89202 Making Cross-Border Banking Work for Africa Thorsten Beck, Michael Fuchs, Dorothe Singer and Makaio Witte Making Cross-Border Banking Work for Africa Thorsten Beck, Michael Fuchs, Dorothe Singer and Makaio Witte Published by Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH Registered offices in Bonn and Eschborn, Germany Dag-Hammarskjöld-Weg 1-5 65760 Eschborn, Germany Telephone: +49 (0) 6196 79 - 0 Fax: +49 (0) 6196 79 - 1115 E-mail: info@giz.de Website: www.giz.de On behalf of: German Federal Ministry for Economic Cooperation and Development (BMZ) Edited by: Deborah Davis, Washington, DC Design and layout: Iris Christmann, Wiesbaden Printed by Metzgerdruck, Obrigheim This volume is a product of the staff of the Association of African Central Banks, the German Federal Ministry for Economic Cooperation and Development, and the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the member central banks of the Association of African Central Banks; the German Federal Ministry for Economic Cooperation and Development; or The World Bank, its Board of Executive Directors, or the governments they represent. The organizations do not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the Association of African Central Banks, the German Federal Ministry for Economic Cooperation and Development or The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. All rights reserved. Reproduction and dissemination of material in this volume for educational or other non-commercial purposes is authorized, without any prior written permission from the copyright holders, provided the source is fully acknowledged. Reproduction of material in this volume for resale or other commercial purposes is prohibited without written permission of the copyright holders. Applications for such permission should be addressed to: Makaio Witte, Dag-Hammarskjöld-Weg 1-5, 65760 Eschborn, Germany; telephone: +49 61 96 79-2860; e-mail: makaio.witte@giz.de. ISBN: 978-3-944152-39-4. © 2014 Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH and International Bank for Reconstruction and Development / The World Bank Contents 05 Contents List of Boxes 07 List of Figures 07 List of Tables 08 List of Acronyms 09 Acknowledgments 11 Preface 13 Executive Summary 15 Chapter 1: Recent Trends in Cross-Border Banking in Africa 23 1.1 Financial Sector Development in Africa 26 1.2 The Expansion of Cross-Border Banks 28 1.3 Reasons for Cross-Border Expansion 50 1.4 Business Models of Cross-Border Banks 54 1.5 Conclusion 59 Annex: List of Cross-Border Banks in Africa 60 Chapter 2: Cross-Border Banking: Benefits and Risks 67 2.1 The Theory 69 2.2 The International Evidence 72 2.3 Implications of Global Experiences for Africa 84 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 89 3.1 Why Cross-Border Regulatory Cooperation? 91 3.2 International Standards as Reference Point for Cross-Border Regulatory Cooperation 99 3.3 Tools and Instruments of Cross-Border Regulatory Cooperation 107 3.4 Looking Beyond “Sunny-Day” Cooperation: New Forms of Regulatory Cooperation 111 3.5 Cross-Border Regulatory Cooperation in the Presence of Heterogeneity Across Countries 119 3.6 Conclusion 122 06 Contents ccccc 4: Current Issues in Cross-Border Regulatory Chapter and Supervisory Cooperation in Africa 123 4.1 Harnessing Financial Integration to Promote Financial Deepening: Policy Choices Faced by Regulatory Authorities 125 4.2 Current Practices and Challenges in Coordination Among African Regulators 130 4.3 Conclusion 143 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 145 5.1 Policy Recommendations 146 5.2 Looking Forward 153 Data Sources 157 Literature 159 List of Boxes / List of Figures 07 List of Boxes Box 1.1: Definitions 25 Box 1.2: Is Barclays Africa Group South African or British? 32 Box 1.3: A Different Kind of International Bank in Africa: The Operations of Rabobank 33 Box 1.4: A Brief History of Banking in Africa 36 Box 1.5: A Counterpoint to the Expansion of African Cross-Border Banks: Access Bank 52 Box 1.6: The Largest African-Based Pan-African Banks at a Glance 57 Box 2.1: Scale Economies 75 Box 2.2: Examples of Serving the Lower-End of the Market by Cross-Border Banks 81 Box 3.1: Complex Structures and Resolution of Cross-Border Banks: Learning From History 92 Box 3.2: The Basel Committee’s Evolving Framework for Cross-Border Banking Supervision 101 Box 3.3: How Relevant are International Banking Standards in the African Context? 105 Box 3.4: Nordic Baltic Cooperation Agreement on Cross-Border Financial Stability, Crisis Management, and Resolution 115 Box 3.5: The European Bank Coordination “Vienna” Initiative (EBCI) 118 Box 4.1: Divided Responsibilities in WAEMU 137 List of Figures Figure 1.1: Number and Share of Foreign Banks in Africa, 1995-2009 28 Figure 1.2: Foreign Bank Ownership by Region, 1995-2009 29 Figure 1.3: Share of Assets Owned by Foreign Banks in Africa, 2011 31 Figure 1.4: Geographical Footprint of Major International Banks in Africa, 2013 34 Figure 1.5: Cross-Border Expansion of African Financial Groups over Time, 1990-2013 39 08 List of Figures / List of Tables Figure 1.6: Geographical Footprint of Major African Banks in Africa, 2013 40 Figure 1.7: Share of Assets in Host Country Banking Systems (%) by Selected African Banks, 2011 44 Figure 1.8: Share of Assets in Host Country Banking Systems by Selected International Banks, 2011 47 Figure 1.9: Ownership Linkages Among African Banks 48 Figure 1.10: Ownership Linkages of International Banks in Africa 49 Figure 2.1: Interest Rate Spreads in the EAC by Foreign and Domestic Bank Ownership Type, 2012 74 Figure 3.1: Bank Failures and Interventions in Selected Countries 94 Figure 4.1: Median Compliance with BCPs in Africa, 2006-2013 132 Figure 4.2: Compliance with Selected BCPs Relevant for Cross-Border Banking Supervision in Africa, 2006-2013 134 List of Tables Table 1.1: List of Major International Cross-Border Banks in Africa 35 Table 1.2: List of Major African Cross-Border Banks 42 Table 1.3: List of Cross-Border Banks in Africa 60 Table 5.1: Analytical Framework for Policy Action 147 List of Acronyms 09 List of Acronyms AACB Association of African Central Banks AfDB African Development Bank AFRITAC African Regional Technical Assistance Center APRA Australian Prudential Regulatory Authority ASIC Australian Securities and Investments Commission BCBS Basel Committee on Banking Supervision BCP Banque Centrale Populaire du Maroc BCPs Basel Core Principles for Effective Banking Supervision BIS Bank for International Settlements BMCE Banque Marocaine du Commerce Extérieur BOA Bank of Africa BSA Bank Supervision Application CABS Community of African Banking Supervisors CCBG Committee of Central Bank Governors CEMAC Economic and Monetary Community of Central Africa COMESA Common Market for Eastern and Southern Africa CoS College of Supervisors EAC East African Community EAMU East African Monetary Union EAP East Asia and Pacific EAPS East African Payments System EBRD European Bank for Reconstruction and Development EBCI European Bank Coordination Initiative ECA Europe and Central Asia ECB European Central Bank ECOWAS Economic Community of West African States EMDE Emerging Markets and Developing Economies ESAF East and Southern Africa Banking Supervisors Group FIP Protocol on Finance and Investment FSAP Financial Sector Assessment Program FSB Financial Stability Board GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit ICBC Industrial and Commercial Bank of China IFRS International Financial Reporting Standards IMF International Monetary Fund 10 List of Acronyms LAC Latin America and Caribbean LIC Low-Income Countries MAC Monetary Affairs Committee MFI Microfinance Institution MFW4A Making Finance Work for Africa MoU Memorandum of Understanding MSMEs Micro, Small, and Medium-Sized Enterprises OECD Organisation for Economic Co-operation and Development RBNZ Reserve Bank of New Zealand REC Regional Economic Community SADC Southern African Development Community SARB South African Reserve Bank SMEs Small and Medium-Sized Enterprises SSBS SADC Subcommittee of Banking Supervisors UBA United Bank of Africa WAEMU West African Economic and Monetary Union WAMU West African Monetary Union WAMZ West African Monetary Zone WBC WAMU Banking Commission Acknowledgments 11 Acknowledgments The principal authors of this report are Thorsten Beck (Cass Business School), Michael Fuchs (GIZ), Dorothe Singer (World Bank) and Makaio Witte (GIZ). The authors are thankful to the peer reviewers of the report, Robert Cull (World Bank), Ralph De Haas (EBRD), Miquel Dijkman (World Bank), Maria Soledad Martinez Peria (World Bank), Mauro Mecagni (IMF), and Cedric Mousset (World Bank), and for the comments provided by Mario Guadamillas (World Bank), Dirk Jan Grolleman (IMF), Paul Matthieu (IMF), Jean Pesme (World Bank) and Jan Rein Pruntel (consultant). Excellent research assistance was provided by Johanna Jagnow and Atilla Yücel. The report benefited from the support and was prepared under the guidance of Gabriela Braun (Program Manager, GIZ), Gaiv Tata (Director, Africa Finance and Private Sector Development and Director, Global Practice on Financial Inclusion, World Bank), and Rundheersing Bheenick and Samuel Méango (respectively Chairman and Executive Secretary of the Association of African Central Banks). Coordination and management support was ably provided by the Secretariat of the Making Finance Work for Africa Partnership, led by Stefan Nalletamby (Coordinator) and his team, notably Hugues Kamewe Tsafack (Stakeholder Relationship Officer). The authors are especially grateful for the insights officials of banks and regula- tory authorities provided during reporting visits to Ghana, Kenya, Mauritius, Morocco, Mozambique, Namibia, Netherlands, Nigeria, South Africa, Tanzania, Togo, Uganda, and the United Kingdom. The report greatly benefited from four regional studies on cross-border banking supervision in the East African Community (EAC); the Economic Community of West African States (ECOWAS); the Southern African Development Com- munity (SADC); and the West African Economic and Monetary Union (WAEMU) produced by GIZ in preparation for this report. The authors are also grateful for the comments received during consultation events with the Working Committee on the Workplan of the Community of African Banking Supervisors (CABS) in Port Louis, and in the context of roundtable discussions with the Association of African Central Banks (AACB), the World Bank, and GIZ in Washington, DC. Preface 13 Preface This report is a joint effort and the result of a long-standing collaboration among the Association of African Central Banks (AACB), the World Bank, and Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH, under the umbrella of the Making Finance Work for Africa (MFW4A) Partnership. African policymakers identi- fied cross-border banking and the challenge of supervising increasingly complex and interconnected financial groups as a key policy priority for the continent at a series of high-level events held between 2011 and 2014.1 The importance of cross-border banking relates to a combination of factors, including its significant increase across Africa in recent years; the severe challenges faced by developed countries’ supervisory author- ities in supervising international bank groups and in dealing with troubled or insolvent banks; and widespread cross-border contagion during the global financial crisis. These issues have revived an age-old debate about the costs and benefits of closer financial integration, and the implications for national and regional financial sector policies and supervisory practices in the African context. Despite the increased attention paid to and the acknowledged relevance of cross- border banking, there continues to be a lack of comprehensive research and analysis of this topic as the basis for developing a coordinated policy agenda. This report aims to fill this gap by providing a comprehensive analysis of the growth of cross-border banking in Africa; assessing the benefits and risks of the expanding cross-border linkages; assess- ing regulatory frameworks and current arrangements for cross-border supervisory cooperation; and discussing advisable supervisory approaches and policies to balance the benefits with the risks of deepening cross-border linkages. The analysis was inspired by an in-depth dialogue with African regulatory and supervisory authorities, as well as the inputs of international policy and regulatory bodies such as the G20’s Financial Stability Board (FSB), the International Monetary Fund (IMF), its African Regional Technical Assistance Centers (AFRITAC) and the German Federal Financial Supervisory Authority (BaFin). 1: Policy Workshop on “The African Approach to the Implementation of International Standards for Banking Supervision”, Kampala, April 2011; Meetings of FSB Regional Consultative Group Sub-Saharan Africa in 2012 and 2013; African Governors‘ Roundtable at the IMF-World Bank Spring Meetings, Washington D.C., April 2012, 2013 and 2014; Inaugural Meeting of the Community of African Banking Supervisors (CABS), Algiers, January 2013; Meeting of the Working Committee on the Worplan of the CABS, Port Louis, January 2014. Executive Summary 15 Executive Summary Cross-border banking has become an increasingly important feature of African finan- cial systems, and this trend has accelerated in the past decade. African banks have not only substantially increased their geographic footprints on the continent, but have also become economically significant beyond their home countries and of systemic impor- tance in a number of jurisdictions. This growth and expansion of African banks has, in recent years, reduced the relative importance of traditional, mostly European, banks on the continent and has shifted the burden of managing the risks and the reaping benefits of cross-border banking from the traditional home countries in Europe towards African policymakers. Financial systems in Africa stand to gain significantly from further cross-border banking integration. The academic literature provides nuanced messages about the impact of cross-border banking on efficiency and competition, financial deepening and outreach, and financial stability. The impact depends on a number of factors, including the financial infrastructure (e.g. efficiency of credit information-sharing systems); market structure (e.g., existence of dominating state-owned banks); and regulatory policies that can either foster or impede market contestability. The potential for reaping the benefits of economies of scale and fostering financial and economic development are likely to be greater in Africa than elsewhere, given that financial markets are still small and shallow. Most cross-border expansion in Africa has come in the form of stand-alone subsidiaries with limited integration across affiliate networks or with parent banks, even in sub- regions with currency unions and joint bank regulatory authorities. Cross-border banks in Africa have deployed different market strategies and degrees of engagement in host countries. To date, cross-border banks in Africa have been primarily servicing their large corporate clients, and especially target growth sectors such as natural resource extraction. Although exceptions exist, cross-border banks are still mostly reluctant to engage in servicing the lower end of the market. Understanding and addressing the reasons for this reluctance will enable African banking systems to better leverage the upsides of cross-border banking. Regional financial integration also entails new potential sources of risks. New channels of contagion will and are indeed already developing, as national banking systems and financial markets become increasingly interwoven, allowing for the transmission of shocks across borders. These new potential risks underscore the importance – for both banks and their supervisors – of having in place adequate provisions for risk manage- ment and mitigation. These risks also call for a greater commitment and adherence 16 Executive Summary to common rules of the game, as embodied in internationally accepted standards and practices, in order to foster greater confidence in the financial sectors on the continent. Given the serious repercussions for both financial stability and financial deepening associated with bank fragility, particularly with regard to larger pan-African banking groups, both the banks and the authorities have a mutual interest in upgrading bank oversight and cross-border supervisory cooperation. Potential stability risks emerging from cross-border banking often need to be addressed collaboratively rather than by individual home or host country supervisors. The rapid expansion especially of pan-African banks therefore requires both a significant upgrade of domestic oversight and an adjustment of the current toolbox of cross-border regulatory cooperation. Consolidated supervision of a bank’s activities through all its affiliates, across sectors and jurisdictions; Memoranda of Understanding (MoUs) between home and host supervisors; and Colleges of Supervisors (CoS) for individual cross-border financial institutions, are the basis for effective cross-border regulatory cooperation. However, as the recent global financial crisis has shown, these tools are not sufficient to safeguard financial stability. Greater efforts are necessary to develop cross-border frameworks for crisis management and bank resolution (thus building on domestic efforts in this area), not only as a means to collaboratively deal with financial sector fragility and failed banks, but also to set the appropriate ex ante incentives to reduce the probability of systemic banking crises. Addressing this agenda requires that authorities carefully assess country and region-specific circumstances, as the scope and intensity of cross-border regulatory cooperation should be in proportion to the strength of the cross-border linkages between home and host country banking systems. Countries across Africa have made progress in upgrading their regulatory frame- works, including elements applied to cross-border banking, but significant gaps remain. Few African countries have effective systems of consolidated supervision in place. This is of concern especially for the supervision of a few large African cross-border financial institutions that are active across the continent and systemically important in a number of countries, but are under neither consolidated nor effective solo (home or host) coun- try supervision. Moreover, only a small but increasing share of cross-border linkages across the continent is covered by proper arrangements for home-host supervisory cooperation, such as MoUs or CoS. Yet even where such arrangements are in place, considerable efforts are required to implement them effectively, in order to enable regular and trusting exchange of relevant information on a bilateral or multilateral basis. In view of the multiple demands posed by the intensification of cross-border banking and the limited supervisory resources, regulatory authorities in Africa face an import- ant trade-off. On the one hand, they are faced with the more traditional task of safe- guarding their financial institutions and banking systems, taking into consideration the Executive Summary 17 risks associated with the growth of cross-border activity in recent years. On the other hand, they are faced with the challenge of reaping the significant gains associated with cross-border banking in the form of more efficient intermediation and deepening of financial markets, which will also contribute to greater resilience and thereby support a virtuous cycle in support of financial system stability and robustness. Policy recommendations The recommendations of this report provide an agenda for supervisory authorities to both reap the benefits and contain the risks of cross-border banking. They are orga- nized along three policy objectives: (a) reaping the benefits of cross-border banking; (b) safeguarding the real and financial sector against risks stemming from cross-border banking; and (c) preparing for cross-border repercussions of idiosyncratic and systemic bank fragility. Actions need to be taken both on the national level and in collaboration among supervisory authorities. Given the number of countries in Africa and their widely varying circumstances, the policy recommendations have to be fine-tuned to suit the context of individual countries, country pairs, and sub-regions. The recommended prioritization and sequencing of policy actions will vary significantly across jurisdictions, depending on factors such as the significance of cross-border banking activities, the degree of integration within banking groups, the complexity and opacity of group structures, and the quality of home and host supervision. The recommendations also come with the recognition that cross-border regulatory cooperation is a process that takes time and involves both soft and hard institution building. It also cannot be expected to be a one-off process, as changes in international standards and in the cross-border banking landscape in Africa will require continuous adjustments to such a framework. Reaping the benefits of cross-border banking If Africa takes advantage of the efficiency gains and innovations that cross-border banks can provide, the continent stands to benefit substantially from cross-border banking in terms of financial deepening and increased outreach to previously unbanked parts of the population. Despite increasing financial integration, the impact of cross-border banking on banking efficiency and financial outreach has so far been limited. One of the constraining factors is the often still rudimentary financial infrastructure. It is there- fore a matter of priority to strengthen financial infrastructure in a consistent manner across countries, particularly those that share strong cross-border linkages. In the context of cross-border banking, the focus would be on improving the comparability 18 Executive Summary of credit information across countries; enhancing the efficiency of payments systems, particularly with regard to cross-border retail payments and the servicing of migrant flows; strengthening mutual recognition of procedures for registration of property and collateral rights; mechanisms for foreclosure on collateral; and improving financial literacy and the availability of comparable information on the cost of financial services. Such upgrades might be best undertaken in a coordinated manner within sub-regions, as is already the case in some Regional Economic Communities. Given the low levels of financial intermediation in most African banking sectors, particularly in the lower end of the market, considerable upside potential lies in transfer of know-how, IT, infrastructure, and risk-management skills relating to low-income retail banking and products suited to small savers and enterprises. Experience indicates that, where banks have successfully developed such banking skills and products in their home market, they are more likely to do so abroad. To promote market deepening, greater competition and innovation, authorities could consider encouraging the entry of banks that are experienced in servicing underserved market segments and can demonstrate success based on tested business models. Authorities could also consider a move towards more integrated banking models based on a sound framework for consolidated supervision, clearly established and func- tioning channels of information exchange between home and host country supervisors, and effective cross-border resolution frameworks. Banks expanding across borders in Africa are almost universally required to establish not only self-standing subsidiaries but also local IT functions; to use predominantly local labor; and to establish independent, local management functions. This “fortress banking” runs directly counter to reaping the potential economic gains from cross-border banking. In particular, more integrated banking models would provide the opportunity for significant cost savings in a traditionally high-cost industry, and could make it cost-efficient and therefore attractive to provide financial services to a broader set of clients. Policies fostering more integrated banking models could, for example, include reducing the complexity and length of the licensing process; reducing initial capital requirements for bank subsidiaries (with requirements designed to grow in line with the foreign bank’s business engagement and risk expo- sures); reducing or doing away with requirements to establish new branches where these exist (leaving, for example, decisions about the structure and security of bank premises to the banks); encouraging full mobility in the use of labor (skills transfer); encouraging the use of centralized, common IT platforms for both internal operations and the provi- sion of client services (such as ATMs, card services, and internet banking); and allowing the establishment of centralized audit and risk management system. A move away from stand-alone subsidiaries towards more integrated subsidiaries and eventually even branching – if certain preconditions are met – could be considered especially for formally integrated regulatory areas such as the Central and West African currency unions. Executive Summary 19 In addition, regulatory harmonization could contribute to greater certainty with regard to predictability and consistency in implementation, significant reduction in compliance costs across the region, and raising standards in more challenged environ- ments. Regulatory harmonization is a huge undertaking, and care needs to be taken to focus convergence efforts on key concerns, especially in environments with severe capacity constraints. Prioritization and sequencing are crucial, and the focus needs to be on those policy areas where harmonization is essential to the integration agenda. For example, in an environment where credit risk is the key risk factor, prioritization might suggest a focus on loan loss classification criteria and provisioning requirements. It is important that policymakers leverage the complementarity between the national and sub-regional agendas in promoting banking efficiency. Regional Economic Communities can play an important role in establishing best practice processes, fostering exchange on these practices, and monitoring progress in their implementation. However, given the complexity of implementing multilateral reform programs, it will be extremely important to prioritize such sub-regional endeavors, and it may be more expedient to implement some reforms on a bilateral basis so as to draw lessons before replicating them at the sub-regional level. Such a bilateral approach could be appropriate in fast-evolving areas, such as regulation of mobile money and banking agents and the related development of payments systems infrastructure. Safeguarding stability of cross-border banking in normal times Consolidated supervision is a critical component of overseeing cross-border banks, yet most African home country supervisors still lack adequate frameworks, implementation capacity, and consolidated accounting data. Establishing or improving frameworks of consolidated supervision and their effective implementation is therefore a high priority for safeguarding financial stability in Africa. To effectively carry out consolidated supervision, authorities require sufficient data on the activities of banks. Information about the size and nature of cross-border banking activities in Africa is currently difficult to come by or unavailable. A key immediate task facing African authorities is to improve the availability and regular exchange of relevant information. To make this task manageable, it is strongly recommended that the smaller group of African home supervisory authorities take the lead in developing the required formats as well as a platform enabling regular information exchange on a basic set of data. This data set should include information on (a) basic qualitative and quantitative characteristics of cross-border banks; (b) supervisory data as it relates to performance; (c) qualitative information on regulatory frameworks and definitions underlying supervisory data; and (d) market intelligence. Making such data publicly available will allow for overseeing and monitoring of ongoing developments in cross-border 20 Executive Summary banking, and could serve as the basis for a risk-based approach to strengthening bank- ing supervision. At the same time, timely exchange of more detailed, institution-specific information based on Memoranda of Understanding among supervisors and as input to supervisory colleges is also necessary for effective cross-border supervision, and an early detection of financial fragility. Effective consolidated supervision will also depend on building trust between home and host supervisory authorities, which is essential to the quality and frequency of information exchange, particularly when it relates to more detailed, institution-specific information. Formal arrangements can be instrumental in building trust and anchoring expectations. Improvements in supervisory cooperation – through signing of appropriate MoUs and formation of properly structured CoS, joint inspections of cross-border banks, and peer learning and capacity building events – is thus an important area of action on both bilateral but also sub-regional levels. Moreover, the fact that there are large African cross-border banks that are not sub- ject to consolidated supervision from their respective home country supervisors needs to be addressed. Given the large geographic footprint of these institutions, coordination on the pan-African level is urgently called for. The potential reputational and stability risks associated with unreported or undisclosed risks of cross-border banks and con- glomerates mean that the authorities and the banks have an important common social responsibility in seeing that regulatory gaps are addressed. A private/public partner- ship could play an important role in advancing this agenda, and thereby ensure a level playing field for banks. At the same time, the Community of African Bank Supervisors (CABS) or the Financial Stability Board’s Regional Consultative Group for Sub-Saharan Africa might serve as a coordination forum to identify systemically important institutions and how best to coordinate policy actions and monitor implementation of consolidated supervision for systemically important African cross-border banks. The CABS could also become a permanent forum for discussing issues related to cross-border cooperation. While such an exchange can play a decisive role in shaping the regional policy dialogue and capacity building agenda, day-to-day cooperation will have to be implemented on either a bilateral level or within smaller groups focusing on specific cross-border institutions. The CABS can, however, serve as an important forum for exchanging ideas and experiences and for driving convergence towards a common set of international standards, while at the same time facilitating the development of Africa-appropriate regulatory frameworks and providing the entry point for more detailed cooperation between individual countries or on individual banks. Executive Summary 21 Preparing for cross-border repercussions of bank fragility Preparing for the cross-border repercussions of bank fragility clearly requires a solid foundation for safeguarding the stability of banking in normal times. However, being prepared for times of distress also requires that authorities be equipped with sound resolution frameworks on a national level so that resolution proceedings can be initiated in a timely manner, with clearly assigned responsibilities among relevant authorities that have sufficient powers regarding transfer of assets and liabilities and implementing bank restructuring. Aside from planning for the eventualities of bank failure, resolution frame- works are also a preventive measure in that they affect the incentives of banks with regard to risk-taking even in normal times. There is also a considerable outstanding agenda in many African countries relating to respecting the hierarchy of creditors in bank resolution and preventing legal actions that constrain the implementation of resolution measures. Cross-border regulatory cooperation also has to look beyond information exchange during normal times and towards preparation for cross-border repercussions of idiosyncratic and systemic bank failures. Joint crisis simulation exercises could be used as the foundation for joint crisis management plans. Where relevant, this could also involve expanding supervisory colleges to include resolution authorities such as ministries of finance in so-called Crisis Management Groups. Given that the orderly resolution of cross-border groups is inevitably complex, every effort should be made to take precautionary measures to avoid the emergence of non-transparent, strongly interwoven international financial groups through strict monitoring, limits on intra-group exposure, and enhanced business continuity planning. While consolidated supervision is the basis for such measures, such a precautionary approach goes beyond collecting the necessary information and towards a more active involvement of supervisors, as is currently done for several systemically important financial institutions in the U.S. and Europe in the context of recovery and resolution plans. Looking forward An overarching conclusion arising from the analysis undertaken in this report, including from discussions with supervisors and bankers throughout Africa, is that information exchange is weak and needs to be strengthened significantly in the face of expanding cross-border banking activity in Africa. Establishing a platform for regular information exchange that makes publicly available a basic set of data about cross-border bank- ing activities in Africa will be an important first step to facilitate better supervision of cross-border activities and foster closer collaboration among authorities. Clearly setting in motion and sustaining a process for compiling and exchanging such information requires commitment and focus. Given that there are more than 40 regulatory 22 Executive Summary authorities in Africa (including regional authorities representing groups of countries), the practical and effective way to address this information vacuum is to place responsibility with the small number of home country supervisors – Kenya, Mauritius, Morocco, Nigeria, and South Africa – to take the lead in this exercise, drawing, as found expedient, on expertise from institutions such as the International Monetary Fund and the World Bank. Once developed, the data should as far as possible be published and in any event be made available to all relevant home and host supervisors in Africa, with a view to enabling an informed exchange among supervisory authorities and improving the quality and effec- tiveness of supervision. Chapter 1: Recent Trends in Cross-Border Banking in Africa 23 1. Recent Trends in Cross-Border Banking in Africa 24 Chapter 1: Recent Trends in Cross-Border Banking in Africa 1. ccccc Recent Trends in Cross-Border Banking in Africa Cross-border banking has been a critical part of Africa’s financial history since colonial times. While the period after independence saw a wave of nationalization across the continent, with many of the colonial banks exiting, this trend was reversed in the 1980s with the arrival of financial liberalization. Failing state-owned and private banks were sold mostly to global investors or multinational banks. Increasing international and regional economic integration, including of financial services, and deregulation further increased the number of foreign banks, and by the mid-2000s many African banking systems were yet again dominated by foreign banks. This publication focuses on cross-border banking through a bank’s commercial presence in countries outside the bank’s home country, or through foreign direct investment (FDI). It does not consider cross-border bank flows between countries or direct service provision without commercial presence.2 See Box 1.1 for definitions. Currently 104 cross-border banks with at least one branch or subsidiary outside their home country are active in Africa, one third of which are cross-border banks of non- African origin (see Annex 1.1). The majority of these institutions (71) have only a limited footprint in one to four African jurisdictions. The remaining 33 institutions are repre- sented through their home country operations and through branches or subsidiaries in five or more African countries. With a view to concentrating on the major players, the following analysis focuses primarily on the cross-border banks with the larger geographic and operational footprint in Africa. This introductory chapter documents trends in cross-border banking in Africa and the increasing shift in the composition of foreign banks in Africa. The next section provides a short overview of financial systems in Africa to set the stage. Section 2 characterizes the population of cross-border banks operating in Africa today, their expansion across the continent, and their importance in the host countries. Section 3 explores the reasons for the expansion of cross-border banking on the continent. Section 4 assesses the different business models banks use to expand across the continent as well as the characteristics of their group structures. Section 5 concludes. 2: In terms of World Trade Organization (WTO) definitions, the focus is on mode 3 (foreign commercial presence) rather than mode 1 (cross-border supply) (United Nations, 2002). Chapter 1: Recent Trends in Cross-Border Banking in Africa 25 Box 1.1: Definitions A cross-border bank is a bank with a commercial presence outside its home country, by way of at least one branch or subsidiary. Branches are entities that operate as a bank but do not have a separate legal status and are therefore part of the same legal entity as the (foreign) parent bank. In contrast, subsidiaries are separate legal entities that may be wholly owned or majority-owned by a bank in another country. Subsidiaries may themselves own subsidiaries in third countries.a Aside from branches and subsidiaries, banks can also open a representative office. While representative offices are generally prohibited from performing any banking opera- tions, they offer the opportunity for banks to facilitate interactions between the parent bank and commercial and financial businesses outside the home country. Cross-border banks are considered to be foreign-owned if they are controlled by a shareholder or group of shareholders from outside the licensing jurisdiction. Control over a bank can be exercised if an individual or entity holds more than 50 percent of shares in a bank, subsidiary or branch. In cases where there is no majority shareholder, the bank is still classified as foreign-owned when a foreign minority shareholder has a controlling stake in a bank; i.e. holds the relative majority of shares and exerts de facto control over the bank’s management.b Majority ownership is attributed to the country where the controlling share- holder(s) are based. To determine the responsibilities and mode of cooperation in the surveillance of cross-border banks, it is crucial to distinguish between home and host supervisors. While this distinction is straightforward in most instances – e.g., Kenya Commercial Bank is based in Kenya, its home country, and has affiliates in a number of jurisdictions in eastern Africa, the host countries – there are a number of examples of banks not being headquartered in the country that holds the majority ownership. The home supervisor is the national or regional authority responsible for supervising the operations of a cross-border bank on a consolidated basis. The decision as to which authority becomes a group’s home supervisor is determined by the applicable banking regulation.c Generally, regulation follows the same logic as described above, meaning that consolidated supervision is to be exercised by the country that is home to the controlling shareholders of a group. By implication, if the own- ership structure of a group changes significantly, the responsibility for conducting consoli- dated supervision will shift accordingly from one authority to another. For example, when the Moroccan Banque Marocaine du Commerce Extérieur (BMCE) took a controlling stake in Bank of Africa (BoA), Bank Al-Maghrib, the Moroccan central bank, assumed the responsibility as home supervisor for all of BoA’s operations. Banks can be part of banking groups, financial conglomerates and mixed conglomer- ates. While banking groups focus on (mostly commercial) banking business across their different subsidiaries, financial conglomerates include non-banking financial subsidiar- ies, such as insurance companies or brokerage firms. Mixed conglomerates also com- 26 Chapter 1: Recent Trends in Cross-Border Banking in Africa ccccc prise non-financial subsidiaries. Conglomerates raise additional challenges for supervisors, as they might be subject to different regulatory frameworks and supervisory authorities. This is often even more striking in the case of mixed conglomerates, where the non-financial part of the company is not subject to specific regulation or supervision. a. As defined by the Basel Concordat (BIS, 1983). A third way for a financial institution to establish a presence outside its home country is by joint venture. Joint ventures are separate legal entities with two or more owners of which at least one is a bank. The owners may or may not be established in the country where the joint venture is established, and there may or may not be one owner with a majority interest. b. For example, it is possible for a minority shareholder owning 30 percent of shares to control a bank if seven other shareholders own 10 percent each. c. Note that in some African countries, banking regulation fails to provide a solid basis for conferring the mandate to supervise a banking group or holding company on a consolidated basis to a particular supervisory authority. To ensure legal certainty and effective supervision of cross-border banks, it is vital to close these regulatory gaps. See discussion in Chapter 4. 1.1. Financial Sector Development in Africa Recent developments in cross-border banking in Africa need to be understood in the context of vast variations in financial sophistication and depth across countries. While countries such as South Africa and Mauritius have fairly developed banking systems and a private credit-to-GDP ratio – a common measure of financial depth – of more than 100 percent in 2012, according to the World Bank’s World Development Indicators, most African coun- tries are characterized by shallow banking systems that often offer little more than basic banking services despite marked improvements over the last decade (Beck and Cull, 2013). A comparison of the median low and lower-middle income countries in Sub-Saharan Africa and outside of Sub-Saharan Africa suggests that the shallowness of financial systems in many African countries is more than just a function of income. While outside of Sub-Saharan Africa the median private credit-to-GDP ratio was 34 percent in 2011, it was only 18 percent in Sub-Saharan Africa. Low levels of intermediation across Africa go hand in hand with high levels of liquidity and reliance mostly on local and thus stable deposit funding, rather than on wholesale and/ or cross-border bank flows. This concerns both domestic but also international banks in Africa. Though most banks have very stable funding, their assets tend to be concentrated on the short end of the yield curve, both with regard to their lending and their investments, which are predominantly in government securities. Shallow financial markets in Africa are often explained by the simultaneous presence of four adverse characteristics of African economies and societies (Honohan and Beck, 2007; Beck and Cull, 2013). First, the small size of many economies does not allow financial ser- vice providers to reap the benefits of scale economies. This results in customers that require smaller transactions, including low and middle-class households and smaller enterprises, Chapter 1: Recent Trends in Cross-Border Banking in Africa 27 being excluded from formal financial services. The highly dispersed populations in many African countries often further limit the effective market size as financial service provision outside urban centers is often not cost-effective under traditional banking business models. Second, a large proportion of economic agents operate in the informal sector and do not have the formal documentation necessary for financial transactions. This increases the costs and risks for financial institutions and, again, excludes large segments of the popu- lation from formal financial services. Third, volatility – both at the individual level, related to fluctuations in the income streams of many microenterprises and households, and at the aggregate level, related to the dependence of many African economies on commodity exports – further increases costs and risks for financial service providers. Finally, gover- nance problems continue to plague many private and government institutions throughout the continent and undermine not only the market-based provision of financial services, but also reform attempts and government interventions aimed at addressing market failures. Aside from shallowness, the financial systems in most African countries are also char- acterized by relatively low intermediation ratios and high cost of financial services. Sample comparison between 307 banks from low and lower-middle income countries in Africa and 720 banks from non-African developing countries shows that African banks are compara- tively well capitalized, over-liquid, and provide only limited lending to the real economy (Beck and Cull, 2013). Moreover, the comparison shows that overheads for this sample of African banks are comparatively high and one of the major drivers of the generally large interest rate spreads on the continent. The shallowness of African financial markets goes hand in hand with fragmented and deficient financial infrastructure, including payments, clearing and settlement, and credit registry systems, all of which contribute to high overheads, which are further augmented by the absence of an efficient contractual framework. The deficient contractual framework severely compromises the ability of financial institutions to enforce contracts and thereby curtails the financial system’s contribution to corporate restructuring and to more efficient allocation of capital in the economy. Few countries in Africa have effective credit information systems, be they private credit bureaus or public credit registries. Overall, payment services, particularly at the retail level, remain fragmented and costly. While shallow, African banking systems are mostly quite stable, thanks to regulatory upgrades over the past two decades in most African countries and the high capitalization and liquidity levels of banks. Aside from a few smaller hidden pockets of fragility, the last 15 years have only seen one systemic banking crisis on the continent, in Nigeria.3 3: The Nigerian crisis of 2009 was largely home-made. Following a hike in minimum capital in 2005 and a significant consolidation among banks in Nigeria there was dramatic growth in the banking sector. Among the factors fuelling rapid expansion in banking assets were margin loans provided by banks to their clients, allowing them to buy bank shares on credit. This fueled an overvaluation of the Nigerian stock market. When the stock market fell, clients defaulted en masse on their margin loans and non-performing loans spiked. 