareholders rights index, and shareholders suits index. An IEG case study found that Bank Group complementarity could have benefited more from the substantial political 302 Appendix F Case Study Overviews for Eight Countries capital the Bank Group has built with the government through its DPL series, rigorous staff engagement, and a prolonged series of economic reforms. The Bank Group was quite effective in leading energy sector restructuring and in engagement on accountability and governance standards, but it was less effective in areas such as privatization, including in the financial sector. For the former, the frequent cross-fertilization between counterparts in ministries and World Bank staff helped promote an effective energy sector dialogue. In addition, through infrastructure investments, the Bank Group helped increase the operational ability of sector utilities and access to energy in Egypt. However, there is still a need for sustained investments to modernize infrastructure. The success of the Bank Group SOE activity in the energy sector can be attributed in part to a greater care shown by the government of Egypt and agents in adhering to environmental and social safeguard measures, even as Bank Group operations faced other shortcomings such as slowness or lack of response to requests from stakeholders. For example, the Giza North project faced significant implementation delays and unrest among the local populations due to lowering water levels and incomplete implementation of resettlement plans. The South Helwan Supercritical Power Plant had to address air and water quality, noise, solid waste, and local socioeconomic impacts, including on employment. Clients still welcomed the Bank Group’s capacity building and other upgrading support even as they often face issues of misalignment between World Bank standards and procedures and local or country-wide realities. Kenya (Field Based) Kenya’s energy sector faces the challenge of continuing to increase access while reducing losses and restoring or preserving the financial integrity of its utilities. On the generation side, Kenya Electricity Generation Company Limited’s balance sheet indicates that it cannot take on significantly more debt to fund expansion; current initiatives to improve balance sheet and 303 Appendix F Case Study Overviews for Eight Countries operational performance may not be enough to bridge the financing gap. Geothermal Development Company’s revenue model is not enough to cover its true costs because of implicit government of Kenya subsidy. Meanwhile, if the government delivers on transmission and distribution targets, the energy sector may exceed 20 percent of the total government debt burden. Kenya Electricity Transmission Company’s inadequate revenue model results in a reliance on government financing rather than its own balance sheet. Kenya Power’s target to reach near-universal access by 2020 may be a high cost, given that a significant increase is estimated in cost per connection past a 70–80 percent connection rate based on trends seen in other countries. Pending issues include improving the regulator’s independence and utilities’ independence of corporate governance in practice, enforcing competitive procurement of independent power producers, reaching cost recovery tariffs, balancing investments in transmission and distribution infrastructure with generation, and stabilizing the grid in the face of growing intermittent renewable energy sources. The Bank Group financed 16 projects during 2008–18, totaling $2 billion in commitments. Most projects (14) were lending operations, and 2 were analytical products. (An additional lending project evaluated by IEG during 2008–18 but approved before the period was included in the analysis.) The Bank Group’s engagement supported the development of the sector by supporting (i) construction and rehabilitation of generation, transmission, and distribution infrastructure; (ii) capacity-building assistance to utilities, including for improving their corporate governance; (iii) strengthening the financing and ability of state-owned utilities to attract long-term private capital to refinance short-term debt; and (iv) upstream and downstream reforms for the development and strengthening of the nascent oil sector’s legal and regulatory framework and SOEs. All large electricity generation projects (that is, all those not coming under feed-in tariff policy) would continue to require long-term Power Purchase Agreements with Kenya Power (KPLC) to raise the necessary debt financing 304 Appendix F Case Study Overviews for Eight Countries as KPLC was still the sole purchaser and distributor of all electricity produced, the cornerstone of the electricity sector in Kenya. Therefore, strengthening KPLC’s financial position was critical for Kenya’s electricity sector. However, the company undertook investments in support of the government’s electrification program that were detrimental to its financial sustainability. The World Bank supported the company’s debt restructuring through International Development Association financing via the KE Electricity Modernization Project. The Bank Group’s engagement was effective at increasing access to electricity but failed to reduce technical and commercial losses. Losses were in part due to unbalanced development of generation based on optimistic demand estimations and unplanned projects in advance of matching transmission and distribution capacity. Such losses are detrimental to the financial sustainability of utilities. The objective of enhancing the policy, institutional, and regulatory environment for sector development, including private sector participation under the Energy Sector Recovery Project, was substantially achieved. Among other things, an energy policy was successfully put in place through the enactment of the Energy Act in 2006, leading to the establishment of the Energy Regulatory Commission and the Rural Electrification Authority. The Energy Regulatory Commission was fully operational (although not always fully independent) by project closure and was able to resolve 76 percent of annual disputes and complaints. The World Bank provided a 2015 Policy Sector Note “Powering Kenya’s Future: Future Role of the Public and Private Sectors (ESW)” (P133675) which assessed power sector institutions including the electricity utilities KenGen and KPLC. The report stressed the importance of ensuring the Energy Regulatory Commission’s independence to approve Power Purchase Agreements by KPLC with KenGen and independent power producers, and the review of tariff applications by KPLC in determining retail tariffs. 305 Appendix F Case Study Overviews for Eight Countries Notable positive factors of success behind the Bank Group’s engagement include persistent government commitment to continue the unbundling process; textbook corporate governance arrangements in two major SOEs (KPLC and Kenya Electricity Generation Company Limited); strong coordination between development partners, including the World Bank; strong cooperation between the World Bank, IFC, and MIGA to attract independent power producers; and strong public sector capacity within the utilities. Negative factors of success include the political economy behind the politicization of long-term power development planning and approval of some specific generation programs, and poor project design in the World Bank’s engagement in the petroleum sector and IFC’s investment in KPLC. Mozambique (Desk Based) Mozambique was not able to sustain the 7.9 percent per year average GDP growth rate it achieved from 1993 to 2014. The revelation of $1.4 billion in previously undisclosed debt undermined confidence in the economy and contributed to weakening its growth. Real GDP growth is estimated at 3.3 percent in 2018, down from 3.7 percent in 2017 and 3.8 percent in 2016. This rate of growth is well below the 7 percent GDP growth achieved on average between 2011 and 2015. Four percent GDP growth is projected for 2019, and the growth may be higher in the medium term if gas production investments are materialized. The debt crisis has also placed the fiscal outlook for Mozambique under immense pressure. The revelations of undisclosed loans triggered a suspension of IMF program and donor support to the budget. The debt revelations pushed public debt to 127 percent of GDP in 2016, shifting the economy into an unsustainable position. Public debt levels are expected to exceed 100 percent of GDP until 2020. Furthermore, contingent liabilities and debt costs continue to emerge from SOEs under operational and financial difficulties. Mozambique has been in default of its Eurobond and the two previously undisclosed loans. 306 Appendix F Case Study Overviews for Eight Countries The recent debt crisis has also emphasized the depth and critical importance of governance challenges facing Mozambique. Governance indicators for Mozambique, including the World Bank Country Policy and Institutional Assessments, reflect a gradual decline of government effectiveness, control of corruption, the rule of law, and voice and accountability over the past several years. The debt crisis has revealed governance weaknesses in public investment management, debt management, and oversight mechanisms for SOEs. Furthermore, it has generated a crisis of confidence in the government’s fiduciary capacity and its ability to responsibly manage natural resource revenues. The Institute for Management of State Property (IGEPE) represents the participation of the state in the Mozambican enterprise sector. IGEPE was created in 2001 with the goal of restructuring loss-making state companies and managing profitable state companies by coordinating state participation in key companies and industries. In 2016, IGEPE held a portfolio of 103 companies, of which 4 were in the process of dissolution. The energy sector in Mozambique is underdeveloped, with major inefficiencies in generation and transmission. Only 25.2 percent of the population had access to the grid as of October 2016 (up from 12 percent in 2005), and less than 2 percent of all rural households can use electricity for lighting. Electricidade de Moçambique (EdM) is a vertically integrated, government-owned electric utility. EdM buys most of its power supply from Hidroeléctrica de Cahora Bassa, owner and operator of the Cahora Bassa hydropower plant, the largest in southern Africa. The government of Mozambique owns 82 percent of Hidroeléctrica de Cahora Bassa, which operates as an independent power producer. The national petroleum company, Empresa Nacional de Hidrocarbonetos (ENH), the state-owned hydrocarbon company, represents the government in petroleum operations. ENH is also operating a gas distribution network to provide households and industry with piped gas in the south of Mozambique. The Mozambique Mining Exploration Company (Empresa Moçambicana de Exploração 307 Appendix F Case Study Overviews for Eight Countries Mineira) is a state-owned corporation formed by the government in 2010 to participate in mining projects, undertake exploration and mining development, and promote value addition to mineral products. The Energy Reform and Access Project (APL 1- P069183) helped establish the Conselho Nacional de Electricidade as an independent advisory regulatory body for the electricity sector. The Implementation Completion and Results Report (ICR) rated the achievement of project development objective as “satisfactory.” The project contributed to strengthening the capacity in the energy sector as witnessed by the increased private sector participation in the electricity sector and the establishment of new sector institutions. An Energy Sector Policy Note (P152677) aimed to support the government of Mozambique in determining priorities for policy decisions to deliver efficiently produced, technically and financially sustainable electricity supply to the Mozambican population. The financial sector in Mozambique is shallow, bank dominated, and foreign owned. There are 19 commercial banks that account for the bulk of financial sector assets. Most of these banks are subsidiaries of foreign banks, primarily from Portugal and South Africa. All banks except for one are privately owned. Two commercial banks with minority government ownership, Banco Commercial de Moçambique and Banco Austral, experienced difficulties after their privatization in 1996 and 1997, respectively. Recapitalization was undertaken by the government in 1999 and 2001. In 2005, the government proved its commitment to returning the banks to private ownership. The government is no longer a shareholder in Banco Austral, and it has only 17.1 percent shares in the Banco Internacional de Moçambique, the successor to Banco Commercial de Moçambique. The unbundling of the electricity utility was initially part of the government’s energy sector reform program, which aimed to improve efficiency and financial viability in the energy sector. However, unsatisfactory results of similar reforms supported in other Sub-Saharan African countries and findings of recent comparative research led the 308 Appendix F Case Study Overviews for Eight Countries government to change its policy (Vagliasindi and Besant-Jones 2013). It put in place a new strategy that no longer sought private participation in EdM. Instead, the focus moved to increasing the role and effectiveness of the national electricity regulator, particularly in monitoring EdM. SOE reform was not specifically mentioned in the Bank Group CPSs for Mozambique for FY08–11 and FY12–15. Increasing access to electricity and improving transparency in extractive industries were among the objectives of Bank Group strategies. Reflecting on the hidden debt crisis in 2016, the Country Partnership Framework (CPF) for Mozambique for FY17–21 highlighted the need to strengthen debt management, fiscal management, and oversight over SOEs. The CPF program proposed to use ASA and budget support financing to support the reforms needed to meet oversight and transparency challenges. Over the evaluation period, the Bank Group supported 17 SOE reform projects, excluding World Bank ASA, with a $1.14 billion commitment in Mozambique. World Bank lending accounted for 94.2 percent of these commitments through 15 projects. World Bank lending was provided through nine adjustment and six investment operations. IFC’s portfolio included one investment and one advisory project in the energy sector. Furthermore, five World Bank ASA projects totaled $1.3 million. The World Bank supported SOE reform mainly through the Macroeconomics, Trade, and Investment and the Energy and Extractives Global Practices, which together account for 90 percent of total SOE reform lending commitments. Almost three-quarters of the projects in the Mozambique portfolio included SOE reform interventions in energy sector. Bank Group support to SOE reform in the energy sector mainly aimed to improve SOEs’ business and operations management through rehabilitating infrastructure (EdM) and strengthening capacity (EdM, ENH, and Empresa Moçambicana de Exploração Mineira). IFC provided financing for the Central Solar de Mocuba S. A. (in which EdM has minority ownership) for the construction of a solar power plant. 309 Appendix F Case Study Overviews for Eight Countries Bank Group support to SOE reform in the financial sector was provided through four lending projects with minor SOE components. Bank Group support for the establishment of a deposit guarantee fund was an important mechanism to reinforce financial stability because the country’s bank safety net and crisis management frameworks were inadequate. World Bank SOE reform interventions in improving corporate governance were mainly provided through support for compliance with the Extractive Industries Transparency Initiative (EITI) standards in the extractives sector. The Poverty Reduction Support Credit (PRSC) series (7th, 8th, 9th, and 11th) included prior actions on the implementation of the EITI process. The support for EITI has continued with the Mining and Gas Technical Assistance and Mining and Gas Technical Assistance Additional Financing Projects. An IFC advisory project provided support to EdM in internal reporting and redesign of the company’s risk management system. At the national level, PRSC 10 provided support to improve SOE oversight by tightening the legal framework for public enterprises. To better manage fiscal risks generated by public enterprises, the government modernized the legislation governing SOEs in 2012 and approved the implementing regulations in 2013. Within 17 lending projects, IEG identified 39 SOE reform interventions, of which 16 interventions aimed to improve SOEs’ business and operations. The interventions on improving business operations are followed by improving corporate governance (10 interventions), changing SOE ownership (5), reforming the regulatory framework at the sectoral level (5), and strengthening financial management (3). The EdM Treasury and Risk Management Support Project (IFC advisory services; 602694) provided support to EdM (state-owned power utility) in resource mobilization and facilitation. The project was expected to result in the improvement of treasury, risk management, and resource mobilization functions of EdM. This would result in EdM improving its overall 310 Appendix F Case Study Overviews for Eight Countries performance and getting access to more financing. The Mining and Gas technical assistance (P129847) project assisted the National Petroleum Company (ENH) in validating its financial needs and evaluating financial options for the Area 1 LNG Project. World Bank ASA provided support for the lending program of the World Bank SOE reform in Mozambique through seven projects. In the energy sector, most of the World Bank ASA projects supported reforms in the regulatory framework at the sectoral level. The Mozambique Gas Master Plan and Policies Project supported the development of a Gas Master Plan and identification of reforms of the sector-related legal framework in the oil and gas sector. Mozambique Energy Sector Policy Work aimed to support the government in determining priorities for policy decisions to deliver efficiently produced, technically and financially sustainable electricity supply. At the national level, interventions supported SOE ownership, improving corporate governance, and reformed the regulatory framework. The Capacity Building for SOE Reform Project provided technical assistance to IGEPE to increase its capacity to rationalize its portfolio of SOEs and improve the performance of remaining state-owned commercial companies. The Enhancing Macroeconomic and Fiscal Policy Making for Inclusive Growth in a Resource-Rich Setting programmatic analytic and advisory activities focused on the management of revenue arising from natural resources, efficient public spending for better public services delivery, and harnessing the commodities boom for inclusive growth. The Strengthening Public Investment, Debt, and Fiscal Risk Management Program provided support to strengthen fiscal management capacity, including debt and SOE fiscal risks of government. The parliament of Mozambique approved a new legal framework for SOEs in 2018 (law) and 2019 (regulations), which was developed with support from this program. The Bank Group strategies and programs on SOE reform in the energy and financial sectors were aligned with the stated objectives of the government’s development plans. The Bank Group has been a key and trusted partner for 311 Appendix F Case Study Overviews for Eight Countries supporting the energy sector in Mozambique. In response to the hidden debt revelations in 2016, the World Bank launched a program to strengthen public investment and fiscal management. The program was designed to respond to government demand for support in these areas and to fill an existing gap in donor support. This program was also aligned with the overarching government public finance management reform program. The components of the program included strengthening public investment management and strengthening fiscal management, including debt and SOE fiscal risks. The Completion and Learning Report Review (CLRR) for FY12–15 concluded that the objective of improved transparency in extractive industries was achieved. However, the CLRR stated that the improved access to electricity outcome was not achieved. Similarly, the CPS Completion Report Review for the CPS FY08–11 reported that Mozambique’s progress in providing electricity services has not been as successful as expected. Continuity of engagement, design, monitoring and evaluation, assessment of the risks, and supervision were important internal factors that affected the implementation of the Bank Group’s SOE reform interventions. The continuity of Bank Group engagement, starting from the late 1980s, has enabled the World Bank to become a leading and trusted development partner, which was behind the success of projects. Project design issues constrained the implementation of SOE reform projects. For instance, the Implementation Completion and Results Report Review (ICRR) of PRSC series 9 to 11 stated that the program design failed to identify the inherent risks of SOEs. The ICRR argued that the design of the development policy operation (DPO) 10 and DPO 11 could have incorporated more measures on SOEs (after the revelation of the hidden debt) and new policy actions to deal with the possibility that further undisclosed borrowing had taken place. Lack of a specific indicator in the monitoring and evaluation framework was mentioned as a factor that impeded assessment of the impact of the intervention for capacity building and training for the state-owned power 312 Appendix F Case Study Overviews for Eight Countries utility. The ICR for the Transmission Upgrade Project acknowledged that potential procurement and implementation delays can be mitigated through the ex ante assessment of government procurement clearance requirements and procedures for large contracts. The ICRR for PRSC series 9 to 11 stated that the World Bank monitored the PRSC program closely through its field presence and engaging with development partners. However, when the loan guarantee for $850 million was revealed, the World Bank did not take strong remedial action and continued with the subsequent two PRSCs without refocusing on SOEs. Major development challenges ahead for Mozambique include maintaining the macroeconomic stability and reestablishing confidence through improved economic governance and increased transparency. Structural reforms are needed to support SOEs and manage fiscal risks. External factors included client commitment, agency coordination, public sector institutional capacity, and external shocks. The government had strong ownership of the program supported by PRSCs 6 to 8. The ICR indicated that a proactive stance led to significant progress in the EITI reform agenda. Similarly, the ICRR for PRSC series 9 to 11 mentioned that initially there was strong government ownership of the reforms that were part of the PRSC series. However, the government failed to disclose the questionable borrowing, and the concealment of the borrowing put the macroeconomic stability of the country at risk by expanding external debt that ultimately led to default. At the initial stage of the Energy Development and Access Project, EdM did not have a designated account, which led to delays in project implementation. Weak institutional capacity, which led to slow progress in project implementation, was one of the factors that is frequently mentioned in ICRs and ICRRs. In the Energy Sector Development and Access Project, the indicator to electrify 250 schools was achieved at 78 percent because of hurricane damage and vandalism. Furthermore, significant appreciation of the US dollar throughout 2014 caused an 313 Appendix F Case Study Overviews for Eight Countries unforeseen situation in which the project amount (denominated in special drawing right) had lost almost $7 million equivalent. Serbia (Desk Based) Serbia is an upper-middle-income and service-led economy that since sustaining armed conflict throughout the 1990s and experiencing frequent changes in its state framework, has made considerable progress in human development, in transitioning to a market economy, and in sustaining democracy. In 2017, Serbia’s GDP per capita was $5,992. 