NIGERIA DEVELOPMENT FINANCE STUDY Lessons from International Experience in Designing the Development Bank of Nigeria OCTOBER 2016 1 Contents LIST OF ACRONYMS ................................................................................................................... i PREFACE ........................................................................................................................................v EXECUTIVE SUMMARY ........................................................................................................... vi Chapter 1—Why a New DFI in Nigeria Now? Problem Identification and Policy Context ...........1 I The Nigerian Context .................................................................................................... 1 II Defining the MSME Financing Gap ........................................................................... 11 III Addressing the MSME Financing Gap ....................................................................... 21 Chapter 2—Choosing an Appropriate Approach to the DBN .......................................................27 I Defining the Role of DFIs........................................................................................... 27 II Core Features of Alternative Approaches to DFIs ...................................................... 36 III International Experience Relating to Alternative DFI Approaches ............................ 42 Chapter 3—Developing and Implementing a Robust Monitoring and Evaluation (M&E) Framework .....................................................................................................................................56 I Motivation for Establishing an M&E Framework for the DBN ................................. 56 II Selection and Application of Performance and Impact Indicators ............................. 56 III Designing an M&E Framework for the DBN ............................................................. 70 Chapter 4—What Makes DFIs Work or Fail? ...............................................................................74 I DFI Governance Arrangements .................................................................................. 74 II. Making Sure DFIs Achieve Development Impact .................................................... 103 Bibliography ................................................................................................................................106 LIST OF ACRONYMS AADFI African Association of Development Finance Institutions ACGS Agricultural Credit Guarantees Scheme ACSS Agricultural Credit Support Scheme ADB Asian Development Bank AFD Agence Française de Développement AFDB African Development Bank AGM Annual General Meeting of Shareholders AIP Access to Information Policy AMCON Asset Management Company of Nigeria ATM Automated Teller Machine BBVA Banco Bilbao Vizcaya Argentaria, S.A. BDC Business Development Bank of Canada BNDES Brazilian Development Bank BOA Bank of Agriculture BOI Bank of Industry BRICS Brazil, Russia, India, China, South Africa CACS Commercial Agriculture Credit Scheme CAR Capital Adequacy Ratio CBN Central Bank of Nigeria CC Creative Commons Attribution CDB China Development Bank CDC Commonwealth Development Corporation CEO Chief Executive Officer CEPAL Comisión Económica para América Latina CFO Chief Financial Officer CG Corporate Governance CODE Committee on Development Effectiveness COFIDES Spain’s Compañía Española de Financiación del Desarrollo S.A. DBN Development Bank of Nigeria DBSA Development Bank of South Africa DEG Deutsche Investitions- und Entwicklungsgesellschaft DFI Development Finance Institution DID Difference-in-Differences EAD Exposure at Default EBRD European Bank for Reconstruction and Development ECB European Central Bank EFIL Export Finance Intermediation Loan EIB European Investment Bank EU European Union EY Ernst & Young FGN Federal Government of Nigeria FINSTATS Financial Statistics i FIRA Mexican Trust Fund for Rural Development FMO Nederlandse Financierings-Maatschappij voor Ontwikkelingslanden N.V. FOGAPE Fondo de Garantía para Pequeños Empresarios (Chilean State Fund) FSB Financial Services Board GDP Gross Domestic Product GEM World Bank’s Growth & Employment Project GRI Global Reporting Initiative IB Infrastructure Bank IDB Inter-American Development Bank IEG World Bank Group’s Independent Evaluation Group IFC International Finance Corporation IFI International Financial Institution IFRS International Financial Reporting Standards IMF International Monetary Fund IPO Initial Public Offering JFC Japan Finance Corporation KBN Kommunalbanken Norway KFW Kreditanstalt für Wiederaufbau (German Development Bank) KPMG Klynveld Peat Marwick Goerdeler (Accounting Firm) KWG German Banking Act LGD Loss Given Default MINT Mexico, Indonesia, Nigeria, Turkey M&E Monitoring and Evaluation MOF Ministry of Finance MOU Memorandum of Understanding MPR Monetary Policy Rate MRR Minimum Discount Rate MSME Micro, Small, and Medium-Sized Enterprises NACRDB Nigerian Agricultural Cooperative and Rural Development Bank NAFIN Mexican Development Bank NBCI Nigerian Bank for Commerce and Industry NBS Nigeria Bureau of Statistics NERFUND National Economic Reconstruction Fund NEXIM Nigerian Export-Import Bank NGN Nigerian Naira NIDB Nigerian Industrial Development Bank NIRSAL Nigerian Incentive-Based Risk Sharing System for Agricultural Lending NPL Nonperforming Loan ODA Official Development Assistance ODI Overseas Development Institute OECD Organization for Economic Co-Operation and Development OI Output Index OLS Ordinary Least Squares OPIC Overseas Private Investment Corporation ii PAIF Power and Airline Intervention Fund PCG Partial Credit Guarantees PD Probability of Default PFI Private Sector