28 Chapter 1: Recent Trends in Cross-Border Banking in Africa 1.2 The Expansion of Cross-Border Banks The presence of foreign banks in countries across Africa has increased significantly over the past two decades. Over the period 1995 to 2009, the number of cross-border bank branches or subsidiaries in Africa almost doubled from 120 to 227 (Figure 1.1). At the same time, the total number of banks stayed virtually the same (421 to 442), resulting in a rise of the share of foreign banks from 29 percent to 51 percent. If instead of the share of foreign banks among the entire bank population in Africa, the average share of foreign banks across countries is considered, an increase from 39 percent in 1995 to 55 percent in 2009 can be observed (Figure 1.2, Panel A). Another way to gauge Figure 1.1: Number and Share of Foreign Banks in Africa, 1995-2009 Panel A: Number and Share of Foreign Banks 250 70% 60% 200 50% 150 40% 100 30% 20% 50 10% 0 0% 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Number of foreign banks (left axis) Share of foreign banks (right axis) Panel B: Total Number of Domestic and Foreign Banks 600 500 400 Number of banks 300 200 100 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 domestic banks foreign banks Source: Claessens and van Horen (2014) Bank Ownership Database. Notes: The total number of foreign and domestic banks is the total number of banks in each country over all countries in Africa. A cross-border bank is considered a domestic bank in its home country but a foreign bank in each country in which it operates a subsidiary or branch. Chapter 1: Recent Trends in Cross-Border Banking in Africa 29 the significance of foreign bank presence is in terms of assets owned by foreign banks relative to total assets in the banking sector. Readily available data on this measure is very limited and only available since 2004 for a larger number of developing countries, and with a significant time-lag. As of 2009, foreign-owned banks hold on average just over half of total banking sector assets in African countries (Figure 1.2, Panel B). Figure 1.2: Foreign Bank Ownership by Region, 1995-2009 Panel A: Average Share of Foreign Banks 70 60 50 40 Percent 30 20 10 0 1995 1997 1999 2001 2003 2005 2007 2009 ECA AFR LAC OECD EAP Panel B: Average Asset Share of Foreign Banks 70 60 50 40 Percent 30 20 10 0 2004 2005 2006 2007 2008 2009 ECA AFR LAC OECD EAP Source: Claessens and van Horen (2014) Bank Ownership Database. Notes: Regional groupings: EAP = East Asia and Pacific, ECA = Eastern Europe and Central Asia, LAC = Latin America and the Caribbean, OECD = Organization for Economic Co-operation and Development, AFR = Africa. 30 Chapter 1: Recent Trends in Cross-Border Banking in Africa In line with global trends of ever closer regional and international economic integration and the globalization of financial services, the presence of foreign banks has increased not only in Africa but in all world regions in the past two decades. Countries in Africa have traditionally had a relatively high share of foreign banks on average (measured both by number of incorporated banks and share of banking assets; see Figure 1.2), compared to other regions of the world, and today their average foreign bank share is second only to countries in the Eastern Europe and Central Asia (ECA) region, which has experienced an unprecedented entry of foreign banks following post-Cold War financial liberalization and integration with the European Union. The only other world region with a comparable foreign bank presence is Latin America and the Caribbean (LAC), with an average foreign bank share of 47 percent. Both countries in the East Asia and Pacific region and the OECD remain below the 30 percent mark for average foreign bank share, and have an even lower share when considering the average share of assets owned by foreign banks. The global perspective, however, masks stark differences across countries within Africa. The map in Figure 1.3 provides an overview of foreign bank ownership as measured by the percentage of total host country banking sector assets held by foreign financial groups. At one extreme, countries such as Ethiopia and Eritrea are still completely closed to foreign capital in the banking system.4 At the other extreme, some smaller financial systems, including Benin, Burkina Faso, Lesotho, Madagascar, Mozambique, and Zambia, are almost completely domi- nated by foreign banks. In between these extremes are a number of countries with a very strong foreign bank presence controlling 60 to 80 percent of total banking sector assets, including Botswana, Chad, Côte d’Ivoire, Guinea-Bissau, Mali, Mauritania, Namibia, Niger, and Senegal. The tendency of smaller financial systems to be domi- nated by foreign banks appears to be independent of income levels, as the examples of Namibia and Madagascar show. Having some of the most developed sectors on the continent and being home to some major African cross-border banks, Morocco, Nigeria, South Africa and Kenya demonstrate a comparatively low share of foreign ownership in their banking systems, in the range of 20 to 35 percent. Part of the variation of foreign bank presence can be explained by historical and country-specific circumstances. In Nigeria, Morocco and Kenya, for example, financial sector reforms at independence resulted in the (partial) nationalization of foreign banks, establishment of state-owned banks, and the growth of local banks due to low entry requirements (see Box 1.4). It is 4: Ethiopia is Africa’s second most populous country and has one of the fastest growing economies on the continent, and despite the closed nature of the Ethiopian financial system, a number of foreign banks have opened representative offices in Ethiopia in recent years. German Commerzbank was the first foreign bank to open such an office in 2007 and in 2013 a number of African banks, including Ecobank, South African Standard Bank and some Kenyan banks, have established – or have announced their intention to establish – representative offices in Addis Ababa. Chapter 1: Recent Trends in Cross-Border Banking in Africa 31 Figure 1.3: Share of Assets Owned by Foreign Banks in Africa, 2011 Share of Assets (%) 0–20 20–40 40–60 60–80 80–100 Sources: Central Bank websites, annual reports of banking groups, Claessens and van Horen (2014) Bank Ownership Data- base, World Bank/IMF country reports, GIZ (2012a, 2012b, 2012c, 2012d). Notes: No data available for countries marked grey. Foreign-owned bank refers to majority foreign-owned banks or to banks that are controlled by a foreign minority shareholder. The reference year is 2011; where 2011 data was not available, figures from 2009-2012 were used instead. also worth noting that two countries in Africa, Mauritius and the Seychelles, have estab- lished themselves as offshore financial centers. Based on their geographical origin, foreign banks in Africa can be grouped into two categories. The first group consists of international banks from outside Africa, particularly Europe, but also from emerging markets such as India and China, which have recently strengthened their presence in the region. The increasing presence of banks from emerging markets is noteworthy for contributing to the rise of South-South banks in Africa. The second group consists of African cross-border banks incorporated in jurisdictions on the African continent, predominantly South Africa, Nigeria, Morocco and 32 Chapter 1: Recent Trends in Cross-Border Banking in Africa Kenya. Note that this is a simplified classification of the bank population, since both the location of a bank’s headquarter and majority shareholding may have changed in the course of time, as shown in Tables 1.1, 1.2, and 1.3. From a regulatory perspective, the relevant criterion for determining a bank’s home supervisor under applicable banking regulations is generally the ownership structure. Consolidated supervision is generally exercised by the supervisory authority of the country that is home to the shareholders who hold a controlling stake in the banking group. Yet even here, the classification is not always clear-cut and sometimes requires judgment calls (Box 1.2). Box 1.2: Is Barclays Africa Group South African or British? Barclays’ operations in Africa illustrate the fact that the home country of a bank cannot always be determined in a straightforward manner. Since purchasing a majority stake in the South African Absa Group in May 2005, Barclays Bank of the United Kingdom (UK) has managed Absa’s African operations alongside its own activities in Africa using the Barclays branding. In mid-2013, Barclays formed Barclays Africa Group by consolidating Absa Group’s and Barclays’ African operations under a single management structure “to create a leading pan-African financial services business,” and took a 62.3 percent stake in the new group (Barclays Bank PLC, 2012).a This was in recognition of the need to strengthen collaboration between Bar- clays and Absa, which had continued to function largely as separate entities across Africa, to strengthen the strategic focus of the combined banking group. If based only on the ownership structure, the operations of Barclays Africa Group on the continent would be attributable to the UK as home country, since it is the majority shareholder. However, the registered head office of Barclays Africa Group is located in Johannesburg and the group is incorporated in South Africa. So even though Barclays Africa Group is majority-owned by Barclays of the UK, it is supervised on a sub-consolidated basis by the South African Reserve Bank (SARB); i.e., the South African supervisor oversees all of Barclays Africa Group’s operations on the continent on a consolidated basis. According to the definition used in this book, Barclays Africa Group is therefore classified as an African-based entity with South Africa as home country. At the same time, Barclays Africa Group is included in the overall consolidated supervision of the Barclays Group by the UK supervisor. a: Barclays Bank continues to have separate operations in Egypt and Zimbabwe. While not part of Barclays Africa Group, they are an integral part of Barclays’ African business and are managed by Barclays Africa Group. The group of international banks comprises, in particular, British and French banks, including Barclays (though Barclays Africa Group is considered African by the definitions used in this report; see Box 1.2), Standard Chartered, BNP Paribas and Société Générale, which have traditionally had a strong regional presence in their former colonies, but have in recent years also expanded beyond their colonial roots. With less regional reach but strong Chapter 1: Recent Trends in Cross-Border Banking in Africa 33 presence in their former colonies, Portuguese banks also fall into this category. US-based Citigroup has established a regional presence in Africa by opening subsidiaries and branches in a large number of countries over the past decades. These international banks continue to have a large footprint in Africa (see Figure 1.4) and are of systemic importance in a number of jurisdictions. For example, before Barclays’ African operations became Africa-based earlier this year, British financial institutions managed more than 30 percent of deposits in Botswana, Mauritius, South Africa, and Zambia, while affiliates of Portuguese banks manage about two-thirds of deposits in Angola and Mozambique and 90 percent in Cape Verde (IMF, 2012). While most of these institutions can look back on decades of oper- ating in Africa, there are also more recent European entrants, such as the Dutch Rabobank (Box 1.3).5 Box 1.3: A Different Kind of International Bank in Africa: The Operations of Rabobank The Netherlands-based Rabobank entered the African market in the mid-2000s through its Rabobank Development division by acquiring a 49 percent stake in Tanzania’s National Micro Finance Bank in 2005 (later dropping to 35 percent). It subsequently acquired equity shares in four more banks in other African countries: Mozambique (49 percent), Rwanda (35 percent), Uganda (27.5 percent), and Zambia (46 percent).a Rabobank’s operations in Africa differ substantially from those of its international peers, which tend to serve the corporate blue-chip market. As a European bank with a long tradition of focusing on agricultural finance, the stated mission of its Rabobank Development division is to invest in established banks in Africa (with the exception of Mozambique, where Rabo Development entered via greenfield investment), with a view to improving access to financial services for underbanked segments of the market and deepening the engagement of African banks in the agriculture sector. Rabobank’s explicit approach is to acquire controlling minority stakes that allow for management control rather than absolute majority stakes, to ensure that these banks “benefit from [its] insights while remaining domestically owned.” a: All equity shares as of September 2013 (Rabobank, 2013) Source: https://www.Rabobank.com/en/rabo_development/about_us/about_us.html. 5: There are also several microfinance institutions (MFIs) that have entered a number of African countries with greenfield investments targeting the low-end of the market, only some of which have bank licenses and are thus captured in our list of cross-border banks. Most of these MFIs are organized in the form of holding companies with little financial connections across countries (see Earne et al., 2014). 34 Chapter 1: Recent Trends in Cross-Border Banking in Africa Figure 1.4: Geographical Footprint of Major International Banks in Africa, 2013 Societé Générale Citigroup Standard Chartered BNP Paribas Source: Annual reports and websites of banks. Notes: Countries marked in dark color: presence through branch or subsidiary. Light color: representative office. Only includes international banks that have a presence through branch or subsidiary in 10 or more African countries. Chapter 1: Recent Trends in Cross-Border Banking in Africa 35 Table 1.1: List of Major International Cross-Border Banks in Africa No. Name Location Majority owner- Number of of headquarters ship/largest African minority countries shareholder 1 Société Générale France France 17 2 Citigroup USA USA 15 3 Standard Chartered UK UK 14 4 BNP Paribas France France 13 5 Bank of Baroda India India 9 6 Access Holding Germany Unknown 5 7 Albaraka Bank Bahrain Bahrain 5 (Group) 8 HBL Pakistan Pakistan Tanzania 5 (Habib Bank Ltd.) 9 International Switzerland Malaysia 5 Commercial Bank (ICB) 10 Rabobank Netherlands Netherlands 5 Notes: Number of countries includes home country (if African) and representation through subsidiaries or branches in African countries; representative offices are not included in the numbers. Rows in darker green indicate banks considered to be pan-African banks, defined here as present in at least 10 African countries. 36 Chapter 1: Recent Trends in Cross-Border Banking in Africa Box 1.4: A Brief History of Banking in Africaa At the time of independence, most African countries had banking systems dominated by foreign-owned banks. In most countries, foreign-owned banks were the only commercial banks. Other banks, if they existed, typically provided a negligible share of total lending. Established during the colonial period, foreign banks mainly provided trade finance and short-term working capital to foreign companies and served the non-African resident community. British banks dominated in the British colonies, while French banks did so in the French and Portuguese banks in the Portuguese colonies. This was partly for political reasons, but partly also due to a deliberate outcome negotiated by banks to preserve their market share and avoid competition in order to extract rents. For example, in West Africa, the Bank of British West Africa (BBWA) entered into an agreement with Banque de l’Afrique Occidentale (BOA) as far back as 1913 to confine each other’s presence to its respective colonial territory and not open branches in the other’s territory (Austin and Ugochukwu Uche, 2007). In the 1960s and 1970s, the governments of the newly independent states undertook a first series of financial sector reforms. Governments felt that the foreign banks did not serve the needs and development goals of the new states. Banks were criticized for discriminating against Africans and African-owned businesses and lending almost exclusively to foreign companies for the purpose of trade finance and other short-term purposes. Thus there was political pressure for governments to intervene in the financial sector. While the extent of intervention varied from country to country, the overall goal was to rectify the perceived financing gaps by issuing directives to reduce the cost of credit and creating development institutions to provide longer-term finance. Most governments established state-owned banks in order to serve those parts of the economy that had been denied loans by the foreign banks. Locally owned private banks also entered the market in some countries, especially Kenya and Nigeria, where entry requirements were low and banks saw an opportunity to compete on service or to make high returns, such as on foreign exchange trading or insider lending. Throughout this period, foreign banks mostly maintained their presence – sometimes with governments taking minority shares, such as in Uganda and Ghana, or majority shares, such as in Nigeria and Malawi – although their relative importance decreased due to the emergence of new state-owned and private local banks. By the 1980s, many African economies were experiencing economic crises due to a combination of external shocks and failing domestic policies. In the financial sector, the interventionist policies came at considerable cost and provided little, if any, benefit. With the exception of Ethiopia, countries with the most extensive government interventions suffered the most damage. Many state-owned banks became insolvent due to political pressure to make unsustainable loans, but also because of lack of technical expertise and Chapter 1: Recent Trends in Cross-Border Banking in Africa 37 poor management. In several state-owned banks across different countries, non-perform- ing loans accounted for up to 80 percent of the loan portfolio. However, not all state-owned banks failed. In countries where government intervention was less heavy handed, resulting in relatively less distorted credit markets, as in Kenya, Ethiopia and Zimbabwe, state-owned banks remained functional. This was often because those banks had private sector par- ticipation or management provided by a foreign bank. As part of a more general economic liberalization agenda imposed on countries as part of the structural adjustment process in the 1990s and early 2000s, the banking sectors in many Africa countries underwent financial liberalization. The objective was to restore credit allocation based on commercial criteria, relying on the credit risk assessment skills of the private sector, and return to market-deter- mined interest rates. The liberalization of the financial sector in turn provided opportunities for entry and expansion of banks across the continent. While many African countries experienced a significant increase in foreign bank partici- pation during this period, Kenya, Nigeria and South Africa saw opposite trends. In Kenya, the relatively prudent management and continuing strong market position of Kenya Commercial Bank, now majority privately owned but still state controlled, coupled with vibrant compe- tition from the private sector, spurred innovation and, in the 2000s, provided a platform for product development and cross-border expansion in East Africa. In Nigeria, at the time of the consolidation caused by the minimum capital hike in 2005, investment by foreign banks was initially discouraged, and when banks were restructured in 2009, foreign banks could only be tempted to make limited investments. In South Africa, the presence of foreign banks declined in the late 1980s due to mounting pressure to disinvest from a country governed by an apartheid regime. Sector volatility in the early 1990s created scope for consolidation through the mergers of several banks to form Absa. The introduction of the Banks Act (1990) led to an industry growth spurt, with a number of new banking licenses being issued, and by the end of 2001 the number of registered banks totaled 43. In early 2002, several banks were placed under curatorship, which resulted in a run on a number of smaller banks. As a result, South Africa today has a highly concentrated banking system with four dominant banks. a. Unless otherwise noted, this box draws especially on the work of Brownbridge and Harvey (1998). Over the past decade, however, the landscape of banking in Africa has changed sub- stantially through significant South-South investment, largely from inside but also from outside Africa, resulting in the diminishing relative importance of traditional foreign banks. As countries throughout Africa entered a phase of liberalization starting in the late 1980s and early 1990s – a reversal of interventionist financial sector policies pursued across the continent by the newly independent countries (see Box 1.4) – new players have entered the market or reinforced their presence. These new players aim to reap the opportunities of an economic growth performance that has steadily outpaced 38 Chapter 1: Recent Trends in Cross-Border Banking in Africa global growth since the early 2000s, including during the global financial crisis; as well as the opportunities of a growing African middle class and, in at least some cases, the large unbanked population.6 From outside of Africa, these non-traditional entrants include banks from India, China, Pakistan, Bahrain and Jordan, among other countries (see Annex 1.1 for a complete list). After a first wave of expansion in the mid-20th century, mainly to serve the Indian trading community, one can currently observe a second wave of expansion of Indian banks in Africa. The Bank of Baroda has the largest network and is currently present in Botswana, Ghana, Kenya, Mauritius, Seychelles, South Africa, Tanzania, Uganda and Zambia (joint venture). In addition, financial institutions such as India’s largest private sector lender, ICICI Bank, and Canara Bank plan to set up operations on the continent. Like other international banks they are eyeing Mauritius as a gateway for their operations in Africa. The Bank of China was the first Chinese bank to enter Africa, opening a subsidiary in Zambia in 1997 to serve Chinese companies operating in Zambia. Subsequently, the Industrial and Commercial Bank of China (ICBC) acquired a strategic stake of 20 percent in South African Standard Bank, the largest African lender by assets and earnings, in 2007. Following this first large-scale investment of a Chinese bank in Africa, which was widely regarded as a success, the Bank of China entered into a strategic partnership with Nedbank Group in 2013 to finance and grow Sino-African business. Given that China has become South Africa’s and the continent’s largest trading partner, in terms of both exports and imports (IMF DOTS, 2012), further Sino-African investments and strategic partnerships can be expected in the future. The entry of non-traditional international banks, however, is dwarfed by the cross-border expansion of African banks, which have bolstered their physical and economic footprint on the African continent in an unprecedented way over the past decade. Ecobank tripled its affiliate network in Africa between 2000 and 2013 from 11 to 32 countries7; Nigeria’s United Bank for Africa increased its footprint from 1 to 19 countries; Morocco’s Attijariwafa Bank increased its footprint from 1 to 12 countries after acquiring the African interests of the French Crédit Agricole; and Morocco’s BMCE went from 2 to 18 countries over the same period, mostly through the strategic acquisition of Bank of Africa. While most financial insti- tutions started or accelerated their expansion in the past 5 years, Ecobank and a number of South African banks, in particular Standard Bank, started their expansion much earlier. Standard Bank was the first African bank to expand into other countries on the continent by acquiring the Africa-based operations of ANZ Grindlays, a British bank, in 1993. With this acquisition, Standard Bank – outside of South Africa known in Africa under the brand 6: According to the Global Findex database, only 23 percent of adults age 15 or older are banked in Africa (Klapper and Singer, forthcoming). 7: This enumeration refers to a this enumeration refers to a bank’s presence in its home country plus representation in other African countries through subsidiaries or branches and does not include representation through representative offices. Chapter 1: Recent Trends in Cross-Border Banking in Africa 39 Figure 1.5: Cross-Border Expansion of African Financial Groups over Time, 1990-2013 35 30 25 Number of countries 20 15 10 5 0 Ecobank United Bank for Africa Standard Bank BMCE / Bank of Africa Attijariwafa Bank Banque Centrale Populaire Access Bank Afriland First Bank Guaranty Trust Bank 1990 2000 2005 2010 2013 Sources: Annual reports and websites of banks. Notes: The graph shows representation of African cross-border banks in Africa through branches or subsidiaries, not including presence through representative offices. Barclays Africa Group is not included here since it was only created in 2013. Prior to the consolidation of Barclays’ Africa operations and South African Absa’s affiliate network, Barclays and Absa operated side by side 2005-2013, thus further complicating the representation in this figure. In 2013 Barclays Africa Group had branches or subsidiaries in 10 countries. For more information see Box 1.2: Is Barclays Africa Group South African or British?. name Stanbic – rapidly expanded its footprint across Southern and Eastern Africa to include operations in East Africa, Sudan and the former Rhodesia (now Zimbabwe). Ecobank first expanded across West Africa in the 1990s and after 2005 expanded aggressively to Central, Eastern, and Southern Africa as part of its vision to become a pan-African bank.8 Figure 1.5 shows the chronological sequence of this fast-paced expansion over the period 1990 to 2013. 8: Ecobank’s expansion post 2005 also coincided with the drastic increase in minimum capital requirements in Nigeria which, together with limited domestic opportunities, propelled many Nigerian banks to expand abroad (see discussion in section 1.3. Reasons for Cross-Border Expansion below). 40 Chapter 1: Recent Trends in Cross-Border Banking in Africa The pattern of expansion has been similar among all these groups. Originating from their home markets, banks have generally expanded first to neighboring countries, then across the region and, in some cases, even across the continent and beyond (see Figure 1.6). In each sub-region, one or two countries have emerged as dominant financial cen- Figure 1.6: Geographical Footprint of Major African Banks in Africa, 2013 Ecobank United Bank for Africa Banque Sahélo-Saharienne Attijariwafa Bank Chapter 1: Recent Trends in Cross-Border Banking in Africa 41 ters. South African banks have the strongest presence in the South African Development Community (SADC), with Standard Bank Group now covering the entire southern and east- ern part of the continent, parts of Central Africa plus Nigeria and Ghana. Moroccan-based Attijariwafa Bank and Banque Centrale Populaire have clustered their presence in West Standard Bank Banque Marocaine du Commerce Extérieur Barclays Africa Group Banque Centrale Populaire Sources: Annual reports and websites of banks. Notes: Stars indicate the headquarter of each bank. Countries marked in dark color: presence through branch or subsidiary. Light color: representative office. Only includes African-headquartered banks that have a presence through branch or subsidiary in 10 or more African countries. 42 Chapter 1: Recent Trends in Cross-Border Banking in Africa Africa, while their Moroccan competitor BMCE has expanded across West, Central and East Africa. Several Nigerian banks have in recent years expanded their presence beyond West Africa to also include Central, Eastern, and Southern Africa. Kenyan banks are con- centrated in East Africa and Libyan banks in Western and Northern Africa. Table 1.2: List of Major African Cross-Border Banks No. Name Location of Majority owner- Number of headquarter ship/largest African minority countries shareholder 1 Ecobank Togo South Africa 32 2 United Bank Nigeria Nigeria 19 for Africa (UBA) 3 Standard Bank South Africa South Africa 18 Group (Stanbic) 4 Banque Marocaine Morocco Morocco 18 du Commerce Extérieur (BMCE) 5 Banque Sahélo- Libya Libya 14 Saharienne pour l’Investissement et le Commerce (BSIC) 6 Attijariwafa Bank Morocco Morocco 12 7 Banque Centrale Morocco Morocco 11 Populaire du Maroc (BCP) 8 Barclays Africa South Africa UK 10 Group 9 Access Bank Nigeria Nigeria 9 10 Guaranty Trust Nigeria Nigeria 9 Bank Ltd. Notes: Number of countries includes home country (if African) and representation through subsidiaries or branches in African countries; representative offices are not included in the numbers. Rows in darker green indicate banks considered to be pan-African banks, defined here as present in at least 10 African countries. Chapter 1: Recent Trends in Cross-Border Banking in Africa 43 In many cases, the cross-border expansion has been accompanied by a fast-paced build-up of local networks of affiliates abroad. Several groups have multiplied branches in host countries. For example, Attijariwafa Bank increased the number of branches in Senegal alone from 40 to 87 in 2009-2010. Over the same period, BMCE’s total number of branches and points of service jumped from 12 to 26 in Burkina Faso and from 17 to 43 in Benin. All in all, these developments have led to an increasing density of cross-border ownership linkages and financial flows within the African sub-regions. In terms of their geographic footprints, eight African and four international banks are represented through branches or subsidiaries in 10 or more African jurisdictions each, and can be considered pan-African (see Table 1.1 and Table 1.2). While African cross-border banks primarily focus on the African market, three of them – Standard Bank, United Bank for Africa, and First National Bank – have also ventured beyond Africa and started to build a significant presence in Europe, the Americas and Asia. The expansion of African banks across the continent has also given them economic significance beyond their home countries. A number of African cross-border banks now hold a significant share of assets in host country banking systems and can be consid- ered systemically important (see Figure 1.7). In addition to having the largest geographical footprint of any banking group in Africa, Ecobank holds more than 10 percent of banking system assets in 13 African host countries. It especially dominates the banking systems of Liberia, the Central African Republic and Guinea, where it holds around 40 percent of total sector assets. Other banks also have a significant asset share of host banking systems across the continent. Standard Bank holds more than 10 percent of banking system assets in 9 countries, BMCE in 5 countries, Barclays Africa Group in 6 coun- tries, First National Bank in 4 countries, and Attijariwafa Bank in 3 countries. There are 9 instances where Ecobank, Standard Bank, Barclays Africa Group, BMCE, or First National Bank individually holds more than 30 percent of banking system assets in a host country. African cross-border banks have thus reached a systemic presence in a number of host countries, even if this presence needs to be judged against the shallow depth of the financial systems that characterizes small economies in Africa. 44 Chapter 1: Recent Trends in Cross-Border Banking in Africa Figure 1.7: Share of Assets in Host Country Banking Systems (%) by Selected African Banks, 2011 Panel A: Ecobank 50 45 40 35 30 25 20 15 10 5 0 Liberia Central African Republic Guinea Guinea-Buissau Togo Chad Burkina Faso Benin Niger Mali Côte d’Ivoire Senegal Rwanda Panel B: Standard Bank 50 45 40 35 30 25 20 15 10 5 0 Lesotho Namibia Uganda Botswana Malawi Swaziland Mozambique Zambia Zimbabwe Chapter 1: Recent Trends in Cross-Border Banking in Africa 45 Panel C: BMCE / Bank of Africa Panel D: Barclays Africa Group 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 Madagascar Benin Niger Djibouti Burkina Faso Seychelles Botswana Zambia Ghana Uganda Mauritius Panel E: First National Bank Panel F: Attijariwafa Bank 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 Namibia Lesotho Botswana Swaziland Senegal Cameroon Mali Sources: Central Bank websites, annual reports of banking groups, Claessens and van Horen (2014) Bank Ownership Data- base, World Bank/IMF country reports, GIZ (2012a, 2012b, 2012c, 2012d). Notes: Includes selected African cross-border banks with presence in at least three African host countries and an asset share of at least 10 percent in the host country banking system in at least three countries. The reference year is 2011; where 2011 data was not available, figures from 2009-2012 were used instead. 46 Chapter 1: Recent Trends in Cross-Border Banking in Africa Notably, not all pan-African banks identified in Table 1.2 have a significant presence abroad, as measured by asset share in the host country banking system: United Bank of Africa (Chad, Burkina Faso), Banque Sahélo-Saharienne (Central African Republic, Chad), Libyan Foreign Bank (Chad), and Banque Centrale Populaire (Central African Republic, Côte d’Ivoire) each only hold asset shares exceeding10 percent of the bank- ing system in only one or two host countries. Thus, while they have a large geographic reach, their stakes are rather thinly spread across the continent and have not yet result- ed in a strong increase in their economic significance.9 However, if the current growth trend in cross-border banking in Africa continues and African-based banks engage more in product development and innovation, their contribution to financial deepening is likely to become more significant, and the number of countries in which African banks reach systemic importance is likely to increase. The systemic importance of a number of African banks across the continent stands in stark contrast to the importance of the international banks. Among the internation- al banks identified in Table 1.1, only four banks hold more than 10 percent of banking system assets in at least three African host countries (see Figure 1.8). Société Générale and Standard Chartered each have a share of about 10 to 20 percent of banking system assets in six African countries. BNP Paribas has an asset share of more than 10 percent of the host country banking system in four countries, including Guinea, where it holds almost half of all banking system assets. Rabobank holds about 16 percent of banking sector assets through its controlling minority shareholdings in three African countries. As noted above, a small number of Portuguese banks also individually account for a sizeable share of banking system assets in Angola, Cape Verde, and Mozambique (IMF, 2012). 9: It is important to note that in the absence of better measures systemic importance is gauged by asset share. Systemic importance is not necessarily correlated in a linear manner with market share, but can also come from having a critical position within payment systems or other elements of infrastructure. Similarly, market share in terms of number of depositors rather than volume of deposits can also make an institution systemically important in socio-economic terms. Chapter 1: Recent Trends in Cross-Border Banking in Africa 47 Figure 1.8: Share of Assets in Host Country Banking Systems (%) by Selected International Banks, 2011 Panel A: Société Générale Panel B: Standard Chartered 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 Equatorial Guinea Senegal Cameroon Côte d’Ivoire Chad Burkina Faso Botswana Zambia Uganda Mauritius Ghana Zimbabwe Panel C: BNP Paribas Panel D: Rabobank 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 Guinea Madagascar Gabon Senegal Zambia Tanzania Rwanda Sources: Central Bank websites, annual reports of banking groups, Claessens and van Horen (2014) Bank Ownership Data- base, World Bank/IMF country reports, GIZ (2012a, 2012b, 2012c, 2012d). Notes: Includes selected international cross-border banks with presence in at least three host countries and an asset share of at least 10 percent in the host country banking system in at least three countries. The reference year is 2011; where 2011 data was not available, figures from 2009-2012 were used instead. 48 Chapter 1: Recent Trends in Cross-Border Banking in Africa Figure 1.9: Ownership Linkages Among African Banks MAR LBY MLI NER SEN TCD GNB BFA GIN BEN TGO NGA CIV GHA SDN LBR CAF CMR GNQ UGA KEN GAB COG COD RWA SYC MWI MOZ ZMB ZWE MDG MUS NAM BWA SWZ LSO Size of national ZAF banking system USD share of assets in host country banking system Ecobank only Sources: Central Bank websites, annual reports of banking groups, Claessens and van Horen (2014) Bank Ownership Database, World Bank/IMF country reports, GIZ (2012a, 2012b, 2012c, 2012d). Notes: The graph only shows ownership linkages between countries if the share of assets held by home countries constitutes at least 10 percent of the host country’s banking system. The size of the bubbles is in proportion to the absolute size of each country’s banking sector. The reference year is 2011; where 2011 data was not available, figures from 2009-2012 were used instead. The substantial cross-border ownership linkages in African banking sectors are illustrated by network graphs from the perspective of the host countries, both from within Africa (Figure 1.9) and from outside of Africa (Figure 1.10). The size of the country bubbles reflects the absolute size of the respective banking sector, with South Africa, Chapter 1: Recent Trends in Cross-Border Banking in Africa 49 Figure 1.10: Ownership Linkages of International Banks in Africa NLD BEL FRA GBR LUX PRT TUN USA NER SEN TCD GNB DJI GIN TGO CIV GHA GNQ UGA STP RWA COD SYC TZA AGO ZMB MOZ ZWE MDG BWA Size of national banking system USD share of assets in host country banking system Sources: Central Bank websites, annual reports of banking groups, Claessens and van Horen (2014) Bank Ownership Database, World Bank/IMF country reports, GIZ (2012a, 2012b, 2012c, 2012d). Notes: The graph only shows ownership linkages between countries if the share of assets held by home countries constitutes at least 10 percent of the host country’s banking system. The size of the bubbles is in proportion to the absolute size of each country’s banking sector. The reference year is 2011; where 2011 data was not available, figures from 2009-2012 were used instead. Nigeria, Morocco, Kenya, Mauritius and Libya visibly being the largest financial centers in Africa. The links among countries reflect the share of the host jurisdiction’s banking sector owned by banks in home countries. The link between South Africa and Lesotho, for example, is very strong because South African banks account for virtually 100 percent 50 Chapter 1: Recent Trends in Cross-Border Banking in Africa of banking system assets in Lesotho. Among African banks (Figure 1.9), South African-owned banks account for the largest number of cross-border linkages, in part because Ecobank’s operations are attributed to South Africa because a South African investment fund (Public Investment Corporation) is the largest minority shareholder in the group’s holding.10 Figure 1.10 illustrates the considerable but rather thinly spread presence of European and American banks across African jurisdictions. 1.3. Reasons for Cross-Border Expansion The cross-border expansion of African banks is due to many factors, but the pursuit of business opportunities abroad, normally led by the banks’ larger corporate clients, is generally the dominant driver. Economists distinguish between pull factors and push factors when assessing the drivers for cross-border business expansion. Push factors are circumstances in the home country that explain why banks decide to move beyond the borders of their home countries. Chief among them are declining opportunities in the home jurisdiction and regulatory requirements. Pull factors, in contrast, are oppor- tunities in host countries that make it attractive for a bank to enter the new market. In other words, pull factors are the expected benefits that banks hope to reap by venturing into a particular foreign market. One of the most powerful push factors propelling banks to expand beyond their home markets are (perceived) declining or smaller profit opportunities in the home economy, especially relative to opportunities in potential host markets. This is also an important factor for African banks. For example, the end of apartheid in South Africa provided the impetus for cross-border expansion by South African banks by opening up investment opportunities that had been precluded and providing opportunities to leverage the depth and capacity of the South African market. More recently, a similar process has developed in East Africa, where innovation and greater depth of the Kenyan market has provided Kenyan banks with the impetus to replicate their innovative business models across neighboring countries. As assessments of profit opportunities are based on a comparison of a bank’s home market with potential host countries, relatively larger profit margins abroad can also be seen as one of the most powerful pull factors for expansion. Another important factor that has pushed African banks into new markets is regu- latory change in the home country. In Nigeria, for example, the central bank drastically increased the minimum capital requirement from Naira 2 billion (around USD 14 million) 10: The ownership share of South African ownership in Ecobank Transnational International will be greatly enhanced should Nedbank choose to exercise an equity conversion option for up to 20 percent of ETI’s share capital to be exercised during the 12 month period ending November, 2014. Chapter 1: Recent Trends in Cross-Border Banking in Africa 51 at end 2004 to Naira 25 billion (around USD 180 million) at end 2005, with the aim of prompting consolidation and transforming the banking system from one dominated by many small and relatively unstable banks to one with a much smaller number of larger and more steadfast lenders. A wave of mergers and acquisitions reduced the number of licensed commercial banks from 89 to 25 within a year. The remaining banks raised large amounts of new capital, with some achieving capital levels of over Naira 100 billion, exceeding the decreed minimum level by a factor of four. The central bank encouraged this development by promising to make banks that accumulated more than Naira 100 billion in equity eligible to manage Nigeria’s international reserves. The Nigerian banks deployed their capital to fund an explosive growth in the banks’ loan portfolios (Berg et al., 2012). The search for yield due to the large amount of “excess capital” available in the domestic banking system was also the determining push factor driving the aggressive expansion of Nigerian banks across the region (see Figure 1.5). In Kenya, the authorities introduced a similar, albeit more gradual and considerably milder regulatory measure raising the minimum core capital requirement from Kenya Shilling 250 million (USD 4 million) in 2008 to Kenya Shilling 1 billion (USD 12 million) in 2012. While this much more measured rise in minimum capital (compared to Nigeria) may to some extent have contributed to the expansion of several Kenyan banks across East Africa, the impact here may well have been more to discourage expansion by banks domiciled in the smaller financial systems in neighboring countries in East Africa from establishing themselves on the Kenyan market. The ambition to become a leading pan-African bank has influenced the strategic plans of some African banks and fuelled their expansion on the continent. Ecobank was founded in 1985 with the vision to be a regional bank in West Africa. Partly for that reason, it was set up as a bank holding company with the status of an international organization (giving it the status as a non-resident financial institution), which was deemed necessary for it to operate as a regional institution rather than a bank in Togo. In the 2000s Ecobank expanded its vision to becoming a pan-African bank. This ambi- tion was shared by a number of Nigerian banks, including Access Bank and UBA, which believed that their expansion across Africa would afford them a competitive advantage (Lukonga and Chung, 2010). It is worth noting that Access Bank has since reversed course and announced to investors and analysts in early 2013 that it plans to dilute or divest its largely unprofitable holdings in six of the nine countries in which it has banking subsidiaries (see Box 1.5). Given their relatively competitive and well-developed home base, limited domestic opportunities for further growth and declining profit margins, as well as an encouraging, positive engagement by Bank Al-Maghrib, Moroccan banks have expanded predominantly into French-speaking West and Central African countries that combine attractive business prospects and a high level of cultural proximity. 