58 According to the IMF, Serbia’s real GDP growth in the first half of 2018 was the fastest in 10 years and continued at 3.8 percent year-to-year in Q3, supported by the strong recovery of private consumption and robust foreign direct investment and exports. SOEs’ weight in the Serbian economy remains high, but their performance remains poor. One of the major economic challenges ahead for Serbia is structural reform of SOEs. By 2017, there were about 1,200 SOEs employing more than 250,000 people, which is 15 percent of formal employment in Serbia (World Bank Group 2015), and the SOEs were overall characterized to perform poorly, have weak governance mechanisms, and be prone to political interference. Additionally, there are several major public utilities and numerous companies of different sizes in several sectors and under different legal frameworks that can be divided into state and socially owned enterprises, and municipally owned enterprises. Public enterprise wages also remain high. Most SOEs in Serbia are loss making and financially nonviable without state support. Half of all losses at the time were concentrated in former conglomerates that had been formally ring-fenced from bankruptcy and exempted from paying taxes and social contributions. SOEs stayed afloat by receiving direct budget subsidies and soft loans, along with indirect support through unpaid taxes and contributions, state guarantees for loans, and arrears to other state entities and public utilities. This support has had negative implications on the fiscal accounts of the state, allowed the 314 Appendix F Case Study Overviews for Eight Countries transparency of allocation of funds to remain low (given the fragmentation of state support to public enterprises), and was not kept in government records in the form of SOE-specific inventory of state support. In 2018, the IMF recommended that Serbia complete a reform of large public enterprises and develop a strategy for SOE governance. The authorities and IMF staff agreed that developing an ownership and governance strategy for SOEs would complement ongoing efforts to reduce and monitor fiscal risks and improve efficiency. The strategy, to be developed during 2019, would provide an integrated approach to oversight and monitoring of SOE operations, financial consolidation, restructuring or divestment for some strategic SOEs, and measures to improve governance and the institutional framework. Energy is one of the main contributors to the Serbian economy, accounting for nearly 10 percent of GDP. Elektroprivreda Srbije (EPS) is the state-owned power utility in Serbia and needs continued restructuring to achieve commercial and financial viability and to comply with European Union (EU) safety and environmental standards. The Serbian energy sector includes the oil and natural gas industry, coal mines, an electric power system, a decentralized municipal district heating system, and industrial energy. According to the Serbian Energy website (https://serbia-energy.eu/energy- sector-serbia), most of the Serbian energy infrastructure is state owned. EPS faces several challenges to meet future demand and succeed in the competitive European power market, including operational inefficiencies such as overemployment; below-cost tariffs for regulated consumers; lack of payment discipline, particularly from public sector entities; and obsolete infrastructure that needs to be replaced to meet EU safety and environmental standards (World Bank Group 2015). However, because of the steps taken to solve financial sustainability constraints of EPS, the financial standing of EPS improved in 2015–17. Organizational restructuring has led to a more streamlined organizational structure and management, and the financial consolidation plan 2015–19 laid out time-bound measures aimed at 315 Appendix F Case Study Overviews for Eight Countries transforming the company into a commercially and financially viable power utility. The most important step in the energy sector reform in Serbia was the adoption of the new energy law in 2014 for the implementation of the EU Third Energy Package. The main goals of the energy law were to increase security of the energy, improve energy efficiency, provide conditions for greater electricity generation from renewable energy sources, and enhance environmental protection in all areas of energy-related activities. JP Srbijagas is the state-owned natural gas utility in the country and needs continued financial stabilization. JP Srbijagas incurred significant annual losses reaching 1.2 percent of GDP in 2014. Its financial stabilization started in 2016, and progress has been made in terms of its financial position in 2016 and 2017. The Second Public Expenditure and Public Utilities DPL supported the continued implementation of reforms laid out in the Srbijagas financial consolidation plan. They also supported increasing electricity tariffs for cost recovery. Under DPL 1, the Council of the Energy Agency approved a 3.8 percent increase of the electricity tariff for guaranteed supply and amended the Energy Vulnerable Customers Program to increase coverage of targeted beneficiaries. Under the second DPL, the Council of the Energy Agency has approved an increase of the electricity tariff for guaranteed electric supply in 2017 and continued to protect vulnerable households from such electric tariff increase by increasing the number of beneficiaries of the Energy Vulnerable Customers Program. The financial sector in Serbia is dominated by the banking sector with 90.6 percent share in assets, of which state-owned banks hold 17 percent. Serbia has direct ownership stakes in four banks in Serbia, namely Komercijalna Banka, Banka Poštanska štedionica (BPS), Srpska Banka, and Jubmes Banka. Serbia has indirect ownership in MTS Banka and Jugobanka- Jugbanka Kosoversuska Mitrovica. Komercijalna Banka and BPS are regarded to have “systemic importance.” 316 Appendix F Case Study Overviews for Eight Countries The Serbian government decided to reform the state-owned banks by reducing the number of them with direct ownership to one, BPS, while divesting the other four banks. The banking sector in Serbia was hit hard by the global financial crisis of 2008 and incurred high NPLs, low profitability, and declining credit conditions. As a result, most of the state-owned banks incurred significant losses in past years, but their performance has recently improved. The major focus of the state-owned bank reform agenda is on ensuring the viability and sustainability of BPS. The Serbian government has prioritized restructuring BPS to minimize fiscal risk and develop a viable financial institution. Along with the state-owned banks, Serbia has stakes in other financial institutions that have an important role in the financial sector. Most recently, the Serbian government developed a reform strategy for state-owned financial institutions and requested the World Bank’s support through a results-based investment project financing to help implement this strategy. Economic reform plans by the government have continuously prioritized structural reform to the SOE and financial sectors. There is also a recent energy sector development strategy. In December 2007, Serbia submitted to the European Commission a Memorandum on the Budget and Economic and Fiscal Policy for the year 2008 with projections for 2009 and 2010. The 2007 memorandum presented a structural reform framework covering a wide range of structural reforms related to, among other things, the enterprise and financial sectors, public administration, and public finance management, and aiming to foster economic restructuring, enhance competition, stimulate employment, and rationalize social spending. Similarly, the Economic Reform Program 2018–20 has three goals: preserving fiscal sustainability and supporting growth, macroeconomic stability and inclusive growth, and continued efforts for EU accession. However, the Energy Sector Development Strategy 2025–30 seeks secure energy supply, its availability under transparent and nondiscriminatory conditions, and generation and use in accordance with principles of sustainable development. 317 Appendix F Case Study Overviews for Eight Countries Priorities in Bank Group country strategies included SOE reform. At the national level, the Bank Group’s priorities in CPS FY08–11 included continuation of SOE restructuring and privatization, in line with the government’s priorities. In turn, CPS FY12–15 sought to reduce the fiscal burden through improved competitiveness, including by reforming public enterprises, and CPF FY16–20 had two high-impact priorities related to SOEs. The first sought continuation of SOE reform, and the second sought, among other things, strengthening labor market institutions to help mitigate the consequences of such reforms. In the energy sector, CPS FY08– 11 aimed at improving financial sustainability and electricity market access for consumers and suppliers, and strengthening institutional capacity to participate in the regional electricity market. CPS FY12–15 was more sector focused and did not target SOEs; it sought improving the sector’s infrastructure to enhance competitiveness. CPF FY16–20 focused on SOEs, seeking a more efficient and sustainable power utility (EPS) and enhancing infrastructure networks. In the financial sector, CPS FY08–11 did not have a focus on SOEs; it sought strengthening financial sector intermediation to ensure incomes converge with Europe. In CPS FY12–15, the Bank Group aimed at strengthening the supervision of the financial sector, focusing on nonbank financial institutions. CPF FY16–20 sought to assist in creating a more stable and accessible financial sector for private sector growth and economic inclusion. Bank Group support to SOE reform in Serbia was delivered mainly through World Bank projects, and most of it was directed toward restructuring, privatizing, and improving corporate governance. IEG identified 31 Bank Group projects totaling $1.5 billion in commitments that were relevant to SOE reform in the financial and energy sectors. Of the 31 projects, 13 were World Bank lending (7 DPLs and 6 investment project financing), 3 were IFC investments, and 4 were IFC advisory services. The remaining 11 projects were World Bank analytical work. Ownership and corporate governance interventions usually went hand-in-hand in these projects, especially in DPLs. They were also combined with efforts to reduce direct and indirect 318 Appendix F Case Study Overviews for Eight Countries subsidies to SOEs to reduce fiscal deficits (First and Second Programmatic Private and Financial Development Policy Credits) with the implementation of the second electricity tariff adjustment set out in the financial consolidation plan (Public Expenditure DPLs), and with assistance to workers that SOEs planned to lay off in restructuring or the privatization process (SOE Reform DPLs, and in Public Expenditure and Public Utilities DPLs). In the energy sector, SOE reform also supported generation and transmission infrastructure, improving the financial management of SOEs, and developing renewable sources of energy. World Bank analytical work supported public financial management and gas sector reform. IFC provided support to Čačanska Banka to strengthen the capital base of the Čačanska Banka through capital increase and support its lending activities to the small and medium enterprise segment alongside the European Bank of Reconstruction and Development. The KBB SubDebt project provided support to the largest locally owned bank in the Serbian banking sector, Komercijalna (KBB) with long-term financing, which was unavailable in the market at that time. By providing KBB with the sub-debt, IFC aimed to help KBB strengthen its capital base at a time when Basel II capital requirements were to be adopted in Serbia. IEG’s CPS Completion Report Review FY08–11 rated the pillar of encouraging private sector–led growth to ensure income convergence with Europe, under which SOE reform was pursued, as moderately satisfactory. For this pillar, the CPS Completion Report Review FY08–11 assessed the specific objective of “bolstering the competitiveness of the enterprise sector” and stated that of the 1,000 SOEs in the government’s portfolio for which divestiture had not been completed, 430 were sold over 2008–09, and 162 were sent into bankruptcy in 2010. IEG’s CLRR of the CPS FY12–15 stated that overall targets for SOEs were mostly achieved. According to the CLRR, financial viability and investment of the power utility EPS remained impaired after the FY08–11 period. However, in the FY12–15 period, the outcome of improving energy 319 Appendix F Case Study Overviews for Eight Countries efficiency was mostly achieved. The support for this objective came through IFC Balkans Renewable Energy advisory project that succeeded in reducing energy use and the emissions from the use of fuels to generate energy. In the financial sector, the financial system was strengthened after the CPS FY08– 11 period, including through restructuring of state-owned banks. The CLRR of CPS FY12–15 stated that in the banking sector, this result was achieved. Important factors of success for the Bank Group’s engagement included close coordination with other development partners, the programmatic nature of the DPLs with support from analytical work, and the design of monitoring and evaluation frameworks. The CPS Completion Reports emphasized the importance of monitoring and evaluation and stated that the monitoring and evaluation framework should be strengthened in the next CPS. Close coordination with the European Commission, other development partners, and international financial institutions (IMF, the European Bank for Reconstruction and Development, and the European Investment Bank) was mentioned as one of the factors contributing to the implementation of the country strategies. The CLR for the CPS FY08–11 highlighted that the use of programmatic DPL series supported by high- quality ASA was a successful means of supporting and sustaining a difficult policy and institutional reform effort. The alignment of CPS activities with the EU integration agenda was also indicated as one of the key factors affecting the implementation of projects. Institutional capacity was also mentioned as one of the factors affecting the project implementation. The CLR for CPS FY12–15 stated that high-level policy dialogue was critical for achieving several outcomes, particularly in state and socially owned enterprise reform and energy sector reform. Ukraine (Field Based) SOEs constitute a substantial portion of the Ukrainian economy with net revenues accounting for almost 25 percent of GDP. SOEs are the largest supplier of public services and dominate the oil and gas sectors, transport, 320 Appendix F Case Study Overviews for Eight Countries electricity generation and distribution, water supply, machine building, chemicals, and coal mining. The state is the largest owner of financial institutions as a result of the nationalization of systemically important banks during the 2008–10 and 2014–16 crises. State-owned banks own more than 50 percent of banking sector assets. During the 1990s there were mass privatizations, but these had no more than mixed success in improving productivity and service. Most companies that remain under government ownership operate at a loss, primarily because of mismanagement, corruption, and vested interests. The Bank Group supported national-level SOE reform during the late 2000s, mainly through DPLs, achieving no significant progress. The World Bank included SOE management reform and improving the quasi-fiscal position of SOEs as prior actions in a 2006–12 DPL series. There was no progress in these areas. The main reason for lack of progress was that the government was not genuinely interested in reforming SOEs. Especially after the 2008 crisis, the appetite for reform within the country waned. After this experience, World Bank support was reoriented toward sectoral reform. The 2014–16 crisis brought the issue of SOE reform to the fore because of the need to reduce the drain on budgetary resources. Weak corporate governance and poor risk management are substantial barriers to improving the performance of SOEs. Furthermore, although the government has paid lip service to the need for reform, as in the former crisis, once the crisis abated, progress slowed markedly because genuine reform commitment is still absent, primarily because powerful oligarchies within Ukraine benefit substantially from the status quo. During the recent crisis, the European Bank for Reconstruction and Development took the lead in supporting the government in SOE reforms (that is, privatization and corporate governance). World Bank involvement in national-level SOE reform has been limited to policy discussions at the national level, and World Bank engagement at the sector level has been more comprehensive and includes lending and technical assistance. 