Financial Institution PIDG Private Infrastructure Development Group PSM Propensity Score Matching PWC PricewaterhouseCoopers RDD Regression Discontinuity Design ROA Return on Asset ROE Return on Equity ROSC Report on Observance of Standards and Codes RRF Refinancing and Restructuring of Banks (CBN scheme) SARB South African Reserve Bank SDI Subsidy Dependence Index SEBRAE Serviço Brasileiro de Apoio às Micro e Pequenas Empresas SIEBAN Sistema de Estímulos a la Banca SIFEM Swiss Investment Fund for Emerging Markets SIMEST Italy’s Società Italiana per le Imprese all’Estero S.p.A. SMECGS Small and Medium Enterprises Credit Guarantee Scheme SOE State-Owned Enterprise SOFARIS Société Française pour l’Assurance du Capital TKSB Turkey’s Türkiye Sınai Kalkınma Bankası UK United Kingdom USD US Dollar VAT Value-Added Tax WDI World Development Indicator WWII World War II iii PREFACE This study summarizes the international experience that was used to lay the foundation for the design of the Development Bank of Nigeria launched by the Nigerian government in 2015. Although it has long been recognized that financing gaps in areas such as for smaller enterprises and for infrastructure programs have persisted, since the 1980s the international community has been reluctant to sponsor the establishment of new development banks. This reluctance reflects the poor postcolonial experience with development finance. The reaction to these experiences was the so-called “Washington Consensus,” whereby faith was placed in the sanctity of market mechanisms. In an African context, the support provided by the World Bank and other donors in establishing a new development bank in Nigeria represents a departure from accepted wisdom and recent years’ policies. This study documents the shift in policy perspectives that underlies this change in approach. This study was prepared under the guidance and oversight of Andrej Popovic (Senior Financial Sector Specialist and Task Team Leader), Michael D. Wong (Lead Private Sector Specialist), and Guillemette Jaffrin (Lead Private Sector Development Specialist). Overall guidance and quality control was provided by Irina Astrakhan (Practice Manager). The study was prepared by a consulting team led by Michael Fuchs working with Luiz Henrique Alcoforado and Jing Zhao. This study was funded by the UK Department for International Development (DFID). The study benefited considerably from the inputs provided by peer reviewers: Alex S. Berg (Senior Financial Sector Specialist), Eva M. Gutierrez (Program Leader), Roland Lomme (Senior Governance Specialist), Carlos Vicente (Senior Financial Sector Specialist), as well as Ceyla Pazarbasioglu-Dutz (Senior Adviser), and Indira Konjhodzic (Country Program Coordinator, Nigeria), who managed the review process. v EXECUTIVE SUMMARY Context Nigeria has recorded robust growth for more than a decade, and its economy has become more diversified with less reliance on oil, although oil revenue remains a critical source of fiscal revenues and foreign exchange. The recent plunge in oil prices demonstrates how exposed economic growth and the fiscal situation remain to changes in oil prices. Reducing the impact of volatility in oil prices will be important in sustaining growth and thereby reducing the unemployment rate on a sustainable basis, especially among young people. Micro, small, and medium-sized enterprises (MSMEs) have a vital role to play in supporting the further diversification of Nigeria’s economy. Compared to peer countries, Nigeria suffers from a low level of credit provision (see Figure 1), and firm-level data reveals that Nigerian firms face severe credit constraints. While banks compete fiercely for the business of large, well-regarded corporate clients, small firms (operating outside the value chains associated with resource extraction) find it very challenging to access bank lending on affordable terms. By limiting their lending predominantly to enterprises within the value chains servicing large “blue chip” corporates, the Nigerian banks have adopted a defensive strategy in managing their credit risks. This is illustrated by data from the World Bank’s Enterprise Survey that shows that most large firms in Nigeria do not identify the availability of finance as a constraint (see Figure 2). Indeed, among all BRICS and MINT countries,1 Nigeria stands out with the widest gap between the percentage of large firms and the percentage of small and medium firms to identify access to finance as a major constraint. Figure 1: Private sector credit as a percentage of GDP (2013) South Africa China Brazil Turkey Russian Federa on India Indonesia Mexico Nigeria 0 20 40 60 80 100 120 140 160 180 Source: The World Bank. WDI. 1 BRICS: an acronym for Brazil, Russia, India, China, and South Africa. MINT: an acronym for Mexico, Indonesia, Nigeria, and Turkey. vi Figure 2: Percent of firms surveyed identifying access to finance as a major constraint — BRICS and MINT countries (by country total and firm size) 50 45 40 35 30 25 20 15 10 5 - Russian Brazil - Small Brazil Nigeria - Small Nigeria Russian - Small Mexico - Small South Africa - Small South Africa India Indonesia - Small Indonesia Turkey Turkey - Small China China - Small Brazil - Medium Russian - Large Mexico - Large Brazil - Large India - Medium Mexico Turkey - Large India - Large Indonesia - Large South Africa - Large Nigeria - Large China - Large India - Small Mexico - Medium South Africa - Medium Turkey - Medium Indonesia - Medium Nigeria - Medium China - Medium Russian - Medium Source: World Bank Enterprise Survey. Latest available survey results for country data: Brazil 2009, China 2012, India 2014, Indonesia 2009, Mexico 2010, Nigeria 2014, Russia Federation 2012, Turkey 2013, South Africa 2007. The