52 Chapter 1: Recent Trends in Cross-Border Banking in Africa Box 1.5: A Counterpoint to the Expansion of African Cross-Border Banks: Access Bank Access Bank provides an interesting counterpoint to the expansion of African cross-border banks in recent years. It first expanded in French-speaking Africa and Zambia following Nigeria’s banking sector consolidation in 2005, with the ambition of becoming a pan- African bank. However, it recently announced its intention to disinvest from a number of countries and consolidate its cross-border operations in Africa, as not all subsidiaries were profitable, and it became more difficult to justify maintaining full-blown subsidiaries in small economies with low levels of intermediation – particularly Sierra Leone, Gambia and Burundi. In Côte d’Ivoire, the bank met difficulties which it attributed to cultural differences. In Zambia, the bank reacted to the hike in minimum capital requirements for foreign-owned banks. In Ghana, one of its core markets, the bank is expanding in part due to a sizeable subsidiary inherited as part of Access bank’s post-crisis merger with Intercontinental Bank (2011). Currently, foreign subsidiaries contribute an 11 percent share to Access Bank’s assets and earnings. The expected re-focus on core markets – Ghana and Rwanda – is expected to increase this share to 15 to 20 percent. Source: Interview with bank officials and Access Bank, 2013. Among pull factors, and probably the most frequently cited reason for banks to expand across borders, is the need to follow their clients abroad. Following clients is how the first foreign banks were established during colonial times and explains why British, French, and Portuguese banks got established in their respective former colonies. Following their client base remains the most significant factor for African banks. For example, when South African Standard Bank acquired the operations of ANZ Grindlays in 1993, it was primarily to serve its South African corporate customers trading in the rest of Africa (Brownbridge and Harvey, 1998). Although intra-regional trade in Africa is still estimated at a modest 10 to 12 percent of the continent’s total trade, regional economic communities such as the Common Market for Eastern and Southern Africa (COMESA), the East African Community (EAC) and the Southern African Development Community (SADC) are facilitating closer (sub)regional economic integration; and following the client is an increasingly important reasons for banks to expand into new markets. For example, in recent years the advantages for Nedbank in providing Ecobank with a $280 million convertible three-year loan in 2011 took the form of a major enhancement of Nedbank’s ability to service its clients across the African continent.11 11: Ecobank used the loan to fund a considerable expansion of its retail network in Nigeria through the purchase of Oceanic bank from the Nigerian Asset Management Company (AMCON). Chapter 1: Recent Trends in Cross-Border Banking in Africa 53 The same logic applies to international banks entering and expanding across the continent. While Europe and the United States continue to be important trading part- ners, their share of African exports declined from 47 to 33 percent and from 17 to 10 percent, respectively, between 2000 and 2011. Emerging economies’ trade with Africa jumped from 8 to 22 percent over the same period, with China alone increasing its share from 3.5 to 13 percent in that period and becoming the single largest country for African exports (AfDB, OECD, UNDP, and UNECA, 2013). While there are, of course, important country-by-country variations, countries across all sub-groups in Sub-Saharan Africa (oil exporters, low-income countries, middle-income countries) exported a smaller share of their products to developed countries and a higher share to China in 2010 than they did in 1990. There has also been a reorientation towards trade with India and, to some degree, Brazil, though this shift is more heterogeneous across country groups and mostly concentrated in oil exporting and resource-rich countries (IMF, 2011). These shifts in the composition of trade have been particularly important in attracting Chinese and Indian banks to Africa. Banks also look to expand across the continent to diversify their risk. Business cycles across Africa are not synchronized, and by expanding into countries with different eco- nomic profiles, bank can minimize their exposure to risks due to business cycles. More generally, banking business across Africa has become increasingly attractive since Africa’s economic development started taking off in the early 2000s. Improving business climates, including a more stable macroeconomic environment, a growing middle class, and large unbanked populations are luring both foreign and local investors to the most promising growth markets. The increasing importance of natural resources in, for example, Angola and Mozambique, has also contributed to attracting cross- border banks to these countries, both to follow their clients and to benefit from improving economies in both countries. There are also increasing opportunities to finance infra- structure projects undertaken by local enterprises in sectors such as energy, water supply and treatment, and transportation. At the same time, relatively slower growth in home markets and improvements in local macroeconomic management, leading to less attractive yields on government securities and the crowding in of bank lending, have pushed banks abroad. Importantly, the expansion of cross-border banking in Africa over the past decade also needs to be understood against the backdrop of financial liberalization in the late 1980s and early 1990s which reversed interventionist financial sector policies enacted at independence in most African states and created the space in which banks could expand across borders. Given the start-up costs associated with greenfield investments, foreign banks have often found it advantageous to enter markets by taking over the restruc- tured operations of former state-owned banks. In particular, this allows banks to make use of an existing branch network and retail market presence that would be difficult 54 Chapter 1: Recent Trends in Cross-Border Banking in Africa to build up otherwise in a short timespan. Particularly given the relatively small size of intra-regional trade within Africa, this also allows for the entry process to be much more closely associated with market development and deepening strategies than in those cases where banks base their business model on traditional pull factors, such as following their clients’ export/import business. An example of such a successful market acquisition strategy was Standard Bank of South Africa’s purchase of Uganda Commer- cial Bank in 2001. More recently, Netherlands-based Rabobank (see Box 1.3) acquired controlling minority stakes in restructured state-owned banks, such as Tanzania’s National Micro Finance Bank (2005), Zambia’s Zanaco Bank (2007) and Rwanda’s Union de Banques Populaires (2008). In 2012, Rabobank also purchased a controlling minority stake in Uganda’s DFCU Bank, a former development bank. Finally, the global financial crisis of 2008 may have affected the expansion plans of both African and international banks. On the one hand, the European banks active in Africa were among the least affected by the global financial crisis; on the other hand, they may have become more reluctant to undertake further expansion. This has given an opening to both non-African emerging market and African banks to take a larger share in the growing financial systems across the continent. However, the crisis also affected the expansion plans of some Nigerian banks, such as United Bank of Africa and Access Bank, which have scaled down their expansion plans in Africa due to the crisis, while Standard Bank, instead of investing outside the continent, reoriented its expansion plans to within Africa (Lukonga and Chung, 2010). Anecdotal evidence suggests, however, that the strategic focus of African cross-border banks on Africa is mostly driven by their assessment of the business opportunities in Africa in comparison to other regions. 1.4. Business Models of Cross-Border Banks The business models that cross-border banks pursue in Africa vary according to the rea- sons for their expansion across Africa. As banks enter new markets, they face decisions about (a) whether to establish a branch or subsidiary; (b) how closely to integrate their affiliates’ operations with the parent; (c) whether to enter by merger and acquisition or greenfield investment; and (d) what market segment to serve and strategy to pursue. Branches vs. subsidiaries Cross-border banks can enter host countries by establishing branches or subsidiaries. In Africa, the expansion of cross-border banks has predominantly taken the form of sub- sidiaries. While 21 out of 48 African countries – at least in theory – allow foreign bank entry by establishing a branch, according to data from the World Bank Banking Regu- lation and Supervision Database (2012) and the Regulatory Framework Database (AfDB Chapter 1: Recent Trends in Cross-Border Banking in Africa 55 2010),12 de-facto entry by branch can only be verified in 10 countries. Entry by branch is most common in the countries with some of the most developed financial sectors on the continent, such as South Africa and the off-shore financial centers Mauritius and the Seychelles. The decision to enter by branch or subsidiary is to a large extent driven by regulatory requirements in Africa, while other considerations, such as differences in tax rates, seem secondary. However, even where the option of branching is available, branch- ing requirements often do not differ substantially from requirements for establishing a subsidiary. For example, in Kenya regulations establish similar minimum capital requirements for branches and subsidiaries, thus eliminating a key reason (lower set-up costs) for establishing a branch rather than a subsidiary. The decision to enter by branch or subsidiary can also be driven by the business model pursued by a cross-bor- der bank. Cross-border banks that aim to serve a narrow corporate clientele or are reluctant to invest sizeable capital as long as the size of their cross-border activity in the relevant market is uncertain, may find the branch model more attractive. In contrast, cross-border banks with the intent to establish a more broad-based retail operation may find the subsidiary model that provides for the host-country operation to be run as an independent bank, including with separate liquidity and capital buffers, and with its own management, board, and committees, more suitable. Degree of intra-group integration Beyond the choice of whether to enter a market by branch or subsidiary, banks also face the decision of how closely to integrate their operations at the group level. The level of intra-group integration in many African countries is limited, at least in part, by regulatory requirements. Not only are banks expanding across borders almost universally required to establish self-standing subsidiaries, but they are also often required to establish local IT functions, to use predominantly local labor, and to establish independent, local management functions (such as boards and risk-management capacity). These regulations are part of so-called “indigenization” policies. Nevertheless, variation in the level of intra-group integration across banks exists. Some banks have pooled functions and resources in the areas of IT, risk management, customer service, or treasury operations, so their subsidiaries depend heavily for their functionality on the parent. Some cross-border banks also have deep financial linkages within the group in the form of placements (e.g., subsidiaries place deposits with their parent institutions), and/or through intra-group swap transactions, joint guarantees, and syndicated lending operations involving the parent and/or foreign subsidiaries. In other instances, subsidiaries of cross-border banks 12: For Sierra Leone and Swaziland the two databases provided conflicting information with regard to whether foreign banks are allowed to enter by branch. A consultation of the respective banking acts suggests that foreign entry by branch is prohibited in Sierra Leone but allowed in Swaziland. 56 Chapter 1: Recent Trends in Cross-Border Banking in Africa are independent entities in that they have their own systems and conduct intra-group financial transactions at arm’s length. In general, cross-border banks from outside of Africa have low levels of integration, while the operations of African-based cross-border banks tend to be more closely integrated. Merger and acquisition vs. greenfield investment Cross-border banks can enter host countries by merging with or acquiring the operation of an existing financial institution, or by establishing a presence from scratch (greenfield investment). The reasons for choosing one mode over the other often depend on a number of factors, includ- ing the degree of difficulty of obtaining a new banking license, whether any suitable banking operations can be acquired, and whether banks plan to establish a retail-focused presence or enter to follow existing corporate customers. Acquiring the operations of an existing financial institution can often give relatively quick and easy access to extensive operations; however, it is not without challenges, such as combining different corporate cultures. On the other hand, banks have also reported significant start-up costs and long break-even periods from entering a host country in the form of greenfield investment. Market segment and strategy Box 1.6 profiles the business models of the eight largest African-based cross-border banks. Most banks serve primarily their corporate customers, and their expansion is driven by following their clients abroad. However, there are also examples of smaller African-based cross-border banks such as Equity Bank and Kenya Commercial Bank, two Kenyan banks, that have success- fully pursued innovative business models based on agency banking in their home market, which they are looking to export into neighboring countries. (Box 2.2) Chapter 1: Recent Trends in Cross-Border Banking in Africa 57 Box 1.6: The Largest African-Based Pan-African Banks at a Glance Ecobank Togo-based Ecobank is represented in 32 countries across Africa through its home oper- ations and subsidiaries, which makes it the bank with the largest geographic footprint in Africa. Nigeria is by far Ecobank’s largest operation, contributing 45 percent to its balance sheet and 41 percent of profits in 2012. Ecobank’s other operations are managed on a regionally clustered basis, including in WAMZ, UEMOA, EAC, SADC, CEMAC. Its non-African operations are managed by its International division. Ecobank aims to be a Top-5 bank in most of the countries where it operates (It is currently among the top 6 Nigerian banks). It provides a standard product set across its subsidiaries, adapted to local circumstances. Overall, the Ecobank business model consists of largely following the value chains of large corporates. United Bank for Africa (UBA) Nigerian UBA has subsidiaries in 17 African countries, all of which were established in the period 2007 to 2011, except for the Ghana subsidiary, established in 2004. UBA predomi- nately serves corporate customers and its expansion strategy is based on following large Nigerian corporations abroad. It also provides wholesale banking, including treasury business (security issuance and placements, and trading in securities and currencies) and government business. On the retail end, UBA has started to provide e-banking solutions and remittance products. Currently, UBA’s subsidiaries contribute about 20 percent to group profits. UBA is targeting 50 percent of deposits and 50 percent of profits from foreign subsidiaries in the medium term to diversify the group’s income base. Standard Bank Group (Stanbic) South-African-based Standard Bank was one of the first African banks to expand across borders by acquiring ANZ Grindlays’ operations, in Botswana, Kenya, Uganda, Zaire, Zambia and Zimbabwe, as well as minority holdings in banks in Ghana and Nigeria in the early 1990s. Standard Bank mainly serves corporate customers and its main criterion for expanding into new countries is to follow existing clients. In November 2007, Standard Bank Group announced a major strategic partnership with Industrial and Commercial Bank of China Limited (ICBC), which resulted in ICBC becoming a 20 percent shareholder in the Standard Bank Group. It claims to be the largest African bank in terms of assets and earnings. Banque Marocaine du Commerce Extérieur (BMCE) BMCE is one of three large banks that dominate the Moroccan banking sector. Across Africa, it mostly operates under the Bank of Africa (BOA) name. BMCE formed an alliance with BOA Group, the holding company for all BOA operations (first established in Mali in 58 Chapter 1: Recent Trends in Cross-Border Banking in Africa 1982), in 2007 when it had ten subsidiaries across Africa. It has since gradually increased its equity stake in the BOA Group from 35 to more than 70 percent. BOA subsidiaries offer standard banking services to mostly corporate customers. The investment in BOA Group has spurred growth and paved the way for an ambitious expansion plan; the bank aims to cover a large proportion of African countries within 10 to 15 years by taking advantage of investment opportunities in many financial areas such as retail banking, investment banking, insurance, and mobile banking. BMCE’s cross-border African operations ac- count today for more than one third of the group’s earnings. Banque Sahélo-Saharienne pour l’Investissement et le Commerce (BSIC) BSIC maintains a large geographic footprint in Africa, with a presence in 14 African coun- tries in Northern and Western African. However, its relevance to the banking landscape is very limited, given that its regional presence grew out of its mandate to provide Libyan development finance to countries in the region under Muammar Gaddafi. Attijariwafa Bank Attijariwafa Bank is a large universal bank in Morocco which is controlled by the holding company SNI, a conglomerate in which the Moroccan royal family is a key shareholder. By acquiring Crédit Agricole’s (France) retail banking network in five West African countries in 2008, Attijariwafa Bank expanded its presence beyond Tunisia, Senegal and Mali, where it was already operating. Attijariwafa Bank’s strategy is to expand its success- ful business lines in it Morocco into the African countries where it has a retail banking presence. These include products for the lower end of the market – Attijariwafa Bank is the largest provider of microfinance in Morocco. Its strategy also includes capturing the growth of fast-growing African countries and the rise of banking penetration in the region, as well as servicing its Moroccan corporate clients throughout the region. The cross-border African operations of Attijariwafa Bank contribute about one quarter of the group’s profits. Banque Centrale Populaire du Maroc (BCP) BCP is large commercial bank in Morocco. It’s expansion into West Africa is the results of its acquisition of Banque Atlantique’s operations in 2012 (Banque Atlantique was first established in Côte d’Ivoire in 1978). BCP signed a partnership agreement with Atlantic Financial Group, the holding company of all Banque Atlantique operations, to establish Atlantic International Business, in which each holds 50 percent but BCP controls manage- ment. As a result, the entire Banque Atlantique network is now supervised on a consolidated basis by Bank Al-Maghrib, BCP’s home supervisor. BCP’s ambitions are wide-ranging, encompassing the servicing of Moroccan companies expanding in West Africa, guaranteeing trade finance, syndication among subsidiaries within the CFA zones, expansion of retail Chapter 1: Recent Trends in Cross-Border Banking in Africa 59 banking and generation of local funding, and microfinance (BCP is the second largest provider of microfinance after Attijariwafa Bank in Morocco). Barclays Africa Group Barclays Bank of the United Kingdom has been providing financial services in Africa since colonial times. In May 2005, it purchased a majority stake in the South African based Absa Group, and in mid-2013 consolidated its African operations (with the exception of Egypt and Zimbabwe) under the brand name Barclays Africa Group to develop a more coherent strategy for its African business. It had previously operated under both the Barclays and Absa brand names. Barclays Africa Group primarily focuses on serving the corporate sector. Source: Interviews with bank officials and information available on bank’s websites. 1.5 Conclusion Cross-border banking is an important part of Africa’s financial landscape. Foreign banks were the first banks to be established on the continent during colonial times. While these traditional Western foreign banks still have a presence in their former colonies, their importance decreased as local and state-owned banks were established follow- ing independence, and decreased further as South-South banks and especially African banks have expanded across the continent over the last two decades. African banks have not only significantly increased their geographic footprint across the continent; they have also become economically significant beyond their home countries and of systemic importance in a number of countries. While the reasons for the expanding role of African banks vary, they can be attributed both to push factors, such as declining profit oppor- tunities in the home market, regulatory change, and pan-African ambitions; and to pull factors, such as following corporate clients, risk diversification, and increasingly attrac- tive business opportunities in other countries. The expansion of cross-border banking in Africa also needs to be understood against the background of financial liberalization in the late 1980s and early 1990s which reversed interventionist financial sector policies enacted at independence in most African states and created the space in which banks could expand across borders. 60 Chapter 1: Recent Trends in Cross-Border Banking in Africa Annex: List of Cross-Border Banks in Africa Notes: This table includes all international and African cross-border banks present in Africa according to the definition provided in Box 1.1. Presence in a country is by way of parent bank, branch or subsidiary that has received a banking license from the competent regulatory authority by December 31, 2013. Presence in a country by way of representative offices is not considered. Table 1.3: List of Cross-Border Banks in Africa No. Name Origin Location Majority Number of ownership/ of headquarters largest African minority countries shareholder 1 Ecobank African Togo South Africa 32 2 United Bank for African Nigeria Nigeria 19 Africa (UBA) 3 Standard Bank African South Africa South Africa 18 Group (Stanbic) 4 Banque African Morocco Morocco 18 Marocaine du Commerce Extérieur (BMCE) 5 Société Générale Non-African France France 17 6 Citigroup Non-African USA USA 15 7 Banque Sahélo- African Libya Libya 14 Saharienne pour l’Investissement et le Commerce (BSIC) 8 Standard Non-Africa UK UK 14 Chartered 9 BNP Paribas Non-African France France 13 10 Attijariwafa Bank African Morocco Morocco 12 11 Banque Centrale African Morocco Morocco 11 Populaire du Maroc (BCP) 12 Barclays Africa Group African South Africa UK 10 Chapter 1: Recent Trends in Cross-Border Banking in Africa 61 No. Name Origin Location Majority Number of ownership/ of headquarters largest African minority countries shareholder 13 Access Bank African Nigeria Nigeria 9 14 Bank of Baroda Non-African India India 9 15 Guaranty Trust African Nigeria Nigeria 9 Bank Ltd. 16 Libyan Foreign African Libya Libya 9 Bank 17 Afriland First African Cameroon Cameroon 8 Bank 18 Banque Régionale African Niger USA 8 de Solidarité (BRS) 19 BGFI Bank African Gabon Gabon 8 20 First National African South Africa South Africa 8 Bank (FNB) 21 First Bank African Nigeria Nigeria 7 of Nigeria 22 Kenya Commercial African Kenya Kenya 6 Bank (KCB) 23 NedBank African South Africa South Africa 6 24 Orabank African Togo USA 6 25 Access Holding Non-African Germany Unknown 5 26 Albaraka Bank Non-African Bahrain Bahrain 5 (Group) 27 BancABC African Botswana Zimbabwe 5 28 Diamond Bank African Nigeria Nigeria 5 29 Equity Bank African Kenya Kenya 5 30 HBL Pakistan Non-African Pakistan Tanzania 5 (Habib Bank Ltd.) 62 Chapter 1: Recent Trends in Cross-Border Banking in Africa No. Name Origin Location Majority Number of ownership/ of headquarters largest African minority countries shareholder 31 International Non-African Switzerland Malaysia 5 Commercial Bank (ICB) 32 Keystone Bank African Nigeria Nigeria 5 Group 33 Rabobank Non-African Netherlands Netherlands 5 34 Diamond Trust African Kenya Switzerland 4 Bank 35 First International African Liberia Unknown 4 Bank Liberia Ltd. 36 HSBC Bank Non-African UK UK 4 37 I&M Bank Group African Kenya Kenya 4 38 Mauritius African Mauritius Mauritius 4 Commercial Bank 39 Skye Bank African Nigeria Nigeria 4 40 Zenith Bank African Nigeria Nigeria 4 41 Arab Bank Plc Non-African Jordan Various 3 42 Banco Espirito Non-African Portugal Portugal 3 Santo 43 Bank of India Non-African India India 3 44 Credit Agricole Non-African France France 3 45 Groupe Banque Non-African France France 3 Populaire 46 NIC Bank Group African Kenya Kenya 3 47 Advans Bank Non-African Luxembourg Luxembourg 2 48 African Banking African Kenya Kenya 2 Corporation (ABC Bank) Chapter 1: Recent Trends in Cross-Border Banking in Africa 63 No. Name Origin Location Majority Number of ownership/ of headquarters largest African minority countries shareholder 49 Banco Africano de African Cape Verde Angola 2 Investimentos (BAI) 50 Bank of China Non-African China China 2 51 Bank of African Sudan Sudan 2 Khartoum Juba 52 Banque pour le African Mauritania USA 2 Commerce et l’Investissement en Mauritanie 53 Barclays Bank Non-African UK UK 2 PLC 54 Byblos Bank Non-African Lebanon Various 2 S.A.L. 55 Capital Bank/FMB African Malawi Malawi 2 56 Commercial Bank African Kenya Kenya 2 of Africa 57 Commercial Bank African Ethiopia Ethiopia 2 of Ethiopia 58 Coris Bank African Burkina Faso Burkina Faso 2 59 Deutsche Bank Non-African Germany Germany 2 60 Exim Bank African Djibouti Tanzania 2 61 Habib Bank AG Non-African Switzerland Switzerland 2 Zurich (HBZ) 62 Imperial Bank African Kenya Various 2 63 Investec Bank African South Africa South Africa 2 (Mauritius) Ltd. 64 Kingdom Bank African Zimbabwe Zimbabwe 2 Africa Ltd. 64 Chapter 1: Recent Trends in Cross-Border Banking in Africa No. Name Origin Location Majority Number of ownership/ of headquarters largest African minority countries shareholder 65 Millenium (bim) African Mozambique Portugal 2 Bank 66 Opportunity Non-African USA USA 2 International 67 ProCredit Non-African Germany Germany 2 68 State Bank of India Non-African India India 2 (SBI) 69 State Bank African Mauritius Mauritius 2 of Mauritius 70 The Hongkong Non-African UK UK 2 and Shanghai Banking Corporation Ltd. 71 Union Bank African Nigeria UK 2 Nigeria 72 *ADIB Egypt African Egypt United Arab 1 Emirates 73 Ahli United Bank Non-African Bahrain Bahrain/ 1 (Egypt) SAE Kuwait 74 Algeria Gulf Bank Non-African Algeria Kuwait 1 75 Arab Banking Non-African Bahrain Libya/Kuwait/ 1 Corporation United Arab Emirates 76 *Arab Tunisian African Tunisia Jordan 1 Bank 77 *Banco Comercial African Cape Verde Portugal 1 do Atlântico 78 *Banco Comercial African Mozambique Portugal 1 e de Investimentos (BCI) Chapter 1: Recent Trends in Cross-Border Banking in Africa 65 No. Name Origin Location Majority Number of ownership/ of headquarters largest African minority countries shareholder 79 *Banco de African Angola Portugal 1 Fomento - Angola (BFA) 80 *Banco African Sao Tome Portugal 1 Internacional de Sao Tome e P rincipe (BISTP) 81 *Banco Millennium African Angola Portugal 1 Angola SA 82 *Banco Totta de African Angola Portugal 1 Angola SARL 83 *Banco Unico African Mozambique Portugal 1 84 *Bank of West African Guinea- Portugal 1 Africa (Banco da Bissau Africa Ocidental) 85 *Bank VTB Africa African Angola Russia 1 86 *Banque de Dépôt African Djibouti Switzerland 1 et Crédit de Djibouti (BDCD) 87 *Banque African Congo, Luxembourg 1 Internationale Democratic pour l’Afrique au Republic Congo (BIAC) 88 *Banque Non-African Burkina Faso Belgium 1 Internationale pour l’Afrique au Niger SA (BIA) 89 *Cairo African Uganda Egypt 1 International Bank (CIB) 90 *China Non-African China China 1 Construction Bank 66 Chapter 1: Recent Trends in Cross-Border Banking in Africa No. Name Origin Location Majority Number of ownership/ of headquarters largest African minority countries shareholder 91 *Commercial African Cameroon Luxembourg 1 Bank Group 92 **Cooperative African Djibouti Yemen 1 Agricultural and Credit Bank 93 Dahabshiil Bank Non-African United Arab United Arab 1 International S.A. Emirates Emirates 94 Dubai Bank Non-African Kenya United Arab 1 Kenya Ltd. Emirates 95 *Finance Bank African Zambia Netherlands 1 Zambia Ltd. 96 *Finibanco Angola African Angola Portugal 1 97 *Islamic Bank of African Senegal Saudi Arabia Senegal (Banque 1 Islamique du Sénégal) 98 JPMorgan Chase Non-African USA USA 1 Bank 99 *Mercantile Bank African South Africa Portugal 1 Ltd. 100 Royal Bank of Non-African UK UK 1 Scotland 101 *Saba Islamic African Djibouti Yemen 1 Bank (SIB) 102 UBS Non-African Switzerland Switzerland 1 103 *Union National African Egypt United Arab Bank - Egypt SAE Emirates 1 104 Warka Bank Non-African Iraq Iraq 1 * Banks of African origin with non-African owners are included as cross-border banks even if only represented in one African jurisdiction.** The Cooperative Agricultural and Credit Bank is a joint venture of several Egyptian banks. 2. Cross-Border Banking: Benefits and Risks 68 Chapter 2: Cross-Border Banking: Benefits and Risks 2. Cross-Border Banking: Benefits and Risks Cross-border banking has been increasing not only in Africa. Other regions have expe- rienced similar increases in cross-border banking activity, including Latin America and, especially, the former transition economies of Central and Eastern Europe. This trend has triggered a rich academic and policy debate on the benefits and risks of cross-bor- der banking. This chapter summarizes those benefits and risks. Cross-border banking is controversial among both academics and policymakers. Advocates of “open borders” for banks point to benefits from economies of scale and increased efficiency and competition. Skeptics point to the foreign banks’ strategic focus on high-end clients, including wealthy individuals and large foreign and domestic companies, and trade finance. Similarly, while advocates of cross-border banking point to the stabilizing impact of multinational banks, skeptics point to the global financial crisis, which reminded us that foreign bank entry may also open channels for financial contagion. While the presence of foreign banks in Sub-Saharan Africa goes back to colonial times, and has been supplemented in recent years by entry of new regional players based in countries such as South Africa, Nigeria and Morocco, the business models deployed by banks based in these centers have often been similar in many ways to the traditional post-colonial model. Although closer to their African client-base than banks based in Europe, this new generation of pan-African banks also follow their corporate client base abroad and/or redeploy traditional banking models from their home-base markets. As discussed below, the experience in Africa has been similar to that of other developing markets, and the benefits of this traditional model for financial deepening and outreach are not yet known. There are, however, a number of promising examples of cross-border banks exporting successful innovations that have benefited the bottom of the pyramid in selected host countries. This chapter first presents the theoretical arguments on the benefits and risks of cross-border banking. Section 2 summarizes the empirical evidence from around the world and from African countries where it is available. Section 3 summarizes the implications of global experiences for cross-border banking in Africa. Chapter 2: Cross-Border Banking: Benefits and Risks 69 2.1. The Theory Theory provides opposing arguments on the effects of cross-border banking on (a) competition and efficiency; (b) financial deepening and outreach; and (c) stability. The arguments for cross-border banking Proponents of cross-border banking argue that foreign banks have a positive overall effect on the host country’s banking sector by increasing competition, fostering credit growth, lowering volatility, and implementing best practices in terms of supervision and regulation from their home country. Foreign banks, the argument goes, have a comparative advantage when entering new markets in terms of better access to capital, economies of scale, risk diversification, lending technologies, skills and management expertise (see, for example, Detragiache, Gupta, and Tressel, 2008; and Clarke et al., 2005). These advantages allow foreign banks to offer new products, introduce new lending technologies and delivery channels, attract deposits at higher rates, and offer lower lending rates. Ultimately, this will increase competitive pressure on the other players in the banking market, with positive repercus- sions for financial deepening. Proponents also argue that foreign banks can increase competition and broaden the borrower base in an economy through improvements they bring to governance structures. By not being tied to incumbent borrowers and into existing networks of entrepreneurs, bankers, regulators, and politicians, foreign banks can push the financial system and the overall economy away from an insider-based lending model towards a model purely based on commercial lending criteria (“arms- length”). This has implications not just for competition, but also for the broader transformation of economies. The former transi- tion economies in Central and Eastern Europe have often been cited as a prime example of such an impact. Foreign bank entry can also play an important role in strengthening the stability of the financial system. Foreign banks often enter after a banking crisis, when domestic shareholders (be they private or government) have exhausted their capital or their management has been discredited (Cull and Martinez Peria, 2008). On a broader level, cross-border banking can bring important diversification benefits, both for the banks themselves and for the clientele. Banks can diversify across economies whose business cycles are not synchronized, while borrowers in the host countries can diversify across domestic and foreign banks if these are subject to non-synchronized funding shocks. The presence of foreign banks allows domestic firms to have multiple lending relation- ships with domestic and foreign banks. When domestic banks are lending constrained, firms can substitute domestic lending with finance from foreign banks. These diversi- 70 Chapter 2: Cross-Border Banking: Benefits and Risks fication benefits for the host economy are especially large if foreign banks come from different home countries and are thus subject to non-synchronized funding shocks. Foreign bank entry can also contribute to financial stability indirectly, if host country supervisory authorities are induced to substantially increase the quality of supervision and the number of skilled staff in response to the more sophisticated activities and products available from foreign banks. By challenging host country supervisors, the new lenders could thus raise the regulatory and supervisory bar for the entire sector. This mechanism was at work in countries such as Brazil and Hungary in the 1990s and early 2000s (Song, 2004). There might also be a general skill-enhancing effect of cross-border banking if new entrants decide to upgrade the skills base of their staff through training, with possible spillover effects as staff move on to domestic banks. Finally, foreign banks are often more removed from the regulators and political class in the host country, which can reduce political pressures and regulatory capture. The arguments against cross-border banking Skeptics of cross-border banking, on the other hand, argue that foreign banks can potentially have an overall negative effect on the host country’s banking sector by crowding the market without increasing effective competition, decreasing system-wide outreach to lower-end customers, increasing contagion risks, and overwhelming super- visors without the necessary capacity or skill level, thereby increasing risks to the stability of financial system. While some countries have encouraged foreign bank entry as a way of stimulating competition, critics point out that foreign banks do not necessarily do so. Instead, an already crowded banking market may become more crowded with aug- mented system-wide overheads as the number of banks increases. This can especially be the case when foreign banks follow the trading activities of their clients and focus on banking the already banked top end of the market. The effects of foreign bank entry on competition might be determined by the market structure into which foreign banks enter. In financial systems with limited competition and contestability, foreign banks might share in the oligopolistic rents of existing incumbents rather than trying to reduce these rents through competition. This may be especially true if a foreign bank enters by taking over a dominant incumbent player. However, a lack of contribution to competition can also occur when foreign banks enter by greenfield investment. While such entry increases the number of players, the new banks may just contribute to crowding the market rather than increasing effective competition. In financial systems with inefficient government-owned banks, foreign banks might be tempted to turn their efficiency advantages into additional profits rather than competing aggressively. Chapter 2: Cross-Border Banking: Benefits and Risks 71 Second, skeptics also point out that entry of foreign banks can result in decreased system-wide outreach to lower-end customers. For example, the competitive advantage of foreign banks, combined with their greater reliance on hard information about borrow- ers, such as formal financial statements (“transactional banking”) as opposed to soft information that leverages client relationships and local market knowledge (“relation- ship banking”), can lead to crowding out of domestic banks and have negative reper- cussions for riskier and more opaque borrowers. To the extent that local banks hedge or even cross-subsidize higher-risk client segments down-market, this crowding-out may ultimately impinge on financial outreach by leaving small and medium enterprises (SMEs) and poorer households without access to financial services, as the viability of local banks will come under threat (for example, see Gormley, 2010; Sengupta, 2007; Detragiache, Tressel and Gupta, 2008). More directly, the lack of hard information such as formal financial reporting may also prevent foreign banks from lending to lower-end customers such as small and – from an arm’s length perspective – opaque firms (see, for example, Mian, 2006). Specifically, the greater reliance of foreign banks on hard information can have negative repercussions for riskier and more opaque borrowers if foreign banks crowd out domestic banks. Thus there is a danger that entry of foreign banks can result in more market segmentation that narrows the business opportunities of domestic banks, thereby negatively impacting system-wide outreach to low-end customers. Critics of foreign bank entry also challenge the more traditional stability benefits associated with foreign bank entry arising from diversification benefits, and point to significant contagion risks associated with cross-border banking. Cross-border banks can propagate shocks from their home countries or other countries in which they are active into the host country. As became obvious during the recent global financial crisis, multinational banks can withdraw liquidity from host countries if they come under pressure in their home country, thus reducing lending in host countries. Worse, the failure of multinational banks can leave host countries with subsidiaries of such banks that cannot stand on their own. Finally, critics questions whether foreign banks contribute to upgrading regulation and supervision. Effective regulation and supervision starts with domestic banks and does not require the presence of foreign banks. And even where supervisory authorities resolve to enhance the effectiveness of supervision, building sufficient skills may take years, and during the transition the host country banking sector will be exposed to risks to the extent that sophisticated new products and services are not fully understood or regulated. 72 Chapter 2: Cross-Border Banking: Benefits and Risks Branches vs. subsidiaries and integrated business models The benefits and risks arising from cross-border banking might also depend on whether cross-border banking takes place in the form of branches or subsidiaries, as well as on the structure of the host country’s financial sector. As discussed in Chapter 1, the expansion of cross-border banks in Africa has predominantly taken the form of stand-alone subsidiar- ies, involving relatively high set-up and fixed costs, exacerbated by regulatory requirements aimed at ring-fencing these subsidiaries in times of crisis. While the global literature has not focused as much on the distinction between branches and subsidiaries in the context of the benefit-risk trade-offs of cross-border banking, this distinction seems important in the African context, where the small scale of markets can make the fixed costs of stand-alone subsidiar- ies more punishing.13 A more integrated banking model which relies on branches or subsidiaries that are closely linked to the parent bank implies that funding, asset allocation, and risk management are centralized in order to maximize returns at the consolidated level (Fiechter et al., 2011). For the same level of risk exposure, this model rationalizes the use of capital and liquidity, as both funding and capital utilization can be more easily optimized over time across the different units. The cost of doing business is therefore, in general, lower for a more integrated banking model. It can also strengthen the ability of branches or subsidiaries to withstand idiosyncratic host country shocks, as additional capital and/or liquidity can be easily mobilized. On the other hand, a more decentralized model that relies on stand-alone subsidiaries, although it might be less cost effective, might be preferred by both home and host country supervisors in terms of the (perceived) ease of resolving failing banks and in terms of their ability to oversee the foreign bank’s activities (monitor and reporting), minimizing contagion costs. 2.2. The International Evidence The past decade has seen an array of empirical studies assessing different hypotheses about foreign bank entry. Cross-country, regional, and country-level studies have used aggregate, firm, household and bank-level data to gauge the effect of foreign bank entry on different outcomes, including efficiency, competition, stability and financial inclusion. These studies show that the outcomes are highly context specific.14 13: A large body of literature on the U.S. branch deregulation episode in the 1970s and 1980s has documented large efficiency gains from allowing banks to move from bank holding and subsidiary structure to cross-state branching structures. 14: For a more complete survey of the evidence on foreign bank entry, see Cull and Martinez Peria (2013). Chapter 2: Cross-Border Banking: Benefits and Risks 73 Efficiency and competition Cross-country comparisons have shown that foreign bank entry is positively associated with efficiency and competition in host countries, as gauged by net interest margins, profitability and cost efficiency (Claessens, Demirgüç-Kunt, and Huizinga, 2000, 2001). The hypothesis that foreign bank entry is associated with a higher degree of competition in host countries has also been confirmed by Claessens and Laeven (2004). Country-specific studies for Argentina and Colombia confirm the positive effect of foreign bank entry on efficiency in local banking markets (Clarke et al., 2000; Barajas et al., 2000). Similarly, country-specific studies across the Central and Eastern European region also show a positive impact of foreign bank entry on competition in local markets (see Kiralyi et al., 2000 for Hungary; Nikiel and Opiela, 2002 for Poland). On the other hand, broader stud- ies across several countries within the Latin American and Central and Eastern European regions show more mixed results, which might be partly due to the use of different indicators of competition and efficiency. Evidence from India and China shows less of a positive impact from foreign bank entry, which can be explained by the rather limited foreign bank participation as well as the domi- nance of state-owned banks in these countries (Sensarma, 2006; Wu, Chen, and Lin, 2007). Foreign bank entry in many East Asian countries, after the 1997 crisis, on the other hand, has been associated with modest but positive effects on efficiency and competition (e.g., Lee, 2003 for Korea; Unite and Sullivan, 2002 for Philippines; and Kubo, 2006 for Thailand). There are few empirical studies of the effects of cross-border banking on efficiency and competition in Africa. Evidence for countries in the East African Community (EAC) shows that the effects vary across countries and across time. For example, Cihak and Podpiera (2005) find that in the early 2000s, foreign banks in Tanzania and Uganda lent more and charged lower spreads than domestic banks, while foreign banks in Kenya lent less than their domestic counterparts. A recent World Bank-led Financial Sector Assessment Pro- gram in the EAC (2013) covering Burundi, Kenya, Rwanda, Tanzania and Uganda provides insights into differences across foreign banks of different origin. Specifically, EAC-head- quartered banks with subsidiaries in the region have lower spreads and are more efficient than other private domestic banks (Figure 2.1.). At the same time, these banks are highly profitable in their home markets, providing a comfortable buffer against losses or low profits in the first years of their subsidiaries’ operations. Similarly, subsidiaries of banks domiciled in the EAC region have lower spreads and overheads compared to subsidiaries of foreign banks from outside the region, exemplifying the differences in business models.15 EAC banks that expanded within the region also introduced innovative business models, such as 15: The higher costs for non-EAC foreign banks might stem from higher staff costs and/or higher compliance costs, in general reflecting a more conservative approach. 74 Chapter 2: Cross-Border Banking: Benefits and Risks agency banking, into their new markets, which allowed them to break even faster than their competitors. However, as Figure 2.1 shows, cross-border banks may not necessarily pass on their efficiency gains to clients, but use those gains to take higher profits. Figure 2.1: Interest Rate Spreads in the EAC by Foreign and Domestic Bank Ownership Type, 2012 16% 14% 2.6 1.8 12% 0.9 1.5 10% 1.5 6.0 8% 10.9 6% 0.6 8.2 8.6 4% 4.2 2% 1.09 1.44 1.18 1.41 0% -2.0 -2% Private domestic Private domestic Foreign subsidiaries Foreign subsidiaries EAC banks with EAC banks without (of banks within (of banks ouside the subsidiairies in subsidiaries in the EAC) EAC) other EAC countries other EAC countries Reserves Overheads Provisions Profit Source: World Bank 2013a. One factor common to all the cited studies is that the measured empirical results are dependent on the structure of the domestic banking systems on which foreign banks “intrude.” This pertains not only to the nature of competition, as reflected by such factors as state ownership and political influence in the banking sector, but also to deficient financial infrastructure such as weaknesses in the contractual framework and credit information, and to the legal/regulatory framework for bank entry and exit. As discussed above, state ownership might undermine incentives for foreign banks to compete vigorously, as argued in the case of Kenya in the early 2000s by Beck and Fuchs (2004). For example, in the absence of effective systems of credit information sharing, Chapter 2: Cross-Border Banking: Benefits and Risks 75 borrowers might be captive to their original lender, who benefits from accumulated private information about them. Whether foreign bank entry will have a beneficial or detrimental impact on efficiency and competition thus hinges on the conditions in the local banking market as well as on the prevailing supervisory framework and practices. In as much as these factors can be influenced by local regulatory and supervisory authorities, this is good news, as the initiative remains with those authorities. Given the strong evidence of scale economies of banking, there is a strong case for positive effects of cross-border banking on the efficiency in African banking (Box 2.1). Box 2.1: Scale Economies Scale economies in banking are hard to measure and their estimation is subject to assump- tions about cost functions, but there is strong evidence that the cost efficiency of banks in- creases with size. Most of the bank-level evidence is based on US data, exploiting the large number of banks needed for meaningful estimates. While studies from the 1990s did not find significant scale economies beyond a threshold of USD 25 billion (Berger and Mester, 1997), more recent studies find scale economies beyond that threshold (e.g. Wheelock and Wilson, 2009; Feng and Serletis, 2010), even when they explicitly take into account that larger banks might be more aggressive in risk-taking, given their too-big-to-fail status (Hughes and Mester, 2013). One factor explaining this difference might be that scale economies in banking have most likely increased in recent years given technological progress in telecom- munication and computing which has led to significant cost savings. An extensive literature exploring the impact of branch deregulation on banks’ performance in the US has also shown that allowing banks to expand across state borders has resulted in cost reductions and lower pricing (Jayaratne and Strahan, 1998; Black and Strahan, 2001). Scale economies are also present at the banking system level. Simple correlations be- tween efficiency measures, such as overhead costs or net interest margins, and the absolute size of banking systems illustrate the scale problems of African banking. Measures of size, such as population or GDP, are also negatively related to indicators of financial development (Beck et al., 2008; Beck and Feyen, 2013). High structural costs at the banking system level are the result of most banks operating on an inefficient scale, fixed infrastructure costs (e.g. payment system and regulatory framework), and lack of competi- tion given the limited number of players. 