321 Appendix F Case Study Overviews for Eight Countries A theme that was repeated in both crises was the opportunity they provided for SOE reform (both at the national and sector levels), but as recovery occurred, the government and oligarchs were able to thwart or even reverse progress. Under such circumstances, it is an open question whether World Bank assistance (and other donors’ support) risks providing additional resources for rent seeking rather than support for sustainable reforms. Before the 2008 crisis, there were two state-owned banks that controlled less than 10 percent of the assets of the banking sector. During that crisis, three private banks were nationalized, although two of them were subsequently declared insolvent. Although the government was able to stabilize the crisis with substantial support from international donors, the factors that led to the crisis—namely, high levels of NPLs, weak corporate governance, widespread related-party lending, and ineffective supervision—were not addressed in a sustainable manner, and the earlier practices that had contributed to the onset of the crisis resumed. The World Bank supported the banking sector during the 2008–10 crisis through DPLs that helped establish the institutional framework for bank recapitalization, broadened the powers of the Deposit Guarantee Fund, and promoted the commercialization of state-owned banks. However, the effectiveness of the assistance to state-owned banks was questionable because the influence of the former owners remained high. When the 2014–16 crisis emerged because of the shock to the economy resulting from political upheaval, the annexation of Crimea, and the separatist war in eastern Ukraine, the weaknesses in the financial system resulted in a large deleveraging of the banking sector and withdrawal of both domestic and foreign currency deposits that threatened a systemic collapse. There were widespread bank failures. Of the 180 banks that existed before the crisis, 92 subsequently failed. In 2016, the largest private bank in the country, PrivatBank, was nationalized because imprudent lending practices to related-party borrowers had depleted its capital; recapitalization by the state required an injection of nearly 5 percent of GDP. After this 322 Appendix F Case Study Overviews for Eight Countries nationalization, the state owns the three largest systemically important banks that control more than half of the assets of the banking system. There was strong reform momentum during and immediately after the 2014–16 crisis, which resulted in the passage of legislation critical to strengthening the institutional foundations of the banking sector. As part of coordinated donor support during the crisis, there was substantial World Bank assistance for restoring the stability of the financial sector in the form of several DPLs and technical assistance. World Bank assistance included bolstering the framework for banking supervision and regulation, strengthening the oversight capacity of the Bank of Ukraine and the Deposit Guarantee Fund, and increasing the power of the Deposit Guarantee Fund to deal with the resolution of insolvent banks. These two institutions are now recognized as among the best in Eastern Europe and the former Soviet Union countries. With the support of international financial institutions (in which the World Bank took the lead), the government prepared a strategy and road map in 2016 for the reform of the state-owned banks. The strategic priorities were improving governance, dealing with NPLs, and developing a policy for their privatization. The strategy was further revised in 2018, but as yet, very limited progress has been made. So far, no mechanism has been developed for the resolution of NPLs of the state-owned banks, which still amount to more than 50 percent of their portfolio compared with 10 percent for privately owned banks. IFC’s involvement was mainly limited to policy-level discussions and the provision of technical assistance to one of the state-owned banks. The World Bank financial sector specialist and IFC’s senior financial market investment officer coordinate closely with respect to promoting privatization, participating in independent supervisory board member selection, and assisting with the resolution of NPLs. IFC has provided technical assistance to Ukrgasbank regarding its small and medium enterprise business and 323 Appendix F Case Study Overviews for Eight Countries corporate governance practices. IFC is also considering supporting the World Bank’s preprivatization through a convertible debt investment. The Ukrainian authorities have been positive regarding World Bank support since the most recent crisis. The proactive engagement of the World Bank’s locally based personnel was highly appreciated by stakeholders, and the ability of current staff to establish relationships and engage with Ukrainian counterparts was regarded as a major factor in maintaining some momentum for reform. This approach to proactive engagement can be described as best practice. Nevertheless, as in the previous crisis, reform momentum was lagging. Additional uncertainty casts doubt on the sustainability of reform and illustrates the ongoing ability of oligarchs to influence the court system. The national prosecutorial system and the court system are not impartial, so the contracting framework is vulnerable to outside pressure. For instance, a recent decision by high courts in Ukraine ruled that the nationalization of PrivatBank was illegal and that it should be returned to its former owner (an oligarch). Nationalization of the bank was highly supported by the World Bank and other international financial institutions and was part of the efforts to stabilize the banking system in the country. Although this decision is not the last legal step in the process, it illustrates that sustained reform is difficult to achieve in Ukraine. Furthermore, interviews conducted during the mission suggest that the real reason for the nonresolution of the NPL issue is a lack of political will, particularly because the largest debts are owed by about 20 major oligarchs, many of whom are members of parliament. Ukraine possesses large energy resources, but this sector has been poorly managed and has a high prevalence of corruption. Furthermore, Ukraine’s highly subsidized energy prices have led to excessive energy consumption and unnecessary current account deficits. Ukraine committed to the EU’s Third Energy Package, legislation aiming to improve the internal gas and electricity market. This initiative guides Ukraine’s current energy reforms. 324 Appendix F Case Study Overviews for Eight Countries The gas sector in Ukraine is largely state owned and dominated by Naftogaz, the national vertically integrated oil and gas company, which relied on importing gas from Gazprom until recently. The monopolistic market structure without adequate regulation has led to large sector inefficiencies. The lack of basic governance structures and transparent internal controls, inconsistent and excessive state intervention, and inadequate metering of gas flows intensified the inefficiency of the sector. Gazprom has been the monopoly in the Ukrainian gas transit market. Over the years, Ukraine and Gazprom have had a substantial number of arbitration cases related to gas transactions, involving alleged arrears, contract and price disputes, and most recently, related to the Russian-Ukrainian conflict beginning in 2014. The country has substantially decreased its gas imports from Gazprom. Gas and district heating tariffs for households have historically been heavily subsidized, which created large fiscal and quasi-fiscal deficits for the country. The IMF assessed the state energy subsidies at 10 percent of GDP in 2014. Ukraine’s domestic household prices for natural gas in 2014 were at only 12 percent of the world market price level or cost recovery prices. Because of Ukraine’s commitments to the IMF, energy prices for households were increased by up to four times for natural gas in between 2014 and 2016. Therefore, the reduction of quasi-fiscal deficits through tariff increases for gas and heating were temporarily achieved. However, as with any other reform in the country, sustainability of such improvement is uncertain. In May 2019, the government announced plans to cut gas prices instead of implementing a planned increase of 15 percent. The World Bank has led the gas sector reform and the unbundling of Naftogaz, which has become an extremely politicized issue. The government, with lead support from the World Bank, developed a gas reform strategy in 2015, and some progress has been achieved in the sector. Furthermore, because of the arbitration issues with Gazprom and Gazprom’s unreliable supply of gas, the World Bank provided timely support through an International Bank for Reconstruction and Development guarantee to Naftogaz to help it mitigate the risk associated with uncertain 325 Appendix F Case Study Overviews for Eight Countries gas supplies. However, the government has been inconsistent in maintaining its priorities with respect to the unbundling of Naftogaz and still has not taken the necessary steps to effectuate it. Whether the unbundling of Naftogaz will occur is an open question because there is disagreement between the government and Naftogaz regarding the unbundling of the gas sector. Furthermore, the donors (United States and European) and the World Bank have divergent views on how to unbundle Naftogaz. The World Bank supported financial management of Naftogaz (state-owned oil and gas company) through Ukraine Gas Supply Security Facility Project (P155111) with a guarantee that sought to reduce its financial liquidity constraints and thereby enhance its capacity to secure gas supply for the country. The facility would thereby allow Naftogaz to increase volumes of gas in storage during summer months and support purchases of gas during winter months. Three DPLs (P150313, P096389, and P107365) included measures for decreasing quasi-fiscal deficits. Residential gas tariffs increased by 450 percent and heating tariffs increased by 250 percent between 2014 and 2016, which allowed the Ukrainian government to achieve full cost recovery at Naftogaz and District Heating companies in 2016, one year ahead of the original schedule. As a result, Naftogaz’s financial deficit of $3.3 billion (2013) declined to zero (on cash basis) in 2016, meeting the key DPL results target. However, in May 2019, Naftogaz and the government announced a cut in gas prices for households and heat producers and to prevent any gas prices increase in the future, although gas prices had been due to increase by 15 percent. The state has strong control of the electricity sector through several SOEs, but it is in the process of opening up the market by aligning with the EU. The electricity market in Ukraine involves SOEs at every level: generation, transmission, and distribution. The existing model is one of a single buyer and seller of electricity. Liberalization of the wholesale electricity market of Ukraine has been on the agenda of the country since 1996. However, policy 326 Appendix F Case Study Overviews for Eight Countries reforms to liberalize the electricity sector only commenced with the adoption of the Wholesale Electricity Market Law in 2008. The electricity law, which was prepared to align with the EU’s Third Energy Package, was approved by the Ukrainian government in 2017, and the implementation of the law was supposed to start in July 2019. However, market participants seemed unprepared for the full-fledged implementation of the market opening. The World Bank supported the establishment of the National Commission for State Energy and Public Utilities Regulation. DPL2 (P151479) included a prior action to support the government’s plans to strengthen the independence of the energy regulator through the enactment of relevant legislation by the Cabinet of Ministers in July 2015. However, approval of this legislation in parliament was delayed until November 2016, requiring substantial donor pressure to get it passed. By March 2017, plans to rotate energy regulator commissioners and to collect licensing fees to provide an independent funding source for the regulator (two key components of the DPL energy regulator law) have yet to become operational. Stakeholders consulted during the mission expressed their concerns regarding the independence of the regulator. Ukraine’s electricity sector struggles with fundamental obstacles. The industry needs to generate enough funds on a sustained basis to support modernizing infrastructure and facilitate the complex task of integration with the EU’s power system. The generation, transmission, and distribution networks each require substantial investments. Most of the power generation assets will reach the end of their life cycle during the next decade. The electricity sector also faces problems related to regulation, pricing, and commercial operations. The World Bank helped the country to start rehabilitating infrastructure of the main SOEs in the electricity sector and achieved some success in improving reliability, safety, and so on. However, it did not achieve its electricity market reform objective. The World Bank engagement 327 Appendix F Case Study Overviews for Eight Countries commenced in the middle of the 1990s, which created a foundation for the investments to rehabilitate the infrastructure for UkrHydroEnergo (the state-owned electric generation company) and UkrEnergo (the state-owned electric transmission company), the main companies in the sector. Furthermore, the World Bank provided financing for an investment program, that is, hydropower rehabilitation, in return for policy reform. Opening the electricity market was a rather ambitious objective for a project that primarily aimed at rehabilitating a number of hydropower plants. The implementation of the wholesale electricity market concept was not achieved with the support of the project. However, in the reform area, the World Bank has been playing a productive and constructive role without driving the reform agenda. Choice of instrument, engagement of Bank Group staff, analytical work, and coordination among Bank Group institutions were the internal factors affecting the implementation of SOE reform interventions in Ukraine. The World Bank’s investment lending was not the best choice of instrument to support the government’s electricity sector reform. An instrument focused specifically on the reform would be a better option. The presence and frequent engagement of task team leaders were useful to clients, especially when the implementing unit’s capacity was low or the project required frequent interactions with the government. Furthermore, good analytical work helps develop trust and helps improve the World Bank’s credibility for future engagements with clients. Close coordination helps the Bank Group provide full support to the client and to potentially be more impactful. Donor coordination, political economy, government commitment, vested interests, and governance culture were the external factors that affected the outcomes of the SOE reform interventions. Having a unified position toward reforms and policy action among donors helped push government reforms. However, the progress in the reforms was highly dependent on the political context in the country and the commitment of the government. Furthermore, vested interests, the power of oligarchs, and corruption have 328 Appendix F Case Study Overviews for Eight Countries slowed SOE reform efforts in the energy and banking sectors. Lack of a governance culture in the country impedes SOE reforms, which is regarded by international financial institutions as a big hurdle in promoting SOE reform. The political events and financial crisis created windows of opportunity to improve the macroeconomic framework and undertake structural reforms, but these periods were relatively short, and the momentum slowed after the crisis. In countries that have corrupt power structures and court systems, sustainable reform is difficult to achieve. Under such circumstances, it is an open question whether World Bank assistance risks providing additional resources for rent seeking rather than support for reforms. Vietnam (Field Based) Vietnam is considered a development success story with notable poverty reduction and economic growth over the past 30 years. Political and economic reforms (known as Doi Moi) launched in 1986 have transformed Vietnam from one of the poorest countries in the world to a lower-middle- income country in 2009. SOEs have historically played a significant role in Vietnam since the country’s independence, but by 1986 their limitations were already evident. Under the central planning regime, until 1986 there were only two types of firm ownership in the economy: state enterprises and collective enterprises. However, recognizing the benefits of reducing state control across the economy, in 1992 the Vietnamese government launched a privatization program to improve the performance of SOEs. The process of privatization moved slowly and gradually, starting with the easier and smaller SOEs and then continuing with the more difficult and larger ones. The Vietnamese government kept a tight rein on major industries such as utilities and banking through large, state-owned economic groups and enterprises. The privatization process in Vietnam was inhibited by factors such as the fear of losing authority and perquisites by those in control, the 329 Appendix F Case Study Overviews for Eight Countries problems of asset valuation, the fear of losing preferential credit from state commercial bank, and the persistence of soft budget constraints leading to loss-making firms. During the period 2001–06, the number of SOEs decreased from an estimated 5,600 to 2,100, and liberalization picked up in 2004 as the focus on the SOEs targeted switched from smaller enterprises to including 100 percent state-owned companies. This was primarily achieved through the equitization of SOEs by selling their shares to create joint-stock companies that operate under the country’s Enterprise Law rather than the SOE Law. However, it is notable that the SOE capital shares sold by 2007 accounted for only 14 percent of state capital, with only 23 percent of companies with chartered capital of more than 10 billion Vietnamese dong (equivalent to about $500,000) having completed the equitization process. In 2005, Vietnam accelerated the process of creating state economic groups—a loose alliance of SOEs with similar business interests. In 2012, the government of Vietnam launched a number of reform initiatives, including a steering committee for the restructuring of SOEs (to be headed by the Ministry of Finance) improved legal framework for equitization, and formulation of a decree to separate the state regulation function from state ownership rights. In Vietnam, state economic groups and SOEs tend to operate in more capital-intensive sectors, including in the energy sector, which is dominated by state ownership. Electricity of Vietnam (EVN) held 60 percent of Vietnam’s 42-gigawatt installed generation capacity directly and through its three generation subsidiary companies. The remaining 40 percent are owned by other SOEs and private investors. EVN also fully owns the National Power Transmission Corporation, which operates and maintains the national transmission grid. The unbundling of EVN’s operations began in 2008 with the setup of a single trading company—Electricity Power Trading Company—and with the creation of the National Power Transmission Corporation. In 2012, EVN unbundled its generation companies into three 330 Appendix F Case Study Overviews for Eight Countries separate generation companies, aligned with the launch of Vietnam’s competitive generation market. This market was the first stage of power sector reform, with independent power producers and generation companies competing in a power pool to sell to the single buyer. In 2015, the pilot phase for a wholesale electricity market was launched with the goal of being fully operational by 2021. As of 2016, 24 percent of installed capacity in Vietnam was privately held. State presence in the banking sector is large and involves both direct and indirect ownership links. The four largest state-controlled banks—namely Bank for Investment and Development of Vietnam, Vietnam Bank for Agriculture and Rural Development, Vietnam Bank for Industry and Trade, and Commercial Bank for Foreign Trade of Vietnam—dominate local banking activity, accounting for nearly half of total sector assets. The total participation of the state in the commercial banking system is larger because the SOEs and SOCBs have equity participation in several of the 31 joint- stock banks. State-controlled banks dominate the market in terms of loans and deposits, 59 although their share has been declining since the 1990s, decreasing to 70 percent in the early 2000s and later to 45 percent in 2017 (Campanaro and Dang 2018, 83–124). The FSAP 2014 report prepared by the Bank Group and the IMF stated that the weak performance of the financial sector was related to a complex array of institutional and regulatory factors. These factors included episodes of interference by central and local authorities on the investment and credit decisions of SOEs and SOCBs, inadequate governance structures and risk management capacity in these institutions, connected lending in several joint-stock banks, weaknesses in financial infrastructure (including poor financial reporting standards), and deficiencies in financial regulation and supervision. The recommendations included recapitalization plans, the workout of NPLs, and regulatory reforms. The Bank Group, excluding the World Bank analytical work, supported 30 SOE reform projects in energy and financial sectors in Vietnam during the 331 Appendix F Case Study Overviews for Eight Countries period FY08–18. Of these, 39 percent of the projects supported SOEs in financial sector, 34 percent in the energy sector, and 27 percent at national level. World Bank lending accounts for the majority with 24 projects, followed by IFC advisory services (4 projects), IFC investment services (1 project), and MIGA (1 project). Development policy loans and investment lending have nearly the same share in the World Bank lending portfolio. Almost three-quarters of the projects in the portfolio were evaluated, 6 projects are still active, and 2 projects are not evaluated. Half of the projects supported SOE reforms at the downstream level, 40 percent addressed SOE reform issues at the upstream level, and 10 percent of the projects assisted SOE reforms at both levels. The World Bank has been strongly supporting EVN’s restructuring and the unbundling of its services. The MIGA guarantee covered a nonshareholder loan for the design, operation, and maintenance of a privatized hydropower plant. Bank Group engagement in SOE reform was strongly aligned with country priorities reflected in socioeconomic development plans of Vietnam and in the World Bank’s country strategy documents. The constraints to the SOE sector highlighted in these documents (that is, governance, transparency, and financial management) are all addressed through the World Bank’s intervention in energy at a national level, promoting sector-specific objectives and those relating to the environment and social inclusion. At the national level, 11 International Bank for Reconstruction and Development and International Development Association projects are identified— 10 of these are World Bank adjustment loans, and just 1 is a World Bank investment. Six interventions of Vietnam’s PRSC series (4–10) fall within this evaluation’s timeline. Under this series, most interventions are upstream and aim to support Vietnam’s transition to a market economy and business development through state sector and SOE reform. Relevant PRSC components began with strategy development to identify and classify SOEs and their performance. There is a level of focus on addressing issues of expenditure management, governance, and transparency to support 332 Appendix F Case Study Overviews for Eight Countries equitization and corporatization of SOEs at a national level. This was also a strong focus of the Economic Management Competitiveness Credit series (1–3), which worked to implement and enact regulatory changes toward the objective of improving corporate governance for more transparency across the board and in the restructuring of SOEs. SOE reform in the energy sector was primarily supported through World Bank investments and adjustment loans. Relevant interventions under the PRSC series (4, 8, and 10) and the Power Sector DPO series (1–3) targeted the sector at an upstream level. The PRSC series broadly supported the enacting of the Electricity Law while working to improve sector strategies for gas and electricity subsectors and supporting the adoption of market-based pricing mechanisms electricity. The Power Sector DPO