incentive for banks to engage with MSMEs is encumbered by weaknesses in the credit infrastructure (outlined below), by their limited capacity in understanding the business needs and managing the risks associated with lending to smaller enterprises, and by public sector borrowing that provides the banks with “risk-free” assets and may be crowding out their lending to the private sector. Weaknesses in the institutional infrastructure for bank lending, including the contractual and collateral environment, and the credit information framework, are major obstacles to banks’ involvement in MSME financing. More specifically, difficulties in contract enforcement and collateral foreclosure are a significant constraint to the development of the country’s financial system. For example, borrowers face high collateral requirements compared to peer countries (see Figure 3). While reforms to the legal and judicial environment for creditor rights, including secured (collateral) rights, have been on the agenda for several years, they have had a tendency to stall. Recently the Central Bank of Nigeria (CBN) has launched several initiatives designed to strengthen the financial infrastructure. In early 2016 the CBN launched an electronic centralized registry for movable collateral, and the CBN’s efforts have also led to improvements in credit information-sharing, particularly as regards the coverage and quality of the information assembled by the credit bureaus. Nonetheless, despite some progress over recent years, there is still room for considerable improvement in Nigeria’s credit infrastructure, and further improvement will depend on the ongoing focus and commitment of the authorities.2 2 For a review of the benefits associated with strengthening, see materials assembled by the World Bank at: http://www.worldbank.org/en/topic/financialsector/brief/credit-infrastructure. More detailed analysis is provided in Inessa Love, Maria Soledad Martinez Peria, and Sandeep Singh, Collateral Registries for Movable Assets: Does vii Figure 3: Proportion of loans requiring collateral and value of required collateral Proportion of loans requiring collateral (%) Value of collateral needed for a loan (% of the loan amount) Nigeria Nigeria India India Russian Federa on Mexico Indonesia Turkey Sub-Saharan Africa China All Countries All Countries China South Africa Sub-Saharan Africa Mexico Russian Federa on Turkey South Africa Brazil Brazil 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 0.0 50.0 100.0 150.0 200.0 250.0 300.0 Source: World Bank Enterprise Survey. Latest available survey results for country data: Brazil 2009, China 2012, India 2014, Indonesia 2009,3 Mexico 2010, Nigeria 2014, Russian Federation 2012, Turkey 2013, South Africa 2007. Why a new DFI in Nigeria now? Policy intervention to address the substantial MSME financing gap is merited, given the Nigerian context of a vulnerable macro-economy, a shallow financial system, weak institutional infrastructure, and the need to promote sustainable economic development. The Development Bank of Nigeria (DBN), a new DFI focusing on SME finance, should contribute to the solution of the MSME finance problem in Nigeria, but its success is not guaranteed unless it is carefully designed and this design is effectively implemented. The purpose of this study is to identify the challenges facing the authorities in successfully establishing the DBN, thus creating broader awareness of the major tasks facing decision-makers in establishing a new development finance institution and in making sure that this new institution successfully meets its objectives in providing effective support to the MSME sector on a market-conform, financially-sustainable basis. Indeed, predominantly due to governance failings, earlier attempts to establish DFIs in Nigeria fell short in effectively addressing market gaps through structural change. They did not manage to facilitate meaningful increases in financial intermediation, or to galvanize sustainable development in their respective sectors. In recent years, rather than rely on direct support for DFI activities funded by the government budget, the Nigerian authorities have resorted to financing drawing on resources provided by the Central Bank. Provision of funding through such CBN “schemes,” usually on highly subsidized terms, can create risks for the DBN, as borrowers and lenders come to rely on relief from market interest rates in the form of subsidies. As a result, marginal borrowers—who might otherwise just be eligible for lending from a formal financial institution—may defer applying for a loan on the expectation that subsidized lending may be or become available. Similarly banks, rather than develop skills in reaching out beyond their core “blue chip” borrowers to more risky segments, may defer the development of such business lines, as the potential market for such lending is subject to disruption depending on the intermittent availability of subsidies. While largely undocumented, the effectiveness of these subsidized their introduction spur firms’ access to bank finance? (World Bank Policy Research Working Paper 6477, June 2013), and Kristin van Zwieten, Debt Resolution and Business Exit (The World Bank Viewpoint Series Number 343, July 2014). 