76 Chapter 2: Cross-Border Banking: Benefits and Risks Financial depth and outreach With regard to the effect of foreign entry on competition, the empirical evidence on the effect of foreign banks on financial outreach and deepening has been ambiguous, with findings varying across countries and regions and across different sources of data (household survey, enterprise survey, and bank-level data). Cross-country evidence On the cross-country level, Detragiache, Gupta and Tressel (2008) show that a higher share of foreign banks is associated with a lower level and growth rate of financial depth across 89 low-income countries. This relationship is robust to controlling for reverse causation (i.e., foreign banks entering countries with less-developed financial systems to exploit expansion opportunities). While this analysis is cross-sectional; i.e., it averages data for each country over longer time periods, a panel analysis that follows countries over time presents a somewhat different picture (Cull and Martinez Peria, 2008). It shows that foreign bank entry from 1995 to 2002 more often than not happened in the context of financial distress in developing countries. Foreign banks were often brought in to acquire failed domestic banks, whether private or state-owned, and thus recapitalize undercapitalized banking sectors. Cleaning the balance sheets of those target banks likely contributed to lower overall credit levels, but this can hardly be blamed on the foreign acquirers. Conversely, when foreign bank presence was not crisis-induced, meaning that the share of foreign banks was relatively high and stable from the beginning of the period, Cull and Martinez Peria (2008) find that private credit to GDP was significantly higher than in other countries, and this was true even during and after crises. There might be important differential effects of foreign bank entry across countries with different characteristics, as shown by Claessens and van Horen (2014). Their anal- ysis shows that foreign banks’ participation has a negative relationship with financial depth in low-income countries; in countries where foreign banks have a limited market share; where enforcing contracts is costly and credit information sharing is limited; and when foreign banks come from distant home countries. This analysis is in line with a study on the real sector outcome of foreign bank entry (Bruno and Hauswald, 2013) that shows that the effect of foreign bank entry on the growth of industries that are most in need of external finance is stronger in developing countries with better contract enforcement and more effective credit information sharing. Both studies suggest that it is the responsibility of local authorities to create an environment conducive to financial sector deepening to reap the benefits of foreign bank entry, including strong credit information and an effective legal recourse system. Chapter 2: Cross-Border Banking: Benefits and Risks 77 While these studies use financial depth data rather than inclusion data, cross- country comparisons using proxy indicators for inclusion show that foreign bank entry is negatively associated with the number of loan and deposit accounts per capita, but not significantly with the ratio of branches and ATMs to population (Beck, Demirgüç- Kunt and Martinez Peria, 2007). On the other hand, a higher share of foreign banks is associated with lower barriers to accessing deposit services (Beck, Demirgüç-Kunt and Martinez Peria, 2008). These contradictory results can be explained by the different samples used by these two studies, but also by the different dimensions of financial inclusion; i.e., the actual number of accounts and barriers to accessing services. Using the World Business Environment Survey (WBES), with data on more than 3,000 firms in 35 developing and transition countries, Clarke, Cull and Martinez Peria (2006) find that self-reported financing obstacles decline with the share of foreign banks in a country. This relationship is stronger for larger enterprises, suggesting that they benefit more from foreign bank entry; yet Clarke, Cull and Martinez Peria (2006) also document it for smaller firms. The evidence on foreign banks’ impact on enterprise credit is therefore so far inconclusive. Although most of the literature focuses on lending to enterprises, there is some evidence on the effect of foreign bank presence at the household level. Using data on 16,500 households across 19 emerging economies in Central and Eastern Europe, Beck and Brown (2014) find that borrowers of foreign banks are richer, more likely to be formally employed, and more likely to have personal assets than borrowers of domes- tic banks. A higher share of foreign banks in a country’s banking system thus benefits the high end of the household market more than the low end, which seems to confirm critics’ accusation of cherry-picking by foreign banks. It is important to note that this analysis, unlike the ones mentioned before, does not refer to the overall level of access to finance in relation to foreign bank participation, but rather to the relative composition of the population using formal financial services. Country-specific studies often paint more nuanced pictures. In one of the first studies to use loan-level data from a credit registry, Mian (2006) shows a large difference in clientele between domestic and foreign banks for the case of Pakistan. Specifically, clients of foreign banks tend to be enterprises that are larger, located in larger cities, foreign owned, and part of business groups. The difference between foreign and domes- tic banks is more marked for those foreign banks whose headquarters are geographi- cally farther away (e.g., Europe or the United States), suggesting that distance between the parent bank and the loan officer on the ground might be critical in the decision as to which clientele to target. Studying branch, loan and deposit account penetration over time in Mexico over the period 1997 to 2005, Beck and Martinez Peria (2010) document a decline in the number of deposit and loan accounts as the share of foreign banks increased from two to over 78 Chapter 2: Cross-Border Banking: Benefits and Risks 80 percent within a few years. While there was an increase in the share of municipalities with bank branches and in the likelihood of bank presence in a given municipality, a more detailed analysis shows that it was only richer and more urban municipalities that benefitted from this trend. Overall, this seems to be evidence suggesting cherry-picking by the foreign banks entering Mexico. Gormley (2010) shows that a strong presence of foreign banks in regions of India is associated with less lending by domestic banks and an absolute decline in enterprise lending. His analysis thus suggests that foreign banks’ cherry-picking behavior may have an overall negative impact on financial outreach, as observed in the Indian case. More recently, the literature has challenged the view that foreign-owned banks cannot cater to the financing needs of local firms, which includes the long-held belief that foreign and large banks only engage in arms-length lending based on hard information. For example, Clarke et al. (2005) show for a sample of four Latin American countries that large foreign banks often have a greater share and higher growth of lending to small businesses than large domestic banks, with the reverse holding for small foreign and domestic banks. Based on firm-level observations of listed and unlisted companies in Eastern Europe, Giannetti and Ongena (2009) find that all firms benefit from foreign bank lending in terms of growth in firm sales, assets, and leverage, although the effect diminishes for small firms. In another study they find that foreign banks do not limit the access to credit for unlisted firms, although foreign banks are more likely to cater to large and foreign-owned firms (Giannetti and Ongena, 2012). Using cross-country data across the world, de la Torre, Martinez Peria, and Schmukler (2010) and Beck et al. (2011) provide evidence from bank surveys that foreign banks are as well-suited as domestic banks to serve small businesses, but apply transaction-based rather than relationship-based lending techniques. Similarly, Beck and Brown (2014) document that foreign and domestic banks use different lending techniques when reaching out to households across the former transition economies in Central and Eastern Europe. Spe- cifically, the retail lending techniques of foreign banks rely more on financial information and collateral than those of domestic banks, which rely more on relationship lending. The literature has also explored the relationship between the origin of cross-border banks and access to credit. Specifically, Mian (2006) relates the geographic distance between headquarters of the parent bank and loan officers in Pakistan to differences in clientele, and shows that the difference between foreign and domestic bank increases in line with the distance between the parent bank’s headquarters and Pakistan. Similar arguments about the differences in lending techniques and hierarchal distance between loan officers on the ground and headquarters across domestic and foreign banks can also be made for different types of foreign banks. Banks that are geographically, culturally or institutionally closer to the host country can be expected to have shorter hierarchal distances, but might also be more accustomed to the clientele in the host Chapter 2: Cross-Border Banking: Benefits and Risks 79 country. Claessens and van Horen (2014) confirm this conjecture on the aggregate level; any positive effect of cross-border banking on the host country’s financial development is lower the more geographically distant the parent bank’s headquarters. From this perspective, the increasing clout of African cross-border banks on the continent could be seen as a potentially beneficial development, a hypothesis that, however, still remains to be substantiated by empirical evidence. Most of this literature has focused on differences in clientele rather than on direct measures of lending techniques. However, different lending techniques might be driven by different clientele. In one of the few studies to separate these two effects, Beck, Ioannidou and Schäfer (2012) use Bolivian credit registry data and focus on a sample of enterprises that borrow from both domestic and foreign banks in the same month. They show that foreign and domestic banks use different lending techniques to reach the same group. Specifically, domestic banks rely on relationship-based lending, while foreign banks rely on collateral and hard information for their lending decisions and pricing. The African experience There is no rigorous research on the specific situation in Africa despite the substantial foreign bank entry across the continent over the past decade. However, there are a number of examples that seem to refute the assertion that foreign banks in Africa by and large engage in cherry picking and adjust their branch network accordingly, as was the case in Mexico. In Uganda, Uganda Commercial Bank (UCB), the largest government-owned bank – and also the largest bank in the system – was successfully privatized (after the first unsuccessful attempt) to the South African Standard Bank. Although an agreement not to close any branches was in place for only two years following the sale of UCB, Standard Bank kept all branches in place and even opened new ones. It also introduced new products and increased agricultural lending (Clarke, Cull and Fuchs, 2009). Similarly, the privatization of the Tunisia-based Banque du Sud in favor of Attijariwafa Bank led to a 78 percent expansion of the rebranded bank’s network and the develop- ment of a range of new products within five years of the privatization. In Tanzania, the National Bank of Commerce was privatized after being split into a holding (NBC Holding Corporation); a commercial bank (NBC Limited) that assumed most of the original bank’s assets and liabilities; and the National Microfinance Bank, which assumed most of the branch network and the mandate to foster access to financial services. The new National Bank of Commerce was sold to South-African based Absa and its profitability and portfolio quality improved, although credit growth was initially slow. While finding a buyer for the National Microfinance Bank proved difficult, profitability eventually improved and lending grew, while the share of non-performing loans remained low (Cull 80 Chapter 2: Cross-Border Banking: Benefits and Risks and Spreng, 2011). In 2005, Rabobank acquired a 49 percent stake (later dropping to 35 percent) and took over management. It is important to note that these African exam- ples reference positive post-privatization experiences of entry by foreign banks in crisis environments, and therefore do not allow for drawing broader lessons about the impact of foreign bank entry on access to credit. Despite these positive examples and despite healthy growth in banking assets across Africa in the last decade, banks have not always demonstrated the ability to bridge the huge gaps between supply and demand that con- tinue to exist in micro, small and medium-sized enterprises (MSMEs), agricultural and infrastructure finance. An innovative lending model prevalent in South Africa and adopted by South African banks, particularly in their business expansion in neighboring countries in Southern Africa, is salary-based lending. By limiting lending to salaried employees and deducting loan repayments “at source,” even before the salary hits the employee’s bank account, this lending model limits banks’ collateral/security and borrower identification con- cerns. However, in less well-developed financial systems, the impact of such lending on closing the identified lending gaps is at best marginal, because the formal sector rep- resents only a tiny share of the economy in most countries on the continent. With only a small minority of workers holding permanent/formal wage employment, the vast major- ity of private sector jobs are in the informal sector. Due to the comparatively large share of formal employment in South Africa (public and private sector formal employment at around 60 percent) the potential impact of salary-based lending could be expected to be largest in that country (World Bank 2013b).16 There are, nonetheless, some notable exceptions where innovation has resulted in financial broadening and deepening, and the current wave of cross-border activity stands to make a significant contribution towards proliferation of these activities. In Kenya, for example, banks such as Equity Bank and Kenya Commercial Bank have developed business models designed to grow their client base by “graduating” small microfinance borrowers to smaller and eventually quite large individual and SME clients. The success and profitability of these new endeavors is resulting not only in enhanced domestic competition and financial deepening, but also in efforts to export the same business model across the East African Community. Similarly, and going somewhat against the grain of traditional European banks, Netherlands-based Rabo- bank (see Box 1.3) is using its experience in cooperative banking to expand its presence as a provider of business models specifically tailored to the needs of the agricultural value chain in Mozambique, Rwanda, Tanzania, Uganda and Zambia. The case of Zambia National Commercial Bank, which was privatized in 2007 in favor of Rabobank, shows 16: However, a lack of due diligence in assessing borrowers’ ability to repay such borrowing has given rise to increasing concerns as regards consumer over-indebtedness (World Bank, 2013b). Chapter 2: Cross-Border Banking: Benefits and Risks 81 how a formerly wholly state-owned financial institution can branch out, establish agency arrangements with local partners (gas stations, retail stores), and introduce mobile phone technology to facilitate low-cost transactions. A number of cross-border as well as domestic banks have grown beyond the notion that microfinance is merely a corporate social responsibility activity, and are increasingly targeting the bottom of the pyramid to expand their client base (see Box 2.2. for examples). Box 2.2: Examples of Cross-Border Banks Serving the Lower-End of the Market African and international cross-border banks have adopted new products and innovative delivery channels in the past few years to broaden their client base by reaching out to the majority of Africans, who have hitherto excluded from formal financial services. After very successfully rolling out a strategy focused on the retail market in Kenya via agent and mobile banking, Equity Bank seeks to replicate its achievements in other EAC markets. For example, the bank entered the Ugandan market by acquiring Uganda Micro- finance Limited, a Ugandan Tier III microfinance company to serve the lower end of the market and the unbanked, with the intention to move up-market. The bank is hoping to engage in agent banking as soon as Bank of Uganda defines its position and develops guide- lines and rules. In Tanzania, on the other hand, Equity Bank entered as greenfield operation and therefore on a much smaller scale. Nairobi is Equity Bank’s regional hub and has both strong financial (loan syndication) and operational linkages (IT system integration) with its EAC subsidiaries. In The Gambia, Ecobank Gambiaa offers deposit collection services, providing traders and other small-scale businesses with a condaneh (meaning “box” in local parlance) in which they store their daily revenue and which is picked up at their work place in the evening by a bank employee. Ecobank’s objective is to target the unbanked people in the market, encouraging them to accumulate micro savings. After piloting the program in the Greater Banjul Area, the program was rolled out to the rural parts of Gambia. At the writing of this report, Ecobank Gambia planned to partner with microfinance institutions (MFIs) to take care of the daily collection in rural areas. In Ghana, Barclays Bankb has piloted the “Susu Collectors Initiative” to make microloans available to susu collectorsc for on-lending to the clients of collectors. The micro banking program in Ghana started in 2005 and by 2007 had already worked with more than 500 susu collectors, providing loans to more than 280,000 people across the country (Barclays, 2008). Barclays Bank Ghana uses the susu collectors as an intermediary to extend loans and mobilize savings to productive rural communities. The susu collectors intermediate on behalf of the bank, conduct appraisals of prospective borrowers, and advise the bank on risky ven- 82 Chapter 2: Cross-Border Banking: Benefits and Risks tures within the communities. Barclays also provides susu lenders with deposit accounts and training sessions, for example on delinquency management, financial management or risk management. When it became clear that the program was well received, Barclays expanded it to other intermediaries, including credit unions, trade associations, church groups and microf- inance institutions. The initiative creates a win-win situation, as susu collectors can build their capacity and benefit from additional legitimacy from their relationship with the bank, while Barclays gains insight into the indigenous microcredit culture and creates awareness and trust among the unbanked population in its formal banking services (Osei, 2007). Another successful example of a commercial bank initiating cooperative efforts with MFIs is the partnership between Ecobank and ACCION International in Cameroon, formed in April 2010. Both partners have launched “EB-Accion Microfinance” to provide microcredit and savings products to currently unbanked Cameroonians. Ecobank also launched one of the biggest microfinance banks in Nigeria in 2007 in partnership with other stakeholders, and opened EB-ACCION Savings & Loans, a microfinance bank, in Ghana in 2008. It subsequently expanded the microfinance bank network to Senegal, Benin and Cameroon. Ecobank also supports 200 MFIs on the continent in wholesale loans and other products (Ecobank, 2010). Absa (now Barclays Africa Group) has moved into South Africa’s microfinance market and, in partnership with CompuScan, a credit bureau, built Microfinance Enterprise Service Centers, which are rural mobile lending outlets operating out of freight containers. To reach rural and remote areas, the centers are fitted with third-generation mobile telecommunications, general packet radio service and satellite connectivity linked to CompuScan’s South African credit bu- reau database. Prior to the rollout, Absa, together with Fin-Mark Trust, invested considerable effort in understanding the market for microfinance products in general by learning from ICICI Bank in India, and the scope of South Africa’s market potential in particular. In Morocco, Banque Populaire du Maroc has, since 1999, operated a microcredit subsidiary, Fondation Banque Populaire pour le Micro-Crédit, which acquired, in 2009, the Fondation Zak- oura, one of the leading players in the market. In the Arab Republic of Egypt, service compa- nies, a new category of microfinance providers, are acting as agents for banks in the provision of microfinance services. a. Oceanic Bank Gambia until Oceanic Bank was taken over by Ecobank b. Since Barclays consolidated its Africa operations in 2013, Barclays Bank Ghana is part of Barclays Africa Group. c. Susu collection is a traditional informal arrangement in several African countries to mobilize deposit savings against a small fee and provide limited access to micro credit. With an estimated 4,000 susu collectors in Ghana and each collector serving between 200 and 850 clients a day (some even up to 1,500), the susu system provides basic savings and microcredit services to a large number of people excluded from the formal banking system (Osei, 2007). Source: Based on Beck, Maimbo, Faye, and Triki (2011). Chapter 2: Cross-Border Banking: Benefits and Risks 83 Stability A key benefit of cross-border banking arises from its effects on risk diversification, for both financial institutions and local economies. When a domestic bank invests abroad, it becomes less exposed to domestic shocks. As long as business cycles are not perfectly synchronized across countries, expanding beyond the home market allows banks to diversify macroeconomic risk. Similarly, countries with diversified banking systems – made up of both domestic banks and foreign banks from different origin countries that are not subject to synchronized funding shocks – provide risk diversification for local users of banking services. There is a large body of pre-global financial crisis evidence that supports the hypothesis that by diversifying risk, cross-border banking brings ben- efits for both financial institutions and local economies. Arena, Reinhart, and Vazquez (2007) find for a sample of 1,565 banks across 20 emerging markets over the period 1989 to 2001 that foreign bank credit levels were less sensitive to monetary conditions in the host country, while their lending and deposits rates were less volatile than those of domestic banks during crisis periods. Bruno and Hauswald (2014) also show that foreign banks can mitigate the impact of banking crises on industries that are dependent on external finance. Recent empirical evidence has also shown that cross-border banking can help miti- gate the impact of local financial shocks, but that it exacerbates global financial shocks. The critical difference between these two types of shocks has been highlighted both by macroeconomic and bank-level analyses. For example, Popov and Udell (2012) and de Haas and Lelyveld (2014) show that foreign banks have propagated shocks from their home into host countries, while Kalemli-Ozcan, Papaioannou and Perri (2013) show that financial shocks such as the global financial crisis lead to a synchronization of business cycles across countries. Similarly, evidence from the 1990s shows that the Japanese banking crisis had a negative impact on the U.S. commercial real estate market and ultimately real sector activity following withdrawal of Japanese banks from the U.S. market (Peek and Rosengren, 2000). However, there have also been important differ- ences across regions in the impact of the global financial crisis, with banking systems in Central and Eastern Europe much more affected by contagion effects from weakened multinational banks than banking systems in Latin America (Cull and Martinez Peria, 2013). This might be explained by the fact that loan-deposit ratios were typically above 1.0 across the former transition economies in the pre-crisis years – a symptom of a high reliance of subsidiaries in these countries on parent funding, while Latin American subsidiaries of foreign banks relied more on local deposit funding. There has been little work on the stability implications of cross-border banking in Africa. While anecdotal evidence suggests that foreign bank entry has in most cases contributed to the stability of host economies by bringing in fresh capital and more 84 Chapter 2: Cross-Border Banking: Benefits and Risks efficient banking practices, not all foreign bank entry has been alike. For example, Mozambique and Uganda have both experienced attempted privatization of state-owned banks to multinational banks that turned out to have limited resources and capacity, resulting in the re-nationalization of these banks. The impact of the global financial crisis on Africa has provided little evidence for contagion effects through banking system channels, which can be explained by the relatively limited intermediation role of African banks and their high liquidity and capitalization (Beck et al., 2011), and by the fact that the large European banks active in Africa fared relatively well during the crisis. 2.3 Implications of Global Experiences for Africa The theoretical and empirical literature on the effects of cross-border banking does not provide clear guidance on any of the three dimensions – efficiency, financial deepening and broadening, and stability – assessed in this chapter. One conclusion from the varying and sometimes contradictory findings of the literature is that foreign bank entry is certainly not a panacea for increasing access to financial services and stability. On the other hand, foreign bank entry is also not the scourge for low-income countries that it has been sometimes made out to be. Cross-border banks can bring important benefits for local financial systems, but these benefits are context specific and depend critically on host country conditions and policies. To reap the benefits of foreign bank entry, local authorities need to take a proactive role in creating framework conditions that make it more likely for foreign banks to contribute to increased efficiency and competition, financial access, and stability in the financial sector. Some of these conditions can be influenced more easily than others by regulatory authorities. While broader conditions that impact the business-enabling environment, such as the ease of enforcing contracts, often exceed the mandate and influence of central banks and supervisory authorities, other important elements are well within their reach. In particular, factors impacting (a) the cost of entry, such as the availability of financial infrastructure, licensing requirements; (b) the cost of doing business for foreign banks e.g. whether foreign subsidiaries are allowed to use home country IT and risk management systems; (c) maintaining an open policy towards employment of expatriate staff; and (d) establishing a level-playing field – e.g. by applying sanctions even-handedly to domestic and foreign banks that fail to live up to prudential requirements – will all be important in determining the extent of foreign entry and whether such entry has limited or more pervasive impact. Further, keeping an open mind towards useful innovation brought in by foreign banks is crucial. This chapter has pointed to some promising innovations and business models that have successfully been exported to host countries. Host authorities could encourage entry by banks that have a proven track record in implementing particularly relevant product lines, such as Chapter 2: Cross-Border Banking: Benefits and Risks 85 experience in servicing typically underserved client groups, including MSMEs and the rural sector. Cross-border banks in Africa have deployed different market strategies and degrees of engagement in host countries. It is difficult to generalize, but, as outlined in four stereotypical business models below, only as the depth of their operations increases do foreign banks move towards deeper engagement in financial intermediation, deploying more innovative business models and products that most likely are tried and tested in their home markets: • Model 1 – Retaining Large Corporate Clients: The major driver of cross-border banking in Africa remains banks’ ambition to service their large corporate clients as they develop their business abroad. This “defensive” policy can take the form of ensuring the capacity to provide trade finance, but as corporate clients deepen their engage- ment abroad, home-country banks seek to retain their clients by being in a position to support their working capital and investment needs abroad. In the first instance, banks may establish their presence in the form of representative offices booking their clients’ foreign activities on their home balance sheets, so-called “suitcase banking,” but as the local engagement of their clients deepens, the need arises to become locally incorporated. Given that the initial impetus for banks to establish their presence across borders comes from following their clients, the rapid expansion of African banks’ cross-bor- der presence in recent years suggests that the financial rewards to be reaped from this business model are still considerable, although bankers suggest spreads in this sector are generally being squeezed. Indeed, the competition in this space appears quite intense, and one possible spin-off may be that the foreign banks will encour- age domestic banks to deepen their involvement in providing service to other less well-serviced segments. • Model 2 – Targeting Host Country Growth Sectors: A second increasingly important reason for the expansion of African banks across borders is the presence of growth opportunities. These opportunities have traditionally been prevalent in the natural resource extraction industries, but bankers are becoming increasingly aware of the potential spin-offs from such enclave growth. While still largely unexploited, possibil- ities exist in financing ancillary services relating to (a) the value chains servicing the natural resource extraction industries; and (b) the development of local infrastructure and service industries that exhibit considerable growth potential. Engagement at this level involves building a local presence and, in particular, attracting local sources of funding so as to avoid exchange rate exposures. 86 Chapter 2: Cross-Border Banking: Benefits and Risks • Model 3 – Banking the Mass Affluent: In expanding their local presence, cross- border banks have deployed services that involve limited risks, such as payments and remittance services, or products, such as salary-based lending, that constrain banks’ risk exposures to specific target and secured groups. • Model 4 – Exporting Successful Innovations; The major potential contribution of cross-border banking towards financial inclusion lies in the transfer of innovative banking models from countries where banks have engaged in both developing outreach, for example in the form of agency banking and collaboration with mobile- money providers, and in deploying new products targeting MSMEs (micro-lending) and graduating members of such groups to become small enterprise clients. The experience to date with cross-border banks in Africa suggests that most banks are comfortable servicing their large corporate clients (Model 1) and maybe targeting growth sectors (Model 2), but are still reluctant to engage in the deepening process suggested by Model 3 and especially Model 4. The question of how banks can be encouraged to deepen their level of engagement, and to pass on the efficiency gains associated with economies of scale to the end-users of financial services, is further discussed in Chapters 4 and 5. The small scale of most African economies, with the consequent high fixed costs of financial service provision, implies that the potential benefits of financial integration are relatively large as a means of spreading and diminishing the burden of these fixed costs. Empirical evidence has shown a negative association of both country and bank size with interest margins and spreads (Beck, 2007; Beck and Hesse, 2009). What is more, many African economies suffer from a relatively large part of their GDP being derived by a few companies involved in natural resource extraction. Promoting economic diversification and ensuring that the economic benefits associated with natural resource extraction are spread more evenly is a major concern to policymakers in such countries, and the financial sector can make an important contribution to achieving this objective (Beck, 2011). Suitably innovative and outreach-focused banks can be an important catalyst in providing access to credit to local producers, strengthening local supply chains, and supporting the development of import-competing suppliers. While certainly no panacea with respect to deepening local production, opening local financial markets to foreign entry in a selective and targeted manner can provide an effective impetus to innovation and can help unlock a process of financial deepening in support of the broader economic diversification process. Chapter 2: Cross-Border Banking: Benefits and Risks 87 Even in those cases where foreign entry falls short in delivering on the financial deepening agenda, foreign banks often contribute to the funding of larger corporations in countries with highly concentrated real sectors. A small number of companies and sectors implies risky bank portfolios, and cross-border banks can contribute to hedging the consequent risks using their international balance sheets. Despite the potential gains from foreign bank entry in Africa, the benefits to date have been rather narrow, and have only occasionally encompassed the innovative banking models alluded to above. Indeed, in many cases, the opening of banking markets has led to the crowding of narrow markets already serviced by existing financial institutions, thereby only adding to the overheads of banking systems that service a rather narrow client base. Several factors are at play here, not least the rather reluctant commitment of country authorities to harmonize their banking regulatory and supervisory frame- works – as exemplified by reliance on subsidiary-based entry even among countries in the Central African CFA Franc and West African CFA Franc currency unions. Creating a (sub)regional financial market with harmonized regulatory frameworks would lessen the costs and encourage banks to operate across national boundaries. Perhaps as important, however, in determining host-country reluctance to enhance the engagement of foreign banks and use their comparative advantage to harness the financial deepening, are concerns about the distribution of economic benefits arising from greater foreign bank penetration. While the gains from local financial deepening and associated enter- prise development and economic growth clearly more than outweigh the profits accruing to foreign banks, there are concerns about whether the new generation of South-South banks are replicating the role of the colonial banks of the 1960s and 1970s. At the same time, as financial integration continues to increase, the burden will be on the authorities to mitigate the risks arising from new channels of contagion. While cross-border banking has enhanced financial integration in recent years, the depth of the large majority of financial systems across Africa remains relatively low, so the potential for contagion remains confined. 3. Cross-Border Bank Regulation: Rationale and Tools 90 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 3. ccccc Cross-Border Bank Regulation: Rationale and Tools This chapter discusses the rationale for cross-border regulatory cooperation in general and the specific reasoning behind such regulation where financial institutions operate across borders. The potentially diverging interests and incentives of home and host country regulators are key challenges for effective supervisory cooperation. As super- visors focus on costs of bank failure within their own jurisdictions, they may fail to take into account those costs that fall outside their geographic perimeter and domestic responsibility, which can lead to distorted decisions both by home and host country supervisors. Traditional tools of cross-border cooperation among regulators have focused on information exchange during normal times. Experience over the past years, especially during the global financial crisis, has shown that these tools are not sufficient. New and more refined instruments that put a stronger focus on bank resolution and crisis management are needed. However, these enhanced tools require the deployment of adequate supervisory resources and often a high degree of regulatory domestic and cross-border cooperation, thus posing challenges for effective implementation. Implementation costs will vary significantly across country pairs and sub-regions. In defining the actual framework for cross-border regulatory cooperation, national authorities therefore have to carefully weigh the costs and benefits. This chapter is structured as follows. Section 1 presents the theoretical arguments for cross-border regulatory cooperation. Section 2 discusses international standards as a reference point for such cooperation, and section 3 assesses traditional tools of cross-border regulatory cooperation, including consolidated supervision, Memoranda of Understanding, and Colleges of Supervisors. Section 4 addresses extended forms of cooperation that focus on bank resolution and crisis management. Section 5 argues for a context-specific adoption of tools and mechanisms for cross-border regulatory cooperation. Section 6 concludes. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 91 3.1 Why Cross-Border Regulatory Cooperation? Until a few years ago, cross-border banking activities in Africa received scant attention from regulatory authorities. Historically, cross-border banking in post-colonial Africa was dominated by European banks, and supervisors in low-income countries with limited capacity were comfortable to cede responsibility for oversight of these large, often complex banking groups to the banks’ home countries. On the whole, this policy worked well except in the cases of the collapse of two international banks with significant operations in Africa in the early 1990s – BCCI and Meridien Bank – which resulted in the closure of their subsidiar- ies across Africa (see Box 3.1). More recently, the global financial crisis that resulted in the failure of numerous banks in the developed world drew attention to the imperative of closer cooperation across countries in the areas of supervisory intervention, crisis management, and provision of taxpayer support to failing banks. In addition, the emergence of regional African cross-border banks in recent years implies a significant increase in the responsibility of African home supervisory authorities. However, the rapid pace of expansion of cross- border banking operations has, in many instances, not been matched by adequate cross-border supervision. This may be due, in part, to the widespread (mis)perception among African banking supervisors that the requirement of establishing stand-alone sub- sidiaries in host countries will keep their banking systems out of harm’s way. The confidence of African supervisors in the protection offered by the incorporation of subsidiaries was severely dented by the experience of large European banks, such as the Benelux Fortis Group, during the global financial crisis. The deepening of South-South investment in banking in Africa is a relatively recent phenomenon. Until just a decade ago, the operations of African cross-border banks were still relatively thinly spread across markets. However, as described in Chapter 1, the banking sector landscape has since changed markedly. The phenomenon of relatively small invest- ments (from the home country perspective) constituting a systemic presence in host countries is no longer limited to international banks. In the larger financial markets in Africa, such as Kenya, Nigeria, Morocco and South Africa, the former dominance of European banks has been replaced by African banks entering each other’s markets and assuming a dominant position in the smaller markets in Africa. The eight pan-African banks alone (see Figure 1.7) account for 20 percent or more of the respective host country banking system assets in 24 cases.17 In some smaller markets, the presence of individual banks in host country systems is very dominant, for example Ecobank in Liberia and Central African Republic (both 40 percent), Standard Bank in Lesotho (47 percent), and Barclays Africa Group in Seychelles (44 percent). There are few, albeit significant, examples where non-African banks play a similarly dominant role in African host markets, including Portuguese banks in some former colonies. 17: Some countries are host to more than one of such large cross-border bank. 92 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools Box 3.1: Complex Structures and Resolution of Cross-Border Banks: Learning From History In the early 1990s, the case of the Bank of Credit and Commerce International (BCCI) focused the attention of banking supervisors on the implications of complex and cross-border structures for effective supervision and problems in the event of liquidation. BCCI had a Luxembourg holding company and two major subsidiaries: one in Luxembourg con- trolling 47 branches and 2 subsidiaries in 15 countries, and one in the Cayman Islands controlling 63 branches in 28 countries. In addition, there was a raft of subsidiaries and associates with 255 offices in 30 countries. However, the real issue was that the Head Office was domiciled in a supervisory jurisdiction (Luxembourg) where only a very small part (about 2 percent) of the group’s activities took place, making it impossible for the Luxembourg supervisor to carry out effective consolidated supervision under the international rules prevailing at that time. The principal operating businesses lay elsewhere. There were substantial businesses in the USA, where several significant abuses occurred, and in the UK with a large treasury operation and in many senses “the heart and mind” of the bank. The senior management resided in Abu Dhabi, where some of the main shareholders also resided. Corporate governance was not assisted by the fact that the bank had more than one auditor. When the liquidator was appointed to BCCI, supervisors soon realized that differing insolvency legislation across jurisdictions was going to make it difficult to ensure even-handed treatment of all the bank’s creditors around the world. Some jurisdictions liquidated on a separate entity basis: that is the assets in that jurisdiction were used first to satisfy the claims of creditors in that country. The United States has such a law. Other countries, such as the United Kingdom and Luxembourg, operated on a single entity basis where the assets of the whole group would be applied evenly across creditor claims in all countries. This left the liquidator with a very complex series of liquidation actions to manage. The second significant case in the 1990s was that of Meridien Bank, which is well known to African supervisors. Meridien International Bank Ltd (the holding company) was registered in the Bahamas and owned 74 percent of Luxembourg based Meridien BIAO which in turn owned a structure of 20 African banks. The whole structure was established by Greek Cypriot Andrew Sardanis who was resident in Zambia. As the operations around Africa collapsed, the Zambian authorities found themselves with a large problem in the Zambian bank, caused by money being channeled out in the attempt to prop up operations in other countries. The result of these two cases was a series of papers from the Basel Committee concerning cross-border banking, and complex and parallel banking structures. The documents draw atten- tion to the importance of effective consolidated supervision of such structures (as a minimum) and the free flow of information between banks and their supervisor(s), and between supervi- sors. One basic principle is that there should be a clear lead supervisor with adequate access to consolidated information and sufficient financing and personnel resources to be able to exercise effective supervision over the whole bank, wherever its operations may be based. Source: Based on Fuchs et al., 2006. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 93 Beyond the size of its operations in the host country, a bank can attain systemic importance in host jurisdictions in a number of other ways. One important factor is the financial conglom- erate nature of many cross-border banks. Besides the general risk of cross-sectoral contagion, there are other potential sources of risk such as inconsistencies in legislation across the banking, securities and insurance sectors that leave loopholes for regulatory arbitrage or double-leveraging of capital. Another trend that has received little attention is that complex banking groups headquartered in and outside Africa are not only expanding their operations within Africa, but also deepening their financial integration. Hitherto, the larger international groups have focused on servicing exporters and importers outside their home markets, but more recently this profile has been changing, and banks are now exporting innovative business models to other markets. This is a welcome development, but as already acknowledged in Chapter 2, it brings with it greater supervisory responsibilities. For example, contagion risks may become more prevalent: as the depositor base expands and technology allows swifter withdrawals, bank runs – so far a limited phenomenon in Africa – might become a real risk in case of bank fragility. As South-South investment and financial deepening progress, greater demands will inevi- tably be placed on banking supervisors in Africa. In particular, African supervisors will be faced with starker choices. Where they have applied forbearance or sought recapitalization of banks from the fiscal authorities, thus largely preventing any banking crises since the early 2000s (with the notable exception of the Nigerian crisis of 2009), banking supervisors will increasingly be required to take more expedient action. This is due to the escalating fiscal costs of bank rescue, as banks expand across borders, grow and reap economies of scale. The need for a proactive role of supervisory authorities will be reinforced by the inevitable rationalization process that will be required in those instances where foreign banks enter a local market, but are unable to deepen their engagement beyond replicating the rather narrow business models of existing domestic players. It is only in recent years that the topic of cross-border banking supervision has received the attention of regulatory authorities in Africa. For instance, at its inaugural meeting in Pretoria in early 2012, the Financial Stability Board’s (FSB) newly established Regional Consultative Group for Sub-Saharan Africa identified the “growing risk of cross-border financial sector spillover effects” and the need to mitigate “contagion and regulatory arbitrage” in Africa as being among the main challenges and strategic priorities for the continent (FSB, 2012). It is important to remember that the regulatory focus on cross-border banks is not an objective in itself, but is rather a reflection of the opportunities and risks offered by financial integration, as discussed in Chapter 2. If the appropriate financial infrastructure and regulatory frameworks are in place, African economies stand to benefit considerably from cross-border banking in terms of financial deepening and outreach. Already the expansion of regional banks across the continent over recent years demonstrates the banks’ commitment to increased real integration (“follow your clients”), and a bet by these banks on improved economic opportuni- 94 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools ties. The challenge now facing these banks and their supervisors is whether enhanced foreign entry will be used as a key driver of financial deepening and integration. Given the potential disruption and costs (both in fiscal terms and for the process of financial deepening) associated with bank failures and systemic banking crises, supervisors have focused primarily on the stability aspects of cross-border banks. However, regulatory tools and instruments can also be used to harness the benefits from cross-border banking. While cross-border branching offers benefits in terms of cost efficiency – as the fixed costs of incorporation, separate internal systems and regulatory compliance are incurred to a lesser extent – African banking supervisors have focused on facilitating the use of stand-alone subsidiaries in most cases, even among those countries operating with a single currency and with integrated banking supervisory structures, such as the West and Central African currency zones. This is in recognition of constrained supervisory capacity and the fact that it is easier to supervise and resolve self-standing entities. However, even with a subsidiary-based model, regulatory tools and instruments can be used to harness the benefits from cross-border banks for host economies, for example by reducing the costs of doing business relating to foreign bank entry. Authorities can work towards: (a) reducing the complexity and length of the licensing process; (b) reducing initial capital requirements as applied to bank subsidiaries (with requirements designed to grow in line with the foreign bank’s business engagement and risk exposures); (c) reducing the requirements for establishing new branches where these exist (leaving, for example, decisions regarding the structure and security of bank premises to the banks); (d) encouraging full mobility of labor (skills transfer); and (e) encouraging usage of centralized, common IT (both for internal operations and the provision of clients services), and audit and risk management systems. These factors will contribute to greater efficiency in the provision of banking services and provide a platform for enhanced financial deepening. Externalities from bank failure The reason why the banking industry is among the most regulated sectors in most economies are the potentially sizeable externalities from bank failure.18 These externalities are related to three problems resulting from bank failure: First, the domino problem that results from banks belonging to a network and the interconnectedness of banks; the failure of one institution can easily result in the failure of other institutions in spite of sound fundamentals in these other banks. This can happen because of direct exposures of one bank to the other, but also indirectly through exposures to the same markets; if a failing bank is forced to sell assets, the consequent price drop might negatively affect the solvency position of another bank holding similar assets. 