series worked in four key policy areas: development of a competitive power market, power sector restructuring, electricity tariff reform, and demand-side energy efficiency. World Bank investments and one MIGA guarantee primarily targeted objectives at the downstream level. These related to infrastructure improvements with an initial focus on transmission and distribution, toward the goal of improving the quality and reliability of services provided. In the financial sector, more than two-thirds of the SOE reform engagements were financed through World Bank lending. Five of the World Bank lending projects were PRSC series aimed at implementing Vietnam’s socioeconomic development plan, and three were engaged through programmatic series of DPOs (Economic Management and Competitiveness Credit 1–3), which contributed to strengthening financial sector governance and fiscal management. Two of the World Bank investment projects (Rural Finance 2 and 3) sought to support the Bank for Investment and Development of Vietnam and Vietnam Bank for Agriculture and Rural Development to enhance effectiveness of supervision, improve risk management, strengthen human resources management and training, and upgrade information technology and banking technology. The third World Bank investment project (Payment System and Bank Modernization 2) aspired to strengthen 333 Appendix F Case Study Overviews for Eight Countries operational efficiency and risk management of the banking sector in Vietnam through implementing management information systems for safe payment services in four participating banks (three of the country’s largest SOCBs and one joint-stock bank). IFC invested in Vietnam Industrial and Commercial Joint-Stock Bank (VietinBank), which is one of four SOCBs and the fourth largest local bank in Vietnam. IFC’s investment aimed to facilitate the privatization of VietinBank and enhance the World Bank’s operational capacity. IFC also provided advisory services to VietinBank, An Binh Bank, and Banknetv mainly to improve business, operations, and risk management processes. Assessing results for the SOE sector at a national level, there has been a reduced role of the state sector in the overall economy, but this may largely be attributed to a rapidly growing private sector. The PRSC series and interventions aiming to improve transparency, governance, and financial management made incremental progress. At a national level, regulations were supported to help reduce the size of the state sector, with quantitative targets for the reduction in the total number of SOEs through successful equitization and corporatization efforts. The Economic Management and Competitiveness Credit project series also supported SOE restructuring to improve the performance and efficiency of public assets. In energy, Bank Group SOE-related interventions have been successful, considering most indicators are rated at mostly achieved or achieved, with substantial and modest improvements achieved along all dimensions of energy projects. Some of the results achieved through the projects identified in the portfolio include the following: infrastructure improvements in distribution and transmission to improve the quality of service and access in rural areas (P099211), strengthening of sector strategies through early DPL interventions, the successful unbundling of EVN’s generation, transmission, and distribution services for improved competition, and the launch of Vietnam’s competitive generation market. However, one specific objective appears repeated across a number of projects. This is the adoption of 334 Appendix F Case Study Overviews for Eight Countries market-based tariffs and pricing mechanisms to improve the financial viability of EVN and attract other players to the market. Regarding the financial sector, IEG’s review of the CPS FY07–11 stated that the revised State Bank of Vietnam and Credit Institution Laws set the basis to improve the governance of financial institutions, and the laws are under implementation. However, by the end of the CPS period, State Bank of Vietnam was not operationally independent. On corporate governance, the government equitized two of the five SOCBs, falling one bank short of the target, and the share of NPLs declined to less than 5 percent, as targeted by the strategy. IFC participated in the equitization of the third largest bank and provided support to strengthen risk management and improve corporate governance. Overall, governance of state-owned banks remains challenging, with no single state shareholders and many state entities able to intervene in decision-making. Outcome targets related to financial sector reform under the PRSC series 6– 10 included the proportion of NPL and the number of banks using qualitative criteria for loan qualification. The ICRR for the PRSC series stated, “The NPL ratio has proven to be quite difficult to measure, and the corresponding indicator has not been met—rising NPLs may have been linked to the financial crisis.” The ICR mentioned that there has been little progress in credibly estimating and disclosing information concerning NPLs in the banking sector. On the achievements side, seven banks started to use qualitative data when deciding loan qualification (compared with the target of five banks). Furthermore, equitization plans were made for two SOCBs under PRSC 7, and regulations were issued to enhance public disclosure and communication of State Bank of Vietnam’s policies and banking sector statistics under PRSC 10. As for the Economic Management and Competitiveness Credit DPO series, which was implemented during the CPS FY12–16 period, the progress in terms of nonperforming loans and debt was more modest than had been expected. 335 Appendix F Case Study Overviews for Eight Countries The SOE reform portfolio review for Vietnam revealed that client commitment, project design, and collaboration with external actors and donors are the top three factors that affect the success or failure of SOE reform interventions. These three factors accounted for almost half of the factors identified and followed by analytical work, identification of risks at appraisal, and the monitoring and evaluation framework. The government demonstrates its willingness to improve the functioning of SOEs while gradually making them more independent. This is more evident in the energy sector than in the financial sector (where the state still has a strong presence in the so-called joint-stock commercial banks that have significant government ownership through the shareholdings of large SOEs). Contrary to the rapid privatization in other transition economies, the privatization process in Vietnam was inhibited by factors such as the fear of losing authority by those in control, the problems of asset valuation, the fear of losing preferential credit from state commercial banks, and the persistence of soft budget constraints leading to loss-making firms. With a few World Bank adjustment loan interventions related to national, finance, and energy SOE reform, the complexity of project design was highlighted across sectors and interventions. In certain cases, a number of implementing agencies were involved, making external collaboration and coordination more difficult to manage. It is critical to retain a certain level of flexibility in design and in the application of funds (particularly in challenging circumstances) for projects to adapt to unexpected changes. Phased or sequenced approaches in project design appear to be more helpful to the achievement of objectives. The PRSC series benefited from strong support from development partners, under the nominal leadership of the World Bank. The program was successfully supported by 14 donors, and total financial flows through budget support over five years were nearly $3 billion. In terms of institutional capacity, the public sector in Vietnam has adequate technical capacity, but finances of its largest entities like EVN and the large state- owned banks are weak, and they face challenges in corporate governance. 336 Appendix F Case Study Overviews for Eight Countries Link between Bank Group Country Diagnostics and Strategies for Case Study Countries Table F.1 shows the link between identification of SOE reform issues in Systematic Country Diagnostics and inclusion of these issues in CPFs in case study countries. Among the case countries, there is a Country Private Sector Diagnostic only for Kenya. Table F.1. Link between World Bank Group Country Diagnostics and Strategies for Case Study Countries SCD and CPSD CPF Bangladesh No CPSD. 2016 (FY16–20) SCD 2015 Financial sector National level • New IDA operations are • Improved public financial planned to strengthen the management (better corporate governance of implementation of public state-owned financial investment through the institutions and to expand Annual Development the use of personal Program and enhanced identification by linking it to fiscal risk management of delivery of public services. SOEs) is needed to • IDA financing will seek to strengthen the link improve financial market between policy priorities infrastructure and the and resource allocation, regulatory and oversight and for efficient use of capacity of Bangladesh resources for service Bank. It will help strengthen delivery. state-owned banks Financial sector (commercial and development banks) • Strengthening SOCBs, through automation and including their improved management governance, and systems and lay the enhancing overall foundation for a well- supervisory and 337 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF regulatory capacity are functioning insurance and critical initiatives to be pension sector. undertaken. Energy sector Energy sector • The CPF aims to increase • Key challenges include power generation capacity constrained electricity and and access to clean energy natural gas supply, to improve growth and unsustainable short-term competitiveness, in solutions, poorly targeted alignment with the subsidies and significant government’s seventh 5- fiscal burden, and year plan and SCD FY15 distorted market signals. priorities. Priority reforms pertain to • The Bank Group will seek to increase generation narrow the gap between capacity and enhance supply and demand by access to those lacking providing efficient and power. reliable energy supply, diversifying into renewables, and increasing access to electricity. • The Bank Group will lead generation interventions by supporting new gas-fired generation capacity and the repowering of existing gas- fired plants to enhance their efficiency. • In the transmission subsector, the Bank Group had ongoing support to improve the financial and operational efficiency of Power Grid Company of Bangladesh. 338 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF China No CPSD. 2019 (FY20–25) (beyond evaluation SCD 2017 period) National level National level • Level the playing field • Given that China wants SOEs between SOEs and non- to retain an important role in SOEs to enhance market China’s economy, fair competition and promote competition between SOEs the private sector. and non-SOEs would be needed to ensure that the • SOEs still account for markets select the most approximately one-third of productive enterprises all investments. Allowing regardless of their greater private sector ownership structure by competition in some key exposing SOEs to sectors, such as oil and competitive pressure. While gas, electric power, China has made progress, finance, and much remains to be done on telecommunications, this complex and wide- through a more level ranging reform agenda, the playing field could success of which has strengthen SOEs’ important implications for performance by exposing the rest of the world. them to greater competition. • Indicators for competition: Competitive neutrality • Relevant reforms could principles (as noted in the include requiring a market Government Work Report, rate of return on state March 2019), implemented equity capital and by 2022. removing perceived government-implicit Energy sector guarantees of SOE • The Bank Group aims to borrowing. It could include support the improvement of equal access to land, the efficiency of energy natural resources, and markets and facilitate the government subsidies as removal of regulatory and well as equal treatment in institutional barriers that regulations, tax, 339 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF government procurement, lock-in the high dependency and administrative on coal. approvals. • Indicator for energy Financial sector efficiency: Development by • Greater private sector 2022 of a market-based EE competition in the trading mechanism. financial sector could help • Indicator for renewable improve the efficiency of energy: Development of financial intermediation. seven new or revised Energy sector policies, regulations, and standards related to grid Use of fossil fuels could be integration of distributed reduced through continued renewable energy by 2023. promotion of energy efficiency, renewable energy, efficiency in the heavy industries, and green transportation. Renewable energy can continue to grow through further reforms in power tariff structure and dispatch rules. Egypt, Arab No CPSD. 2015 (FY15–19) Rep. SCD 2015 Energy sector Energy sector • The CPF aims to improve • Continued energy subsidy governance of energy reform would pay a triple sector SOEs. dividend by improving the • The Bank Group seeks to country’s fiscal position, strengthen energy sector incentivizing labor- governance through a DPF intensive production and on fiscal stabilization, reducing insider privileges. sustainable energy and • To recover from competitiveness that will unsustainable financial support actions to ensure situation and to meet sustainable energy supply energy demand efficiently through continued reduction and sustainably, the SCD of energy subsidies, 340 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF recommended governance improvements corporatizing the sector in the power and gas including through sectors, opening of the gas improved corporate sector to private governance, regulations, investments, establishment competition, privatization of an independent gas of segments of the sector, sector regulator, and and restoration of the enabling of private sector borrowing capacity of the investments in renewable holding companies. and cleaner forms of energy. • Indicators: reduction of energy subsidies as proportion of GDP and increase in average electricity tariff. • The CPF aims to improve energy generation capacity and energy efficiency. The World Bank, IFC, and MIGA propose to continue to finance private sector–led renewable and nonrenewable energy generation and transmission infrastructure, and the World Bank will support governance improvements (through the DPF) to increase private sector investments and enhance regulatory oversight. Kenya No SCD. The new CPF is planned to be CPSD 2019 completed by December 2020. • The CPSD finds that direct 2014 (CPS FY14–18). competition from SOEs, Energy sector links between competing 341 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF firms through partial • The CPF aims to support the government shareholding, government’s effort to boost and a lack of competitive private sector participation neutrality given limited de and strengthen the financial facto separation of position of Kenya Power and regulatory and Lighting Company. commercial activities in • The World Bank aims to sectors such as electricity, support the government’s air transport, telecom, and oversight in the oil and gas agriculture can crowd out sector and maximize the private sector. responsible private sector • SOEs generate a involvement and shared significant burden on fiscal benefits across key accounts, amounting to segments of the population. 7 percent of GDP in 2016. • The World Bank is • CPSD suggests supporting the development strengthening competition of a Petroleum Sector policy and removing Master Plan and Kenya regulatory barriers and Petroleum Technical government interventions Assistance Project. that restrict entry and competition in various key sectors, including electricity generation and insurance. • The report recommends improving governance and market discipline mechanisms toward SOEs to increase their efficiency and to help crowd-in the private sector. • In the energy sector, the CPSD recommends encouraging private participation through PPPs 342 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF (particularly in transmission); building capacity of the Energy Regulatory Commission and the Ministry of Energy; and implementing a wholesale electricity market. Mozambique No CPSD. 2017 (FY17–21) SCD 2016 National level National level • The CPF highlighted the • Public corporations are need to strengthen debt increasingly engaged in management and manage public investment, fiscal risks better. Reforms to elevating the risk of develop effective oversight contingent liabilities. over SOEs and other public entities were considered as • The recently disclosed urgent, along with reforms loans highlight the need to to overhaul the framework further strengthen the for managing guarantees. public financial management system, • The CPF program will particularly fiscal risk feature advisory support and management. possible policy-based lending focused on fiscal • SCD recommends sustainability and debt strengthening the management. institutional framework for managing public • The CPF proposes technical resources and improving assistance and development fiscal risk management. policy lending aimed at strengthening the legal Energy sector framework for SOEs. • The energy sector is Energy sector underdeveloped and suffers from major • The CPF aims to build inefficiencies in generation stronger capacity for and transmission. oversight and increase transparency in the sector 343 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF • Electrification rates in through more rigorous Mozambique are very low disclosure requirements for and represent a sizeable public corporations. drag on the economy. • Bank Group will continue to • The SCD recommends support Mozambique’s increasing energy membership and adherence generation from both to the EITI. renewable and • Through the ongoing Mining nonrenewable sources. and Gas Technical • The physical condition of Assistance Project, the CPF the system is poor, with will aim to strengthen frequent breakdowns and institutions that are high rates of electricity responsible for losses. management of natural • Transparent decision- resources. making processes and the • Outcome indicator: “Mining effective implementation and gas operations subject of anticorruption to fiscal controls in line with regulations would send a the established fiscal strong signal that the regime.” government is committed • CPF aims to expand access to minimizing the potential and improve reliability of for rent seeking in the electricity. extractive industries. IDA financing will focus on improving electricity service through grid rehabilitation and reinforcement as well as strengthening of the financial and operational functioning of the utility along with public sector investment in the transmission system. Serbia No CPSD. 2015 (FY16–20) SCD 2015 National level National level • SOE reform priority aims to reduce the state’s footprint 344 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF • The state has a large in the economy and make share in the economy. the public sector more There are about 1,300 efficient by privatizing SOEs that employ about commercially oriented 16 percent of the formal enterprises, restructuring Serbian workforce. large public utilities, and • SOEs have net annual rightsizing the public sector. losses of an estimated 3– • The labor market institutions 4 percent of GDP and priority seeks to strengthen require significant state these institutions to facilitate support to stay afloat. formal employment, create • SCD recommends earnings for the less well- reducing the state’s off, and help mitigate the footprint in the economy negative consequences of and making the public SOE reform. sector more efficient by • Under the focus area of privatizing commercially economic governance and oriented enterprises, the role of the state, the CPF restructuring large public aims to support the transfer utilities, and rightsizing the of productive SOE assets to public sector. private ownership. Energy sector Energy sector • Serbia’s power and gas • The CPF will seek to sectors are still corporatize and enhance the characterized by high performance of the state- losses, below-cost tariffs, owned power utility EPS. and lack of payment • Outcome indicators: discipline. “Increase in EPS collection • For Srbijagas, the critical rates” and “decrease in EPS steps are (i) resolution of distribution losses.” debts from large SOEs; (ii) • The CPF aims to support a amendment of the Law on more efficient and Payment Terms for sustainable power utility. Settlement of Financial • The Bank Group plans to Obligations in Commercial continue providing support Transactions to stop SOEs 345 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF and budgetary institutions for the introduction of from accumulating new energy efficiency measures. debt; (iii) corporatization; and (iv) financial consolidation. • For EPS, the priority actions are (i) facilitate collections from SOEs and budgetary institutions; (ii) commercialize EPS, centering on a tariff adjustment; and (iii) reinforce the independence of the supervisory board and put in place accountability mechanisms to monitor performance. Ukraine No CPSD. 2017 (FY17–21) SCD 2017 National level National level • The CPF aims to reform the • The SCD identifies SOE large and inefficient SOE reform as critical for sector, which traps Ukraine to create a level resources in firms that are playing field for the private able to operate at low levels sector and thus increase of productivity and its productivity. Within transparency, as they have private sector productivity, preferential access to SOE reform