3 Value of collateral data is missing for Indonesia. viii lending schemes may well not be much stronger than the direct financing provided earlier through the DFIs. The history of public sector DFIs in Nigeria, a history shared by DFIs in many other countries, reminds us that, despite the imperative to address financing gaps and correct market failures, DFIs may have difficulty fulfilling their mandates in an impactful and sustainable way. The broad mandates of the existing DFIs resulted in direct investment and risk- taking, whereby the DFIs came to substitute rather than catalyze the capacity of the private sector in assessing credit risks and financing investments. As the DFIs lacked expertise in assessing credit risk and were often obliged to undertake lending according to political imperatives rather than on commercial terms, they were quickly confronted with high levels of NPLs. This situation was further exacerbated by the lack of independent and effective oversight. Lessons learnt from these Nigerian experiences—supplemented by lessons garnered from international experience—serve to underline how the success of the DBN will crucially depend on effective implementation of the institution’s specific mandate, strong governance arrangements, and professional risk- management capacity of the new institution. This study lays out the rationale behind the design parameters of the DBN and the process by which the DBN can become successful. However, it is important to recognize from the outset that the DBN’s shorter-term efforts to support the growth of MSME financing will need to be supported by parallel efforts to achieve longer-term structural reforms. Through its activities, the DBN will be able to demonstrate how pilots can be made to work even in the current environment through deployment of instruments such as partial credit guarantees, or by making available term finance to SMEs where term finance is not otherwise provided by private sector intermediaries. In line with implementation of the reforms of the broader legal and institutional environment that facilitate credit provision by private intermediaries to marginal, more risky borrowers, the DBN will gradually be able to “graduate” from servicing one cohort of borrowers to another—always ensuring that the provision of DBN funding through instruments, such as credit lines and partial credit guarantees, is used to tackle a new spectrum of marginalized borrowers, thereby constantly pushing the frontier of the banks’ engagement. This graduation process inevitably depends on the coordinated focus and efforts of the authorities in implementing broader reform of the legal and regulatory environment and in upgrading the financial infrastructure, reforms that go way beyond the immediate authority of the DBN. Choosing an appropriate approach to the DBN According to global experience, DFIs are generally relatively small institutions when measured in relation to the size of the economies in which they operate, or to the assets of local banking systems, indicating that their main role is catalytic. Very few DFIs play a role in actually filling financing gaps. Thus, rather than measuring their impact in terms of the funding they can muster, the focus needs to be on how effectively they use their catalytic role in supporting pilots or providing funding/risk-sharing arrangements that will encourage the private sector to participate in the financing of activities that would otherwise remain unfunded or underfunded. Despite growth in the size of the DFI sector in recent years, DFI assets only account for a very small proportion of banking system assets in the vast majority of countries. The combined total assets to GDP ratio of DFIs in a recent international survey increased from below ix 10 percent in 2006 to 14 percent in 2009. If five major DFIs4 that are large in size both in absolute and relative terms are excluded, the combined total assets to GDP ratio of DFIs increased from 6 percent to 8 percent between 2006 and 2009. In addition, in most countries the market share (DFI assets as a percentage of banking system assets) of DFIs was less than 3 percent (see Figure 4). Figure 4: The relative size and market share of DFIs Rela ve size of na onal DFIs 2006-2009 Combined total assets as % of GDP Combined total assets as % of GDP (excl. major DBs) 16% 14% 12% 10% 8% 6% 4% 2% 0% 2006 2007 2008 2009 Source: The World Bank. WDI and Global Survey of Development Banks (2010). There are three commonly accepted approaches to analyzing the role of DFIs: public sector, laissez faire, and public/private approach. The purely Public Sector approach that seeks to fill financing gaps through government funding, which was popular in post-WWII and postcolonial times and until the late 1980s.The Laissez-Faire approach that promotes the downscaling of DFIs and instead relies on improving financial market infrastructure with a view to providing an environment conducive to greater private sector financing, which gained popularity in the early 1990s but has been widely questioned in recent years. Finally, the more recently developed Public/Private approach that sees the role of DFIs as undertaking targeted interventions designed to catalyze involvement of the private sector. The emerging Public/Private approach, rather than reflecting a consensus on a defined set of policy prescriptions, is more of a trend focused on rebalancing the Public Sector approach. The role of DFIs is redefined so that, instead of replacing the private sector, DFIs’ core function revolves around becoming a catalyst for greater private sector involvement. The table below highlights the key characteristics of this Public/Private approach compared with the Public Sector approach prevalent in Nigeria today. Switching from DFIs built around the Public Sector approach to the philosophy of the Public/Private approach, as illustrated by the elements in Table 1, is the cornerstone of the change in approach embodied by the DBN. 4 The five DFIs excluded are BNDES, China Development Bank (CDB), KfW, Kommunalbanken Norway (KBN), and Agriculture Bank of Turkey (Ziraat). x Table 1: Comparing features of the Public/Private approach with the Public Sector approach Public Sector approach Public/Private