18: The concept of externalities refers to a situation where the costs or benefits of an activity are incurred by an unrelated third party. In unregulated markets, goods with positive externalities tend to be in short supply. Conversely, goods with negative externalities tend to be oversupplied since the full costs of the activity (e.g. pollution, systemic risk) are not borne by the suppliers alone but by society at large. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 95 Figure 3.1: Bank Failures and Interventions in Selected Countries Failures and interventions as percent of total banking assets over the period August 2007 to August 2009 100 80 60 40 20 0 Austria Belgium China Denmark France Germany Greece Iceland Ireland Italy Japan Kazakhstan Korea, Rep. of Kuwait Latvia Luxembourg Netherlands Portugal Spain Sweden Switzerland Ukraine United Kingdom United States Failed banks, fraction total banking assets (%) Government-assisted banks, fraction of total banking assets (%) Source: Laeven and Valencia, 2010. Notes: The column height indicates the sum of failed and government-assisted banks, fraction of total banking assets (%). Second, the hostage problem that results from the maturity mismatch and the inca- pacity of banks to satisfy the liquidity needs of all their deposit customers in the case of a bank run. This might lead to contagion effects throughout the financial system. Such bank runs can be on the retail level, as seen so often throughout banking history, but also on the wholesale level, as seen during the recent global financial crisis, where wholesale funding dried up rapidly and left many banks with nowhere to go for funding other than to their central banks. Third, the refrigeration problem that results from deteriorating lender-borrower rela- tionships due to the loss of information when a bank fails. Soft information is linked to a loan officer and the bank’s institutional structure, which quickly deteriorates if the institution fails. Particularly where banks engage in relationship lending, the disruption of the relationship caused by the failure of the bank can have a detrimental effect on credit provi- sion to individual clients and businesses, with negative repercussions for the real economy. While this discussion seems theoretical, history has shown how bank failures and systemic banking crises can cause havoc on real economies, resulting in both high costs for taxpayers and lost growth (Figure 3.1). While economic growth generally returns to its pre-crisis path some years after the crisis, it does so at a permanently lower level of GDP, thus making output losses permanent. 96 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools These three problems create costs that are external to the stakeholders that cause them, including risk decision-takers in banks, and are thus not, or not fully, taken into account in risk decisions. Even in the absence of bailouts or any financial safety net, banks will therefore take more risks than socially desirable. These external costs thus provide a strong case for a regulatory framework to guard against financial instability caused by bank failure and the negative impact of financial turmoil on economic growth and employment. It is important to note that these externalities do not constitute an argument for preventing bank failure at any price. Failure is part of the market process. Indeed, given the recent entry of new foreign banks in many small African markets, orderly exit of banks may be required, especially in markets where financial deepening disappoints and markets are overcrowded. Rather, the existence of externalities justifies a frame- work that helps minimize the external costs to the rest of the financial system and the real economy, stemming from such failure. This implies that it is critical to design the financial safety net, including the resolution framework, in a way that enables author- ities to efficiently and relatively expediently resolve banks. A well-crafted resolution framework also strengthens the hand of supervisors in sanctioning banks should they fail to live up to prudential requirements, and reduces the necessity for forbearance and bailouts. All too often, African banking supervisors take recourse in regulatory forbear- ance because local resolution frameworks are untried or untested. By instituting more robust resolution frameworks, banking supervisors mitigate the banks’ incentives for excessive risk-taking during normal times, as banks become aware of the readiness of supervisors to implement prompt corrective actions. Cross-border externalities from bank failure The failure of banks in one country can also cause substantial externalities beyond its borders for stakeholders in other countries – and increasingly so, due to the fact that the financial systems of countries have become more interconnected in recent decades. The extent to which cross-border externalities are systemic is much higher in financially and economically integrated areas, because the likelihood of banks facing stress simultaneously is larger in integrated areas. The costs of bank failure imposed on banking systems, economies, and potentially taxpayers in other countries through these cross-border externalities are not taken into account by domestic regulators and supervisors. They can be grouped into four categories (Beck and Wagner, 2013): First, externalities arise from cross-border activities of specific financial institutions. The failure of a bank that has foreign assets will impose costs on borrowers abroad by leading to lower credit availability to foreign firms. Similarly, the cost of foreign depos- Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 97 itors losing access to savings is not internalized by home country supervisors, leading to inefficient decisions. A case in point is Iceland, whose banks, from the perspective of the Icelandic supervisor, had a high share of foreign assets and deposits. With Ice- landic banks becoming increasingly reliant on foreign deposits at the time of liquidity problems in the months before their collapse, it can be argued that Icelandic supervi- sors had insufficient incentives to control their banks’ risk taking. Beck, Todorov and Wagner (2013) show that banks’ cross-border activities distort supervisory incentives, as evidenced by actual intervention decisions during the recent global financial crisis. Specifically, cross-border banks with a high share of foreign deposits and assets experi- enced intervention at a later, more fragile state by their home country supervisors, while cross-border banks with a high share of foreign equity were intervened earlier at a less fragile state. These findings are consistent with the costs of bank failure being borne by foreign depositors and borrowers, thereby providing the incentive for home country supervisors to delay intervention by exercising forbearance. One can use this analysis to also gauge externalities stemming from branches versus subsidiaries. If one considers integrated branches for which the home country super- visor is completely responsible, this can cause larger distortions, as these operations grow outside the control of host country supervisors and put therefore a premium on cross-border regulatory cooperation. On the other hand, the externalities are less in the case of subsidiaries if host country supervisors can intervene independently, although experiences from situations where host country supervisors have sought to intervene to safeguard the finances of subsidiaries – e.g., by preventing a foreign bank from withdrawing funds – have demonstrated how difficult it is in practice for host country supervisors to effectively ring-fence local subsidiaries. The high-level implication for international regulation is straightforward: in order to avoid these distortions, the geographic perimeter of the responsible supervisor should match the geographic footprint of the bank. Obviously, this is difficult if not impossible to implement in practical terms, given the size of banks’ geographic footprints, their vari- ability over time, and the fact that banks based in the same home country can be active in different countries and regions. On a more general level, these findings imply that the benefits from moving to supranational regulation and supervision are higher for regions with significant intra-regional cross-border banking activities. The emphasis here is on the degree of intra-regional activities, since a regional supervisor would not internalize costs arising from activities outside the region. On the bank level, these externalities are also influenced by the integration of central functions, such as liquidity and treasury management. The discussion so far is relevant for African supervisors from two different perspec- tives. First, as host country supervisors for large non-African banks, their concerns with respect to local borrowers and local depositors may not be taken into account by 98 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools home country supervisors in Europe, the United States or Asia. Second, in the case of large regional banks, African home country supervisors will be guided by home country interests and not necessarily by the interests of borrowers, depositors, or equity holders in other African countries. It is likely that these distortions will increase further in the coming years as financial integration increases. Second, externalities can arise even if there is no direct cross-border bank presence in a country. This includes externalities such as cross-border spillovers due to fire sales and common asset exposures, informational contagion among investors,19 direct inter- bank exposures, or counterparty risk. Exposure by banks to the same asset markets as the failing bank in another country is sufficient for this type of externality to occur. The more financially integrated the financial system, the higher the exposure. In the context of African financial systems, this second channel might be somewhat less relevant, as there is still relatively limited reliance on wholesale funding. African banks fund them- selves predominantly through deposits. Third, externalities arise from regulatory arbitrage. Banks have incentives to be regulated and supervised in jurisdictions with lighter regulation, and such jurisdic- tions benefit from an inflow of banking business. However, this can result in negative externalities for other countries, if and when lighter regulation leads to bank fragility or failure. Altogether, circumvention of supervisory oversight due to regulatory arbitrage – e.g. regarding licensing requirements, reporting standards, and observance of pruden- tial regulations – can have a pervasive impact on the solidity of the banking sector, and is a major concern, particularly in low-income countries where supervisory capacity is limited. Those cases where supervisors have had to intervene in Africa relate to the consequences of regulatory arbitrage of this kind, as discussed in Box 3.1 above. Fourth, specific externalities arise within a monetary union because a country cannot simply devalue its currency to regain competitiveness following a shock, but needs to rely on adjustment in domestic prices and productivity. While this process is inevitable, the fall-out of any significant shock will most likely impact bank balance sheets in the affected country rather immediately, resulting in the need to tap the resources of other countries in some form or other. This cross-country subsidization happens in the inter- 19: Bédard (2013, p. 2) explains informational contagion as follows: “According to informational contagion theory, contagion spreads because the financial difficulties of the initial bankrupt firm reveal information on a risk shared with other firms. Contagion occurs because the information needed to determine how similar firms, or securities, are affected by this third party risk is not immediately available, requires a costly analysis, and the creditors of these subsequent firms are risk averse. This type of contagion is manifested by bank runs, panics, and confidence crisis. It can lead to significant losses in the financial system without necessarily triggering bankruptcies, and affects solvent and insolvent institutions alike.” Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 99 ests of avoiding bank failures that would further exacerbate the impact of the shock. The fiscal costs from asymmetric shocks that affect different countries to a different extent are thus potentially much higher in monetary unions, as the experience of several countries in the Eurozone has shown over the past few years. A similar need to tap common resources might arise if the banking system is too large relative to fiscal revenue, and thus becomes too-big-to-save, as again the examples of several countries within the Eurozone have shown (Bertay, Demirgüç-Kunt and Huizinga, 2011). Further, relying on a common lender of last resort might result in tragedy of commons problems, as it is in the interest of every member government with fragile banks to “share the burden” with the other members through, for example, drawing on liquidity support by the joint lender of last resort. It is important to note that this externality applies on the systemic level, rather than just for individual institutions. The costs arising from this potential burden sharing, or rather burden shifting, across countries in monetary unions increases in line with the overall size of the bank- ing systems and the inter-linkages across borders within the union. Indeed, the poten- tial costs of such burden-shifting are important in explaining why the country authorities in the two CFA monetary unions have been reluctant to entertain closer integration of their financial systems. These financial systems operate with mostly small and segre- gated banking systems. Clearly, this imposes considerable opportunity costs, because the advantages of the currency unions in creating a larger economic space hinge on exploiting the economies and synergies associated with more integrated financial markets. Not all cross-border externalities are of equal importance. A crucial distinction arises between externalities related to specific financial institutions, and systemic externalities related to the overall banking system. It is important to identify the financial institutions for which substantial cross-border activities and thus externalities exist. Even if they might not be considered systemically important financial institutions in individual countries, their cross-border linkages might make them systemically important on a sub-regional or regional level. In the African context, this might relate to institutions that do not have a market-dominating position in any African country, but are present in a large number of countries. 3.2 International Standards as Reference Point for Cross-Border Regulatory Cooperation Given the strong theoretical case for cross-border regulation and supervision, as well as historical experience with the failure of large and complex cross-border financial insti- tutions, the Basel Committee on Banking Supervision (BCBS or Basel Committee) has developed a framework of international standards, best practices and guidelines aimed 100 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools at increasing the effectiveness of cross-border supervision. In fact, the establishment of the Basel Committee itself took place in the context of the cross-jurisdictional impacts of the failure of the cross-border activities (foreign exchange exposures) of the German Herstatt Bank in 1974. While there are a number of relevant Bank for International Settlements (BIS) publications, as outlined in Box 3.2, the key elements of cross-border supervision are enshrined in the Basel Core Principles for Effective Banking Super- vision (BCPs) (BIS 2006, 2012). The BCPs describe a set of best practices covering a comprehensive range of topics in banking regulation and supervision. They represent universally recognized standards for banking supervision and are used by countries as a benchmark for assessing the quality of their supervisory systems. Some of the BCPs are more directly aimed at banks’ cross-border activities. Compliance with the later (2012) set of BCPs could therefore be considered a reference point for a more detailed discus- sion of tools and instruments for regulatory cooperation.20 While there can be no doubt that international standards and best practices for banking supervision provide an important point of reference for the long-term reform agenda in Africa, it is also important to acknowledge that those standards have been developed mostly with advanced financial systems in mind. In Africa, the structure of financial markets, supervisory capacity, and the specific risks to financial stability differ significantly from those in advanced financial systems and financial deepening and access to financial services continue to be policy priorities. In recognition of this, the Basel Committee advocates a proportionate approach that allows for the BCPs and their assessment criteria to accommodate a diverse range of banking systems. While this proportionality applies to the BCPs across the board rather than to cross-border banking specifically, it is also important to recognize the impact of capacity constraints as they relate to implementation of those BCPs that are most closely related to cross-border banking. Box 3.3 discusses in more detail the relevance and applicability of interna- tionally accepted standards developed by standard-setting bodies such as the Basel Committee and the FSB for Africa. 20: While not the primary goal of the BCPs, convergence to certain standards can also facilitate cooperation between supervisors. Having similar regulatory and supervisory structures and processes in place facilitates mutual understanding and communication. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 101 Box 3.2: The Basel Committee’s Evolving Framework for Cross-Border Banking Supervision In 1975, the Basel Committee released its “Principles for the Supervision of Banks’ Foreign Establishments” (BIS, 1975), which became known as the “Basel Concordat,” in an effort to improve cooperation among supervisors and close existing regulatory gaps. The original Concordat set out the two fundamental principles: (a) no foreign banking establishment should escape supervision; and (b) supervision should be “adequate” and consistent across member jurisdictions. The latter principle establishes a responsibility both for the super- visor of the parent institution and for the supervisors of its affiliates abroad to ascertain whether the home (host) authority is able to undertake adequate supervision, and take remedial measures should that not be the case. A few years after the Concordat was issued, the Basel Committee reformulated some of its provisions (BIS, 1983) and added the principle that “banking supervisory authorities can- not be fully satisfied about the soundness of individual banks unless they can examine the totality of each bank’s business worldwide through the technique of consolidation.” In 1992, the Basel Committee issued guidance for the implementation of these general principles through its “Minimum Standards for the Supervision of International Banking Groups and their Cross-Border Establishments” (BIS, 1992), for the first time also involving non-Basel Committee member countries in a more systematic manner. In October 1996, the Basel Committee released a report on “The Supervision of Cross-Border Banking” (BIS, 1996), which advanced recommendations to overcome obstacles to effective consolidated super- vision. This report was subsequently endorsed by supervisors from 140 countries. In 2001, upon request of central banks and supervisors, the Basel Committee released the “Essential Elements of a Statement of Cooperation between Banking Supervisors” (BIS, 2001), which provided guidance on the content of bilateral Memoranda of Understand- ing (MoUs) between home and host supervisory authorities. It recommended convergence towards the essential criteria relating to information sharing, on-site inspections, confidenti- ality of information, and ongoing coordination. The next important step in creating a framework for cross-border supervision was the “High-level Principles for the Cross-border Implementation of the New Accord” (BIS, 2003), made necessary by the development of Basel II. The new accord accentuated the need for cooperation and coordination between home country and host country supervisors, since its provisions are applied at each level of a banking group, involving both home and host country supervisors in Pillar 1 (capital requirements) and Pillar 2 (supervisory review) assessments. The six principles set out in that report complemented the general framework set out in the Concordat in the specific context of the Basel II framework, and were followed by guidance on information sharing processes in the report on “Home-Host Information Sharing for Effective Basel II Implementation” (BIS, 2006). 102 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools The global financial crisis revealed gaps in intervention techniques and the absence of appropriate resolution tools in many countries. The Basel Committee’s Cross-Border Bank Resolution Group found that “actions taken to resolve cross-border institutions during the crisis tended to be ad hoc, severely limited by time constraints, and to involve a significant amount of public support” (BIS, 2010). The “Report and Recommendations of the Cross- border Bank Resolution Group” then set out a number of recommendations to bolster national resolution powers and their cross-border implementation; encourage mutual rec- ognition of national arrangements among relevant supervisors; reduce the complexity and interconnectedness of group structures and operations; and ensure firm-specific contingen- cy planning for all systemically important cross-border financial institutions. The recom- mendations also aim to reduce contagion by advocating the use of a set of risk mitigation mechanisms. In 2010, the Basel Committee issued the “Good Practice Principles on Supervisory Colleges” (BIS, 2010), intending to promote the use and strengthen the operation of super- visory colleges. In addition to explaining the background of the principles, the paper also provided some detailed implementation guidance. Finally, by end of 2011, the Financial Stability Board published the “The Key Attributes of Effective Resolution Regimes for Finan- cial Institutions” (the “Key Attributes”) to define the core elements that the FSB considers necessary for an effective resolution regime. Implementation of the Key Attributes aims to empower authorities to resolve financial institutions in an orderly manner without taxpayer exposure to loss from solvency support, while maintaining continuity of their vital economic functions. The document sets out twelve essential features that should be part of the res- olution regimes of all jurisdictions, including resolution authority and powers, safeguards, legal framework conditions for cross-border cooperation, Crisis Management Groups, and recovery and resolution planning. Sources: BIS website and d’Hulster, 2011. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 103 The BCPs (2012) most relevant for the supervision of cross-border banks include the following six principles: • Principle 3 stipulates the need for a framework to support cooperation with both relevant domestic authorities and foreign supervisors. Specifically, it requires that “laws and regulations or other arrangements provide a framework for cooperation and collaboration with relevant domestic authorities and foreign supervisors. These arrangements reflect the need to protect confidential information.” Essential crite- rion 2 states that: “arrangements, formal and informal, are in place for cooperation, including analysis and sharing of information, and undertaking collaborative work, with relevant foreign supervisors of banks and banking groups. There is evidence that these arrangements work in practice, where necessary.” Essential criterion 3 indicates that “the supervisor may provide confidential information to another domestic authority or foreign supervisor but must take reasonable steps to determine that any confidential information so released will be used for bank-specific or system-wide supervisory purposes and will be treated as confidential by the receiving party.” • Principle 5 establishes the licensing criteria: “The licensing authority has the power to set criteria and reject applications for establishments that do not meet the criteria. At a minimum, the licensing process consists of an assessment of the ownership structure and governance (including the fitness and propriety of Board members and senior management) of the bank and its wider group, and its strate- gic and operating plan, internal controls, risk management and projected financial condition (including capital base). Where the proposed owner or parent organi- zation is a foreign bank, the prior consent of its home supervisor is obtained.” In addition, according to essential criterion 10, “[f]or cross-border operations in its country, the host supervisor determines whether the home supervisor practices global consolidated supervision.” • According to Principle 7, “The supervisor has the power to approve or reject (or recommend to the responsible authority the approval or rejection of), and impose prudential conditions on major acquisitions or investments by a bank, against prescribed criteria, including the establishment of cross-border operations, and to determine that corporate affiliations or structures do not expose the bank to undue risk or hinder effective supervision.” Essential criteria make explicit how this applies to cross-border banking. “The supervisor can prohibit banks from making major acquisitions/investments (including the establishment of cross-border operations) in countries with laws and regulations prohibiting information flows deemed necessary for consolidated supervision. The supervisor takes into consideration the effective- 104 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools ness of supervision in the host country and its own ability to exercise supervision on a consolidated basis.” • Principle 10 sets out the requirements for supervisory reporting. “The supervisor collects, reviews and analyses prudential reports and statistical returns from banks on both a solo and a consolidated basis, and independently verifies these reports through either on-site examinations or use of external experts.” • Principle 12 establishes consolidated supervision. “An essential element of banking supervision is that the supervisor supervises the banking group on a consolidated basis, adequately monitoring and, as appropriate, applying prudential standards to all aspects of the business conducted by the banking group worldwide.” Essential criteria add that the home supervisor takes into account the effectiveness of supervision conducted in the host country where the bank has material operations, and that the home supervisor visits the foreign offices periodically and during these visits meets with the host supervisor. • Principle 13 defines home-host relationships. “Home and host supervisors of cross-border banking groups share information and cooperate for effective super- vision of the group and group entities, and effective handling of crisis situations. Supervisors require the local operations of foreign banks to be conducted to the same standards as those required of domestic banks.” Essential criteria indicate that: (a) the home supervisor establishes bank-specific supervisory colleges for banking groups with material cross-border operations; (b) the home and host super- visors share information on a timely basis; (c) home and host supervisors coordinate and plan supervisory activities or undertake collaborative work; (d) the home super- visor develops an agreed communication strategy with the relevant host supervisors; (e) where appropriate, the home supervisor working with relevant host authorities develops a framework for cross-border crisis cooperation and coordination and a group resolution plan; and (f) the home supervisor is given on-site access to local offices and subsidiaries of a banking group in order to facilitate the assessment of the group’s safety and soundness. African banking supervisors have made progress in compliance with these six princi- ples, although this progress is somewhat uneven across countries (see Chapter 4). Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 105 Box 3.3: How Relevant are International Banking Standards in the African Context? International banking standards aim to improve the quality of bank regulation and su- pervision and include the Basel I, II, and III regulatory standards on capital and liquid- ity requirements and the Basel Core Principles (BCPs) for Effective Banking Super- vision. Regulatory standards are essentially a common set of principles on the books and supervisory standards as best-practice guidelines.a Assessments of regulatory and supervisory practices are often undertaken in the context of the Financial Sector Assessment Program (FSAP) by joint World Bank and IMF missions and/or peers from other regulatory authorities. These assessments face challenges when applied to less developed financial systems. While they are no doubt comprehensive, they also set the bar very high for small and under- developed financial systems where not all standards are equally important. The Financial Stability Board and Basel Committee on Banking Supervision recognize this and have emphasized that the various regulatory standards such as for capital, liquidity, and leverage should primarily be applied to internationally active banks, and that the BCPs should be ap- plied proportionally across countries. However, so far little if any guidance has been provided as to what this means in practical terms. The October 2011 FSB-IMF-World Bank report on financial stability issues in Emerging Markets and Developing Economies (EMDEs) noted that “the more financially-integrated EMDEs – especially those that belong to the G20/FSB – should adopt the [Basel II/III] framework according to the agreed timetable. Other countries with less internationally integrated financial systems and/or with substantial supervisory capacity constraints should first focus on reforms to ensure compliance with the Basel Core Principles and only move to more advanced capital standards at a pace tailored to their circumstances”. To accommodate these countries, Basel II and III provide less demanding standardized approaches to calculating required capital levels. African supervisors in general agree that some of the Basel III regulatory tools might not be appropriate for low-income countries, and point to the need to develop other tools to further address specific risks inherent in low-income countries (Kasekende, Bagyenda, and Brownbridge, 2012). Basel II and III might also have unintended direct negative effects on the development of financial systems when applied prematurely. Specifically, they could have detrimental effects on the volume and conditions of international banks’ lending portfolios to developing countries. For example, by favoring larger firms over smaller ones, the Basel II and Basel III risk-weighting systems may induce banks to reduce exposure to riskier assets, including lending to SMEs in developing countries. Similarly, the new Basel III liquidity re- quirements, the Liquidity Coverage Ratio, and the Net Stable Funding Ratio provide incentives for banks to shift from productive longer-term investments in developing coun- tries into assets with a shorter maturity that are eligible for the fulfillment of the liquidity 106 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools ccccc ratios (DFID, 2013). These regulatory changes can have a profound impact on foreign banks’ lending decisions and may be imported to developing countries through their foreign affiliates. At the same time, critical observers have pointed to the limitations of these standards. First, are these standards that have been developed and evolved in reaction to the develop- ment of banking systems in advanced markets correctly calibrated to the needs of low-in- come countries? As guidance for low-income country supervisors, in countries where there will naturally be limited capacity and significant gaps in achieving best international practice, the BCP assessment tool provides an across-the-board measure of achievement (or lack of the same), but insufficient guidance on which gaps need to be filled or prioritization among the gaps. So while helpful, the BCP assessment tool fails in this aspect, which is probably where low-income supervisors are most in need of guidance (Fuchs, Hands, and Jaeggi, 2010; Fuchs, Losse-Mueller, and Witte, 2012). Second, do the assessments undertaken based on these standards properly capture actual implementation and are they applied similarly in different countries? In monitoring the effects of agreed regulatory reforms on emerging market and develop- ing economies, the Financial Stability Board (2013) notes that “a cross-cutting theme is the lack of adequate resources and expertise in EMDEs to adequately respond to the numerous post-crisis global regulatory initiatives.” Third, are the same standards equally relevant for all countries? While the FSB references the need for “targeted, well-coordinated technical assistance,” as yet little guidance is available as to how the efforts of supervisors in emerging market and developing economies best sequence efforts. This skeptical view towards international standards does not mean that peer review of the regulatory and supervisory framework is not important. It rather means that one has to look beyond adherence to the principles (as indicated by the compliance scores resulting from standard assessments) to the actual implementation and functioning of banking supervision. It also implies that a stronger focus has to be put on identi- fying the relevant elements of the regulatory and supervisory framework within each country (as examples, see Kasekende, Bagyenda, and Brownbridge, 2012; Beck, Fuchs, and Witte, 2013). a. Similar principles as the BCPs exist for other segments of the financial system, including insurance, pension, and capital markets. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 107 3.3 Tools and Instruments of Cross-Border Regulatory Cooperation Based on the description of the BCPs most relevant for cross-border regulatory coop- eration, the traditional tools for regulating and supervising cross-border banks have been the consolidated supervision of banks, the signing of Memoranda of Understanding (MoUs), and establishment of Colleges of Supervisors (CoS). All three tools focus mainly on information – both collection of the necessary information about the financial health of financial institutions by the relevant supervisory authorities, and exchange of information among home and host country supervisors. Consolidated supervision Consolidated banking supervision is among the most demanding functions which a supervisor has to perform. The difficulty stems as much from the complexity as from the sheer size of some groups’ activities. These challenges can be compounded in mixed activity groups or conglomerates whose parents or controllers engage in significant non-banking (e.g., insurance and capital markets) and other commercial activities. The cross-border dimension further adds to these difficulties. Consolidated supervision entails the supervision of (banking/financial/mixed) groups that include at least one bank. Such groups may be dominated by one or several banks, or may be a conglom- erate (a group that consists of at least two corporations, one of which is a bank and the other a non-bank financial institution or a commercial corporation). Consolidated supervision requires that (a) the structure of a group can be supervised; (b) its capital is effective and readily available to cover losses from risks elsewhere in its corporate group; and (c) its governance processes and systems are reliable and adequate to manage the risks of its entities on a solo and a group basis. The discussion of consolidated supervision is relevant at both the national and the cross-border level. At the national level, consolidated supervision is necessary to account for any domestic non-banking activities, including financial activities such as capital markets or insurance. At the cross-border level, consolidated supervision is necessary to account for all of the group’s international activities. Given the existence of intra-group exposures, the speed with which resources can be shifted across different parts of banks, and the general opacity of banking, it is insufficient to rely on information about the financial health of individual bank branches or subsidiaries. Even when operations are ring-fenced (see discussion below), risks can arise due to operational integration. Consolidated supervision builds on the mapping of groups, which requires the tracking of ultimate beneficial owners and their non-banking financial interests, collection of information on non-financial entities, and establishment of reporting requirements. Mixed groups are especially difficult to supervise, as they provide more opportunities to 108 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools leverage capital, obscure risk concentration, and circumvent prudential regulation. It is also quite common that groups do not necessarily present themselves as such and that incomplete information on ultimate beneficial owners, in combination with opaque group structures, greatly complicates the task of effective supervision. An effective mitigation of the risks associated with cross-border banking, such as double gearing of capital, accumulation of intra-group exposures and other types of hidden risk transfers, places considerable demands on home-host cooperation and coordination. Identification of groups and ultimate beneficial shareholders are key to consolidated supervision and banking supervisors therefore need to look beyond the level of bank and banking groups. If banking supervisory authorities limit their enquiry to only the direct shareholders of banks or their holding company, they may well not discover the “puppet-master(s)” and their collaborators who may have hidden their presence and interests in an impenetrable forest of intermediate holding companies and other struc- tures (such as trusts and foundations). Given that the purpose behind complex corporate structures is to conceal ultimate ownership and responsibility, a prerequisite for sound consolidated supervision is a flexible regulatory framework which empowers supervisors to exercise discretion where necessary. Accordingly, a risk-based approach is required – a formulaic or compliance-based approach to consolidated supervision will be inadequate, and may result in a false sense of comfort. The assessment of the effectiveness of capital for banks requires understanding how capital is being financed by significant ultimate beneficiary owners, and whether they use intermediary vehicles to borrow and leverage their own capital resources. For bank members of a broader group, whether local or cross-border, the supervisory authorities would need to ascertain where capital is located, and that it is available where the risks are ultimately borne, including that capital in a domestic subsidiary is not compromised due to undisclosed inter-group exposures. Memoranda of Understanding (MoUs) Memoranda of Understanding are legally non-binding declarations of intent to cooperate on certain issues. MoUs can be time limited or not; they can refer to general cooperation agreements or to cooperation on specific banks. They are typically established between supervisory authorities and cover “material supervisory concerns” related to branches or subsidiaries of one country’s banks operating in the other country. “Material supervi- sory concerns” are those related to compliance with relevant laws and prudential regu- lations, or those that pose a risk to the banking institution. Supervisors may cooperate on: (a) overseeing applications to open branches or subsidiaries in the host country, with the potential host supervisor informing the potential home supervisor, who will provide information on the applicant’s capacity to effectively operate a cross-border operation, including its solvency and fitness of proposed directors; (b) exchanging information on Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 109 events that may endanger the stability of banking institutions in the other jurisdiction, and on any penalties or other actions taken that would be relevant to the other supervisor; (c) cooperating in investigations of financial crime; and (d) cooperation in crisis situations. MoUs typically also include agreements on confidentiality and possibly arrangements for regular meetings between supervisors. An important aspect of MoUs is to establish arrangements to facilitate the flow of infor- mation on a continuous rather than periodic basis, and authorize supervisors to exchange confidential information. They should also sanction joint inspections and collaborative work, including local and group-wide supervisory risk assessments. MoUs are an important basis for cooperation between supervisors, both within as across countries. The main challenge of cross-border MoUs is the non-binding nature, as either party may decline to fulfill its obligations without penalty. This challenge is particularly relevant in times of crisis, where the incentives for the host and home country supervisors to share information diverge.21 The effects of such divergent incentives can be mitigated, but only if provisions from the MoU are translated into binding national legislation. However, even that might still raise constitutional issues, given that bank regulators and supervisors are ultimately accountable to national governments and taxpayers. Colleges of Supervisors (CoS) Colleges of supervisors are “multilateral working groups of relevant supervisors that are formed for the collective purpose of enhancing effective consolidated supervision of an international banking group on an ongoing basis” (BIS, 2010:1). Colleges are not meant to be decisionmaking bodies, but mechanisms for increasing cooperation, coordination, and flow of information to enhance the effectiveness of consolidated supervision of cross-border banks. CoS are typically established between countries with significant cross-border bank inte- gration. As in the case of MoUs, they can be for specific banks or for general cooperation. Colleges meet on a regular basis (once or twice a year) to exchange information and coordi- nate supervisory activities. According to Basel Committee (BIS, 2010) guidance, supervisory colleges should be structured to allow the home supervisor to exercise meaningful oversight of groups on a consolidated basis, while allowing host country authorities to be sufficiently represented to enable the home supervisor to benefit from their in-depth assessment of local subsidiaries. The level of representation should be a function of the economic importance of subsidiaries for the wider group, and of the systemic relevance of the bank for the host country. CoS should also provide effective platforms for information sharing and collaborative work. 21: See discussion below: Section 3.4 “Looking beyond sunny-day cooperation.” 