is identified as resources, markets, and a priority, and the SCD influence. recommends improving • The Bank Group will corporate governance and continue to provide strengthening accounting technical assistance in this and financial reporting of area. IFC advisory services ineffective SOEs. will help ensure a Financial sector transparent process of 346 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF • The SCD recommends attracting investors and strengthening the financial effective operators, act as an sector, including through adviser on divestiture, and interventions in the participate in pre- banking sector that affect privatization investments, state-owned banks. equity of privatized SOEs, • The reforms include and post-privatization implementing a projects. framework to recapitalize Financial sector and resolve banks and • The CPF aims to strengthen strengthen supervision; financial sector stability restoring credit growth by through further enhancing putting in place effective bank resolution, completing NPL resolution framework; the restructuring of the and improving banking system, and further governance of state- strengthening supervision. owned banks. • A resumption of credit Energy sector growth will also require • The SCD recommends action to resolve NPLs, while that immediate actions reforming and strengthening should be taken toward the corporate governance of strengthening energy the large state-owned sector governance and banking sector. accountability and Energy sector improving its financial • The CPF aims to improve viability. infrastructure services in • The reforms include energy sector. reducing household and • The Bank Group will district heating prices to continue to focus its support import parity price levels on hydropower, by 2017; mitigating the transmission, district heating, impact of price increases and energy efficiency on vulnerable households interventions, while with social assistance; engaging in policy dialogue supporting the Naftogaz and providing advice for restructuring to reduce energy sector reforms to 347 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF losses and improve strengthen sector governance; and governance, competition, promoting efforts to and sustainability. enhance energy efficiency • The reforms to be and raise investment and supported are gas sector domestic production. restructuring, establishment of wholesale electricity power market, establishment of a national energy efficiency fund, and optimization and sustainability of the Housing and Utilities Subsidy (HUS) program. Vietnam No CPSD. 2017 (FY18–22) SCD 2016 National level • Reform of the SOE sector is National level one of the priorities in the • The SCD advises CPF. Necessary reforms deepening reforms of the include (i) further progress in state-owned sector equitization, (ii) divestment through separation of from noncore assets, (iii) ownership and regulatory enhanced governance and functions, further transparency, and (iv) divestment, and better elimination of any remaining corporate governance. preferential treatments of Energy sector SOEs. • The SCD recommends Energy sector scaling up power • The Bank Group aims to generation capacity, support (i) increasing energy including in renewables, efficiency (demand and while promoting energy supply sides), including efficiency. targeting enterprises to upgrade inefficient production systems and 348 Appendix F Case Study Overviews for Eight Countries SCD and CPSD CPF introduce new technologies; (ii) scaling up nonhydropower renewable energy with particularly focus on solar, wind and gas-to-power; (iii) promoting the financial viability of the power sector and Electricity of Vietnam; (iv) introducing competition in gas and electricity markets; and (v) improving sector governance. Note: CPF = Country Partnership Framework; CPS = Country Partnership Strategy; CPSD = Country Private Sector Diagnostic; DPF = development policy financing; EITI = Extractive Industries Transparency Initiative; EPS = Elektroprivreda Srbije; FY = fiscal year; GDP = gross domestic product; IDA = International Development Association; NPL = nonperforming loan; PPP = public-private partnership; SCD = Systematic Country Diagnostic; SOCB = state-owned commercial bank; SOE = state-owned enterprise. References Bangladesh Ministry of Finance. 2018. Bangladesh Economic Review 2018. Dhaka: Ministry of Finance. Campanaro, Alessandra, and Cuong Duc Dang. 2018. “A Supply-Side Analysis of the Vietnamese Banking Sector.” In Mobilizing Finance for Local Infrastructure Development in Vietnam: A City Infrastructure Financing Facility, by Alessandra Campanaro and Cuong Duc Dang, 83–124. Washington, DC: World Bank. Council of European Energy Regulators. 2016. Status Review on the Implementation of Distribution System Operators’ Unbundling Provisions of the 3rd Energy Package. CEER Status Review C15-LTF-43-03. Brussels: CEER. https://www.ceer.eu/documents/104400/-/-/882514d5-c57f-86f8-50a3- 90185e270f15. 349 Appendix F Case Study Overviews for Eight Countries Egypt Independent. 2018. “Egypt Government Bill Prepares Companies for Renationalization.” Egypt Independent, October 30. IMF (International Monetary Fund). 2018. “Bangladesh—Staff Report for the 2018 Article IV Consultation.” IMF Country Report no. 18/158, IMF, Washington, DC. Vagliasindi, Maria, and John Besant-Jones. 2013. Power Market Structure: Revisiting Policy Options. Washington, DC: World Bank. http://documents1.worldbank.org/curated/en/795791468314701057/pdf/Po wer-market-structure-revisiting-policy-options.pdf. World Bank. 2016. “Bangladesh—Modernization of State-Owned Financial Institutions Project.” Project Appraisal Document PAD1576, World Bank, Washington, DC. World Bank Group. 2006. China Country Assistance Strategy Completion Report for FY03–05. Washington, DC: World Bank World Bank Group. 2015. Serbia—Public Finance Review 2015. Report 96451-YF. Washington, DC: World Bank. Notes 56 Evaluation resources allowed for 5 field missions which were purposively selected using a selectivity framework to assure regional distribution and maximum learning about diverse country conditions and Bank Group instruments. For example, allowing for regional representation, Vietnam was selected in East Asia because it offered a larger number and more diverse set of SOE reform projects in the two focal sectors than did other countries. For example, for the evaluation period, Vietnam had 15 national-level SOE reform projects; 21 in Financial Sector and 17 in Energy, reflecting engagement of all institutions and major instruments of the WBG. In comparison, China had 7 qualifying national-level SOE reform projects, 13 in the financial sector and just one in the energy sector, with no MIGA activity and only one World Bank lending activity. However, due to the substantial importance of China for learning, China was selected for a desk study. 57 “Companies that have handed down judicial rulings on their de-privatization include the following: Nile Cotton Ginning, Tanta Flax, Omar Effendi, Steam Boilers, Shebin El Koum Spinning and Weaving, Arab Foreign Trade, and Simo Paper. …[the Arab Republic of] Egypt’s government adopted a program of privatization in the 1990s. The program started slowly in 350 Appendix F Case Study Overviews for Eight Countries 1993, but then accelerated later on, as dozens of companies were privatized. In 2011, the Egyptian judiciary ordered various of those enterprises to be returned to the state. The same year, the government announced its abolishment of the privatization scheme altogether” (Egypt Independent 2018). 58 Constant 2010 dollars. World Development Indicators, 2018. 59 BIDV, VietinBank, and Vietcombank were transformed into Joint Stock Commercial Bank, and they are no longer classified as state-owned commercial banks by the State Bank of Vietnam. However, the state still has majority ownership in these three banks. Together with the Vietnam Bank for Agriculture and Rural Development, which is a state-owned commercial bank, the four state-controlled banks account for almost half of total sector assets. 351 Appendix G. Summary of Structured Literature Review and Bibliography Literature Review for State-Owned Enterprise Reform The Independent Evaluation Group conducted a structured literature review of relevant academic, World Bank Group, and other literature on state- owned enterprise (SOE) reform in general, and for specific sectors (energy and finance) and topics regarding SOE reform. (See appendix A on methodology for a description of the structured literature review methodology.) The literature review has two main objectives: to understand the characteristics of SOE reform, and to assess the role and effectiveness of different interventions, including complementary or sequential interventions that may influence SOE reform results. The review intends to provide a theoretical basis to be used in the evaluation to establish the relationship between interventions policy and performance regarding SOE reform. It helps in formulating and interpreting complementary evaluative tools and is one source for triangulation in validating patterns and relationships between interventions, contextual factors, and outcomes while controlling for various relevant explanatory variables. Typology of SOE Reform Drivers Economic growth, poverty alleviation, social welfare, environmental protection, and governance are general drivers for reform of SOEs. Ineffective SOEs create budgetary pressure and bring high financial and economic costs to the financial system. In many countries, underperforming enterprises in particular generate contingent liabilities and impose a fiscal burden and a source of fiscal risk (World Bank 2014b). Furthermore, SOEs are generally required to pursue both financial and additional developmental 352 Appendix G Summary of Structured Literature Review and Bibliography objectives, such as employment improvement, infrastructure creation, and service delivery to designated groups or regions, all of which can later lead to inconsistent or uneconomic decision-making (Vagliasindi 2008a). SOEs also have challenges in energy efficiency. For example, Andersson and Khalid (2018) have concluded that private firms emit less carbon dioxide than SOEs, and thus privatization of SOEs can bring mutual benefits of economic efficiency and energy efficiency. The literature also provides evidence that SOE underperformance is directly linked to poor corporate governance such as cumbersome decision-making processes, mixed incentives among managers and boards, inefficient manager selection, political patronage and rent seeking, and SOEs’ lack of accountability (Bacon 2018; Vagliasindi 2008). The drivers behind energy sector reform include lack of competition in the market, hidden costs, inefficient cross-subsidization, inadequate service quality, insufficient regulatory framework, and relatively high pollution level. The issue of monopoly is particularly acute in the reform in the electricity market. Along with the emergence of informational asymmetry and market distortion, deadweight loss increases in relationship to the potential for a competitive market (Sen, Nepal, and Jamasb 2016). Reform in this sector aims to transform a monopoly market to competitive wholesale and retail markets. The relevant policy measures include the unbundling and reorganization of vertically integrated state-owned utilities, the privatization of SOEs, the establishment of independent regulators, and the introduction of fair competition in the market (OECD 2016a; Popovici 2011). Hidden costs in the energy sector arise from underpricing, lack of collection, and unaccounted losses. The literature provides evidence that tariff reforms in SOEs resulted in significant reduction of hidden costs (Vagliasindi 2012). SOEs in the energy sector also tend to be the source of many serious pollution incidents and often fail to comply with environmental regulations, particularly for SOEs in the electricity generation and extractive industries. 353 Appendix G Summary of Structured Literature Review and Bibliography SOEs in the financial sector suffer from a series of vulnerabilities, including low profitability and stability, related restrictions on the entry of new private or foreign institutions, and liability to political risks (World Bank 2014). Many researchers have argued that non-state-owned commercial banks are more efficient than SOEs (Stewart, Matousek, and Nguyen 2016). The efficiency of state-owned banks is distorted because their goals are not solely to maximize profits but also to fulfill social development goals (Lam, Rodlauer, and Schipke 2017). However, there is also evidence that banking reform’s impact on banks’ profitability is ambiguous, and it depends on ownership structural changes during the reform process (Fu and Heffernan 2009). Typology of SOE Reform Interventions Direct support for SOE reform is provided at the upstream and downstream levels. Upstream interventions to improve fiscal policy and public finance management aim to reduce preferential access to SOE finance and manage the fiscal burden and fiscal risks associated with SOEs. Ensuring that private firms receive the same credit terms as SOEs in competition is one of the policies to support this objective. For instance, in Chile, state-owned banks are not even allowed to lend to the government or SOEs (World Bank 2014). Establishing debt management policies is also important—a government’s general fiscal policy can incorporate SOE debt management and monitoring mechanisms. The International Monetary Fund emphasized the importance of legislation on all debt transactions and government guarantees (Lam et al. 2017). Furthermore, international good practices have suggested that all guarantee proposals, including on SOE debt, should be subject to scrutiny. Possible approaches include guarantee fees and quantitative ceilings on guarantees (Allen and Alves 2016). The definition and cost of public service obligations for SOEs should also be defined clearly and separated from other regular commercial activities. Potential alternative mechanisms to replace public service obligations include direct subsidies, (conditional) cash 354 Appendix G Summary of Structured Literature Review and Bibliography transfers to targeted groups, and contracting services to private providers (World Bank 2014b). Enhancing transparency and accountability is essential for improving the corporate governance of SOEs. Setting objectives, reviewing and auditing performance, reporting on performance, and ensuring adequate disclosure at the company level are among the mechanisms to improve transparency in SOEs (OECD 2010). Centralized institutional arrangements for oversight and financial monitoring offer SOEs more operational independence and less political intervention (World Bank 2014a). SOEs’ own audit committees as internal auditors play the role of primary quality controller, and other external and independent auditors are complementary, without much duplication. Regarding reporting, reports presented to the public should be brief and easy to read and provide information on financial performance (OECD 2010). Regulations to promote SOEs disclosing and reporting transparently are also crucial. For instance, in Argentina, all SOEs are required to make financial information and annual performance reports public. Similarly, in Brazil, public institutions (including SOEs) must disclose information on their internal hierarchic structure and procurement processes. Furthermore, feedback loops are required for SOEs to receive and respond to public requests (World Bank 2014a). Effective legal and regulatory frameworks are critical for improving SOE performance (OECD 2005). The ideal legal framework should provide an equal condition for SOEs and private enterprises regarding access to financing and market competitiveness. It is also important for the law to define the minimum liabilities of the state clearly. Such liabilities include attending shareholders meeting and maintaining dialogues with oversight bodies (World Bank 2014b). Longer-term legal and institutional framework reform is built based on fundamental changes on state ownership frameworks, company codes, and corporate governance (Kikeri 2018). Measures aiming to strengthen courts’ independence and the effectiveness of judicial systems are also important (Molnar 2017). 355 Appendix G Summary of Structured Literature Review and Bibliography Governments can exercise their ownership rights in SOEs through the centralization model or agency model. The centralization of effective state ownership and shareholder role can help improve transparency, consistency, and accountability of SOEs and strengthen their independency from political interferences (OECD 2015). In this model, strategic functions are unified in a single agency, and daily management responsibility is given to line ministries (World Bank 2014a). In the agency model, the key shareholders should each be assigned to specific roles accordingly—for example, a policy- making department can be granted the authority to design policy; a financial ministry can provide strategic guidance and support to SOEs’ portfolios; and parliament can provide constitutional oversight of enterprise performance and policy compliance (World Bank 2014a). Privatization, public-private partnership (PPP) arrangements, and corporatization are among the mechanisms to reform SOEs’ ownership structure. Ownership change would rarely stand out as a single reform but would as a series of policies aiming to increase free competition (Nellis and Birdsall 2005). Governments share three common goals when implementing privatization policies: reducing national deficits and debt, fostering financial markets development, and increasing efficiency (Kikeri and Phipps 2008). One common practice in the privatization process is the establishment of a centralized privatization institution. This institution aims to separate the state’s ownership policy-making functions from supervisory functions to reduce conflicts of interest, minimize potential political interference and improve the resource distribution efficiency, and promote consistency in the application of corporate governance standards (Vagliasindi 2008a). SOE reforms at the downstream level aim to improve and strengthen SOE ownership structure, corporate governance, financial management, and business and operations. Privatization reform at the enterprise level often involves transactional actions, either regarding financial transaction (sale of shares, award of PPP concessions) or technical analysis (World Bank Group 356 Appendix G Summary of Structured Literature Review and Bibliography 2018). Enterprises should be analyzed individually on whether they need to be privatized, merged, consolidated, or liquidated (Vagliasindi 2008). In 2005, the Organisation for Economic Co-operation and Development adopted the Guidelines on Corporate Governance of SOEs, which later became a critical and effective guidebook for SOE governance reform. A solid corporate governance is crucial, especially for large and visible SOEs that are strategically important or risky to the economy (Vagliasindi 2008b). Regarding financial reporting, a set of high-quality accounting standards is the key, and for nonfinancial reporting, it is the effectiveness and information publicity that should be improved. Countries such as Chile and India have made significant progress in their nonfinancial reporting. They have established guidelines for SOEs to release information on related-party transactions, policy commitments, ownership and governance structures, or risk management (Habib, Jiang, and Zhou 2014). A World Bank study used six Latin American countries as case studies to demonstrate that using accounting standards, financial reports, and external audits can lead to SOEs’ good risk management (World Bank 2014b). The International Finance Corporation developed a six-step plan outlining how an SOE could analyze its current corporate governance before the reform. These six steps start with a summary of first impressions, followed by self-assessment, review, the establishment of a corporate governance improvement program, program implementation, and a supervisory system (McMahon 2010). Boards of SOEs should be allowed to carry out delegated responsibilities independently (Crane-Charef 2015). Slovenia has prohibited high-level official from serving on SOE boards (Hamilton and Berg 2008). Countries such as South Africa explicitly require SOEs’ boards to include at least one- third nonexecutive members (Kikeri 2018). India also requires SOEs to have no more than two government representatives on the board. Malaysia removed all regulatory-related officials from government-linked corporations. Furthermore, many countries have already required the presence of independent directors on SOE boards. For example, in 357 Appendix G Summary of Structured Literature Review and Bibliography Mozambique, the majority of directors (including the chairs) must be independent (World Bank 2014). Especially in low-capacity countries, significant technical assistance on corporate governance and human resource capacity building are necessary at the initial phase (World Bank 2014a). Countries such as Ghana have followed international best practices and developed an SOE practical curriculum to help their boards achieve their goals (Kikeri 2018). Appointing qualified staff for the management level is critical for sound SOE performance (Christiansen 2013). The World Bank argued that “empowering the board to appoint and, subject to clear terms, remove the CEO... reduces the scope for government interference in operational decision-making” (World Bank 2014a). To enhance the financing capacity of SOEs, the literature provides the following recommendations: rationalize state holdings; consolidate the funding model of commercial SOEs and provide them with bargaining power; have a diversified mix of funding for SOEs by including debt and equity and possibly PPP; and address the problem of inappropriate capital investment