approach Ownership and o State ownership o Mixed government/nongovernmental Funding o Board and management appointed ownership by government o Board and management selected according o Government funding that includes to transparent, professional criteria fiscal transfers/subsidies o Initial capital funded by government/donors supplemented with wholesale market funding Mandate and o Social policy mandate supplemented o Strategy tailored to addressing specific Strategy with countercyclical interventions market failures/financing gaps o Projects selected according to public o Emphasis on market conformity so as to sector’s assessment of needs encourage learning, adoption, and o Projects characterized by high supplementary funding from the private risk/low return activities and poor sector viability o DFIs assessed on whether they achieve:  financial sustainability;  meaningful scale;  crowding-in of private sector funding Instruments and o Direct lending provided on the o Second-tier lending relying on private Clients balance sheets of DFIs sector to establish/maintain client o Targeting underserved client groups relationships and assess credit risks o Subsidized credits and guarantees o Focus on catalytic instrument design (providing financing otherwise unavailable on the market) and targeting financing gaps (excluded client groups) o Targeted client groups adjusted as private sector becomes conversant and takes them on o Market-conform pricing—leveraging public-/donor-funded initial capital (with zero required return) to provide limited discount to market rates The following are the defining features of the Public/Private approach: Ownership and funding: The Public/Private approach reflects the recent trend of moving away from 100 percent state ownership to increasing use of a mixed public-private capital structure. Minority nongovernmental ownership from so-called impact investors, such as international financial intermediaries (IFIs) and/or bilateral foreign government donor agencies, will significantly strengthen governance and facilitate transfer of expertise. To ensure a highly professional board, it will be important that board members are selected according to transparent criteria that focus on ensuring a high level of professional capacity and technical expertise in those areas required of the DFI. To ensure leverage and to enable the DFI to reach scale in the activities it supports, a diversified funding structure is envisaged with one-off government capitalization, investment by outside shareholders, and internal funds generated being combined with financing xi from capital markets. While investment on the part of government and donors may allow DFIs to provide services at a small discount compared to commercial terms, the focus of this approach is primarily on providing funding that complements funds provided by private sector intermediaries (catalytic) and on terms that are similar to those offered by private sector intermediaries (market- conform). Mandate and strategy: Adopting a Public/Private approach, DFIs seek to leverage the private sector’s capacity in funding, risk assessment, etc., while balancing the discipline that this private sector involvement implies with fulfilling their development mandate in a sustainable way. The role of DFIs under this approach is to focus on identifying and resolving specific market failures. DFIs address such market failures with instruments that encourage learning, adoption, and supplementary funding by private intermediaries. Thus, under this approach, the primary aim of DFIs is to ensure that their funding (“start capital”) is leveraged. The aim is to ensure that scalability is reached while also maintaining financial sustainability. This can only be achieved if: (a) the instruments and funding terms provided by DFIs are market-conform; that is they are consistent with and encourage private sector financing on similar/parallel terms; and (b) the DFI devotes attention to remaining on the frontier of financial service provision—seeking to test market segments and/or instruments that the private sector has as yet been unwilling to adopt or is not conversant with. With the aim of encouraging private sector participation, financial viability, and scalability under the Public/Private approach, it is found to be advisable to restrict the mandate of DFIs to performing the role of second-tier, wholesale intermediation. Advantages associated with establishing DFIs as wholesale rather than retail intermediaries are that: (a) second-tier DFIs will be perceived as supporting retail-level institutions rather than competing with or looking to replace them; (b) second-tier DFIs will need to rely on private retail intermediaries as regards expertise in managing client relationships and assessing credit risks. Instruments and clients: Under the Public/Private approach, DFIs target specific market failures using tailored instruments. Simultaneously they seek to encourage private sector participation so that they can adjust their focus to new markets and clients as soon as the private sector becomes conversant with and is willing to target those markets/market segments currently being serviced. By providing longer-term finance that is otherwise unavailable in the market or by sharing the risk assumed by private financial intermediaries, the Public/Private approach encourages private sector intermediaries to engage in the provision of finance in areas where the private sector lacks capacity and/or which the private sector finds too risky. Developing and implementing a robust M&E framework Developing and implementing a results-oriented approach to monitoring and evaluation (M&E) not only helps to ensure that results are measured and communicated, but