110 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools While supervisory colleges can contribute to better cooperation among supervisors, they are no panacea: as financial systems become more intertwined, supervisory and regulatory weaknesses in one jurisdiction affect the quality of group-wide supervision, so the colleges will prove to be no stronger than their weakest link. Establishing an effective CoS comes with a number of challenges. One potential problem is that in the case of large multinational banks, not necessar- ily all host country supervisors are invited to participate. Home country supervisors are primarily interested in inviting host country supervisors with branches and subsidiaries that are significant for the bank’s operations. This might leave out host country supervisors with subsidiaries that are dominant in the host market, but not of material importance to the overall bank.22 Moreover, home country priorities shape the makeup of the CoS, which makes estab- lishing a level playing field among countries participating in supervisory colleges extremely difficult, and can weaken the incentives of participating countries. However, home country supervisors can assist in leveling the playing field by supporting supervisory capacity build- ing and regulatory strengthening in host countries. Another challenge for CoS is whether the right people are sitting around the table. While supervisors are the most relevant persons for day-to-day supervision during normal times, resolution and fiscal authorities are critical in the case of crisis management, be it in relation to idiosyncratic bank failures or systemic bank fragility. Expanding CoS towards Crisis Management Groups that include resolution and fiscal authorities might therefore be important in a crisis situation. Finally, there is the issue of committee decisions. Given that colleges of supervisors are informal rather than sanctioned by legal agreement, the accountability of supervisors to their countries, and the difficulties of taking and enforcing decisions in a group that lacks statutory authority, each supervisor is in effect free to take his/her own decision, even if not in line with the decisions of the committee or the interests of other supervisors. Ultimately, the final decision whether to intervene in the parent bank, with repercussions for subsidiar- ies elsewhere, lies with the home country supervisor. Consolidated supervision, MoUs about information sharing, and CoS each play an important role in supervising cross-border financial institutions and are closely interlinked. MoUs with guidelines for information sharing can be an important basis for consolidated supervision as can be CoS. MoUs can also provide the formal underpinnings for organizing CoS and such colleges can serve to put life into MoUs, especially when it concerns institution-spe- cific MoUs rather than general bilateral MoUs. Central America, a sub-region with highly integrated ownership of banking groups, provides an example for such interlinkages. The 22: A similar issue may arise with bilateral or multilateral MoUs. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 111 Council of Central American Banking Supervisors, Securities Supervisors and Other Finan- cial Institutions Supervisors, consisting of eight country members, used an MoU to set up a liaison committee in 2007, in order to facilitate effective consolidated supervision and enable the Council to coordinate action relating to financial conglomerates. Similar arrangements in Africa are still in their infancy. For example, a recently created master MoU among the five English-speaking West African countries covers home-host rela- tionships and consolidated supervision; and a CoS of these countries lays the ground- work for joint on-site examinations, with training for the supervisors being offered by the Central Bank of Nigeria. As mentioned above, these tools refer mostly to the exchange of relevant information between the home and host country supervisor concerning the health of the parent and subsidiary, respectively. The main challenge here is that only the exchange of hard infor- mation can be mandated, while it is often the soft information about a bank’s health, not necessarily reflected in balance sheet ratios, that is relevant for supervisors. As we will discuss below, experience suggests that supervisors will be particularly wary of exchanging information in times of evolving difficulties or potential crisis. 3.4 Looking Beyond “Sunny-Day” Cooperation: New Forms of Regulatory Cooperation Cross-border banking adds additional potential conflicts of interest to a regulatory and resolution process that is already, at the domestic level, characterized by conflicts of interest among different parties, including bank owners/management, bank creditors and depositors, bank regulators, and taxpayers. While interests of the home and host country supervisors align in normal times – if (a) the parent bank is of systemic importance in the home country, (b) the subsidiary is of systemic importance for the parent bank, and (c) the subsidiary is of systemic importance in the host country – there can be conflicts of interest when the financial situation of the bank deteriorates (D’Hulster, 2011). If the host country supervisor considers the survival of the parent bank to be endangered, the host authorities might consider ring-fencing the bank’s locally domiciled subsidiary, possibly resulting in withholding of necessary liquidity support to the parent to guarantee the survival of the bank under scrutiny. The externali- ties are likely to be higher among financially more integrated countries, since the hurdles to moving business across borders are lower. As long as the residual claimholders (i.e., equity and possibly junior debt-holders) are in the home country’s jurisdiction, the home country supervisor has incentives to “delay, deny and minimize” in cases where problems arise in the parent bank. As a result, the home country supervisor may limit information sharing with host country supervisors to avoid remedial actions by the host country, including ring-fencing, which would restrict 112 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools the parent bank from drawing liquidity from its subsidiary. Likewise, were problems to arise in the parent bank, the host country supervisor will, as soon as these become apparent, have every incentive to ring-fence the local subsidiary, especially if the sub- sidiary is of systemic importance in the host country. Similarly, the host country super- visor has every incentive to overstate problems arising in the subsidiary vis-à-vis the home country supervisor so as to attract fresh capital and/or liquidity and/or to justify ring-fencing to salvage whatever possible. Especially where the subsidiary is of material importance to the parent, the home county supervisor might have an incentive to force the parent bank to cut the subsidiary loose in terms of squeezing it of fresh liquidity, thus avoiding eventual contagion effects impacting the parent. As discussed by D’Hulster (2011), in situations where banks get into trouble, lack of specific ex-ante planning and commitment can lead to unilateral regulatory actions by home and host country regula- tors that exacerbate fragility. Specifically, a “regulatory run”, in which each supervisor tries to ensure that liquidity stays either with the parent or with the subsidiary and ring- fences the subsidiary, can exacerbate the financial costs of salvaging the bank. In addition to these theoretical considerations, the recent crisis in the developed world has cast doubt on whether the standard tools of cooperation are sufficient. During the global financial crisis, several large multinational banks and domestic banks with large cross-border exposure failed. While the standard tools of MoUs and CoS were often in place and these arrangements seemed appropriate for cooperation in normal times, they failed in times of fragility, where swift, decisive and collaborative intervention was needed. A striking example is that of Fortis, a financial conglomerate with significant interests in Belgium (where it was headquartered), the Netherlands and Luxembourg. Given the longstanding cooperation between the Benelux countries, one would have expected smooth cooperation in the fall of 2008, when Fortis got into serious problems. While in the first round of liquidity support, Belgian and Dutch supervisors still seemed to cooperate well, this cooperation failed when it became clear that solvency support was needed and national ministers of finance had to become involved. This not only gave the resolution a political dimension, but led to recriminations on the split-up of Fortis along national lines and cost-sharing between Belgium and the Netherlands. It is important to note that a subsidiary structure, such as is common across Africa, will not help in this situation. In the case of the Icelandic banks, which operated through both branches and subsidiaries across Europe, the impact of the crisis in 2008 was similar irrespective of how the Icelandic banks’ cross-border activities were organized. These events provide an important message for African supervisors, as they illustrate the pitfalls of placing too much reliance on ring-fencing as a means of defining the exposures and responsibilities of relevant authorities in crisis situations. One lesson from the global financial crisis is therefore that more emphasis has to be put on the specifics of cross-border cooperation in the resolution phase, beyond cooperation Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 113 arrangements used in normal times. High capitalization and liquidity levels across most of Africa’s banking systems,23 the increasing but currently still limited interconnectedness among the banking systems, and therefore the somewhat more limited immediate impact that bank failures would have on financial systems and real economies, all seem to suggest that resolution frameworks should be of less immediate concerns for legislators and regu- lators across Africa. However, it is crucial to remember that resolution frameworks also set incentives for banks and their supervisors during normal times. Provided certain trigger conditions are met, bank resolution regimes allow the supervisor to overrule shareholders during the pre-insolvency stage. In turn, this allows for good-bank– bad-bank solutions, whereby a set of liabilities and a corresponding set of performing assets are transferred to an acquiring bank, while the remainder can be liquidated. It also allows for continuity, thus avoiding the loss of private information inside the bank. The advantage relative to a scenario in which no special bank resolution regime is in place is that the decision for the authorities to intervene is no longer a binary one (that is, forbearance vs. liquidation). It thereby serves the public interest of financial stability at lower overall cost to taxpayers. This increased focus on resolution is reflected in the policy response by the G20 and the Financial Stability Board (BIS 2010). Specifically, there is an increased emphasis on the importance of improved cooperation, primarily through supervisory colleges; on enhancing flexibility in composition and design of college structures (“variable geometry approach”); and on the role of colleges in providing inputs to Crisis Management Groups, encompass- ing also central banks and finance ministries of home and host countries. The movement towards establishing Crisis Management Groups reflects the advice provided by the FSB’s Key Attributes paper (2011), which places increased emphasis on cross-border aspects of crisis management and includes resolution planning as part of institution-specific cross-border cooperation agreements. There are different ways of achieving closer cross-border cooperation between regulators with regard to resolution frameworks. The five options discussed below differ in the intensity and degree of legal commitment and level of cooperation. Focus cooperation towards crisis preparation The increased focus on resolution on the national level has resulted in crisis simulation exercises and the development of contingency plans. Fire drills can help to develop the necessary capacities of supervisory staff. They can also help expose shortcomings in communication among regulators and with the fiscal authorities, and in the existing 23: Supervisors in Africa have come to regard high capital levels as an extra cushion of defense in warding off crisis situa- tions. It is likely that in a more integrated and competitive banking environment banks will become increasingly unwilling to maintain capital levels that are higher than internationally accepted best practice due to the impact that this has on their cost structure. 114 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools bank insolvency framework, and thus guide the process of reforming the bank resolution system. Fire drills can also lead to the preparation of contingency plans for a major bank failure or a systemic banking crisis. Expanding such exercises to cross-border simu- lation exercises could be a first important step towards closer cooperation. Simulation exercises could also inform the process of developing cross-border contingency plans. Several African countries, including Namibia, South Africa, and Uganda have under- taken such crisis simulation exercises on a national level. While acknowledging the complexity of decisionmaking involving a number of country authorities, particularly under acute time pressure in crisis situations, an important next step in building aware- ness – not least to avoid a crisis situation – would be to undertake such exercises within sub-regions that are closely financially integrated or between countries that share responsibility for large cross-border banks. Expand MoUs and CoS to include bank resolution One of the major stakeholders in bank resolution, the Minister of Finance as the taxpayers’ representative, is invariably missing as a signatory to MoUs and as a participant in CoS. While cooperation between supervisors is important during normal times, it is also critical to involve all relevant resolution authorities in normal times, as part of the process of preparing for times of stress. Particularly for systemically important financial institutions, Crisis Management Groups should be formed that include the supervisory authorities, central banks, and finance ministries of jurisdictions that are home or host to entities that are material to resolving the crisis, and these crisis groups should cooperate closely with authorities in other jurisdictions where banks have a systemic presence.24 The tasks of such groups would include preparing for the recovery and resolution planning process for systemically important institutions under institution-specific cooperation agreements, and ensuring the resolvability of systemically important institutions. One example in this respect is the Nordic-Baltic cooperation (see Box 3.4), which could serve as an example for regions that already have interconnected financial systems. It is important to note, however, that widening the stakeholders in MoUs and colleges does not make the agreements more binding. Nonetheless, including ex-ante burden sharing on the agendas of such fora can provide a framework for discussing actions to be taken in the case of bank failures, and for facilitating expedient restructuring and resolution of weak cross-border banks. 24: See FSB (2011) Key Attributes paper. Where separate public authorities responsible for guarantee schemes or resolution authorities exist, these should also be included. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 115 Box 3.4: Nordic Baltic Cooperation Agreement on Cross-Border Financial Stability, Crisis Management, and Resolution Denmark, Estonia, Finland, Iceland, Latvia, Lithuania, Norway and Sweden recently signed a Memorandum of Understanding on cross-border cooperation reflecting the interconnected- ness of their banking systems. The notable difference between this agreement and other agreements is that it includes ministries of finance as tax payer representatives, an explicit focus on crisis management and resolution, as well as specific burden sharing agreements. The need for such an agree- ment was evident after the global financial crisis, which severely affected some of the Nordic countries, and in light of close cross-border banking links. The enhanced Nordic cooperation recognizes that cross-border banks have brought great benefits (but also risks) to both home and host countries. Historic links across the signatory countries and a tradition of generally good cooperation facilitate such a pro-active approach to joint crisis management and resolution planning. Legally binding cooperation As discussed above, MoUs are legally non-binding declarations of intent, the value of which drops dramatically in times of distress. For cooperation to become legally binding, it has to be mandated in banking legislation. However, concerns may arise as to whether such legally mandated cooperation might violate the sovereignty of the collaborating authorities. The cooperation among regulators in Australia and New Zealand is an example of legally binding cooperation. The two countries are closely integrated financially, with branches and subsidiaries of Australian banks dominating New Zealand’s financial sec- tor. This has led to extensive cooperation and information sharing between the Australian Prudential Regulatory Authority (APRA) and the Reserve Bank of New Zea- land (RBNZ), facilitated by similar levels of regulatory development, joint history, and a common legal tradition. The cooperation is further strengthened by a 2006 amendment to the Reserve Bank of New Zealand Act, which legally obliges the RBNZ to cooperate and consult with Australia’s financial supervisory authorities to try to avoid actions that may negatively affect financial system stability in Australia. The Australian Banking Act was amended in similar manner. There is also a Trans-Tasman Council on Banking Supervision (which includes the RBA and the Australian and New Zealand Treasuries) that meets on a regular basis. Recently, a Memorandum of Cooperation on Trans- Tasman Bank Distress Management was drafted. 116 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools Supranational supervisor A supranational supervisor is the strongest form of regulatory cooperation across borders if it is set up to match the geographic perimeters of banks and regulators, as it effectively delegates the regulation and supervision of banks to a regulatory authority responsible for several countries. Such a supranational supervisor internalizes costs of bank failures that are external to domestic supervisors and thus helps to overcome coordination problems. There are several examples of supranational supervisors both in and outside of Africa. The most recent example is the Eurozone, in which the European Central Bank (ECB) is responsible for the regulation and supervision of the largest banks in the Eurozone as part of the Single Supervisory Mechanism. It is important to note, however, that the existence of a supranational supervisor does not, in itself, resolve the problem of cross-border externalities. Other important factors include the mandate of the supervisor within the financial safety net, in terms of inde- pendence and powers, and the degree to which the overall financial safety net, including the resolution powers and resources, are moved to the supranational level and linked to supranational supervision. The discussions within the Eurozone have been exactly along these dimensions, complementing the Single Supervisory Mechanism with the nec- essary resolution mechanisms and resources. In the context of the Eurozone, this has taken on increasing urgency, as the supervisory responsibility has been partly shifted towards the ECB, while – at the time of writing this report – resolution authority remains on the national level, though coordinated across countries. Given limited fiscal space, particularly in many peripheral countries, but also given the incentives for burden-shift- ing that occur when members of a currency union are confronted with bank fragility (see discussion above), many economists fear that this split of responsibility will neither resolve the current crisis nor lay the foundations for a sustainable financial safety net for the Eurozone. The CFA zones in West and Central Africa are similar examples of situations where formal supranational structures exist and supervisory authority has been assembled centrally, but responsibility for bank resolution continues to lie with the respective country authorities. This unstable constellation has not given rise to the same political discussions as in Europe in recent years. This is due partly to the relatively low depth of the financial systems across the two CFA zones, which in itself limits the fiscal ramifi- cations of potential banking crises; and partly to political pressures from local country authorities, which oblige supranational supervisory authorities to sanction forbearance for protracted periods, thereby postponing the realization of the escalating fiscal costs of bank restructuring and eventual resolution. In such circumstances, the continued existence of weakly capitalized financial institutions will inevitably put upward pressure Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 117 on banking spreads (as banks delay rationalization of their operations and seek to re- cuperate high provisioning costs), and thereby contribute to further delaying the process of financial sector deepening. Stand-alone subsidiaries and ring-fencing Especially in the case of asymmetric interests, such as a market-dominating sub- sidiary in the host country that is not of material importance to the parent bank and thus to the home country regulator, it is important to explore more robust supervisory arrangements. Given these asymmetric interests and the limited or negligible influence that host country supervisors have vis-à-vis home country supervisors, many African supervisors have opted for stand-alone subsidiaries, with a view to being able to impose ring-fencing should the parent bank get into trouble. This is in spite of the efficiency losses that this approach entails. From the viewpoint of host country supervisors focusing on preserving the stability of their local financial system, ring-fencing is a sub-optimal but rational response to the lack of better international cooperation. As noted above, in recent cases during the global financial crisis in Europe, ring-fencing proved to be inef- fective, and there is no guarantee that ring-fencing in case of fragility will shield the host country effectively against considerable negative repercussions. Most importantly, the host country will suffer significant, sustained efficiency losses from forcing subsidiaries to be stand-alone, with negative repercussions for financial deepening and inclusion. Finally, in times of crisis, authorities might consider establishing a regional coordi- nation group similar to the European Bank Coordination Initiative (EBCI) or Vienna Initia- tive (Box 3.5) to bring together supervisory authorities and the private sector to address problems in a quick and coordinated manner. The impetus for the establishment of the Vienna Initiative was the fallout from the global financial crisis, which severely impacted the cross-border banking activities of Western European banks in Central and Eastern Europe. The Initiative successfully leveraged the support provided by international financial institutions (IFIs) and large, regionally important international banks. In Africa, building on the initiative already taken by the Association of African Central Banks (AACB) to form the Community of African Bank Supervisors (CABS), a similar African initiative could bring together Cen- tral Banks from affected countries, the World Bank, the IMF, the IFC, the African Devel- opment Bank (AfDB), as well as representatives (CEOs) of relevant cross-border banks. 118 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools ccccc Box 3.5: The European Bank Coordination “Vienna” Initiative (EBCI) In November 2008, a number of banks with large presence in emerging Europe sent a letter to the European Commission to call for a quick and coordinated response to the problems in emerging Europe and, more specifically, to ensure sufficient funding for banks operating in the region. In response, the Vienna Initiative was created as a coordination platform for multinational banks, their home- and host-country supervisors, fiscal authorities, the IMF, and development institutions to safeguard a continued commitment of parent banks to their subsidiaries and to guarantee macroeconomic stability in emerging Europe. The EBCI was launched at the height of the financial crisis to provide a framework for coordinating the crisis management and crisis resolution of financial sector issues that were highlighted by the economic downturn and involved large cross-border bank groups systemically important in the emerging Europe region. Created in January 2009, it brought together public and private sector stakeholders of EU-based cross-border bank groups present in emerging Europe, including international financial institutions (IMF, the EBRD, European Investment Bank (EIB), and the World Bank (WB)); European institutions (European Commission and the European Central Bank as observer); home and host country regulatory and fiscal authorities of large bank groups; and the largest bank groups operating in the region. The Vienna Initiative’s aims are to: • Prevent a large-scale and uncoordinated withdrawal of cross-border bank groups from the region––this could have triggered a systemic bank crisis not only in individual countries but in the region as a whole. • Ensure that parent bank groups publicly commit to maintain their exposures and recapitalize their subsidiaries, as part of the overall balance-of-payments support to countries where IMF/EC macroeconomic support programs have become necessary (Bosnia and Herzegovina, Hungary, Latvia, Romania, and Serbia). • Ensure that national support packages of cross-border bank groups benefit their subsidiaries in emerging Europe and avoid a home bias in dealing with Europe’s banks. • Agree on basic crisis management and crisis resolution principles in the region. Host countries authorities are responsible for appropriate macroeconomic policies; liquidity support in local currency irrespective of bank ownership; and supporting their deposit insurance schemes. Parent bank groups–and the home country authorities behind them– are responsible for providing funding in foreign exchange and recapitalizing subsidiaries. • Strengthen cross-border regulatory cooperation and information sharing in the context of IMC/EC-supported programs and beyond. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 119 The EBCI has successfully completed its crisis-management phase: • It has helped resolve the “prisoner’s dilemma.” For crisis cases, external sustainability could only be assured if foreign banks remained engaged in the countries in which their subsidiaries work. • Commitments were upheld. Parent banks maintained the agreed exposure limits. This was critical particularly as the crisis proved to be worse and the recovery took longer than expected. Subsidiaries were also recapitalized according to stress-testing results. • Over time consultations allowed for “controlled” deleveraging. During regular reviews exposure commitments were reduced on a country-by-country basis. This permitted banks to increase room to manage liquidity internationally while still supporting exter- nal sustainability. • The Vienna Initiative informed as well as supported policy decisions in both home and host countries. For example, banking sector support packages of home country au- thorities were allowed for use in the subsidiaries of bank groups; monetary policy tools in host countries such as reserve requirements could be loosened to address weak demand with the assurances that additional liquidity will not be used for capital flight, creating pressure on the exchange rate. The Vienna Initiative was also used to examine issues that benefit from joint private-public sector assessments. The first two subjects were the development of local currency and capital markets and the role banks can play in helping with the absorption of EU structural funds. Recent research has shown the Vienna Initiative to be quite successful in preventing uncontrolled deleveraging of West European Banks in Central and Eastern Europe. Specifically, de Haas et al. (2014) find that while both domestic and foreign banks sharply curtailed credit during the financial crisis, foreign banks that participated in the Vienna Initia- tive were relatively stable lenders. As important, they find no evidence of negative spillovers from countries where banks signed commitment letters to countries where they did not. Source: Based on EBRD, 2012. 3.5 Cross-Border Regulatory Cooperation in the Presence of Heterogeneity Across Countries While the discussion so far has pointed to the need for cross-border regulatory and supervisory cooperation and offered different options, it is to be expected that such cooperation will reveal differences in incentives among supervisory authorities. Cross-border regulatory cooperation will need to take into account significant hetero- geneity across countries in terms of legal frameworks and policy preferences, as well 120 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools the often asymmetric importance assigned to banks by home and host country super- visors. These differences will give rise different feasible forms of cooperation. One size will not fit all, and tailored solutions are required. One source of heterogeneity is that countries differ in their legal and regulatory systems. The mandates and independence of regulatory authorities, their regulatory and supervisory tools, and supervisory assessment frameworks may vary significantly. This makes it hard to specify a common set of rules and standards in bank resolution, forcing adaptation of general principles to local circumstances. For example, anecdotal evidence suggests that conflicts between Belgian and Dutch supervisors over the Fortis resolution came about as a result of different organizational cultures across the respec- tive supervisory authorities. There may also be differences in the development of regu- latory and supervisory structures, including competence, sophistication, and resources that make cooperation more difficult. Different levels of economic development may also lead to different objective functions, with regulators in low-income countries potentially more worried about financial inclusion and financial deepening than regulators in middle-income countries (D’Hulster, 2011). In addition, these objective functions might vary over the financial cycle. A second source of heterogeneity arises from preferences. Countries may differ, for example, in how they view the role of government in the economy (one consequence being differences in state ownership), their focus on fiscal independence, or their risk tolerance. For example, a basic trade-off in banking (and finance more generally) is between risk and return; e.g., lightly regulated institutions may perform better under normal conditions but may be more prone to fragility, while heavy-handed regulation reduces the risk but may also depress banks’ profitability and their contribution to economic growth. Differences in risk tolerance can also lead to differences in the (potential) costs of bank failure. Third, there can be important asymmetries between countries, which lead to different objective functions and ultimately conflicts of interest between home and host country supervisors. Only if (a) the parent bank is of systemic importance in the home country, (b) the subsidiary is of systemic importance for the parent bank, and (c) the subsidiary is of systemic importance in the host country, are the interests of home and host countries aligned.25 For the home country supervisor, the most challenging case is if only the first two conditions are met, so that the host country supervisor will put less emphasis on the supervision of the subsidiary. For the host country supervisor, the challenging case is if either (a) or (b) are not met, the most prominent case being that of a market-dominating 25: Obviously, the interests are also aligned if none of these three conditions hold, but then there is no immediate supervisory concern anyway, unless both countries are part of a monetary union or have other forms of close financial integration, as argued above. Chapter 3: Cross-Border Bank Regulation: Rationale and Tools 121 subsidiary which is thus systemically important for the host country, while this subsid- iary is not systemically important for the overall bank and thus for the home country supervisor. This is a rather common situation for many supervisors across Africa with respect to the subsidiaries of cross-border banks. How much integration, how much cooperation? Different sources of cross-border banking externalities, heterogeneity in countries’ legal frameworks and policy preferences, and the often asymmetric importance assigned to banks by home and host country imply that appropriate forms of cooperation between countries necessarily differ depending on the circumstances. A taxonomy of five differ- ent forms of cooperation most adequate and feasible between different types of country pairs and within different types of sub-regions in Africa can be derived: First, in the case of countries or country pairs for which there are low externalities of cross-border banking, no heavy institutional solutions are necessary. Countries with a limited share of foreign ownership within their banking systems and limited integration with global financial markets, require only a limited degree of cooperation and can focus on harmonizing their regulatory and supervisory frameworks with international stan- dards, taking into account domestic priorities and sources of financial fragility. Second, in the case of countries that have been increasing their financial integration, there may be a high degree of heterogeneity. This is the case in West Africa, where the financial systems have developed differently in Anglophone and Francophone countries, while Nigerian cross-border banks are active in countries of both groups. These differ- ent developments would suggest a closer regulatory cooperation within than across the two different sub-regions, even though cross-border banking stretches across them. Given the high heterogeneity, a first step across West Africa would be to create colleges of supervisors and (preferably) resolution authorities, and undertake joint crisis simu- lation exercises. As discussed in more detail in Chapter 4, a first step has been taken towards establishing such a college of supervisors. Third, there is the case of asymmetric interests between the small host country and the large home country supervisor. This is the case for most African countries that are host to large European banks, such as Standard Chartered, Société Générale or BNP Paribas. Given that African supervisors are often excluded from CoS established by home country supervisors, host country supervisors have often, correctly, reacted by trying to force these subsidiaries to take on stand-alone structures. One approach might be for host country supervisors across Africa to join forces in being represented in home country CoS. Note that the case of asymmetric interests may also be applicable for countries that are hosts of pan-African banks. 122 Chapter 3: Cross-Border Bank Regulation: Rationale and Tools Fourth, there are more closely integrated sub-regions with lower heterogeneity, such as East Africa. Here, CoS and resolution authorities should be a first step, ideally leading to a situation where cooperative agreements become part of national banking legislation. Obviously, this would depend on upgrading and converging regulatory frameworks across the countries, especially in the area of resolution. As this sub-region moves toward a currency union, the establishment of a supranational financial safety net should be considered. Fifth, there are regions, such as the two CFA zones, that already have joint currencies and bank regulatory authorities, but with supervision still on the level of national bank- ing systems. These are, by definition, areas where changes in the exchange rate cannot be used to ease real sector adjustment, but as yet little effort has been made to exploit the possibilities of financial integration. Here, a push for more integration, including harmonized bank resolution practices and cross-border branching, accompanied by consolidated supervision, could be considered. This taxonomy is of course not static. Both the degree of externalities faced by coun- tries, country pairs and regions, as well as the degree of heterogeneity between countries, vary over time. As financial and economic integration between countries and within region changes, so does their exposure to externalities from cross-border banking. Similarly, heterogeneity across countries varies over time, which affects the level of cooperation. 3.6 Conclusion Cross-border regulatory and supervisory cooperation is essential for reaping the benefits of closer financial integration while managing its risks to stability. As much as African supervisors should be concerned about adjusting their regulatory frameworks to reap more benefits from cross-border banking, further expansion, especially of pan-African banks, requires both a significantly stronger focus on the necessary tools to properly supervise these banks and an adjustment of the current toolbox of cross- border regulatory cooperation. Consolidated supervision, MoUs and CoS are important tools and the basis for effective cross-border regulatory cooperation. The recent global financial crisis, however, has shown that these tools are not sufficient and that a greater focus on cross-border frameworks for bank resolution and crisis management is necessary, both in planning for such incidents and in setting the appropriate incentives. 4. Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 124 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 4. Current Issues in Cross-Border Regulatory ccccc and Supervisory Cooperation in Africa Financial systems across Africa are highly bank-centric and characterized by small size, a lack of depth and low intermediation efficiency, and therefore stand to benefit considerably from greater cross-border integration. In the African context, the key potential of regional financial integration lies in the ability of small financial systems to overcome the lack of scale that hampers their development and deepening. Through the exploitation of economies of scale and scope, regional financial integration can play an important role in advancing financial development and growth, if appropriate policies are pursued by the authorities. At the same time, new channels of contagion will and are indeed already developing as national banking systems and financial markets become increasingly interwoven, allowing for the transmission of shocks across borders. Under- development of markets and scarce supervisory capacity can both constrain the devel- opment of opportunities arising from closer integration and exacerbate contagion risks. Increasing regional financial integration thus underscores the importance of having in place adequate provisions for risk management and mitigation. It also calls for a greater commitment and adherence to common rules of the game, as embodied in international standards and sound practices, to foster greater confidence in the financial sector. Absent such commitment and adherence, cross-border integration would imply greater contagion potential and higher risks to financial stability. Thus, authorities in Africa face an important trade-off. On the one hand, they are faced with the traditional task of safeguarding their banking systems, taking into consid- eration the risks associated with the growth of cross-border banking activity in recent years. On the other hand, they are faced with the task of encouraging more efficient intermediation and deepening of financial markets, which will contribute to greater resilience and to a virtuous cycle in support of financial system stability and robustness. This trade-off is stronger in Africa than in other regions of the world, given the small size and low intermediation capacity of the majority of financial systems in Africa and therefore the relatively large scope for financial institutions and markets to grow. Impor- tantly, greater financial integration, if undertaken responsibly, can contribute significantly to achieving such enhanced efficiency and market-deepening. This chapter first discusses the policy options available to African banking supervi- sors for fostering both financial deepening and stability. Section 2 describes the current status of cross-border regulatory cooperation. Section 3 concludes. Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 125 4.1 Harnessing Financial Integration to Promote Financial Deepening: Policy Choices Faced by Regulatory Authorities Banking supervisors in Africa have reacted differently to the dual challenges of main- taining stability and fostering more efficiency to promote market deepening. Two of the traditional tools in the supervisors’ arsenal have proven to be crucial in this respect. The first is licensing policy. Licensing regulations stipulate a set of requirements that apply to newly established banks, whether capitalized domestically or as subsidiaries of foreign banks. Countries in Africa diverge quite markedly as to how they apply such requirements. On the one hand, supervisors in some countries feel obliged to license all those applicants that live up to the stipulated requirements. Supervisors in other countries approve banking licenses based on their assessment of the need for more participants in the domestic market, and a qualitative evaluation of what the prospective entrant has to offer in the form of new or innovative banking services. There is scope for such assessment, as licensing requirements normally include an evaluation by the licensing authority of the applicant’s strategic and operating plans,26 which may be taken to include the business plan. This approach is based on the realization that providing licenses to prospective entrants that merely conform to a set of licensing requirements can result in entry by a number of institutions with similar business profiles. The more subjective approach can be more conducive to financial market development, though it is also open to abuse. Given the dangers associated with a more subjective approach and the fact that entry of foreign banks will result in enhanced competition, it may be more expedient to rely on achievement of stipulated entry requirements. The trade-off, particularly in an African context, is that supervisors are generally overburdened and the entry of new banks stretches their capacity yet further. They also have little experi- ence in sanctioning and eventually withdrawing bank licenses, and liquidation of banks is a lengthy and often untested process. The second important regulatory tool is capital requirements. Increases in the minimum capital requirements can both discourage entry by foreign banks and force rationalization of the number of banks in the market. At the same time, a hike in min- imum capital may encourage banks to expand abroad, as there are only limited new business opportunities in the domestic market and banks will be eager to find opportu- nities for employing their capital efficiently. The Nigerian experience demonstrates that forcing consolidation by raising minimum capital requirements can be a risky option. The scramble to maintain traditionally high returns on equity led Nigerian banks to fund high-risk activities, resulting in the 2009 banking crisis as well as considerable fiscal costs and reduced financial depth. 26: This, at least, is what Basel Core Principle 5 prescribes. 126 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa Thus, while bank supervisors face the challenge of building capacity to manage risks associated with banks’ increasing cross-border presence, they also can exert consider- able influence on market forces through their administration of the regulations at their disposal. The question posed by this chapter is whether the pressures towards banking integration have been adequately matched by an upgrade in banking regulation, supervi- sory capacity, and cross-border cooperation among banking supervisors. African regulatory authorities face a broad array of policy choices beyond licensing and capital requirements, as well as a regulatory agenda relating directly to cross- border supervision, described in more detail below. In broad terms, the policies can be reactive, as is typically the case despite the generally higher costs of such an approach, or more proactively directed at harnessing the benefits of integration while managing its risks. The latter approach includes moving from policies characterized by forbearance to the effective sanctioning of banks and preparing for the eventuality of financial fragility in a proactive manner. Policymakers can adopt policies designed to leverage regional integration, such as (a) dismantling capital and exchange controls so as to promote more effective use of capital; (b) pursuing regulatory harmonization to reduce transaction costs; (c) establishing regional payments and settlement systems to encourage cross-border payments; and (d) removing any restrictions on the free mobility of staff and skills to encourage knowledge transfer. However, even in the politically more cohesive sub-regions of Africa, there has been only patchy progress in implementing such policies with a view to facilitating cross-border financial integration. For example, while efforts have been made to harmonize regulations across the EAC, progress in implementation has been slow, and removal of restrictions on the mobility of labor is still a highly contentious issue for some EAC member states. However, an encouraging recent development is that two regional payments and settlement systems were launched in 2013 – the East African Payments System (EAPS) and the payments system of the Southern African Development Community (SADC), the latter on a pilot basis. This slow progress within sub-regions may in part reflect wariness about the effects of greater regional financial integration on domestic markets – fears that more open- ness will result in greater market share for the already dominant banks from neigh- boring countries. In addition, supervisors’ efforts are hampered by limited capacity to undertake consolidated supervision and to manage complex group structures. As a result, national supervisors often continue to focus on safeguarding assets and increasing capitalization levels at domestic banks. While recognizing the potential upsides, national authorities may with reason also look at the expansion of cross-border banking with some degree of suspicion, as the immediate benefits of integration may not be equally distributed. In particular, they may have difficulty convincing themselves of the efficiency gains associated with greater Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 127 integration and scale. This comes in spite of recent analysis, for example, of the expansion of predominantly Kenyan banks within the EAC, which reveals that those banks actively pursuing expansion within the sub-region are also the more innovative banks, particularly in providing credit to the small and medium-sized enterprises cru- cial for sound economic growth (World Bank, 2013a). The same analysis also revealed – as one might suspect – that those banks expanding across the EAC are also among the more profitable Kenyan banks. However, looking beyond these immediate benefits of financial integration, the gains from trade in financial services for local economic growth and productivity can be expected to far outweigh the immediate benefits to particular pioneering banking groups. Across Africa, interest rate spreads, which are largely determined by bank over- heads and profits, remain high, and pressures to enhance efficiency appear to be weak. More importantly, expansion in cross-border activity has not as yet resulted in sus- tained pressure to reduce these spreads. On the one hand, this could be an argument for reducing risk premiums in banking through sound macroeconomic management, thereby reducing the “risk-free” interest rate on government securities (crowding-in the private sector); and for addressing weaknesses in the financial sector infrastructure. The latter might entail enhancing credit information-sharing and instituting less costly/ lengthy mechanisms for registration and foreclosure on fixed and movable property. On the other hand, efforts need to be made with regard to both domestic banks and international banks to enhance competition through more effective banking supervision, including policies that replace regulatory forbearance with effective exit policies. Such an approach would not only ease the burden on supervisors in overseeing a dispropor- tionate number of financial institutions (relative to the size of the financial sector), but could also contribute to fostering competition among those banks that remain, with knock-on effects on efficiency and interest rate spreads. Indeed, in addressing inefficiencies in banking intermediation in Africa, efforts will be required across this rather broad agenda. However, it is important to emphasize the complementarity between the national and regional agendas in promoting bank- ing efficiency and thereby the potential benefits of closer regional collaboration, as for example in exploiting a common approach to credit information sharing and agreeing on sanctioning of banks and exit policies. The scope for such collaboration varies across Africa and is taking hold as part of collaboration at the sub-regional level. Collaboration among the two CFA zones offers good opportunities in this regard, as these countries already enjoy a common legal framework and common currency. Also promising in this regard are efforts being undertaken by the EAC and within SADC. These have already resulted in collaboration on core elements of the financial infrastructure, not least the payments systems, and efforts are underway to broaden the collaboration to include credit information sharing and enhanced coordination on supervisory issues. 