decision-making by setting clear policies for the treatment of noncommercial objectives (Kikeri 2018). Multilateral financial institutions, including the World Bank, have been advocating energy sector reform to improve economic efficiency and attract private investment through four distinct reform actions: regulation, restructuring (unbundling), private sector participation, and introduction of competition (Eberhard et al. 2011). Regulation can help redistribute the gains between producers and consumers (Estrin and Pelletier 2018). When monopoly SOEs in the energy sector operate under insufficient regulation, it tends to result in inefficiency. The reform in regulatory frameworks involves the establishment of an independent regulatory entity that can oversee utilities regarding their operational and financial performance. Regulations are most effective when private sector participation and competition have been introduced at least to some extent (Foster et al. 2017). According to Bacon (2018), the greater market power private participants hold, the higher 358 Appendix G Summary of Structured Literature Review and Bibliography the chances that independent regulator can redistribute the profit gains and improve efficiency. Balza, Jimenez, and Mercado (2013) analyzed different countries’ regulatory changes in the electricity sector and found that a new regulatory program tends to come with competition in generation, accessibility of generators to transmission systems, and unregulated prices for large consumers. Furthermore, the authors identified that price setting mechanisms could be quite diverse from country to country, ranging from price caps to free markets. Unbundling is the pathway to competition because it helps avoid any potential conflicts of interest and reduces the concentration of market power, particularly when generation is involved. Unbundling can also be costly, especially in smaller countries where unbundling might result in losing the scale economies or reducing management capacity. As of 2017, 74 percent of developed countries have conducted unbundling to some extent, and in developing countries, only 43 percent have unbundled (Foster et al. 2017). Private sector participation and the adoption of contracts will bring in private management and investment to the energy sector to improve efficiency and reduce hidden costs (Eberhard et al. 2011; Foster et al. 2017). The coverage of private sector participation can be partial. For instance, only generation can be privatized while transmission remains a monopoly, or part of the generation can be privatized while the rest remains state owned (Foster et al. 2017). Competition, often followed by unbundling and privatization, allows generation companies to compete in supply markets and consumers to negotiate contracts with generators through an energy market (Foster et al. 2017). There is a spectrum in introducing competition to the electricity market, starting from monopoly and moving to a single buyer model, where a single entity holds a vertical monopoly throughout the value chain, but some independent power producers are allowed in generation. Moving to the 359 Appendix G Summary of Structured Literature Review and Bibliography next stage allows competition on the wholesale level, where direct, bilateral contracts of power purchase can be designed between generators and large customers. A final stage of competition is the retail competition market with complete vertical unbundling, and all consumers are free to choose the suppliers. As of 2015, 70 percent of Asian countries have introduced some degree of competition in their energy markets (Foster et al. 2017; McMahon 2010). Competition also plays a key role in the extractives industry. Victor et al. (2012) argued that the lack of competitiveness leads to inefficient SOEs. Reforms in the energy sector often entail different sets of policies, but policies are often more effective when combined (Bacon 2018). Researchers Zhang, Parker, and Kirkpatrick (2008) concluded from their research in electricity sector reform in developing countries that “competition in electricity generation is more important than privatization or the establishment of independent regulation in bringing about performance improvements.” They argued that privatization can be effective in increasing productivity and efficiency only when combined with independent regulation. Regarding reform of SOEs in the financial sector, Rudolph (2009) emphasized three key points: (i) The mandates of SOEs should be specific but flexible, allowing directors to adopt different financial products to fit the market needs; (ii) mandates of SOEs should be reviewed and revised on a regular basis; and (iii) some types of returns are expected for SOEs to be financially sustainable. Separating the ownership and supervision functions clearly is very important to avoid any potential conflicts of interest. State- owned financial institutions and other financial institutions should have the same supervision on pricing and risk management. Furthermore, it is argued that under a proper bank regulation, development banks tend to develop better risk management systems and experience an efficiency boost (OECD 2010). A close monitoring mechanism for SOEs in the financial sector is crucial because state-owned financial institutions are often expected to fulfill functions that private institutions do not perform, such as projects 360 Appendix G Summary of Structured Literature Review and Bibliography that might generate positive externalities instead of high rates of return. Assessing the financing practices of state-owned banks concerning the public sector is key to pursuing fiscal discipline (Gonzalez-Garcia and Grigoli 2013). It is critical to moderate government financial support or intervention in the financial sector to prevent market distortion. National development banks should provide their services at market prices to remain sustainable. Management of these financial institutions should be independent from political influence to reduce unnecessary related costs. Furthermore, development banks need a separate system of performance evaluation, which includes development-oriented indicators (Cull, Martínez Pería, and Verrier 2018). Effectiveness of SOE Reform Because of regulatory reform, restructurings, improved governance practices, and higher competitiveness in the market, many countries have witnessed a boost in SOE performance after their reforms. For instance, in Indonesia, SOEs’ profits increased at an annual rate of 19 percent between 2004 and 2009 (Abubakar 2010). However, SOEs’ performance is not always positive, and some SOEs might not experience a significant performance improvement, or their profit gains still come from limited competition (World Bank 2014a). Financial performance of SOEs is empirically linked to the composition of SOE boards (Vagliasindi 2008b). The empirical research of SOEs in the Republic of Korea has shown that board size, corporatization, and information transparency were positively related to the efficiency of SOEs. Curi et al. (2016) studied SOEs in Lithuania and concluded that good corporate governance can enhance firm efficiency. 361 Appendix G Summary of Structured Literature Review and Bibliography Critical Variables and Contextual Factors in SOE Reform Complementarity of reforms is crucial in the effectiveness and sustainability of SOE reform interventions. Although SOE reforms have covered a wide range of topics, all of them aim to insulate SOEs from political interference to certain degrees (Wong 2018). An integrated approach is critical to SOE reform, and a single change is hardly sufficient to sustain performance improvement (Sen, Nepal, and Jamasb 2016). Private ownership is argued to enhance profitability of firms (Tran, Nonneman, and Jorissen 2015). However, private ownership does not necessarily lead to economic gains. Factors such as the degree of reform, regulatory framework, and competitive environment should also be considered (Balza, Jimenez, and Mercado 2013). Furthermore, the degree of regulators’ independence plays an important role. If regulators are not independent from government control, the effectiveness of regulation reform cannot be guaranteed (Bacon 2018). Preconditions, particularly the regulatory framework and appropriate privatization process, are critical factors to SOE reforms. These include well- designed sequencing, complementary policies implementation, and regulatory capacity creation (Estrin and Pelletier 2018). Country characteristics such as geography, political system, income distribution, and system size can significantly affect the cost and benefit of reform. For instance, in the energy sector, the Latin America and the Caribbean Region, with rather competitive political dynamics and middle- range income bracket, tends to perform better than others in SOE reforms (Foster et al. 2017). Doidge, Karolyi, and Stulz (2007) showed theoretically and empirically that country characteristics are an important determinant of firm-level governance. Furthermore, countries with weak capital markets and institutions might offer smaller benefits, and introducing good governance could be more costly to the firm (Vagliasindi 2008a). Political economy and vested interests are also affecting the implementation of SOE reforms. Political interference might result in operational constraints 362 Appendix G Summary of Structured Literature Review and Bibliography and distorted objectives (Boubakri et al. 2012). Corrupted elites might take advantages of privatization to take state assets for themselves (Estrin and Pelletier 2018). Macroeconomic environment is also a critical variable. It has been observed that the performance of privatized firms might improve after the reform under the period of accelerating economic growth (Tran, Nonneman, and Jorissen 2015). The effectiveness of SOE reform depends heavily on the extent of policy design and implementation. For instance, in the electricity sector, partial vertical unbundling when transmission is separated only from generation (but not from distribution) tends not to improve performance because of conflicts of interest (Vagliasindi 2012). Munawarah, Zainuddin, and Muharam (2017) examined the role of auditing after SOE reforms and concluded that it can enhance the credibility of SOE business management and improve the effectiveness of corporate governance. Stakeholders Ever since the first wave of SOE reforms in 1960s, decision makers, donors, and international financial institutions have attempted to improve SOE performance (Nellis 2006). Different players are involved in the SOE reforms, such as governments, the judiciary, regulators, the public, shareholders, and other stakeholders. The complexity of players usually results in a mixed system of interests in SOE reforms. For instance, public and minority shareholders may capitalize on SOEs in indirect ways; however, they tend to have little control in running SOEs. Political ministers, who control SOEs directly, also have little stake in improving SOE performance because of their short-term perspective (Crane-Charef 2015). The state as the controller of SOEs pays significant attention to the firm’s performance, especially concerning the trade-off between states’ decreasing controlling right and 363 Appendix G Summary of Structured Literature Review and Bibliography potential revenue gains from increased efficiency (Ivanova and Bikeeva 2016). Legislative bodies also influence the effectiveness of corporate governance implementation. A greater share of legislators on the board can lead to an effective supervisory function and higher financial performance (Munawarah, Zainuddin, and Muharam 2017). Outcomes and Gains Choosing the correct measurement for performance, benchmarks, and statistical methods is one important aspect of measuring the outcomes of SOE reforms. Additionally, limited data availability and consistency, and selection effects may add to the challenges of analyzing the economic effects of privatization (Gupta, Ham, and Svejnar 2008). Some measurements for the utility sector include generation capacity, transmission and distribution losses, the number and quality of connections, and charged tariffs (Bacon 2018). Performance measurements include rate of return on financial investment, output growth, or productivity growth. Some researchers would compare the performances of SOEs to those of private firms, and some focus on the pre-and postreform performance changes performance for companies privatized through share issues (Megginson, Ullah, and Wei 2014). Pre- and postcomparison is the most common approach in studying the outcome of SOE reform (Tran, Nonneman, and Jorissen 2015). The empirical literature on the effects of SOE reforms in emerging economies offers mixed results. Differences in methodologies and control variables might have contributed to this inconclusiveness. Many hypotheses are raised in studies—for example, the shift to private ownership is likely to improve profitability and a more efficient capital and labor distribution (Tran, Nonneman, and Jorissen 2015). The SOE performance analysis has been studied at various levels, such as single-country, cross-country, and cross-industry comparison. Significant correlations between ownership structure and productivity growth have 364 Appendix G Summary of Structured Literature Review and Bibliography been verified in many single-country comparison studies. One study on the Pakistani cement industry concluded that labor use and employment risks were reduced after the reform (Ghulam 2017). Brissimis, Delis, and Papanikolaou (2008) found a positive impact of banking reform on efficiency improvement. One comparison of utilities based on 35 governance indicators is one of many examples proving that governance reform can improve SOEs’ performance (Vagliasindi 2008). Dinc and Gupta (2011) found a positive impact on SOEs’ postprivatization performance in India after addressing the selection bias. Zhang, Parker, and Kirkpatrick (2008) provided an econometric assessment with panel data for 36 developing and transition countries over 18 years. They argued that the gains in economic performance mainly came from the introduction of competition as basis. Privatization and regulations are not as effective when lacking the establishment of a competitive market. Similarly, Balza, Jimenez, and Mercado (2013) also examined the electricity sector in 18 Latin American countries on private sector participation, institutional reform, and firm performance. They found that stable sectoral institutions are key to improving electricity sector performance. Clarke, Cull, and Fuchs (2007) focused on developing countries and found that bank performance usually improved after privatization. They studied the impacts of privatization of the Uganda Commercial Bank and showed that the bank improved its profitability while improving financial access (Rabiei and Rezaie 2013). Summary Tables The following tables define variables used in identifying drivers of SOE reform (table G.1), types of SOE reforms (table G.2), and effectiveness of SOE reforms (table G.3). 365 Appendix G Summary of Structured Literature Review and Bibliography Table G.1. Drivers of SOE Reform Driver Type Description Count General drivers Economic growth The comparably low productivity and profitability of 118 SOEs is harmful to economic growth: ineffective SOEs create budgetary pressure and bring high financial and economic costs to the financial system as a whole. Social welfare Three different views have shed lights on SOEs’ role in social welfare. The agency view focuses on the discrepancy between the objectives of managers (the agents) and of owners (the principals); the social view considers SOEs as institutions created by government aiming to resolve actual or perceived market failures and to maximize social welfare; the political economy view argues that SOEs are a mechanism for politicians to pursue their individual goals. Environmental Studies have investigated whether the private protection sector is also more carbon efficient through functional aspect and institutional aspect. Governance SOE underperformance is directly linked to poor corporate governance such as cumbersome decision-making processes, mixed incentives among managers and board, inefficient manager selection, political patronage and rent seeking, and the lack of accountability in SOEs. Regulations Failed external regulatory systems of SOEs, such as insufficient regulation and oversight systems, can drive SOE reform. Drivers specific to the energy sector Monopoly Along with the emergence of informational 46 asymmetry and market distortion, the deadweight 366 Appendix G Summary of Structured Literature Review and Bibliography Driver Type Description Count loss increases to where the need for a competitive market emerges. Planning and Particularly for developing countries, few regulation governments explicitly monitor the reliability of energy supply or require SOEs to disclose information to the public regularly. Thus, it is hard for SOEs or even the whole sector to attract investment for further development. Hidden costs Hidden costs are abundant in the energy sector: network losses, underpricing, cross-subsidization, and mismanagement. Environment SOEs tend to be the source of many serious pollution incidents and often fail to comply with environmental regulations, particularly in the electricity generation and extractive industries. Drivers specific to the financial sector Profitability The efficiency of state-owned banks is distorted 52 because their goals are not solely to maximize profits but also to fulfill social development goals. Hinder new and Financial SOEs in emerging markets often displace private entry commercial financial services, hinder new private entry, and reduce market competitiveness. Stability, State ownership in financial institutions is usually vulnerability, and more risk-avoiding, and inefficient financial SOEs risk taking can generate contingent liabilities, increase fiscal risk, and weaken the financial system as a whole. Political risk Financial SOEs have less concern about political risk of the host country because they are controlling state equity and are less likely to default. Note: SOE = state-owned enterprise. 367 Appendix G Summary of Structured Literature Review and Bibliography Table G.2. Typology of SOE Reform Interventions Driver Type Description (subcounts in parentheses) Count General The interventions for SOE reforms can generally be 258 divided into four groups: upstream, downstream, indirect support, and financial leverage. Upstream Macro/fiscal policy/ Applying financial fiscal policy targeting SOEs can 136 public finance help improve SOE governance and performance management and reduce governments’ liabilities. (19) Governance Both transparency and accountability are crucial to the governance of SOEs. They offer substance to shareholders’ rights, provide remedy choice for potential manipulation, and serve as a foundation for public trust. (9) National regulatory Effective legal and regulatory frameworks are framework critical for improving SOE performance. (22) National SOE There are two ways to build institutional capacity to ownership exercise ownership rights: the centralization model and the agency model. Centralization in the state’s role in ownership improves the independence of the state’s policy-making and regulatory role. The agency model implements institutional changes. (38) Privatization Privatization (the deliberate sale of SOEs to private agents) initiates both ownership and corporate governance changes. Privatization on the upstream level usually has an impact on multiple SOEs and consists of a series of policies aiming to increase free competition. (48) Downstream SOE ownership Privatization reform at the enterprise level often 118 involves transactional actions, either regarding financial transactions (sale of shares, award of 368 Appendix G Summary of Structured Literature Review and Bibliography Driver Type Description (subcounts in parentheses) Count public-private partnership concessions) or technical analysis. Corporate Transparency and disclosure are crucial to the governance accountability of SOEs. Under an effective reporting regime, SOEs should abide by the same reporting, control, and audit programs as other public entities. Financial A sound control environment for SOEs entails governance and effective systems, standards, and procedures and control can protect the efficiency and integrity of the firm’s operation. Good practice on internal controls includes safeguarding assets from misuse, maintaining accounting records, and ensuring financial information reliability. Performance Different institutions have different systems of management performance management, and some of the crucial elements include the state ownership function, an independent but authorized board, and accountable executive management. Environment and Reforms such as helping excluded groups and social aspects support to improve the environment can help SOEs achieve social or environmental objectives. Energy sector Regulation The responsibilities of SOEs should be clearly 62 identified and separated by clarifying government’s various roles in designing policy, price setting, and managing utility assets and services. Restructuring Large SOEs or the energy sector can be unbundled vertically and horizontally into separate smaller but state-owned companies. Unbundling can happen to different extents, resulting in different improvements such as in transparency and accountability. 