also contributes to strengthening DFI effectiveness and efficiency. Understanding and defining the criteria of success at an early stage is crucial to being able to measure results. Moreover, being able to provide information on monitoring and assessment of results is a critical part of the feedback cycle. Despite the important role of robust DFI M&E frameworks, the monitoring and xii evaluation of DFI performance is still an evolving science for several reasons. Firstly, while DFIs may have had limited capacity and faced technical challenges relating to developing appropriate M&E methodologies, the more likely explanation is that insufficient attention was paid to M&E by DFI management. The foundation for any M&E system is solid and comprehensive data, so it is crucial to establish routines from the outset that ensure collection of relevant data. Secondly, the public sector mind-set that was pervasive in earlier generations of DFIs paid little heed to establishing an environment of transparency and accountability. In effect, the seeds of the failure associated with the Public Sector approach were sown from the outset. If DFIs are allowed to function without appropriately designed M&E systems, they will be unable to provide the M&E evidence required to assess their performance. In the absence of any background information, due diligence by the DFI’s board and management becomes very difficult, and designing and taking corrective action to strengthen DFI efficiency and effectiveness becomes as good as impossible. While different evaluation methods have their strengths and weaknesses, scoring matrices provide a useful framework for assembling both qualitative and quantitative information about DFI performance. By weighting the scores given to the various elements in a scoring matrix and allocating higher scores to quantitative elements, the evaluator can compensate for the subjectivity of qualitative scores, while still preserving the scope of the overall assessment methodology. Similarly, the scoring methodology also provides the evaluator the opportunity to place greater emphasis on core aspects of the DFI’s operations, thereby tilting the evaluation outcomes toward those aspects of the assessment that are regarded as more central to fulfilling the DBN’s mandate. The use of scoring matrices is supported by the African Association of Development Finance Institutions and the Inter-American Development Bank. It has been applied in evaluating KfW, the German development bank, and several Latin American/Caribbean programs, including the Bank of Foreign Trade of Colombia (Bancóldex), a state-owned commercial bank that operates as Colombia’s entrepreneurial development and export-import bank. Given that there are subjective and qualitative elements to the evaluation process, it will be important that the evaluation is undertaken by an adequately staffed independent unit within the DBN. In a similar way to the autonomy usually assigned to an enterprise ’s internal audit committee, the evaluation unit should be established as an independent unit reporting directly to the chairman of DBN’s board. Independence is critical for M&E at the institution level, as any internal division can hardly provide a third-party assessment unless the evaluation division has a separate budget, personnel, and reporting lines from the business divisions. Hiring reputable external consultants may be a way to introduce technical expertise in specific areas and even help achieve a more objective assessment, but it is unlikely that external consultants could replace the first-hand understanding of the evaluation context of an internal evaluation unit. The table below assembles some core components and suggested indicators of an M&E framework suitable to the DBN. The focus of the suggested M&E framework is on monitoring inputs, processes, products, and outputs. The majority of the suggested indicators are quantitative, although there are also qualitative indicators (such as on corporate governance and internal cost control systems). xiii Suggested key elements of an M&E framework for the DBN Evaluation items Key indicators Process/inputs - Corporate governance review - Review of internal control systems - Operational efficiency measures: costs per employee, costs per loans extended Products - Measure of compliance of activities with intentions of DBN lending programs - Financial sustainability indicators: capital adequacy ratio (CAR), nonperforming loan ratio (NPL), provisions over NPLs, return on equity (ROE) - Subsidy dependence index (SDI) Outputs - Percentage of firms supported by DBN programs - Share of total MSME credit supported by DBN programs - Share of long-term finance supported by DBN programs - The output index (OI): weighted sum of loans extended to different income groups as the proportion of the sum of loans extended Outcomes - Growth in productivity and employment of firms supported compared with economy-wide trends Notes: The subsidy dependence index (SDI) is defined as annual subsidy received by the DFI as the proportion of the average annual yield earned by the outstanding loan portfolio of the DFI. This ratio measures to what extent the DFI relies on subsidies. The output index (OI) is defined as the weighted sum of loans extended to different income groups as the proportion of the sum of loans extended. The output index is designed to reflect whether the DFI lends comparatively more to lower-income groups, indicating the level to which the DFI fulfills its social objectives. Full support by DBN’s shareholders and DBN leadership (the board of directors and management) will be critical to undertaking effective M&E. In giving credibility to the M&E process, it will be important that DBN’s shareholders (especially the principle shareholder, the Federal Ministry of Finance) commit to using the outcome of the M&E process as a key performance indicator in evaluating the performance of DBN’s board and management. As in many other countries, existing Nigerian DFIs have not defined a rigorous analytical approach to M&E, nor assembled the data required to perform systematic analysis, nor made public the outcomes of such analysis. It will therefore be essential that DBN’s shareholders and board pay particular attention to (i) defining the institution’s approach to M&E at the time of its inception, and (ii) setting up a rigorous results framework, specific data collection requirements, and requirements as to the frequency and disclosure of the outcomes of the evaluation analysis. What makes DFIs work or fail? Too little attention has hitherto been paid to the governance arrangements, related safeguards, and design of development finance initiatives. Given the decision to establish a new DFI, how should it be governed to minimize risks and maximize chances of success? A good corporate governance framework not only provides proper incentives for the board and management to pursue key objectives that are in the interest of the institution and its shareholders, but also facilitates effective monitoring of the institution’s performance and xiv operations. A robust corporate governance framework is one that focuses on: (i) the strategic guidance of the company, (ii) the effective monitoring of management by the board, and (iii) the board’s accountability to the company and the shareholders. A major challenge for DFIs is to find a balance between the State’s responsibility for actively exercising its ownership functions (such as the nomination and election of the board) and refraining from imposing undue interference on the management of the company. Many of the problems commonly recognized to afflict DFIs can be associated with, if not attributed directly to, weaknesses in corporate governance. Identified examples5 of such weaknesses include:  Government officials acting in the capacity of shareholders directly intervening in day-to-day operational decisions, but also government in all its forms (even without a formal role) asking DFIs for special favors (e.g., to whom to lend, on what terms to lend, and when to forgive indebtedness): “political intervention” or “political capture”;  Executives acting almost autonomously (without clear reporting lines), pursuing unintended objectives: “mission creep”; taking decisions contrary to commercial and/or financial management principles thus eroding the institution’s “self-sustainability”;  Board members lacking the necessary experience, skills, and capacities to effectively and properly exercise their functions according to the institution’s objectives; and  Lack of accurate and complete reporting (on financial and nonfinancial matters alike), giving rise to inadequate decision-making by those who rely on reporting, and thereby misleading shareholders, investors, legislatures, and society in general. The following outlines briefly the core aspects of corporate governance to be applied to the Development Bank of Nigeria: Ownership and control: Although the majority of DFIs are government-owned, the DBN’s mixed public–private ownership structure provides a number of safeguards and has emerged as good practice. The advantage of mixed ownership is that it reduces the risk of governance failures because it provides protection to the DFI in the form of safeguards from political intervention in the DFI’s day-to-day operations while maintaining the DFI’s focus on public policy goals. In the case of investment by so-called “impact investors,” themselves “soft investors” with broader socioeconomic objectives and/or multilateral development institutions, DFIs in developing and emerging markets stand to benefit considerably from the experience and technical expertise that such investors have to offer, both as regards governance and management practices, but also in designing and implementing development programs. Currently the federal government of Nigeria has provided NGN 20 billion as paid-up capital to the DBN. In addition, African Development Bank and European Investment Bank have committed to provide equity of US $50 million and US $20 million, respectively. In addition to direct private ownership, “control arrangements” can also be put in place as a way to ensure that private sector shareholders (equity investors or institutional lenders) 5 David H. Scott, “Strengthening the Governance and Performance of State-Owned Financial Institutions,” Policy Research Working Paper, The World Bank, August 2007. xv exercise control on key management and governance issues. These mechanisms provide assurance that the DFI is properly run; fulfills its performance, financial, and public policy objectives; maintains its sustainability; and is as far as possible insulated from political interference. In this context, the Memorandum and Articles of Association (MemArt) for DBN sets out a number of safeguards and provides for a list of “reserved matters,” which require approval by 75 percent of the total voting power of members. As on-lenders such as the World Bank and the African Development Bank will be providing much of the initial funding for the DBN’s operations, donors (whether providing equity as impact investors or loans) will have—at least initially—considerable influence over these matters. Most SOEs use some kind of “performance contract” between the ownership entity/ministry and the DFI/board. This contract should be negotiated with the board (ideally on an annual basis), and should include key performance indicators that allow for performance to be clearly understood and