128 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa Reducing the costs of cross-border banking Across Africa, banking supervisors require that cross-border banking takes place pre- dominantly in the form of self-standing subsidiaries. Even in the two monetary zones in West and Central Africa, cross-border branching is quite rare. Where branches exist, they often face quite stringent regulatory requirements, including having the same capital levels as a subsidiary. The reasons for requiring self-standing subsidiaries are often well-founded and reflect the desire of bank supervisors to make oversight manageable both for themselves and for risk-managers within banking groups. Setting, overseeing and implementing reporting requirements for a self-standing legal entity is significantly more straightforward than doing so for a branch. In addition, the ring-fenced liquidity and capi- tal buffers of the subsidiary are expected to offer greater protection against contagion for the host country in times of crisis, although experience during the recent global financial crisis calls into question the strength of this protection.27 The preference for subsidiary-driven integration comes at a cost. By requiring banks to establish separately capitalized entities, each with their own board and risk-management functions, the subsidiary model limits the capacity of financial intermediaries to reap the full benefits of regional integration. While a subsidiary model does not preclude the deployment of shared IT systems and risk management methodologies, national super- visors can and often do require that locally domiciled subsidiaries set up self-standing IT systems and board and oversight committees. Such “indigenization” policies have, for example recently been implemented in Namibia.28 In addition cross-border expansion by banking groups through self-standing subsidiaries is likely to be associated with higher funding costs due to constraints on flexibility in the management of group commitments and on the flexibility to move funds to where they are most needed. 27: As was demonstrated very forcefully by the Icelandic banking crisis, in the absence of local incorporation requirements, small countries can become the source of major and systemic financial disruption. While foreign-owned banks typically operate as branches of the home-country bank in the European Union, in Africa most foreign-owned banks have hitherto been required to operate as locally-incorporated subsidiaries, thereby limiting the risk that a small country’s banks could assume obligations that its government would be unable or unwilling to repay. In addition, many countries impose additional requirements that supposedly enable national supervisors to ring-fence subsidiaries in times of crisis. However, the protection offered by subsidiaries is open to debate; experience with some banks during the global financial crisis, such as Fortis Bank, suggests that subsidiarization provides false comfort and only weak protection against contagion. 28: In Namibia regulations determine the localization in the country of all core banking systems so as to support effective banking supervision and payment system oversight. Concerns relate to security and operational reliability together with acceptable contingency arrangements for the timely completion of daily processing and observance of local payment system clearing and settlement cut-off times. The Bank of Namibia argues that with the implementation of the National Payment System, it was realized that shared IT infrastructure and banking systems with parent companies in South Africa introduced constraints contributing to operational and systemic risks for banking institutions operating in the Namibian payment system. It was also observed that Namibian commercial banks had problems to process Namibian domestic payments on South African public holidays and constraints experienced in adhering to Namibian payment system clearing and settlement cut-off times. It was also noted that IT governance, decision making processes and prioritization of IT projects with the parent companies in South Africa remained an obstacle to the progression of the payment system reform objectives of the Bank of Namibia. Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 129 Given the potential role of cross-border banking in forcing the pace of innovation and cost reduction through economies of scale, consideration should be given to whether measures could be taken to reduce the cost to foreign banks of establishing a foothold in new markets. Minimum capital requirements can be a significant barrier to entry in the case of banks with ambitions to expand across markets, particularly when considering developing activities in smaller, frontier markets where the opportunities for capital accumulation are limited and the allocation of capital is difficult to justify. In addition, many countries in Africa impose other costs on banks’ cross-border activities as part of their indigenization policies, such as requirements regarding use of local rather than expatriate employees, establishing local IT systems, developing local risk management and board structures, and even requiring foreign banks to seek significant local owner- ship. Such requirements result in “fortress banking” and can significantly limit achieve- ment of economies of scale. Indeed, rather than protecting their domestic banking systems, the authorities might look for ways of encouraging entry of those banks with a proven track-record in providing financial services targeted towards underserved potential client groups. This could, for example, take place in conjunction with the privatization of domestic banks. While entry by banks possessing relevant expertise is important, harnessing that expertise depends on a raft of financial sector policies designed to strengthen the local regulatory and financial infrastructure applied to foreign and domestic bank entry alike. This includes reducing the complexity and length of the licensing process, reducing initial capital requirements for bank subsidiaries (with requirements designed to grow in line with the bank’s business engagement and risk exposures), reducing or doing away with require- ments to establish new branches where these exist (leaving, for example, decisions about the structure and security of bank premises to the banks), encouraging full mobil- ity in the use of labor (skills transfer), and encouraging usage of centralized, common IT (for both internal operations and the provision of clients services), and establishing audit and risk management systems. These factors will contribute to greater efficiency in the provision of banking services and provide a platform for enhanced financial deepening. Another way of reducing the cost of cross-border banking would be to sanction cross-border branching with lower regulatory requirements than those for subsidiaries. However, as experienced in Europe, where bank branch “passporting” across national boundaries was introduced in the early 1990s, a deepening of cross-border banking integration via branches needs to be matched by a commensurate harmonization of regulatory and supervisory frameworks and financial sector safety nets to effectively contain the attendant stability risks. This is not yet in place even in more integrated areas of Africa. Cross-border branching may seem especially appropriate within currency unions, given that the necessary institutional infrastructure and macro- economic conditions are in place. Nonetheless, the CFA currency unions in West and 130 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa Central Africa have not resulted in integrated banking or regulatory systems. While the currency unions have been in place for some time, the overall regulatory framework still seems relatively weak, with key supervisory decisions (entry and exit) taken by national ministers of finance rather than by the regional banking commissions, and supervision even of cross-border banking groups done strictly along national lines. The policy agenda might therefore benefit from focusing on further bank integration accompanied by upgrading of the regulatory framework. The EAC has adopted a somewhat different approach and is planning to move towards a currency union at the end of a longer process of real and financial sector integration. In this case, it is important to accompany the existing financial integration process, as driven by the private sector, with accelerating integra- tion of cross-border regulatory frameworks, with the possible goal of a supranational regulator. 4.2 Current Practices and Challenges in Coordination Among African Regulators Current practices reveal a wide variation in cooperation among bank regulators and supervisors across the African continent, with some sub-regions and country pairs much more advanced than others. Convergence to international standards and regulatory upgrade One important condition to facilitate effective cross-border regulatory cooperation is a consistent regulatory and supervisory framework across countries. As already discussed in the previous chapter, this does not and should not imply copy-paste of regulations and supervisory practices across countries, which, given different historic development and legal traditions across the continent, would also be very hard to achieve. There are many reasons to encourage countries to harmonize their regulatory frameworks: to create a level playing field and to avoid the situation that banks operating in a country with a comparatively lax/stringent regulatory framework gain/incur an undue advantage/disadvantage compared to their peers; to create greater certainty as regards predictability and consistency in implementation; to reduce compliance costs, and provide the impetus to raise standards in more challenged environments; and to ensure that the risks associated with cross-border banking integration are effectively managed. As capital and investment flows are increasingly liberalized, harmonization becomes more important. Realistically, harmonization is a longer-term goal even within sub- regional groupings such as SADC and EAC, and in order to maximize mitigating risks and reaping benefits from the harmonization process, the authorities will need to pay attention to the sequencing and coordination of steps towards harmonization. Even Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 131 within the two sub-regional CFA currency zones, the Economic and Monetary Com- munity of Central Africa (CEMAC) and West African Economic and Monetary Union (WAEMU), challenges remain in implementing regulations in a harmonized fashion due to the lack of a coordinated bank resolution process. The Basel Core Principles for Effective Banking Supervision (BCPs) stipulate global best practices for banking supervisory authorities and are frequently used as a refer- ence for regional regulatory convergence. BCP assessments measure the gap vis-à-vis international standards in banking supervision and regulation, and offer the basis for an indicative assessment on which to measure progress. However, as already discussed in Chapter 3, in countries with underdeveloped banking systems and that face pervasive weaknesses and capacity constrained supervisors, care needs to be taken how these standards inform the regulatory reform agenda. In such countries, not all BCPs may be equally important in the short term, and supervisory effectiveness may benefit from pri- oritization and sequencing of reform efforts, with a view to focusing scarce supervisory resources on managing specific risks in the respective financial sectors and on closing urgent regulatory and supervisory gaps. Compliance with the BCPs is assessed in the context of IMF/World Bank Financial Sector Assessment Programs across the world, including for 18 African supervisory authorities over the period 2006 to 2013. Figure 4.1 illustrates the median compliance across those 18 supervisory authorities. The highlighted BCPs correspond to the six principles identified in Chapter 3 as most relevant for the regulation and supervision of cross-border banks: Principle 3 (cooperation and collaboration), Principle 5 (licensing criteria), Principle 7 (major acquisitions), Principle 10 (supervisory reporting), Principle 12 (consolidated supervision), and Principle 13 (home-host relationships. Note that the numbering of the BCPs in the Figure 4.1 refers to the old categorization based on the 2006 version of the BCPs, as assessments in Africa over this period are based on this older version. 132 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa Figure 4.1: Median Compliance with BCPs in Africa, 2006-2013 Compliant Largely Compliant CONSOLIDATED SUPERVISION HOME-HOST RELATIONSHIP SUPERVISORY REPORTING Materially LICENSING CRITERIA Noncompliant Noncompliant 1 1.1 1.2 1.3 1.4 1.5 1.6 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Source: World Bank. The values in the graph reflect the median compliance of 18 African supervisory authorities with the BCPs as assessed between 2006 and 2013 by IMF/World Bank assessors. The graph does not consider self-assessments carried out by African supervisory authorities. Overall, the median African supervisory authority of this sample is mostly “largely compliant” with the complete set of BCPs, as it is with five out of the six BCPs identified above as most relevant for the regulation and supervision of cross-border banks. How- ever, for the six BCPs of particular interest here, Figure 4.2 shows that the median can conceal large variations: • On Principle 3, cooperation and collaboration, the median African country is largely compliant. The large majority of African countries are considered to be either fully or largely compliant, with only four supervisory authorities materially non-compliant. In the wake of the experience from the global financial crisis, however, as well as given expanding regional cross-border banking within Africa, it is important to note that the standards have been raised over the past years, as we have discussed extensively in Chapter 3. • On Principle 5, licensing requirements, the assessment shows the median African country again largely compliant. Behind the median, however, is a wide variation, with only four of the 18 countries compliant and eight countries – almost half of the sample – materially non-compliant. This matches anecdotal evidence that most host country supervisors undertake a relatively superficial consultation process with home country supervisors before licensing subsidiaries of cross-border banks into their banking system. Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 133 Further, Principle 5 requires supervisors to assess whether a foreign bank setting up in their jurisdiction is subject to effective consolidated supervision by a compliant home supervisor. In reality, the capacity of banking supervisors to undertake effective consol- idated supervision is still limited and underdeveloped. While home country supervisors are increasingly aware of their responsibilities in regard to providing consolidated supervision, the emergence of cross-border banking groups is a fast-evolving phe- nomenon, and consolidated supervision is a relatively new, highly complex and evolving area of expertise in countries such as Kenya, Nigeria and Morocco and with WAEMU Banking Commission. There is a tendency for host jurisdictions to believe that, in the absence of consolidated supervision, adequate protection is provided for their own jurisdiction by requiring the capitalization of a subsidiary. While this gives some com- fort, it does not necessarily take account of all the risks that may arise and certainly does not negate the need for a host supervisor to continue ongoing liaison with the home supervisor of the bank, or indeed the supervisors in any other jurisdiction where it has significant operations, as already discussed above. • According to Principle 7, major acquisitions, supervisory authorities need to have the “power to approve or reject, and impose prudential conditions on, major acquisitions or investments by a bank, against prescribed criteria, including the establishment of cross-border operations, and to determine that corporate affiliations or structures do not expose the bank to undue risks or hinder effective supervision.” The median African country is again largely compliant, although only three countries are fully compliant. Five countries are materially non-compliant. • Principle 10, supervisory reporting, refers to the BCPs included in this list with which the assessed supervisory authorities demonstrate the highest compliance. Fourteen countries are found to be compliant or largely compliant, and only four countries materially non-compliant. • In contrast, Principle 12, consolidated supervision, confronts African supervisors with the greatest challenges; the median African country is materially non-com- pliant. Out of the 18 countries for which BCP assessments are available, only three countries were assessed as fully compliant and two countries were assessed as non-compliant. Nine countries were deemed materially non-compliant. • On Principle 13, home-host relationship, the 18 African countries are largely compliant. However, even here, seven countries are materially non-compliant. 134 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa Figure 4.2: Compliance with Selected BCPs Relevant for Cross-Border Banking Supervision in Africa, 2006-2013 12 10 Number of countries 8 6 4 2 0 Cooperation Licensing Major Supervisory Consolidated Home-Host Criteria Acquisitions Reporting Supersions Relationship Compliant Largely Compliant Materially Non-Compliant Non Compliant Not assessed Source: World Bank. The values in the graph reflect the median compliance of 18 African supervisory authorities with the BCPs as assessed between 2006 and 2013 by IMF/World Bank assessors. The graph does not consider self-assessments carried out by African supervisory authorities. Regulatory harmonization is a huge undertaking, and care will need to be taken to focus on key concerns, especially in an environment with severe capacity constraints. Prioriti- zation and sequencing are crucial, and the focus needs to be on those policy areas where harmonization is essential to the integration agenda. Effective implementation will inevitably require more time in the more challenged countries, and care will need to be taken not to rush through passage of legislation and regulations without a commensurate increase in supervisory and enforcement capacity. There has been progress in upgrading supervisory capacity and substantial attempts at convergence. For example, the EAC member countries are undertaking systematic efforts to harmonize prudential supervisory rules and practices in the region. As a first step, EAC central banks are encouraged to accelerate the development of national frameworks. Further, it is envisaged to develop a common set of definitions, concepts and organiz- ing principles governing EAC central banks’ stability function. In addition, the Banking Supervision Sub-Committee of the EAC has focused, among other things, on promoting credit reference bureaus, laying out a roadmap towards Basel II adoption, encouraging the development of a consolidated financial services regulatory framework, ensuring that licensing is vested with the central bank (not with the Ministry of Finance), and ensuring mutual consultation in the licensing process for new financial institutions and representative offices. Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 135 In SADC, the Protocol on Finance and Investment (FIP) sets out to enhance cooper- ation and coordination of regulatory and supervisory policies among the central banks in the region. It also encourages SADC member states to develop harmonized stan- dards of practice and regulations, thereby ensuring that all SADC central banks follow common procedures and operational frameworks. The long-term vision is to achieve a coherent and convergent approach to banking in the region as a foundation for SADC to move toward monetary union and a single SADC-wide central bank. The Committee of Central Bank Governors (CCBG) has developed a Model Central Bank Law. The objective of this model law is to harmonize central bank legislation across SADC member states, including the adoption of general principles facilitating the operational independence of central banks as well as accountability and transparency standards in the legal and operational frameworks of central banks. Another dimension of convergence is the coordinated strengthening of compliance with international standards for banking regulation and supervision, particularly as regards consolidated supervision. For example, in the context of the FIP, SADC mem- ber states agreed to work towards full compliance with the BCPs and conduct regular external and self-assessments to review progress towards this objective. Notwithstand- ing the regulatory convergence process in several sub-regions across Africa, significant differences in supervisory cultures continue to exist. It will take significantly more time for such a convergence to be achieved. Remaining challenges include revising respective national legal and regulatory frameworks to incorporate consolidated supervision, and developing relationships with other international supervisors by signing MoUs with both bank and non-bank regulators. Consolidated supervision Few countries in Africa have in place proper regulatory frameworks for consolidated supervision. This is becoming an increasing challenge as several African countries take on a more prominent role as home countries for large pan-African banks. Even where such frameworks are formally in place, it is crucial to look beyond the rulebook and consider effective implementation. One major concern is the question of who is being supervised under which regime. One sub-region where consolidated supervision, including identification of ultimate beneficial owners and wider group interests, causes serious concerns is in WAEMU, where most cross-border banking groups are managed by commercial holding compa- nies, as opposed to credit establishments. These holding companies do not fall under the purview of the banking law. While the WAMU Banking Commission (WBC) has some rights vis-à-vis these holding companies as owners of credit institutions, this is a grey area and WBC actions in this respect have been disputed. Even though WBC decisions 136 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa prevail in each of the WAEMU member states, it has limited power outside the region, even if a credit institution has subsidiaries or a parent inside the zone. De facto, WBC supervises each bank individually at the country level, even if it is a member of a group and WBC is the home supervisor. Consolidated data are not collected and even less analyzed. There is no systematic review of the holding group, its financial situation and operations, except when the holding group requests a license for a new member of its group (see Box 4.1). On a broader level, the complex corporate structures of many of the financial groups make them inherently difficult to supervise. For example, Standard Bank of South Africa and FirstRand both have complex holding company structures as well as many subsidiaries, which makes risk assessment very difficult. In addition, several of the big financial groups, including Standard Bank Group and Ecobank, reportedly have high intra-group exposures. The difficulty of acquiring data on intra-group exposures makes them much more difficult to monitor. Further, the large regional presence of these groups means that contagion from any one country of operation could potentially spread across borders. One country that has recently upgraded its system of consolidated banking supervi- sion is Kenya. Working with other regulators in Kenya and host regulators across East Africa, the Central Bank of Kenya has established and is implementing a framework designed to identify ultimate owners/beneficiaries in complex group structures in the financial sector. It is using the supervisory college for Kenya Commercial Bank, the most active Kenyan-based cross-border bank, as a pilot for implementing this framework. In summary, consolidated supervision is critical to minimize risks arising from the increasing bank integration across the region, but is also an area where significant upgrades are necessary in both legislation/regulation and implementation. Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 137 Box 4.1: Divided Responsibilities in WAEMU Based on international and African experience, delegating all major supervisory decisions, including licensing and intervention, to independent supervisors can strengthen bank stabil- ity. Despite the centralization of banking supervision within WAEMU, supervisory decisions by the WAEMU Banking Commission (WBC) are often still subject to approval by national government authorities. With regard to cross-border banks, supervision is still essentially performed on a national basis, with little reference to the fact that a bank under review may belong to a cross-border group. And despite the WBC being the single supervisory authority, formal responsibility for bank resolution continues to reside with national authorities, many of which are reluctant to take decisions to restructure and/or liquidate failing banks. The split between responsibility for undertaking banking supervision and the authority and resources to rescue or resolve banks is of particular concern with regard to cross-border banks, as these larger banks are often arranged in groups and can be affiliated with banking or non-banking group structures that pose particular difficulties for consolidated supervi- sion. Indeed, as of now, the WBC does not have the authority to supervise the holding com- panies of banking groups when the holding company is a non-bank. Given the existence of complex banking structures, opportunities do arise for malfeasance, and it is important that the authority of the WAEMU Banking Commission be expanded so that groups with banking interests are supervised on a fully consolidated basis. This focus on strengthening consolidated supervision is reciprocal, and also applies, for example, to the authorities in Morocco, Nigeria, and Libya, where several banks active in WAEMU countries are now headquartered. In these cases, the WBC as host supervisor needs to ensure that the home country authority devotes adequate capacity to effectively supervising the structures and activities of these banking groups. Memoranda of Understanding (MoUs) There has been an increasing use of MoUs across the region and between host country supervisors within and home country supervisors outside the region. There are at least 30 bilateral MoUs between bank regulators within Africa, more than 20 bilateral MoUs with countries outside Africa, and more than a dozen new MoUs currently under prepa- ration. In addition, there are several multilateral MoUs, including one among the EAC countries (discussed below in more detail) and one in the West African Monetary Zone (WAMZ) among Nigeria, Ghana, Sierra Leone, The Gambia and Liberia. Even where they are in place, however, often no formal protocols governing the frequency and methods of communication are in place. Most importantly, these MoUs cover only a small part of cross-border banking linkages. 138 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa To illustrate the shortfall in coverage, the WAEMU Banking Commission has signed five MoUs with other supervisors, whereas cross-border WAEMU banks operate in 30 countries. Bilateral cooperation agreements have been concluded with the supervisory authorities of France, Morocco, Nigeria, Guinea and the CEMAC. However, there are still 19 relevant countries with which WBC has neither a formal nor informal form of col- laboration. It should be recognized, however, that supervisory authorities in Africa have made significant progress in closing gaps in MoU coverage in recent years. In the SADC region, bilateral MoUs are the primary mechanism for facilitating information sharing. There are nine MoUs in which both countries are SADC members. In addition, South Africa has signed MoUs with three other African countries and 11 countries outside of Africa; Mauritius with seven non-African regulatory authorities. Importantly, the large majority of SADC MoUs fails to address resolution of cross-border institutions in the event of insolvency; Bank of Mauritius has started to include this type of provisions in its MoUs only recently (e.g., with Malawian and Indian authorities). South Africa has not yet devel- oped a cross-border resolution strategy, and is indeed currently looking to strengthen its legal framework governing insolvency and bank resolution – an effort directed in the first instance at establishing a coordinated framework for resolving group structures with the financial sector in South Africa. Notably, authorities in SADC appear unwilling to contem- plate handing over part of their powers to a regional institution, arguing that such a step should be preceded by further legal and regulatory harmonization. Colleges of Supervisors (CoS) Colleges of supervisors assemble supervisors from countries where a particular cross-border bank has its activities, and are usually convened by the home country supervisor. In discussing the use of colleges of supervisors in Africa, a distinction has to be made between the participation of African supervisors in colleges created by home country supervisors outside the continent and colleges within the continent. CoS of large non-African cross-border banks have been around for quite some time, but few African supervisors participate – mostly because they are not invited. This reflects the limited importance their subsidiaries have for the parent banks and consequently for the home country supervisors. There are only a few examples of African participation in supervisory colleges for non-African banks. These include Bank of Mauritius’ participation in the col- leges of State Bank of India, Bank of Baroda, Deutsche Bank, HSBC, Standard Chartered Bank, Barclays Bank Plc, and Standard Bank of South Africa. The Reserve Bank of South Africa participates in the Barclays CoS, which is not surprising given the importance of Barclays Africa Group as a growth driver for Barclays; as well as in the supervisory college for China Construction Bank. Another example is the Bank of Mozambique’s participation in the college for the Portuguese Grupo Banco Comercial Portugues. Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 139 There has been an increasing use of CoS for African cross-border banks, often supported by MoUs, but much more can be done. Kenya has started setting up CoS for its largest cross-border banks. As a precursor to these efforts to managing banking risks on a consolidated basis, the Central Bank of Kenya signed a MoU with other regu- latory authorities (for the insurance, capital markets and pensions industries) in Kenya in 2009. It is planning to institutionalize cooperation with relevant host country super- visors, and is being assisted by the IMF East AFRITAC in setting up supervisory colleges. The first one, the Kenya Commercial Bank Supervisory College, was launched in Nairobi in October 2012. Further colleges are planned in the near future. Currently, the only existing supervisory colleges in SADC are the two for Mauritius Commercial Bank and the State Bank of Mauritius, which convened for the first time at end-2013, and the college for Standard Bank of South Africa. There are regular meet- ings within ECOWAS, where central bankers and regulators discuss regulatory harmoni- zation and exchange experiences. While progress has been made in establishing supervisory colleges within the West African community (WAMZ) region, the impact is as yet rather patchy. There is a general “college of supervisors” in the WAMZ, which meets on a regular basis. However, this is a generic college and participation is by country representation rather than at a bank- by-bank level. The college is aimed at enhancing coordination, cooperation, and information exchange among supervisors in the WAMZ area in general, rather than at strengthening the supervision of a specific bank. At issue here is the fact that the geographical mapping of banks active in WAMZ may or may not match the composition of the WAMZ-based college. As yet there are no specific colleges of supervisors for Nigerian banks with foreign subsidiaries, nor is there a supervisory college for Eco- bank, which is headquartered in Togo and has subsidiaries across Africa. As a result, supervisors have limited capacity to verify information regarding intra-group liquidity and capital flows across borders. While subsidiaries may not be thought to be systemic from the parents’ point of view, they may nevertheless be systemic in the host countries such as Sierra Leone, Ghana and Liberia where they are located. Recognizing a common interest in managing home/host risks, the Central Bank of Nigeria and relevant host country central banks have commenced joint examinations of Nigerian banks in West African countries (The Gambia, Ghana, Guinea, and Sierra Leone). A general conclusion is that actual implementation is far behind formal agreements, which in turn are insufficient as a basis for effective supervision of cross-border banks in Africa. Cultural and historic factors can make cooperation between certain countries or within sub-regions more difficult. This may, for example, be the case for coopera- tion with South Africa, mostly due to historic reasons linked to the apartheid regime. However, there is also an issue of asymmetric size and hence economic importance, as is the case with Kenya vis-à-vis its neighbors or Nigeria vis-à-vis other countries in the 140 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa continent. This is important in terms of materiality of host country operations for the parent bank. Gaps in resolution frameworks and crisis preparedness There is increasing recognition that effective resolution frameworks are critical not only in addressing idiosyncratic and systemic banking distress, but also in influencing risk-taking by financial institutions in normal times. However, few countries in Africa have a separate bank resolution framework, and in most countries, either supervisors or courts can intervene in banks (Beck et al., 2011). Even in South Africa, supervisors have to obtain approval from the minister of finance to intervene in a bank. The lack of legal clarity in terms of the power of intervention and, more critically, the lack of a clear framework for resolving weak banks, undermine not only market discipline, but also supervisory independence with respect to banks. The main challenges in the domestic agenda thus include the lack of an appropriate legal framework to deal with failing banks. In addition, there is a need to establish and rehearse effective crisis response mech- anisms. Several countries in the region, including South Africa, Namibia and Uganda, have undertaken holistic reviews of the crisis management and bank resolution frame- works. Some countries have gone through crisis simulation exercises, which in turn is a good starting point for a more systematic crisis planning. As discussed in section 3.4, improvements in bank resolution on the national level should be used as basis for closer cross-border cooperation on crisis preparedness. Regional and trans-continental cooperation fora Efforts across the African continent to create sub-regional organizations and platforms for collaboration are numerous, and memberships are often overlapping. Experience with collaboration within these blocks is heterogeneous. In some instances – frustrated by the slow pace of change – groups of countries within the blocks are attempting to forge closer collaboration, while in other instances, countries are keen to disassociate themselves from working more closely together. SADC has two regional institutions that serve as coordinating fora for cross-border regulatory cooperation: the Committee of Central Bank Governors (CCBG) and the SADC Subcommittee of Banking Supervisors (SSBS). While the CCBG was established for the broader purpose of fostering the process of regional economic co-operation and integration, the SSBS was specifically established in 2005 to deal with coordination of activities relating to banking supervision. The SSBS was established on the basis of a Protocol on Finance and Investment and more specifically Annex 8, which “aims at Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 141 facilitating the effective application of international regulatory and supervisory stan- dards to banking systems in the region, and [at] harmoniz[ing] banking regulatory and supervisory matters across member states.” The SSBS, which includes two banking supervision representatives from each SADC central bank, meets once or twice per year to carry out the work program approved by the CCBG. The primary initiatives of the CCBG and SSBS relate to harmonization of regulatory standards, including (a) designing information technology platform designed to promote harmonization of banking supervi- sion; (b) drafting a Model Central Bank Act; and (c) drafting a Model Payment Systems Act and Model Banking Regulation Act. While it is clear that progress has been made, SADC countries are at very different stages in the process of adopting international good practices with respect to banking supervision. It is difficult to imagine that a home-country supervisor such as South Africa or Mauritius would be comfortable relying upon information from a country that is still attempting to apply Basel I, does not apply International Financial Reporting Stan- dards (IFRS), and is non-compliant with several relevant BCPs. Similarly, a host country supervisor in such a country would struggle to analyze and incorporate information from a home country with a more developed supervisory framework. Another regional effort to promote banking supervision harmonization is the Bank Supervision Application (BSA). Originally proposed by the East and Southern Africa Banking Supervisors Group (ESAF), the BSA was developed with CCBG support from 2002-2003. Hosted initially by the South African Reserve Bank in Pretoria, the BSA has since moved to Maputo. The BSA serves as a generic platform for a variety of banking supervision functions, such as licensing, on-site and off-site supervision, maintenance of supervisory databases, and enforcement actions. The BSA facilitates automated capture and validation of data from commercial banks and other reporting financial institutions. It also facilitates supervisory assessments, including assessments of BCP compliance. The BSA has been adopted by all SADC members except Botswana, Mauritius, Seychelles, and Tanzania. One of the main obstacles to strengthening regional financial integration is the existence of a multiplicity of legislative and institutional models among member states in SADC. The SADC Model Central Bank Law was approved by the SADC Ministers of Finance in July 2009. From the perspective of strengthening cross-border supervision, perhaps the most important provisions are those that establish the legal and operational autonomy of the Central Bank. In the absence of both de jure and de facto independence, efforts to harmonize banking supervision can be derailed due to political interference. The CCBG emphasized that the process of developing a Model Law that all parties could agree upon was more difficult than expected. Furthermore, it was challenging to estab- lish a timeline for compliance, since each country was permitted to proceed at its own pace (GIZ, 2012c). 142 Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa In East Africa, the common colonial history (in the case of the three original EAC members Kenya, Tanzania and Uganda), and close cooperation among the different central banks, facilitates the regulatory integration process; and the objective of a common cur- rency area imposes even higher requirements for regulatory integration than in the case of other sub-regions. At the same time, this integration also provides additional opportu- nities for reaping benefits in terms of further financial deepening and competition within the EAC. The cross-border cooperation among central banks on bank regulation and supervi- sion is part of a larger cooperation agenda that also includes monetary and exchange rate policy. Specifically, Article 85 of the EAC Treaty (on banking and capital market develop- ment) envisages the development and integration of financial markets and the associated legal and regulatory frameworks as prelude to the long-term goal of establishing the East African Monetary Union (EAMU). Several items in the article refer directly to bank regula- tion and supervision, including the call for harmonization of Banking Acts (item b), and the call for harmonization of regulatory and legislative frameworks and regulatory structures (item d). While the agenda on financial market integration goes well beyond banking and also covers the capital markets, insurance and pension sectors, the banking sector natu- rally features high on the agenda, given the importance of the banking sector within EAC countries’ financial systems and the role of central banks as the driving force behind the integration process. Subsequently, the EAC has undertaken several steps to institutional- ize supervisory cooperation among the five member countries. The Monetary Affairs Committee (MAC), formed in 1997, is composed of the member states’ central banks and is at the heart of regional supervisory cooperation. The main task of the MAC is to advance implementation of EAC decisions towards the envisaged full integration of the member states. MAC meetings are attended by the central bank governors as well as representatives of various departments of the central banks, including supervisory departments, and cover a wide range of areas and topics including monetary policy, bank regulation and financial inclusion. Working groups across the five central banks prepare these meetings and do follow-up work on the technical level. In addition, EAC central banks signed a multilateral MoU in 2008 to facilitate collabo- ration in supervision and information sharing with a view to increasing regional financial integration. The MoU has facilitated closer cooperation among the EAC authorities, including joint on-site examinations of banks and training opportunities. However, the MoU does not address crisis management issues. Further, it is important to note that the joint supervision activities currently do not serve the purpose of cross-border supervision, though typically they are undertaken for cross-border banks. The current more limited purpose is to facilitate exchange of experiences and foster a mutual learning process. Given the growing prevalence of pan-African banking groups that operate across countries without respect to their sub-regional affiliations, there is broad recognition Chapter 4: Current Issues in Cross-Border Regulatory and Supervisory Cooperation in Africa 143 that some financial institutions may be content to exploit potential opportunities that this could provide for regulatory arbitrage. On the other hand, if properly engaged, commercial entities might be a powerful lobby group to persuade the government authorities of the advantages of closer regional cooperation in establishing a level play- ing field and more open market access. Country authorities in Africa would be well- advised to leverage such potential private initiative and engage proactively in a private/ public partnership designed to strengthen the prudent expansion of banking integration. Finally, the Community of African Bank Supervisors (CABS) was established in January 2013 as a subsidiary body of the Association of African Central Banks (AACB). The CABS is intended to become a platform for maintaining and deepening the dialogue among African banking supervisors at the level of heads of banking supervision. As stated in a the communiqué of the inaugural meeting of the CABS, which took place in Algeria, the “CABS shall become a platform for exchanging views at the level of bank supervisors, learning from peers, reflecting on relevant global discussions and help in voicing the concerns of the continent. Its main objective is to contribute to ongoing efforts to strengthen banking regulatory and supervisory frameworks on the continent.” Going forward, the CABS may become a key forum on the continent to discuss issues related to cross-border cooperation. While such an exchange can play a decisive role in shaping the regional policy dialogue and capacity building agenda, day-to-day cooperation has to be implemented on either a bilateral level or within smaller groups focusing on specific cross-border institutions. The CABS can, however, serve as important forum to exchange ideas and experiences, drive the convergence to a common set of international standards, while at the same time facilitating the development of Africa-appropriate regulatory frameworks and providing the entry point for more detailed cooperation between individual countries or on individual banks. 4.3 Conclusion This chapter has discussed the current state of cross-border regulatory cooperation in Africa. Considerable benefits are still to be reaped from cross-border banking as a means of fostering financial deepening, while additional safeguards are needed. Current practices reveal that cooperation among regulators is still nascent and evolving. Despite progress in upgrading supervisory capacity and towards regulatory convergence in recent years, there is still a considerable gap between the international best practices as they relate to consolidated supervision, MoUs and CoS and the reality on the ground. In addition, the goal posts are changing, both in terms of what is considered good or best practice in cross-border regulatory cooperation, and in terms of the needed regulatory response, given the rapid development of cross-border banking integration across the continent. 5. Cross-Border Regulatory Cooperation in Africa: Looking Forward 146 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 5. Cross-Border Regulatory Cooperation in Africa: Looking Forward The integration of banking markets is on the rise across Africa. This report has made the case that there are significant benefits to be reaped from this trend. The potential benefits for financial and economic development are likely to be greater in Africa than elsewhere, given the still small and shallow financial markets. However, cross-border banking carries risks related to contagion and coordination failures among regulators. These risks will likely increase over time, as African financial systems deepen further and the importance of regional banks increases. This concluding chapter first offers a (far from exhaustive) list of policy recommendations that can help harness the benefits and mitigate the risks of cross-border banking. It then discusses some general observations regarding the cooperation of African authorities on cross-border banking issues going forward and the overarching conclusion that information exchange among authorities in Africa needs to be considerably strength- ened to effectively harness the benefits and mitigate the risks of cross-border banking. 