369 Appendix G Summary of Structured Literature Review and Bibliography Driver Type Description (subcounts in parentheses) Count Private sector Private sector participation and the adoption of participation contracts will bring private management and investment to the energy sector to improve performance efficiency and reduce hidden costs. Competition Competition, often followed by unbundling and privatization, allows generation companies to compete in supply markets and consumers to negotiate contracts with generators through an energy market. Financial sector Regulatory Rudolph (2009) has summarized three key points 78 framework that should be included in financial SOEs reform: (i) The mandates of SOEs should be specific but flexible, allowing directors to adopt different financial products to fit the market needs; (ii) mandates of SOEs should be reviewed and revised regularly; and (iii) some types of returns are expected for SOEs to be financially sustainable. Monitoring A close monitoring mechanism for finance SOEs is of utmost importance because state-owned financial institutions are often expected to fulfill functions that private institutions do not perform. Monitoring can be divided into two parts: internal and external. Ownership and Policies that include improving the supply of credit privatization information, increasing (or ensuring) market contestability (competition), and improving contract enforcement can strengthen competitiveness in the finance sector. Note: SOE = state-owned enterprise. 370 Appendix G Summary of Structured Literature Review and Bibliography Table G.3. Effectiveness of SOE Reform Variable/Factor Description Count Critical variables Ownership structure changes and corporate 74 governance (quality of the board, quality of transparency, and strategic planning) are critical to SOE postreform performances. Stakeholders Different players are involved in SOE reforms—such as 10 governments, the judiciary, regulators, the public, shareholders, and other stakeholders—because different stakeholders tend to have different incentives in firm development. Contextual factors Preconditions, including the regulatory framework and 47 appropriate privatization process, country characteristics, political economic factors, and electoral systems, are critical factors to SOE reforms. Outcomes and gains The empirical literature on the effects of SOE reforms 34 in emerging economies have offered mixed results. The SOE performance analysis has been studied at various levels, such as single-country, cross-country, and cross-industry comparisons. Note: The count number is the simple summation of each subtopic. There are some overlaps among each subtopic. SOE = state-owned enterprise. References Abubakar, Mustafa. 2010. “Ministry of State-Owned Enterprises, Indonesia: Towards World Class Corporations.” PowerPoint presentation. Allen, Richard, and Alves Miguel. 2016. How to Improve the Financial Oversight of Public Corporations. Fiscal Policy How-To Notes. Washington, DC: International Monetary Fund. https://www.imf.org/external/pubs/ft/howtonotes/2016/howtonote1605.pdf. Andersson, Fredrik N.G. & Opper, Sonja & Khalid, Usman, 2018. "Are capitalists green? Firm ownership and provincial CO2 emissions in China," Energy Policy, Elsevier, vol. 123(C), pages 349-359. 371 Appendix G Summary of Structured Literature Review and Bibliography Bacon, Robert. 2018. “Taking Stock of the Impact of Power Utility Reform in Developing Countries: A Literature Review.” Policy Research Working Paper 8460, Rethinking Power Sector Reform, World Bank Group, Washington, DC. https://doi.org/10.1596/1813-9450-8460. Balza, Lenin, Raul Jimenez, and Jorge Mercado. 2013. “Privatization, Institutional Reform, and Performance in the Latin American Electricity Sector.” Energy Division Technical Note IDB-TN-500, Inter-American Development Bank, Washington, DC. https://publications.iadb.org/publications/english/document/Privatization- Institutional-Reform-and-Performance-in-the-Latin-American-Electricity- Sector.pdf. Boubakri, Narjess, Omrane Guedhami, Dev Mishra, and Walid Saffar. 2012. “Political Connections and the Cost of Equity Capital.” Journal of Corporate Finance 18 (3): 541–59. doi:10.2139/ssrn.1589688. Brissimis, Sophocles, Manthos Delis, and Nikolaos Papanikolaou. 2008. “Exploring the Nexus between Banking Sector Reform and Performance: Evidence from Newly Acceded EU Countries.” Journal of Banking & Finance 32 (12): 2674– 83. https://doi.org/10.1016/j.jbankfin.2008.07.002. Christiansen, Hans. 2013. “Balancing Commercial and Non-Commercial Priorities of State-Owned Enterprises.” OECD Corporate Governance Working Papers no. 6, OECD Publishing, Paris. https://doi.org/10.1787/5k4dkhztkp9r-en. Clarke, George R. G., Robert Cull, and Michael Fuchs. 2007. “Bank Privatization in Sub-Saharan Africa: The Case of Uganda Commercial Bank.” Policy Research Working Paper 4407, World Bank, Washington, DC. Crane-Charef, Mary. 2015. “Stocktaking of Anti-Corruption and Business Integrity Measures for Southern African SOEs.” OECD Corporate Governance Working Papers no. 18, OECD Publishing, Paris. https://doi.org/10.1787/5jrtd6fghvf1-en. Cull, Robert, María Soledad Martínez Pería, and Jeanne Verrier. 2018. “Bank Ownership—Trends and Implications.” Policy Research Working Paper 8297, World Bank, Washington, DC. http://documents1.worldbank.org/curated/en/810621515444012541/pdf/WP S8297.pdf. 372 Appendix G Summary of Structured Literature Review and Bibliography Curi, Claudia, Justas Gedvilas, and Ana Lozano-Vivas. 2016. “Corporate Governance of SOEs and Performance in Transition Countries. Evidence from Lithuania.” Modern Economy 7 (12). De Luna-Martínez, D., C. L. Vicente, A. B. Arshad, R. Tatucu, and J. Song. 2018. 2017 Survey of National Development Banks. Washington, DC: World Bank Group. http://documents.worldbank.org/curated/en/977821525438071799/2017- Survey-of-National-development-banks. Dinc, Serdar, and Nandini Gupta. 2011. “The Decision to Privatise: Finance and Politics.” The Journal of Finance 66 (1): 241–69. https://doi.org/10.1111/j.1540-6261.2010.01631.x. Doidge, Craig, G. Andrew Karolyi, and Rene M. Stulz. 2007. “Why Do Countries Matter So Much for Corporate Governance?” Journal of Financial Economics 86 (1): 1–39. https://doi.org/10.1016/j.jfineco.2006.09.002. Eberhard, Anton, Orvika Rosnes, Maria Shkaratan, and Haakon Vennemo. 2011. Africa’s Power Infrastructure: Investment, Integration, Efficiency. Directions in Development: Infrastructure. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/2290. Estrin, Saul, and Adeline Pelletier. 2018. “Privatization in Developing Countries: What Are the Lessons of Recent Experience?” The World Bank Research Observer 33 (1): 65–102. https://doi.org/10.1093/wbro/lkx007. Foster, Vivien, Samantha Witte, Sudeshna Ghosh Banerjee, and Alejandro Moreno. 2017. “Charting the Diffusion of Power Sector Reforms across the Developing World.” Policy Research Working Paper 8235, Rethinking Power Sector Reform, World Bank Group, Washington, DC. http://documents.worldbank.org/curated/en/576801510076208252/Charting -the-diffusion-of-power-sector-reforms-across-the-developing-world. Fu, Xiaoqing, and Shelagh Heffernan. 2009. “The Effects of Reform on China’s Bank Structure and Performance.” Journal of Banking and Finance 33 (1): 39–52. https://doi.org/10.1016/j.jbankfin.2006.11.023. Ghulam, Yaseen. 2017. “Effect of Privatization and Broader Reforms on Labor Use Adjustments and Efficiency.” Annals of Public and Cooperative Economics 88 (4): 499–518. https://doi.org/10.1111/apce.12164. 373 Appendix G Summary of Structured Literature Review and Bibliography Gonzalez-Garcia, Jesus, and Francesco Grigoli. 2013. “State-Owned Banks and Fiscal Discipline.” IMF Working Paper 13/206, International Monetary Fund, Washington, DC. https://www.imf.org/external/pubs/ft/wp/2013/wp13206.pdf. Gupta, Nandini, Jhon C. Ham, and Jan Svejnar. 2008. “Priorities and Sequencing in Privatization: Evidence from Czech Firm Panel Data.” European Economic Review 52 (2): 183–208. https://doi.org/10.1016/j.euroecorev.2007.05.004. Habib, Ahsan, Haiyan Jiang, and Donghua Zhou. 2014. “Audit Quality and Market Pricing of Earnings and Earnings Components in China.” Asian Review of Accounting 22 (1): 20–34. https://doi.org/10.1108/ARA-05-2013-0034. Hamilton, Ron, and Alexander Berg. 2008. “Corporate Governance of SOEs in Zambia.” Unpublished Manuscript. Washington, DC: World Bank. Ivanova, Irina A., and M. V. Bikeeva. 2016. “Corporate Social Responsibility: Specificity, Formation Mechanism, Estimation of Management Efficiency.” European Research Studies Journal 19 (3A): 167–84. Kikeri, Sunita. 2018. Corporate Governance in South African State-Owned Enterprises. Washington, DC: World Bank. https://doi.org/10.1596/30029. Kikeri, Sunita, and Verena Phipps. 2008. “Privatization Trends.” Public Policy for the Private Sector Note no. 317, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/965751468182641229/Privatiza tion-trends. Lam, W. Raphael, Markus Rodlauer, and Alfred Schipke, eds. 2017. Modernizing China: Investing in Soft Infrastructure. Washington, DC: International Monetary Fund. Lam, W. Raphael, A. Schipke, Y. Tan, and Z. Tan. 2017. “Resolving China's Zombies: Tackling Debt and Raising Productivity.” IMF Working Paper 17/266, International Monetary Fund, Washington, DC. McMahon, Gary. 2010. The World Bank’s Evolutionary Approach to Mining Sector Reform. Extractive Industries and Development Series 19. Washington, DC: World Bank. http://hdl.handle.net/10986/18288. Megginson, William L., Barkat Ullah, and Zuobao Wei. 2014. “State Ownership, Soft- Budget Constraints, and Cash Holdings: Evidence from China’s Privatized 374 Appendix G Summary of Structured Literature Review and Bibliography Firms.” Journal of Banking and Finance 48 (November): 276–91. https://doi.org/10.1016/j.jbankfin.2014.06.011. Molnar, Margit. 2017. “Revitalizing China’s Economy by Improved Corporate Governance and State-Owned Enterprise Reforms.” Journal of International Commerce, Economics and Policy 8 (3): 1–23. https://doi.org/10.1142/S1793993317500156. Munawarah, Muhammad Din, Fatlina Zainuddin, and Harmum Muharam. 2017. “What Effects Do Privatisation Policies Have on Corporate Governance of State-Owned Enterprises?” European Research Studies 20 (4A): 124–32. Nellis, John. 2006. “Back to the Future for African Infrastructure? Why State- Ownership is No More Promising the Second Time Around.” Working Paper 84, Center for Global Development, Washington, DC. https://www.cgdev.org/sites/default/files/6352_file_WP_84.pdf. Nellis, John, and Nancy Birdsall. 2005. Reality Check: The Distributional Impact of Privatization in Developing Countries. Washington, DC: Center for Global Development. OECD (Organisation for Economic Co-operation and Development). 2005. OECD Guidelines on Corporate Governance of State-Owned Companies. 2005 ed. Paris: OECD Publishing. OECD. 2010. Corporate Governance: Accountability and Transparency: A Guide for State Ownership. Paris: OECD Publishing. OECD. 2015. OECD Guidelines on Corporate Governance of State-Owned Enterprises. 2015 ed. Paris: OECD Publishing. OECD. 2016a. “Global Knowledge Sharing Network on Corporate Governance of State-Owned Enterprises.” Proceedings from the First Meeting, Mexico City, Mexico, June 7–8. http://www.oecd.org/daf/ca/Global-Knowledge-Sharing- Network-on-SOEs-Proceedings-June-2016.pdf. OECD. 2016b. State-Owned Enterprises as Global Competitors: A Challenge or an Opportunity? Paris: OECD Publishing. Popovici, Vlad. 2011. “Power Generation Sector Restructuring in Romania—A Critical Assessment.” Energy Policy 39 (3): 1845–56. https://doi.org/10.1016/j.enpol.2011.01.023. 375 Appendix G Summary of Structured Literature Review and Bibliography Rabiei, Mahnaz, and Parmida Rezaie. 2013. “Does Privatization Effect Profitability of Banking Sector? (Case Study of Iran).” Actual Problems of Economics 139 (1): 479–85. Rudolph, Heinz P. 2009. “State Financial Institutions: Mandates, Governance, and Beyond.” Policy Research Working Paper 5141, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/114531468159602791/State- financial-institutions-mandates-governance-and-beyond. Sen, Anupama, Rabrindra Nepal, and Tooraj Jamasb. 2016. Reforming Electricity Reforms? Empirical Evidence from Asian. Oxford, U.K.: The Oxford Institute for Energy Studies. https://doi.org/10.26889/9781784670504. Stewart, Chris, Roman Matousek, and Thao Ngoc Nguyen. 2016. “Efficiency in the Vietnamese Banking System: A DEA Double Bootstrap Approach.” Research in International Business and Finance 36: 96–111. https://doi.org/10.1016/j.ribaf.2015.09.006. Tran, Ngo My, W. Nonneman, and A. Jorissen. 2015. “Privatization of Vietnamese Firms and Its Effects on Firm Performance.” Asian Economic and Financial Review 5 (2): 202–17. Vagliasindi, Maria. 2008a. “Governance Arrangements for State Owned Enterprises.” Policy Research Working Paper 4542, World Bank, Washington, DC. Vagliasindi, Maria. 2008b. “The Effectiveness of Boards of Directors of State-Owned Enterprises in Developing Countries.” Policy Research Working Paper 4579, World Bank, Washington, DC. Vagliasindi, Maria. 2012. Implementing Energy Subsidy Reforms: Evidence from Developing Countries. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/566201468177234043/Impleme nting-energy-subsidy-reforms-evidence-from-developing-countries. Victor, David G., David R. Hults, and Marl C. Thurber, eds. 2012. Oil and Governance: State-Owned Enterprises and the World Energy Supply. Cambridge, U.K.: Cambridge University Press. Wong, Simon C. Y. 2018. “The State of Governance at State-Owned Enterprises.” Private Sector Opinion 40, International Finance Corporation, Washington, 376 Appendix G Summary of Structured Literature Review and Bibliography DC. https://www.ifc.org/wps/wcm/connect/b1adde06-267d-4d79-bfaf- 62f17de51f4a/PSO40.pdf?MOD=AJPERES&CVID=m7T0xLQ. World Bank. 2012. Global Financial Development Report 2013: Rethinking the Role of the State in Finance. Washington, DC: World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/11848/Globa lFinancialDevelopmentReport2013.pdf?sequence=1. World Bank Group. 2014. Corporate Governance of State-Owned Enterprises: A Toolkit. Washington, DC: World Bank. World Bank. 2014a. Corporate Governance of State-Owned Enterprises in Latin America: Current Trends and Country Cases. Washington, DC: World Bank. http://documents1.worldbank.org/curated/en/634451468088457556/pdf/894 680WP00P1260orate0Governance0LAC.pdf. World Bank. 2014b. “Tajikistan: Fiscal Risks from State-Owned Enterprises.” Tajikistan Policy Notes on Public Expenditures no. 5, World Bank, Washington, DC. http://documents1.worldbank.org/curated/en/571421468119375495/pdf/891 830NWP0P1475020Box385273B00OUO090.pdf. World Bank. 2014c. Transparency of State-Owned Enterprises in Vietnam: Current Status and Ideas for Reform. Washington, DC: World Bank. Zhang, Yin-Fang, David Parker, and Colin Kirkpatrick. 2008. “Electricity Sector Reform in Developing Countries: An Econometric Assessment of the Effects of Privatization, Competition and Regulation.” Journal of Regulatory Economics 33 (2): 159–78. 377 Appendix H. Summary of Analysis of Country Strategies and Diagnostics and FSAPs World Bank Group Country Strategies The review of country strategies reveals that the vast majority of country strategies (96 percent) substantially discuss state-owned enterprise (SOE) reform priorities or propose a work program to address these priorities. 60 SOE reform priorities are mentioned briefly in only two country strategies (El Salvador and Mauritius), and 38 of them substantially discuss these priorities and include a work program (figure H.1). Figure H.1. Degree of Discussion of SOE Reform Issues in Country Strategies (percent of documents) Source: Independent Evaluation Group review. Note: SOE = state-owned enterprise; WBG = World Bank Group. Among the country strategies that substantially discuss or propose a work program on SOE reform, 89 percent address priorities in the energy sector, 23 percent address priorities in the financial sector, and 11 percent discuss SOE reforms at the national level. Commercial banking is the focus of 80 percent of country strategies that address financial sector SOEs. More than three-quarters of the country strategy documents that proposed a work 378 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs plan included interventions supported by World Bank adjustment programs. Sixty percent of the documents planned interventions financed through World Bank investment projects, and 38 percent involved World Bank advisory services and analytics. Regarding International Finance Corporation (IFC), 38 percent of strategies planned IFC investment services, and 21 percent planned IFC advisory services. Seven percent of strategies planned for Multilateral Investment Guarantee Agency guarantees. SOE reform priorities at the upstream and downstream level are discussed equally. More than half of the strategy documents discuss both upstream and downstream interventions. At the upstream level, strengthening the regulatory framework is the most frequently discussed engagement area in strategy documents (70 percent), followed by improving governance, SOE ownership, and macro and fiscal policy. At the downstream level, more than two-thirds of the strategy documents discuss improving SOEs’ business and operations. Interventions on SOE ownership and improving environmental and social aspects of SOEs, corporate governance, and financial management are the other interventions that are mentioned to reform SOEs at the firm level. Strengthening regulatory frameworks and business and operations are the two most commonly covered SOE reform area in all Regions. Thirty-eight percent of the country strategies for the Latin America and the Caribbean Region proposed interventions in improving the regulatory framework for SOEs. This Region is followed by Middle East and North Africa (33 percent), Europe and Central Asia (32 percent), and Sub-Saharan Africa (32 percent). In Latin America and the Caribbean, 46 percent of the proposed SOE reform interventions are on improving the business and operations of SOEs. In the East Asia and Pacific and Sub-Saharan Africa Regions, more than 17 percent of the SOE reforms proposed aim to improve corporate governance. The country strategies reviewed in South Asia and in Middle East and North Africa Regions do not address corporate governance interventions at all. In the upper-middle-income countries, interventions on improving the 379 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs regulatory framework and business and operations account for more than two-thirds of the SOE reform interventions proposed in these countries. The share of corporate governance interventions is highest in low-income countries (21 percent), compared with 10 percent of the share in upper- middle-income countries. Strengthening the regulatory framework and improving business and operations represent 55 percent of all SOE reform interventions proposed in countries affected by fragility, conflict, and violence (FCV), compared with more than two-thirds of the SOE reform interventions in non-FCV countries. In countries financed through the International Bank for Reconstruction and Development, the interventions on SOE ownership account for almost one- quarter of the SOE reform interventions, and they represent 15 percent of interventions in International Development Association countries. Support in rehabilitation of infrastructure, preparation of sector strategies, and improving process efficiency are the three most frequently discussed SOE reform mechanisms in the country strategy documents. More than half of the strategy documents reviewed address support for rehabilitating infrastructure and preparation of sector strategies, and almost half of the documents cover support to improving process efficiency in SOEs. Other mechanisms discussed in the strategy documents include setting up regulatory agencies, reform of laws and regulations, privatization of SOEs, reform of environmental aspects in SOEs, compliance with international standards, improving service quality, capacity building, and adoption of pricing mechanisms. The review of the Completion and Learning Review and Country Partnership Strategy Completion Report Reviews reveals that two-thirds of the objectives for the SOE reform interventions proposed in the country strategies are mostly achieved or achieved. More than one-fifth of the SOE reform objectives set in the strategy documents have mixed outcomes, and 7 percent are mostly not achieved or not achieved. 380 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs Country-Level Diagnostics Bank Group Systematic Country Diagnostics The review of Systematic Country Diagnostics (SCD) documents reveals that SCDs substantially discussed and/or proposed SOE reform priorities (figure H.2). 61 SOEs are substantially discussed and are not considered a top priority for reform in only three countries (Colombia, El Salvador, and Mali). However, SOEs are substantially discussed and reform actions proposed in the remaining 36 countries. Figure H.2. Degree of Discussion of SOE Reform Issues in SCDs (percent of documents) Source: Independent Evaluation Group review. Note: SCD = Systematic Country Diagnostic; SOE = state-owned enterprise. Among the SCDs that substantially discussed and/or proposed SOE reforms, 90 percent address priorities in the energy sector. SOE reform priorities in finance accounted for 21 percent, and 5 percent of the discussion was on SOE reforms at the national level. Commercial banking is the most frequently discussed subsector in the financial sector, and the majority of the discussion in the energy sector is on nonrenewable energy SOEs. SOE reform priorities at the upstream and downstream level are discussed almost evenly in the SCDs. At the upstream level, strengthening the regulatory framework is the most frequently discussed engagement area 381 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs (48 percent), followed by improving governance, SOE ownership, financial management, and macro and fiscal policy. At the downstream level, 54 percent of SCDs discuss improving SOEs’ business and operations. Policy actions on corporate governance, SOE ownership, environmental and social aspects of SOEs, and financial management are the other areas mentioned at the firm level. Strengthening the regulatory framework and business and operations are the two most commonly covered SOE reform areas in all Regions. Forty- three percent of the documents for the Middle East and North Africa Region proposed policy actions to improve the regulatory framework for SOEs, followed by Latin America and the Caribbean (42 percent), East Asia and Pacific (33 percent), South Asia (25 percent), Europe and Central Asia (23 percent), and Sub-Saharan Africa (22 percent). In South Asia, 50 percent of the proposed SOE reforms are on improving the business and operations of SOEs. In the East Asia and Pacific and in Latin America and the Caribbean Regions, 17 percent of the SOE reforms proposed aim to improve corporate governance, and the SCDs reviewed in South Asia and in Middle East and North Africa Regions do not address corporate governance interventions at all. In the lower-income countries, interventions on improving the regulatory framework and business and operations account for 52 percent of the SOE reforms proposed. The share of corporate governance interventions is higher in lower-middle-income countries (8 percent) and low-income countries (6 percent) than in upper-middle-income countries at 5 percent. Most of the SOE reforms proposed concentrate on five areas in FCV countries (83 percent) and non-FCV countries (75 percent). In FCV countries, these areas are split almost evenly among strengthening regulatory frameworks (23 percent), improving business and operations (15 percent), governance (15 percent), financial management (15 percent), and macro and fiscal policy (15 percent). In non-FCV countries, there is greater emphasis on strengthening regulatory frameworks (30 percent) and improving business and operations (25 percent) and, to a lesser degree, on 382 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs governance (14 percent), financial management (4 percent), and macro and fiscal policy (3 percent). Reform of laws and regulations at the sectoral level, process efficiency, preparation and strengthening of sector strategies, and price tariffs are the four most frequently discussed SOE reform mechanisms in the SCD documents. One-third of the SCD documents reviewed address process efficiency, preparation and strengthening of sector strategies, and price tariffs, and 41 percent of the documents cover reforms of laws and regulations in SOEs. Other mechanisms discussed in SCD documents include rehabilitation of infrastructure, public-private partnership (PPP) arrangements, budget practices, setting up regulatory agencies, privatization of SOEs, auditing and oversight, compliance with international standards, and anticorruption measures. The analysis of the SCDs reveals that the most common type of energy sector indicator is the price tariff (dollars per kilowatt-hour), found in 18 percent of the documents. This indicator was used to measure performance and served as benchmark for SOE reform in Argentina, Bangladesh, Belarus, Botswana, Cameroon, Honduras, and Kosovo. Access to electricity as a percentage of urban or rural population and electricity as a percentage of gross domestic product (GDP) are the other indicators mentioned in the SCDs. For the financial sector, a variety of indicators were used, including state-owned commercial banks as a percentage of GDP, state-owned commercial bank assets as a percentage of banking sector assets, state-owned bank leverage ratio, and state-owned bank indebtedness. Country Private Sector Diagnostics The Country Private Sector Diagnostic (CPSD) is a new tool prepared jointly by IFC and the World Bank. CPSDs aim to assess constraints to and opportunities for private sector–led growth and provide recommendations for cross-cutting and sector-specific reforms and policy actions. SOE reform 383 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs concerns are discussed in the CPSDs, and some of them list SOE reform as a priority policy action. The findings of the CPSDs in the context of SOE reform in the energy and financial sectors are summarized in the next sections. Kazakhstan CPSD The CPSD for Kazakhstan states that Kazakhstan’s economy is “out of balance and heavily tilted toward the public sector, with large SOEs directly involved in production activities, a comprehensive welfare state and, conversely, a limited role for truly private enterprise” (Fengler et al. 2017, 4). Increasing private investment will depend on reforms in the areas, including SOE reform and competition policy. The CPSD suggests that the Bank Group may provide support to the government’s strategy on SOE and subsidy reform. The state is the largest depositor and borrower through fully and quasi-state entities. Furthermore, owners of the largest banks are closely related to the country’s political leadership. The CPSD suggests that the government of Kazakhstan needs to take coordinated action to move the privatization agenda forward. Moreover, the government needs to resolve the issue of insolvent banks and seek least-cost solutions in dealing with the overhang of distressed assets. Nepal CPSD The CPSD includes a section on the energy sector that lists the findings from the Infrastructure Sector Assessment Program. The energy sector is currently undergoing restructuring, in which the Nepal Electricity Authority is in the process of being unbundled from a vertically integrated utility into state-owned companies for generation, transmission, and distribution. Ghana CPSD Contingent liabilities from energy sector SOEs led to debt sustainability risks when the country’s public debt increased to 74 percent of GDP in 2016. 384 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs Ghana has 36 SOEs that are wholly owned by the state, concentrated largely in critical sectors of the economy, such as energy, finance, and infrastructure. Many of these SOEs underperform compared with the private sector, incur financial losses, and present an increasing burden on the budget and the banking system. In the energy sector, management inefficiencies, lack of timely utility tariff adjustment, and the accumulation of arrears have translated into severe power shortages, which further undermines investor confidence and adds to the economic slowdown. Ethiopia CPSD The private sector in Ethiopia remains nascent. The state continues to play a heavy role through SOEs in key areas of the economy, including telecommunications, finance, energy, logistics and transport, and manufacturing. In the financial sector, the two state-owned banks dominate the banking sector, and SOEs have a considerable share in the credit market. The CPSD recommends modernizing the state-owned banks and upgrading the regulatory framework in the financial sector. In the energy sector, the CPSD suggested that continued legal, regulatory, and institutional reform is needed. The institutions and state-owned utilities will require support in technical and financial planning as the sector moves from a public sector– dominated model toward a market with greater private sector participation. The report emphasized the importance of restoring financial sustainability to the energy sector (through tariff reform, debt restructuring plans, and so on) in attracting private sector participation. The CPSD also suggested considering pilot projects in downstream power and distribution segments to have private sector engagement. Build-operate-transfer financing and off- grid minigrids are the tools recommended for promoting private sector engagement. 385 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs Uzbekistan CPSD The CPSD proposed to support corporatization of airports and policy engagement in SOE reform and privatization in the chemicals sector. Angola CPSD Angola is home to the oil and gas company Sonangol, Africa’s largest SOE. Despite several waves of privatizations in the late 1990s and 2000s, SOE assets in the portfolio of the Institute for Management of State Assets and Shareholdings account for 78 percent of the country’s GDP today. SOEs have a dominant or substantial presence in agriculture, transport, construction, and banking. Their financial performance is poor on average and has deteriorated over the years. The CPSD suggests that changing the government’s role in the productive economy can be addressed by reducing public shareholding in SOEs and increasing PPPs for service delivery. In the power sector, the main issues listed are energy utilities’ capitalization levels, electricity market prices that are not economically sustainable, and questions about the regulator’s long-term independence. The CPSD suggests strengthening the energy utilities’ capacity as power purchasers by improving their technical and financial performance. Furthermore, the CPSD recommends reviewing tariffs to enable cost recovery, reforming the distribution company to reduce technical and commercial losses, and implementation of a strategic master plan. In the financial sector, the CPSD suggests reducing the state presence in the sector and restructuring public banks with high nonperforming loans. Burkina Faso CPSD Burkina Faso has only 13 firms under government ownership in strategic sectors after three waves of privatizations in the late 1990s and early 2000s. The CPSD indicates that although the public sector still represents more than one-third of GDP, this does not seem to crowd out the private sector. In the energy sector, Burkina Faso’s installed capacity is extremely limited and 386 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs highly dependent on neighboring countries to meet its demand. The national utility, Société Nationale d'électricité du Burkina Faso, or SONABEL, has limited generation and technical capacities. The CPSD recommends the following interventions in the energy sector: (i) Issue the required decrees to implement full provisions of 2017 Law on Energy; (ii) develop a least-cost energy generation plan and grid integration study, considering regional interconnections and shared infrastructure with mining companies; (iii) improve planning, execution, technical, and transactional capacities, including for renewable energy projects undertaken by independent power producers; and (iv) decentralize the independent power producers’ procurement process from the Ministry of Finance to the Ministry of Energy. Morocco CPSD SOEs are leading the investment and finance in infrastructure and account for just over half of total public infrastructure investment. SOEs also borrow from domestic banks and issue local currency bonds, mostly held by pension funds. The CPSD suggests optimizing the use of state guarantees to ensure that they catalyze rather than crowd out commercial finance. Outsourcing to the private sector the operation of infrastructure services through PPP arrangements and encouraging the origination of transactions suitable for private sector investment are also listed as other important measures. Rwanda CPSD SOEs occupy a prominent position in Rwanda’s enterprise sector. Public sector financing sought to make up for the dearth of formal private sector entities in the aftermath of the genocide, and SOEs played a useful role at the recovery and reconstruction stage, when medium- and large-sized private firms were largely absent. However, if the government still sees a role for SOEs to play, a good balance with the private sector will have to be found, for example the government withdrawing gradually from productive 387 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs activity through ongoing privatization of SOEs such as community processing centers. SOEs, including public-private investment groups in which government has invested along with private investors, are present in many sectors of the Rwandan economy. Allocating sufficient resources to the Rwanda Inspectorate and Competition Authority and extending its mandate to monitor the impact of SOEs on competition, promoting the principle of competitive neutrality to ensure equal treatment of all investors, and removing regulatory barriers to entry and rivalry in input sectors would improve the competition framework. In addition, there are questions about SOE performance in terms of adequacy of investments in needed production capacity, consistency with good commercial practices, and negative fiscal effects from treasury subsidies. Information gaps exist on which firms are SOEs, where they operate, and whether they have preferential access to key inputs such as land and finance. The CPSD provides the following recommendations for SOE reform: (i) Continue reforming SOEs and redefining their role in the economy by strengthening corporate governance through the adoption of the Organisation for Economic Co-operation and Development’s Guidelines on Corporate Governance of SOEs and (ii) decide on appropriate SOE involvement within each sector. For each sector with one or more SOEs, the government should evaluate the extent of competition within a sector, the relative competitiveness of SOEs versus private enterprises, significant social considerations, and long-term economic development goals. Kenya CPSD The Kenya CPSD finds that Kenya has a broad presence of SOEs including in sectors where private participation is viable. Kenya registers SOEs in at least 17 sectors compared with an average of 15.4 in Organisation for Economic Co-operation and Development countries, including in sectors where there 388 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs is active private sector participation such as retail, accommodation, manufacturing, banking, insurance, and agri-processing. In enabling sectors that provide essential inputs to the rest of the economy (electricity, transport, finance, telecommunications, and education) and where natural monopolies and SOEs are important, the effectiveness in achieving policy goals in terms of affordability and access to quality services is limited, affecting the costs for enterprises in traded sectors. It finds direct competition from SOEs, links between competing firms through partial government shareholding, and a lack of competitive neutrality given limited de facto separation of regulatory and commercial activities in sectors such as electricity, air transport, telecommunications, and agriculture can crowd out the private sector (particularly new investors), further limiting opportunities for socially impactful market creation. SOEs also generate a significant burden on fiscal accounts and debt, particularly on railways and electricity, amounting to 7 percent of GDP in 2016. The CPSD suggests strengthening competition policy and removing regulatory barriers and government interventions that restrict entry and competition in various key sectors, including electricity generation and insurance. Furthermore, the report recommends improving governance and market discipline mechanisms toward SOEs to increase their efficiency and to help crowd-in the private sector. In the energy sector, the CPSD recommends encouraging private participation through PPPs (particularly in transmission), building the capacity of the Energy Regulatory Commission and the Ministry of Energy, and implement a wholesale electricity market. Review of Bank Group Financial Sector Assessment Program Reports The review of Financial Sector Assessment Program (FSAP) documents reveals that most FSAPs substantially discussed and proposed recommendations for reforming state-owned financial institutions (SOFIs; 389 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs figure H.3). 62 SOFIs are substantially discussed and/or recommendations are proposed in 22 countries. However, SOFIs are briefly discussed in two countries (Kazakhstan and Mauritius) and are not mentioned in five countries (Colombia, Djibouti, Kosovo, Morocco, and Tanzania). Figure H.3. Degree of Discussion of SOFI Reform Issues in FSAPs Source: Independent Evaluation Group review. Note: FSAP = Financial Sector Assessment Program; SOFI = state-owned financial institution. Among the FSAPs that substantially discussed and/or proposed recommendations on SOFI reforms, 66 percent addressed priorities in the state commercial bank subsector, 28 percent address priorities in nonbank financial institutions, and 24 percent discussed state development banks. FSAPs focused on the state commercial banking subsector more than any other subsector in South Asia, East Asia and Pacific, and Europe and Central Asia Regions. However, they did not discuss and/or propose recommendations on SOFI reforms on the Middle East and North Africa Region. Similarly, there is more focus on state commercial banking in lower- middle-income and upper-middle-income countries, and most of the SOFI reforms in low-income countries are geared toward nonbank financial institutions. At the upstream level, governance is the most frequently discussed reform (46 percent), followed by regulatory framework, macro and fiscal policy, and SOFI ownership. At the downstream level, 45 percent of FSAPs identify 390 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs corporate governance as the main issue. Challenges in financial management, business and operational management, and SOFI ownership are also discussed in the FSAPs at the firm level. Issues in governance and regulatory frameworks are mentioned in all Regions and, except in Middle East and North Africa, were the most common. In the Latin America and the Caribbean Region, 67 percent of documents discussed governance issues. This Region is followed by East Asia and Pacific (50 percent), Europe and Central Asia and Sub-Saharan Africa (40 percent each), and South Asia (33 percent). In the East Asia and Pacific Region, regulatory framework challenges were present in 25 percent of the FSAPs. In the Europe and Central Asia and Sub-Saharan Africa Regions, they were mentioned in 20 percent of the documents. SOFI challenges in the Middle East and North Africa Region were not discussed. SOFI challenges concentrate on governance in FCV countries (100 percent). In non-FCV countries, regulatory framework (21 percent), macro and fiscal policy (15 percent), and financial management (12 percent) are also important issues. In terms of the recommendations in FSAPs, at the upstream level, strengthening regulatory framework is the most frequently discussed area (41 percent), followed by improving governance, SOFI ownership, and macro and fiscal policy. At the downstream level, 50 percent of FSAPs provide recommendations in SOFIs’ ownership. Policy actions on financial management, corporate governance, and business and operational management are discussed at the firm level. Strengthening regulatory frameworks and governance are the two most commonly recommended SOFI reform areas in all Regions except for Middle East and North Africa. Fifty percent of the documents for the East Asia and Pacific Region proposed actions to improve the regulatory framework for SOFIs. This Region is followed by South Asia (43 percent), Europe and Central Asia (27 percent), Latin America and the Caribbean (22 percent), and 391 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs Africa (20 percent). In Latin America and the Caribbean, 44 percent of the proposed SOFI reforms are on improving governance of SOFIs. In Sub- Saharan Africa and in Europe and Central Asia, 40 percent and 36 percent, respectively, of the SOFI reforms proposed address SOFI ownership, and the FSAPs reviewed in Latin America and the Caribbean only address 11 percent of SOFI reforms and do not address any in the Middle East and North Africa Region. In the lower-middle-income and upper-middle-income countries, interventions on regulatory framework and governance account for 69 percent and 62 percent, respectively, of the SOFI reforms proposed. By contrast, they account for only 44 percent in low-income countries. The share of SOFI ownership interventions is higher in low-income (44 percent) and upper-middle-income countries (31 percent) than in lower-middle- income countries (16 percent). Most of the SOFI reforms proposed concentrate on three areas in FCV countries (87 percent) and only two areas non-FCV countries (100 percent). In FCV countries, most of the reforms are split among SOFI ownership (32 percent), governance (29 percent), and regulatory frameworks (26 percent). In non-FCV countries, there is more emphasis on governance (67 percent) than on regulatory frameworks (33 percent). Privatization of SOFIs, professionalization of boards of directors, and setting up or reform of regulatory agencies are the three most frequently discussed SOFI reform mechanisms in the FSAP documents. Of the FSAP documents reviewed, 41 percent address privatization of SOFIs, 24 percent cover professionalization of boards of directors, and 21 percent address setting up or reforming regulatory agencies. The mechanisms that were the least discussed include financial reporting, improvements on the service, improving debt management, restructuring state banks, and adopting pricing mechanisms. 392 Appendix H Summary of Analysis of Country Strategies and Diagnostics and FSAPs References Fengler, Wolfgang, Christopher David Miller, Daniel Coutinho, and Indermit S. Gill. 2017. Creating Markets in Kazakhstan. Country Private Sector Diagnostic. Washington, DC: World Bank. Notes 60 This analysis uses the most recent country strategy documents prepared within the evaluation period and reviewed by the Independent Evaluation Group. The population is 114 countries with a reviewed Country Partnership Strategy (CPS). Of these 114 countries, the team drew a stratified random sample of 46 (40 percent) for review. The team rated each country strategy in the sample based on its coverage of state-owned enterprise reform issues. 61 This analysis covers recent Systematic Country Diagnostic (SCD) documents prepared within the evaluation period. The stratified random sample drawn for CPS analysis is also used. A total of 39 SCDs are reviewed because the remaining seven countries do not have an SCD yet. 62 This analysis covers recent Financial Sector Assessment Program reports prepared within the evaluation period. The stratified random sample drawn for CPS analysis is also used. A total of 29 Financial Sector Assessment Program reports are reviewed. 393 The World Bank 1818 H Street NW Washington, DC 20433