5.1 Policy Recommendations Given the high level of aggregation, the general recommendations that follow have to be differentiated and tailored to match the country-specific circumstances and reflect the enormous variation in regulatory frameworks, stability needs across the continent, and the level of integration and regulatory harmonization within sub-regions. They also have to be adapted according to the intensity of cross-border banking linkages across country pairs and within sub-regions. These different areas for policy action are summarized in the following matrix (Table 5.1), which distinguishes among three policy objectives and three levels of decisionmaking. The three policy objectives are (a) reaping the benefits of cross-border banking; (b) safeguarding the real and financial sector against risks stemming from cross-border banking; and (c) preparing for cross-border repercussions of idiosyncratic and systemic bank fragility. With regard to decision-making, the matrix distinguishes among policy actions taken on (a) the national level; (b) the bi-lateral or sub-regional level, as often represented by the African Regional Economic Communities, or by multilateral groups of countries reflecting the scope of operation of individual cross-border banks; and (c) the African level. While the classifica- tion by level of decisionmaking seems clear-cut in theory, in practice these different levels can depend on each other, and incentives to take decisions on one level can be influenced by the ability to take decisions on other levels. The classification of policy recommendations according to policy objectives is also somewhat blurred, as the objectives are interrelated. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 147 For example, the benefits of cross-border banking can only be realized in a stable financial environment with credible crisis management and bank resolution practices. Hence, to the extent that certain regulatory and supervisory policies can achieve more than one of these three objectives, there might be an overlap. Table 5.1: Analytical Framework for Policy Action Reaping benefits Safeguarding Preparing for of cross-border stability of cross-border banking cross-border repercussions of banking in potential bank normal times fragility National level • Strengthen • Introduce or • Upgrade crisis financial improve management and infrastructure consolidated bank resolution • Move towards supervision frameworks more integrated banking models • Encourage entry of banks with innovative busi- ness models Bilateral/ • Enhance • Improve infor- • Extend subregional/ regulatory mation exchange cooperation on multilateral level harmonization between home crisis • Coordinate and host countries preparedness upgrade of • Build trust financial between home infrastructure and host country authorities African level • Enhance • Address gaps in • Take measures to regulatory consolidated avoid emergence of harmonization supervision of non-transparent, large African strongly banks interwoven groups with financial activities 148 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward Reaping the benefits of cross-border banking If Africa takes advantage of the efficiency gains and innovations that cross-border banks can provide, the continent stands to benefit substantially from cross-border banking in terms of financial deepening and increased outreach to previously unbanked parts of the population. Across Africa, bank spreads – largely explained by high overheads and profits – remain high, and pressure to enhance efficiency remains weak. Thus an important consideration for policymakers is how the benefits of cross-border banking – rather than predominantly benefiting corporate clients based on a profitable, but rather narrow business model – can be expanded and leveraged to the benefit of SMEs and households more broadly. A range of policy measures could be deployed to enhance the engagement of foreign banks and encourage them to engage in the deepening agenda. Many of these are germane to financial deepening, such as the strengthening of financial infrastructure in a consistent manner across countries that share strong cross-border links. In the context of cross-border banking, the focus would be on improving the comparability of credit information across countries; enhancing the efficiency of payments systems, par- ticularly as relates to cross-border retail payments and the servicing of migrant flows; strengthening mutual recognition of procedures for registration of property/collateral rights and mechanisms for foreclosure on collateral; and improving financial literacy and the availability of comparable information to consumers. Banks deliver highly dif- ferentiated services, and unbundling the true costs and benefits of packages of services can both enhance competition among banks and provide an important component of consumer protection. While many African countries still have to improve their financial infrastructure from very rudimentary levels, such upgrades might be best undertaken in a coordinated manner within sub-regions, as is already the case in some Regional Economic Communities. Bank clients would benefit from efficiency gains resulting from stronger competition from domestic and cross-border banks as a result of improved regional financial infrastructure. Other policies are more narrowly associated with improving the service delivery of foreign banks. The traditional model of cross-border banking in Africa – i.e., of banks following their large corporate clients across borders – provides the impetus for rather shallow involvement in host countries. In many instances, bank subsidiaries and branch networks remain small and banks are obliged to syndicate their lending or, more likely, book the loans to their larger clients on the their home balance sheets – a practice that is referred to as “suitcase banking.” Particularly given the rather shallow depth of most African financial markets, considerable upside potential lies in transfer of know-how, IT, infrastructure, and risk-management skills relating to low-income retail banking and products suited to small savers and enterprises. Experience indicates that, where banks Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 149 have successfully developed such banking skills and products on their home market, they are more likely to do so abroad. Given these challenges, African authorities would be well advised to promote market deepening, and greater competition and innovation, by encouraging the entry of banks that are experienced in servicing underserved mar- ket segments and can demonstrate success based on tested business models. The authorities could also consider a move towards more integrated banking models predicated on a sound framework for consolidated supervision, clearly established and functioning channels of information exchange between home and host country supervi- sors, and effective cross-border resolution frameworks. This would also help to further reduce the cost of doing business. Banks expanding across borders in Africa are almost universally required to establish not only self-standing subsidiaries, but also local IT functions; to use predominantly local labor; and to establish independent, local manage- ment functions (such as boards and risk-management capacity). This “fortress banking” runs directly counter to reaping the potential economic gains from financial deepening associated with cross-border banking. In particular, more integrated banking models would provide the opportunity for significant cost-savings in a traditionally high-cost industry and could make it cost-effi- cient and therefore attractive to provide financial services to a broader set of customers. Policies fostering more integrated banking models could, for example, include reducing the complexity and length of the licensing process; reducing initial capital requirements as applied to bank subsidiaries (with requirements designed to grow in line with the foreign bank’s business engagement and risk exposures); reducing or doing away with requirements to establish new branches where these exist (leaving, for example, deci- sions about the structure and security of bank premises to the banks); encouraging full mobility in the use of labor (skills transfer); encouraging usage of centralized, common IT for both internal operations and client services (such as ATMs, card services, and internet banking); and allowing the establishment of centralized audit and risk man- agement systems. These factors will contribute to greater efficiency in the provision of banking services and provide a platform for enhanced financial deepening. A move away from stand-alone subsidiaries towards more integrated subsidiaries and eventually even branching – if certain pre-conditions are met, including consolidated supervision and delegation of licensing and intervention powers to supranational bank supervisors – could be especially considered for formally integrated regulatory areas such as the Central and West African currency unions. In addition, regulatory harmonization could contribute to greater certainty as regards predictability and consistency in implementation, significant reduction in compliance costs across the region, as well as raising standards in more challenged environments. Regulatory harmonization is a huge undertaking and care will need to be taken that efforts to promote convergence are focused on key concerns, especially in 150 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward environments with severe capacity constraints. Prioritization and sequencing are crucial, and the focus needs to be on those policy areas where harmonization is essential to the integration agenda. For example, in an environment where credit risk is the key risk factor prioritization might suggest a focus on loan loss classification criteria and provisioning requirements. Even in these areas, however, there is considerable technical detail involved in establishing an effective harmonization arrangement. Harmonization of loan loss classification and provisioning requirements also requires supervisors to coordinate policies regarding collateral recognition, appraisal practices, and eligibility criteria for restructuring loans. It is important that policymakers leverage the complementarity between the national and sub-regional agendas in promoting banking efficiency. Regional Economic Communities can play an important role in establishing best practice processes, fostering exchange on these practices, and monitoring progress in their implementation. However, given the complexity in implementing multilateral reform programs, it will be extremely important to prioritize such sub-regional endeavors, and it may be more expedient to implement some reforms on a bilateral basis so as to draw lessons before replicating them at the sub-regional level. A bilateral approach could be appropriate in fast-evolving areas, such as regulation of mobile money and banking agents and the related development of payments systems infrastructure. Safeguarding stability of cross-border banking in normal times Consolidated supervision is a critical component of overseeing cross-border banks, yet most African home country supervisors still lack adequate frameworks, implementation capacity and consolidated accounting data. Establishing or improving frameworks of consolidated supervision and their effective implementation is therefore a high priority for safeguarding financial stability in Africa. While cross-border consolidated supervi- sion is largely the responsibility of the home country supervisors, host country supervi- sors do have the responsibility of sharing relevant information. To effectively carry out consolidated supervision, authorities require sufficient data on the activities of banks. Available information about the size and nature of cross-border banking activities in Africa is currently difficult to come by or unavail- able. As discussed in Chapter 1, efforts to assemble such information are arduous and the outcomes are too often partial and in some cases based on sources which are not current. A key immediate task facing African authorities is to improve the availability and regular exchange of relevant information. To make this task manageable, it is strongly recommended that the smaller group of African home supervisory authorities take the lead in developing the required formats as well as a platform enabling regular informa- tion exchange on a basic set of data among African supervisors. This basic set should Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 151 include information on (a) basic qualitative and quantitative characteristics of cross- border banks; (b) supervisory data as it relates to performance; (c) qualitative informa- tion on regulatory frameworks and definitions underlying supervisory data; and (d) mar- ket intelligence (see more details below). Making this data publicly available will allow for overseeing and monitoring ongoing developments in cross-border banking and could serve as the basis for a risk-based approach to strengthening banking supervision. Beyond this broader effort to assemble, exchange, and make publicly available a basic set of data about cross-border banking activities in Africa, timely exchange of more detailed, institution-specific information based on MoUs among supervisors and as input to supervisory colleges is necessary for effective cross-border supervision and early detection of fragility. Whether or not formal agreements or institutions are in place, the quality and fre- quency of information exchange, particularly when it relates to more detailed, institu- tion-specific information, is a function of trust between home and host supervisory authorities. Formal arrangements can be instrumental in building trust and anchoring expectations. Improvements in supervisory cooperation – through signing of appropri- ate MoUs and formation of properly structured CoS – is a priority area of action on both bilateral and sub-regional levels. Nonetheless, it needs to be recognized that in view of escalating incentive conflicts between home and host supervisors in times of crisis, these arrangements are a necessary but not a sufficient condition to safeguard stability. Besides informational challenges, establishing effective consolidated supervision in Africa also needs to address the fact that there are large African cross-border banks that are not subject to consolidated supervision and are thus not under appropriate home country supervision. Given the large geographic footprint of these institutions, coordination on the pan-African level is urgently called for. Given the potential reputa- tional and stability risks associated with unreported or undisclosed risks of cross-bor- der banking groups, the authorities and the banks have an important common social responsibility in seeing that regulatory gaps are addressed. A private/public partnership could play an important role in advancing this agenda to ensure a level playing field for banks. At the same time, the Community of African Banking Supervisors (CABS) or the Financial Stability Board’s Regional Consultative Group for Sub-Saharan Africa might serve as a coordination forum to identify systemically important institutions and how best to coordinate policy actions and monitor implementation of consolidated supervi- sion for systemically important African cross-border banks. The CABS may also become a permanent forum for discussing issues related to cross-border cooperation. While such an exchange can play a decisive role in shaping the regional policy dialogue and capacity building agenda, day-to-day cooperation will have to be implemented on either a bilateral level or within smaller groups focusing on specific cross-border institutions. The CABS can, however, serve as an important forum 152 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward for exchange of ideas and experiences, and drive convergence towards a common set of international standards while at the same time facilitating the development of Africa- appropriate regulatory frameworks and providing the entry point for more detailed cooperation between individual countries or on individual banks. Preparing for cross-border repercussions of bank fragility The management of both idiosyncratic and systemic banking sector distress in many African countries is characterized by forbearance, ad-hoc measures, and bailouts even of non-systemic institutions. Preparing for the cross-border repercussions of bank fragility clearly requires a solid foundation for safeguarding the stability of banking in normal times, including prompt corrective action in the sanctioning of banks. However, in being prepared for times of distress, authorities need to be equipped with sound resolution frameworks on a national level so that resolution proceedings can be initiated in a timely manner, with clearly assigned responsibilities among relevant authorities (Central Bank, Ministry of Finance, other relevant authorities), and with sufficient powers regarding transfer of assets and liabilities, establishing bridge banks, and implementing bank restructuring. Overall, there is a considerable outstanding agenda in many African countries relating to respecting the hierarchy of creditors in bank resolution and preventing legal actions that constrain the implementation of resolution measures. The establishment of resolution frameworks and/or their upgrading is important not only in its own right, to help supervisory authorities prepare for situations of failure, but as importantly, as a preventive measure. The existence of robust and credible exit policies and procedures to be implemented, should a situation arise where the winding down even of larger institutions, becomes necessary, will contribute to changing banks’ incentives for risk-taking even in normal times. Cross-border regulatory cooperation also has to look beyond information exchange during normal times, towards preparation for cross-border repercussions of idiosyncratic and systemic bank failures. One way to prepare for such eventualities would be to undertake joint crisis simulation exercises that could be used as the foundation for joint crisis management plans. Where relevant, this could also involve expanding supervisory colleges to include resolution authorities such as ministries of finance in so-called Crisis Management Groups. Given that the orderly resolution of cross-border groups is inevitably complex, caution should be taken to avoid the emergence of non-transparent, strongly inter- woven international financial groups through strict monitoring, limits on intra-group exposure, and enhanced business continuity planning. While consolidated supervision is the basis for such measures, such a precautionary approach goes beyond collecting the Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 153 necessary information towards a more active involvement of the supervisory authorities, as currently done for several systemically important financial institutions in the United States and Europe in the context of recovery and resolution plans (also known as living wills). While these policies can help to contain the impact of crises, they need to be carefully calibrated so as not to undo some of the very advantages and efficiency gains associated with cross-border banking. These general policy recommendations have to be fine-tuned to suit the context of individual countries, country pairs, and sub-regions. Critically, given the severe resource constraints faced by supervisory authorities, rec- ommended prioritization and sequencing of these actions will vary significantly across the continent among home and host jurisdictions, depending on factors such as the significance of cross-border banking activities, the degree of integration within banking groups, the complexity of group structures, and the quality of home/host supervision This calls for tailor-made policy agendas suited to different sub-regions and countries, including the appropriate prioritization and sequencing of reforms. However, despite differences across countries, African supervisors do stand to bene- fit from exchanging experiences with other countries in the region in terms of cross-bor- der banking and the supervision of cross-border financial institutions. As financial systems often have similar characteristics and supervisors are faced with similar challenges and deal with some of the same large multinational or African cross-border banks, more regular and structured exchange of experiences will be helpful in informing national regulatory processes as to how to harness the benefits while mitigating the risks of cross-border banking. Learning from each other will also have the additional benefit of helping to build trust among supervisors on the continent, a crucial factor for effective cross-border supervision. 5.2 Looking Forward The increase in cross-border banking activities in Africa presents both developmental opportunities, if the increasing impetus towards financial integration is to be har- nessed to the benefit of financial deepening and outreach, and potential stability risks, especially as such growth is often undertaken by complex banking, financial, or mixed conglomerate groups. Neglecting these opportunities and risks will be to the detriment both of financial system development and economic growth going forward. The recom- mendations discussed in this chapter provide an agenda for supervisors to reap both benefits and contain risks. The recommendations come with the recognition that cross-border regulatory cooperation is as much an objective as it is a process. While ultimately the objective is a cross-border regulatory framework that internalizes the externalities discussed in 154 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward Chapter 3, it is clear that such a framework cannot be created overnight. Rather, such a framework requires a long process of soft and hard institution building. It is also not a one-off process, as changes in international standards and in the cross-border banking landscape in Africa will require continuous adjustments to such a framework. An overarching conclusion arising from the analysis undertaken in this report, including from discussions with supervisors and bankers throughout Africa, is that information exchange is weak and needs to be strengthened significantly in the face of expanding cross-border banking activity in Africa. More detailed, institution-specific information exchange is necessary for effective cross-border supervision and early detection of fragility, based on MoUs among supervisors and as input to supervisory colleges. At the same time, establishing a platform for regular information exchange that makes publicly available a basic set of data about cross-border banking activities in Africa will be an important first step to facilitate better supervision of cross-bor- der activities and foster closer collaboration among authorities. This basic set should include the following four categories of data: (i) Basic qualitative and quantitative information on cross-border banks, such as group and ownership structure; risk management; governance; group exposures across Africa, including the proportion of assets deployed and loans extended in various markets; the types of activities pursued; and the market shares in those markets. (ii) Supervisory data including capital, liquidity, non-performing loans, provisioning rate, share of rescheduled loans and other asset quality indicators, overall capital level, ratio of liquid assets/liquid liabilities, and loan-to-deposit ratios; i.e., data of the kind that is regularly published by supervisory authorities in developed mar- kets. This would facilitate development of scoring/ evaluation methodologies for cross-border banks’ most significant subsidiaries and as input to measuring risk profiles on a consolidated basis. (iii) Qualitative regulatory information required to interpret and meaningfully compare supervisory data across jurisdictions, including the regulatory framework applied (Basel I, II, III or elements from different frameworks) and the definitions underlying key supervisory data (definition of capital, non-performing loans, provisioning and loan classification rules). (iv) Market intelligence that would alert home and host country supervisors to the risks associated with banks’ cross-border activities. This could take the form of country risk reports, including risk ratings, and could be used to encourage home country authorities to develop bank risk profiles on a consolidated basis. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 155 Clearly setting in motion and sustaining a process for compiling and exchanging such information requires commitment and focus. Given that there are more than 40 regulatory authorities in Africa (including regional authorities representing groups of countries), the practical and effective way to address this information vacuum is to place responsibility with the relatively small number of home country supervisors: Kenya, Mauritius, Morocco, Nigeria, and South Africa.29 This small quorum of countries could develop the required formats for information exchange, drawing, as found expedient, on expertise from institutions such as the International Monetary Fund and the World Bank, which already have experience in the field of data assembly. Once developed, the data should as far as possible be published and made available to all relevant home and host supervisors in Africa, with a view to enabling an informed exchange among supervisory authorities and improving the quality and effectiveness of supervision. Financial systems across Africa have deepened significantly over the past decade, and cross-border banking has been an important part of this development. Judging from the past decade, the next decade promises more entry but also exit of banks across Africa as new opportunities and risks arise and as authorities sharpen their focus on augmenting the depth and outreach of the financial systems. It is critical for supervisors to contribute to and be prepared for these developments with the necessary national tools and cross-border cooperation. Establishing a platform for regular exchange of information relating to the scope and nature of cross-border banking activities provides an important foundation and first step towards achieving this objective. 29: Bank Al-Maghrib, the Moroccan central bank, is currently the only home country central bank that actively encourages information exchange along the lines outlined above for the banks under its supervision engaged in cross-border activities. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 157 Data Sources The data used throughout this report come from a variety of sources, including: • unpublished regional studies on cross-border banking supervision in four African regions (SADC, WAMU, EAC, and ECOWAS) produced by GIZ in preparation for this report; • information provided by African central banks (banking supervision reports/ financial stability reports, data from websites); • information provided by commercial banks (annual reports, financial statements, websites); • World Bank and IMF reports; • and the Bank Ownership Database (Claessens and van Horen, 2014), which contains bank ownership information over the period 1995-2009 for more than 5,300 banks active in 137 countries; and the Regulatory Framework Database (AfDB, 2010) and the World Bank Banking Regulation and Supervision Database (2012), which contain information on bank regulations. The availability and quality of data varied greatly across countries and regions. To ensure comparability of data across African jurisdictions, a reference year was defined (usually 2011). If no data was available for the reference year, data for earlier or later years, if available, was used for best coverage. The publication also draws on information received in the context of consultation events with African stakeholders and insights officials of banks and regulatory authorities provided during reporting trips to Ghana, Kenya, Mauritius, Morocco, Mozambique, Namibia, Nether- lands, Nigeria, South Africa, Tanzania, Togo, Uganda, and the United Kingdom. While extensive efforts have been undertaken to cross-check the data, the authors do not guarantee the accuracy of the data included in this report. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 159 Literature Access Bank. 2013. “Full Year 2012 & First Quarter 2013 Results: Presentation to Investors and Analysts.” April. Available at http://www.accessbankplc.com. African Development Bank (AfDB). 2010. Regulatory Framework Database. Available at http://www.mfw4a.org/fileadmin/data_storage/documents/MFW4A-documents/Update%20 Regulatory%20Framework.xls. African Development Bank (AfDB), Organisation for Economic Co-operation and Devel- opment (OECD), United Nations Development Programme (UNDP) and United Nations Economic Commission for Africa (UNECA). 2013. African Economic Outlook 2013. Available at http://www.africaneconomicoutlook.org/en/outlook/. Arena, Marco, Carmen Reinhart, and Francisco Vazquez. 2007. “The Lending Channel in Emerging Economies: Are Foreign Banks Different?” IMF Working Papers 07/48. Austin, Gareth and Chibuike Ugochukwu Uche. 2007. “Collusion and Competition in Colonial Economies: Banking in British West Africa, 1916-1960.” Business History Review, 81(1), pp. 1-26. Bank for International Settlements (BIS). 1975. “Concordat.” Available at http://www.bis.org. Bank for International Settlements (BIS). 1983. “Principles for the supervision of banks’ foreign establishments (Basel Concordat).” Available at http://www.bis.org. Bank for International Settlements (BIS). 1992. “Minimum standards for the supervision of international banking groups and their cross-border establishments.” Available at http://www.bis.org. Bank for International Settlements (BIS). 1996. “Supervision of Cross Border Banking.” Available at http://www.bis.org. Bank for International Settlements (BIS). 2001.”Essential elements of a statement of cooperation between banking supervisors.” Available at http://www.bis.org. Bank for International Settlements (BIS). 2003. “High-level principles for the cross-border implementation of the New Accord.” Available at http://www.bis.org. 160 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward Bank for International Settlements (BIS). 2006. “Home-Host Information Sharing for Effective Basel II Implementation.” Available at http://www.bis.org. Bank for International Settlements (BIS). 2006. Core Principles. Available at http://www.bis.org. Bank for International Settlements (BIS). 2010. “Good Practice Principles on Supervisory Colleges.” Available at http://www.bis.org. Bank for International Settlements (BIS). 2010. “Report and. Recommendations of the Cross-border Bank. Resolution Group.” Available at http://www.bis.org. Bank for International Settlements (BIS). 2012. “Core Principles.” Available at http://www.bis.org. Barajas, Adolfo, Roberto Steiner, and Natalia Salazar. 2000. “The impact of liberalization and foreign investment in Colombia’s financial sector.” Journal of Development Economics, 63(1), pp. 157–196. Barclays Bank PLC. 2012. Annual Report 2012. Available at http://group.barclays.com/ Beck, Thorsten. 2007. “Efficiency in financial intermediation: Theory and empirical measurement.” Open Access publications from Tilburg University. Beck, Thorsten. 2011. “Finance and Oil: Is there a Resource Curse in Financial Develop- ment?” Discussion Paper 2011-017, Tilburg University, Center for Economic Research. Beck, Thorsten and Martin Brown. 2014. “Foreign Bank Ownership and Household Credit.” Journal of Financial Intermediation. Beck, Thorsten and Robert Cull. 2013. “Banking in Africa.” Policy Research Working Paper Series 6684, The World Bank. Beck, Thorsten, Asli Demirgüç-Kunt, and Maria Soledad Martinez Peria. 2007. “Reaching Out: Access to and Use of Banking Services Across Countries.” Journal of Financial Eco- nomics, 85(1), pp. 234-66. Beck, Thorsten, Asli Demirgüç-Kunt, and Maria Soledad Martinez Peria. 2008. “Banking Services for Everyone? Barriers to Bank Access and Use around the World.” World Bank Economic Review, 22(3), pp. 397-430. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 161 Beck, Thorsten and Erik Feyen. 2013. “Benchmarking financial systems: introducing the financial possibility frontier.” Policy Research Working Paper Series 6615, The World Bank. Beck, Thorsten, Erik Feyen, Alain Ize, and Florencia Moizeszowicz. 2008. “Benchmarking financial development.” Policy Research Working Paper Series 4638, The World Bank. Beck, Thorsten and Michael Fuchs. 2004. “Structural Issues in the Kenyan Financial System: Improving Competition and Access.” Policy Research Working Paper Series 3363, The World Bank. Beck, Thorsten, Michael Fuchs, and Makaio Witte. 2013. “Strengthening Banking Regulation and Supervision: A Framework for Developing Reform Priorities.” Mimeo. Beck, Thorsten and Heiko Hesse. 2009. “Why are interest spreads so high in Uganda?” Journal of Development Economics, 88(2), pp. 192-204. Beck, Thorsten, Vasso Ioannidou and Larissa Schäfer. 2012. “Foreigners vs. Natives: Bank Lending Technologies and Loan Pricing.” Discussion Paper 2012-055, Tilburg University, Center for Economic Research. Beck, Thorsten, Samuel Munzele Maimbo, Issa Faye, and Thouraya Triki. 2011. Financing Africa: Through the Crisis and Beyond. The World Bank: Washington, DC. Beck, Thorsten and Maria Soledad Martinez Peria. 2010. “Foreign Bank Participation and Outreach: Evidence from Mexico.” Journal of Financial Intermediation, 19(1), pp. 52-73. Beck, Thorsten and Wolf Wagner. 2013. “Supranational Supervision - How Much and For Whom?” CEPR Discussion Paper 9546. Beck, Thorsten, Radomir Todorov, and Wolf Wagner. 2013. “Supervising cross-border banks: theory, evidence and policy.” Economic Policy, 28(1), pp. 5–44. Bédard, Mathieu. 2013. “Informational Contagion and the Entrepreneurial Production of Informational Remedies.” CAE Working Papers 96, Aix-Marseille Université, CERGAM. Berg, Gunhild, Michael Fuchs, Leonardo Iacovone, Thomas Jaeggi, Andrew Lovegrove, and Carolina Sanchez Villegas. 2012. “Financing Small- and Medium-Sized Enterprises in Nigeria.” Unpublished working paper, The World Bank, Washington, DC. 162 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward Berger, Allen and Loretta Mester. 1997. “Inside the Black Box: What Explains Differences in the Efficiencies of Financial Institutions?” Journal of Banking and Finance, 21(7), pp. 895–947. Bertay, Ata Can, Asli Demirguc-Kunt, and Harry Huizinga. 2012. “Is the financial safety net a barrier to cross-border banking?” Policy Research Working Paper Series 5947, The World Bank. Black, Sandra and Philip Strahan, 2001. “The Division of Spoils: Rent-Sharing and Discrimination in a Regulated Industry.” American Economic Review, 91(4), pp. 814-831. Brownbridge, Martin and Charles Harvey. 1998. Banking in Africa. African World Press: Trenton, NJ. Bruno, Valentia and Robert Hauswald. 2013. “The Real Effect of Foreign Banks.” Review of Finance. pp.1-34. Cihak, Martin and Richard Podpiera. 2005. “Bank Behavior in Developing Countries: Evidence from East Africa.” IMF Working Papers 05/129. Claessens, Stijn, Asli Demirgüç-Kunt, and Harry Huizinga. 2000. “The Role of Foreign Banks in Domestic Banking Systems.” In Claessens, Stijn and Marion Jansen, (eds.), The Interna- tionalization of Financial Services: Issues and Lessons for Developing Countries, Kluwer Academic Press: Boston, MA. Claessens, Stijn, Asli Demirgüç-Kunt, and Harry Huizinga. 2001. “How Does Foreign Entry Affect Domestic Banking Markets?” Journal of Banking and Finance, 25(5), pp. 891-911. Claessens, Stijn, and Luc Laeven. 2004. “What Drives Bank Competition? Some International Evidence.” Journal of Money, Credit, and Banking, 36(3), pp. 563-83. Claessens, Stijn and Neeltje van Horen. 2014. “Foreign banks: Trends and Impact.” Journal of Money, Credit and Banking, 46(1), pp. 295-326. Clarke, George, Robert Cull, Laura D’Amato, and Andrea Molinari. 2000. “On the Kindness of Strangers? The Impact of Foreign Entry on Domestic Banks in Argentina.” In Claessens, Stijn and Marion Jansen, (eds.), The Internationalization of Financial Services: Issues and Lessons for Developing Countries, Kluwer Academic Press: Boston, MA. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 163 Clarke, George, Robert Cull, and Michael Fuchs. 2009. “Bank Privatization in Sub-Saharan Africa: The Case of Uganda Commercial Bank.” World Development, 37(9), pp. 1506-1521. Clarke, George, Robert Cull, and Maria Soledad Martinez Peria. 2006. “Foreign bank participation and access to credit across firms in developing countries.” Journal of Comparative Economics, 34(4), pp. 774-795. Clarke, George, Robert Cull, Maria Soledad Martinez Peria, and Susana Sanchez. 2005. “Bank Lending to Small Businesses in Latin America: Does Bank Origin Matter?” Journal of Money, Credit and Banking, 37(1), pp. 83-118. Cull, Robert and Maria Soledad Martinez Peria. 2008. “Crises as Catalysts for Foreign Bank Activity in Emerging Markets.” In Justin Robertson, (ed.), Power and Politics After Financial Crisis: Rethinking Foreign Opportunism in Emerging Markets, Palgrave Macmillan: New York, NY. Cull, Robert and Maria Soledad Martinez Peria. 2013. “Foreign Bank Participation in Developing Countries: What Do We Know about the Drivers and Consequences of this Phenomenon?” in Gerard Caprio, ed., Encyclopedia of Financial Globalization. Elsevier: Oxford, UK. Cull, Robert and Connor Spreng. 2011. “Pursuing Efficiency While Maintaining Outreach: Bank Privatization in Tanzania.” Journal of Development Economics, 94(2), pp. 254-261. De Haas, Ralph, Yevgeniya Korniyenko, Alexander Pivovarsky and Teodora Tsankova. 2014 “Taming the Herd, the Vienna Initiative and Crisis Transmission” Journal of Financial Inter- mediation. De Haas, Ralph and Iman van Lelyveld. 2014. “Multinational banks and the global financial crisis: weathering the perfect storm?” Journal of Money, Credit, and Banking. 46(1), pp. 333-364. D’Hulster, Katia. 2011. “Cross border banking supervision: incentive conflicts in supervisory information sharing between home and host supervisors.” Policy Research Working Paper Series 5871, The World Bank. De la Torre, Augusto, Maria Soledad Martinez Peria, and Sergio Schmukler. 2010. “Bank Involvement with SMEs: Beyond Relationship Lending.” Journal of Banking and Finance. 34(9), pp. 2280-2293. 164 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward Department for International Development (DFID). 2013. “Global Financial Sector Regulatory Reform: Understanding the Impact on Developing Countries.” London. Detragiache, Enrica, Poonam Gupta, and Thierry Tressel. 2008. “Foreign Banks in Poor Countries: Theory and Evidence.” Journal of Finance, 63(5), pp. 2123-60. Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ). 2012a. “Cross-Border Banking Regulation and Supervision in East Africa – Trends and Perspectives.” Unpublished report. Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ). 2012b. “Cross-Border Banking Supervision in ECOWAS: Cases of Ghana and Nigeria.” Unpublished report. Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ). 2012c. “Cross-Border Supervision: SADC Region.” Unpublished report. Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ). 2012d. “Cross-Border Banking Supervision in the West African Monetary Union (WAMU).” Unpublished report. Earne, Julie, Tor Jansson, Antonique Koning, and Mark Flaming. 2014. “Greenfield MFIs in Sub-Saharan Africa: A Business Model for Advancing Access to Finance.” Consultative Group to Assist the Poor (CGAP) and International Finance Corporation (IFC): Washington, DC. Ecobank. 2010. “Microfinance and Poverty Alleviation.” Ecobank, Lomé, Togo. Available at http://www.ecobank.com/microfinance.aspx. European Bank for Reconstruction and Development (EBRD). 2012. “Vienna Initiative – Moving to a New Phase.” EBRD: London, UK. Feng, Guohua and Apostolos Serletis. 2010. “Efficiency, Technical Change, and Returns to Scale in Large U.S. Banks: Panel Data Evidence from an Output Distance Function Satisfying Theoretical Regularity.” Journal of Banking and Finance, 34(1), pp. 127–38. Fiechter, Jonathan, Inci Ötker-Robe, Anna Ilyina, Michael Hsu, André Santos, and Jay Surti. 2011. “Subsidiaries or Branches: Does One Size Fit All?” IMF Staff Discussion Note 11/04. Financial Stability Board (FSB). 2011. “Key Attributes of Effective Resolution Regimes for Financial Institutions.” Available at http:// http://www.financialstabilityboard.org. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 165 Financial Stability Board (FSB). 2012. “Meeting of the Financial Stability Board Regional Consultative Group for Sub-Saharan Africa.” FSB press release no. 12/2012, February 3, 2012. Available at http:// http://www.financialstabilityboard.org. Financial Stability Board (FSB). 2013. “Overview of Progress in the Implementation of the G20 Recommendations for Strengthening Financial Stablility.” Available at http:// http://www.financialstabilityboard.org. Financial Stability Board (FSB), International Monetary Fund (IMF), and World Bank. 2011. “Financial Stability Issues in Emerging Market and Developing Economies: Report to the G-20 Finance Ministers and Central Bank Governors.” Available at http://www.financialstabilityboard.org. Fuchs, Michael, Richard Hands, and Thomas Jaeggi. 2010. “Recent international reform initiatives to strengthen post-crisis banking supervision and the international regulatory architecture – priorities and potential pitfalls for LICs.” Unpublished Working Paper, The World Bank. Fuchs, Michael, Djibrilla Issa, Richard Hands, and Jeremy Denton-Clark. 2006. “Financial Sector Integration in Sub-Saharan Africa: Can Creating Scale in Financial Markets Support Growth and Development? The Economic Community of West African States.” Making Finance Work for Africa mimeo. Fuchs, Michael, Thomas Losse-Mueller, and Makaio Witte. 2013. “The Reform Agenda for Financial Regulation and Supervision in Africa.” In Thorsten Beck and Samuel Maimbo, (eds.), Financial Sector Development in Africa: Opportunities and Challenge, The World Bank: Washington, DC. Giannetti, Mariassunta and Steven Ongena. 2009. “Financial Integration and Firm Performance: Evidence from Foreign Bank Entry in Emerging Markets.” Review of Finance, 13(2), pp. 181-223. Giannetti, Mariassunta and Steven Ongena. 2012. ““Lending by example”: Direct and indirect effects of foreign banks in emerging markets.” Journal of International Economics, 86(1), pp. 167-180. Gormley, Todd. 2010. “Banking Competition in Developing Countries: Does Foreign Bank Entry Improve Credit Access?” Journal of Financial Intermediation, 19(1), pp. 26-51. 166 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward Honohan, Patrick and Thorsten Beck. 2007. Making Finance Work for Africa. The World Bank: Washington, DC. Hughes, Joseph and Loretta Mester. 2013. ”Who said large banks don’t experience scale economies? Evidence from a risk-return-driven cost function.” Working Papers 13-13, Federal Reserve Bank of Philadelphia, revised 04 Feb 2014. International Monetary Fund (IMF). 2011. Regional Economic Outlook: Sub-Saharan Africa. October. Washington, DC. International Monetary Fund (IMF). 2012. Regional Economic Outlook: Sub-Saharan Africa. April 2012. Washington, DC. International Monetary Fund (IMF). Various Years. Direction of Trade Statistics (DOTS). Available at https://www.imf.org/external/data.htm. Jayaratne, Jith and Philip Strahan. 1998. “Entry Restrictions, Industry Evolution, and Dynamic Efficiency: Evidence from Commercial Banking.” Journal of Law and Economics, 41(1), pp. 239-73. Kalemli-Ozcan, Sebnem, Elias Papaioannou, and Fabrizio Perri. 2013. “Global banks and crisis transmission.” Journal of International Economics, 89(2), pp. 495-510. Kasekende, Louis, Justine Bagyenda, and Martin Brownbridge. 2012, “Basel III and the global reform of financial regulation: how should Africa respond? A bank regulator’s perspective.” Bank of Uganda mimeo. Kiraly, Julia, Bea Majer, Laszlo Matyas, Bela Ocsi, Andras Sugar, and Eva Varhegyi. 2000. “Experience with Internationalization of Financial Service Providers – Case Study: Hungary.” In Claessens, Stijn and Marion Jansen, (eds.), The Internationalization of Financial Services: Issues and Lessons for Developing Countries, Kluwer Academic Press: Boston, MA. Klapper, Leora and Dorothe Singer. Forthcoming. “Financial Inclusion in Africa: The Role of Informality.” Journal of African Economics. Kubo, Koji. 2006. “The Degree of Competition in the Thai Banking Industry Before and After the East Asian Crisis.” ASEAN Economic Bulletin, 23(3), pp. 325-40. Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward 167 Laeven, Luc and Fabian Valencia. 2010. “Resolution of Banking Crises: The Good, the Bad, and the Ugly.” IMF Working Paper WP/10/146. Lee, Byungyoon. 2003. “The Determinants of Foreign Banks’ Entry in Korea and Its Effects on Korean Domestic Banks’ Performance.” Economic Papers, 6(1), pp. 42-65. Lukonga, Inutu and Kay Chung. 2010. “The Cross-Border Expansion of African LCFIs.” IMF Research Paper, Washington, DC. Mian, Atif. 2006. “Distance Constraints: The Limits of Foreign Lending in Poor Economies.” Journal of Finance, 61(3), pp. 1465-1505. Nikiel, Ewa and Timothy Opiela. 2002. “Customer Type and Bank Efficiency in Poland: Implications for Emerging Market Banking.” Contemporary Economic Policy, 20(3), pp. 255–271. Osei, Robert Darko. 2007. “Linking Traditional Banking with Modern Finance: Barclays Microbanking - Susu Collectors Initiative.” UNDP Growing Inclusive Markets Case Study. Peek, Joe and Eric Rosengren. 2000. “Collateral Damage: Effects of the Japanese Bank Crisis on Real Activity in the United States.” American Economic Review, 90(1), pp. 30-45. Popov, Alexander and Gregory Udell. 2012. “Cross-border banking, credit access, and the financial crisis.” Journal of International Economics, 87(1), pp.147-161. Rabobank. 2013. “Looking for Delta: How Global Companies Can Help Sub-Saharan Africa Reach Its F&A Potential.” Available at http://www.rabobank.com. Sengupta, Rajdeep. 2007. “Foreign Entry and Bank Competition.” Journal of Financial Economics, 84(2), pp. 502-528. Sensarma, Rudra. 2006. “Are foreign banks always the best? Comparison of state-owned, private and foreign banks in India.” Economic Modelling, 23(4), pp. 717-735. Song, Inwon. 2004. “Foreign Bank Supervision and Challenges to Emerging Market Supervisors.” IMF Working Papers 04/82. Unite, Angelo and Michael Sullivan. 2002. “The Impact of Liberalization of Foreign Bank Entry on the Philippine Domestic Banking Market.” Philippine APEC Study Center Network Discussion Paper No. 2001-08. 168 Chapter 5: Cross-Border Regulatory Cooperation in Africa: Looking Forward United Nations. 2002. “Manual on Statistics of International Trade in Services.” Available at http://unstats.un.org/unsd/publication/Seriesm/Seriesm_86e.pdf. Wheelock, David and Paul Wilson. 2009. “Are U.S. Banks too Large?” Working Paper 2009- 054B, Federal Reserve Bank of St. Louis. Revised December 2009. World Bank. 2012. Banking Regulation and Supervision Database. Available at http://go.worldbank.org/SNUSW978P0. World Bank. 2013a. Financial System Stability Assessment (FSSA): East African Community. World Bank. 2013b. South Africa Economic Update (June). Wu, Hsiu-Ling, Chien-Hsun Chen and Mei-Hsuan Lin. 2007. “The Effect of Foreign Bank Entry on the Operational Performance of Commercial Banks in the Chinese Transitional Economy.” Post-Communist Economies, 19(3), pp. 343-357.