evaluated. As regards implementation of performance contracts, the involvement of nongovernmental owners will encourage compliance with such performance contracting procedures. Board of directors and management: Considering that the DBN will initially remain a majority government-owned institution, the board appointment process is one of the key factors to ensure that the DFI fulfills its objectives. A key challenge in board composition is to ensure that the board collectively has the mix of skills, experience, and capacity needed to conduct the business of the DFI in an efficient and professional way. The most effective way to control potential weaknesses, and prevent abuses, is to put in place measures—as currently enshrined in the Memorandum of Association of the DBN—aimed at strengthening the governance arrangements around board and management, including: (i) a formal, merit-based, transparent process for appointing independent directors to the board as well as performance-based assessment and salaries of DBN management; (ii) majority of the board to be comprised of independent, highly qualified, professional, and experienced directors who are competitively selected; (iii) encouragement of minority shareholder representation and reputable international investors’ participation on the board of the DBN; and (iv) competitive selection and appointment of key executives (including the CEO) by the board with the participation, to the extent possible, of all shareholders. Mandate: DFI mandates need to be tight enough so as to prevent “mission creep” and flexible enough to give room to adjust the path of the DFI should its mission become less relevant. The latter is particularly important to the Public/Private approach to DFIs—making sure that DFIs preserve their “additionality” by continually seeking to push the frontier of the production possibilities of the financial system. The Memorandum and Articles of Association provides the DBN with quite a narrow development finance mandate. The DBN’s main objective is “[t]o alleviate financing constraints faced by MSMEs and small corporates in Nigeria through providing financing, partial credit guarantees and technical assistance to eligible financial intermediaries on a market-conforming and fully financially sustainable basis.” Furthermore, DBN is mandated only to on-lend its funds on a wholesale/second-tier basis through “eligible financial intermediaries” in xvi order to fulfill its purpose of “increas[ing] availability and access to finance” to “MSMEs and small corporates in Nigeria.” Operational and financial sustainability: Confining DFIs to second-tier operations contributes to improved performance and leverages the expertise of participating private sector financial institutions (PFIs) in managing credit risk. This contributes both to more cost-effective utilization of scarce public funds and to reducing the risk that DFIs expand their activities into areas, such as managing credit risk, where they would in effect be duplicating rather than supplementing the capacity of private sector intermediaries. Risk management is a critical component of the overall accountability framework and ultimately an essential determinant of performance. It is also an important governance tool with a direct impact on sustainability. Risk management will be an integral part of the accountability framework, whereby the performance of the DBN’s board will be assessed according to a performance contract established between the shareholders and the board. Disclosure and transparency: This is a key aspect of governance involving instituting (i) efficient internal audit procedures and audit function monitored by the board and the audit committee; (ii) an annual independent external audit based on international standards—namely, the same high-quality accounting and auditing standards as for listed companies; and (iii) disclosure of financial and nonfinancial information according to high-quality internationally recognized standards. While the MemArt for DBN requires keeping of accounting and other records and distribution of balance sheets (as required by law) as well as the recording of minutes regarding the board workings, ultimate responsibility for ensuring that the DBN lives up to these standards rests with the DBN’s board. Regulation and supervision: Independent and effective regulation and supervision of DFIs is a basic condition for sound governance and for ensuring good performance and financial sustainability. It is critical that DFIs henceforth be regulated and supervised by the CBN. The CBN has developed a set of draft guidelines outlining the supervisory principles that it intends to adopt and apply to all Nigerian DFIs. These arrangements are very similar to standards applied to commercial banks, but modified to take into consideration the business model of the specific DFI. In the case of the DBN, the risk exposure will be considerably lower than that of commercial banks and other existing DFIs: on the asset side, the DBN will not undertake any direct lending and will only be performing second-tier, wholesale functions; and on the liability side, the risk exposure of the DBN will be considerably less than that of commercial banks because it will not be soliciting deposits from the public. Although the current framework proposed by the CBN distinguishes between wholesale and retail DFIs, it applies higher minimum capital requirements to wholesale DFIs. Specifically, it sets the minimum capital requirement for wholesale DFIs at ten times the NGN 10 billion minimum capital requirement applied to retail DFIs and four times the minimum capital of NGN 25 billion applied to commercial banks. Given the lower level of risk assumed by wholesale DFIs, it may be advisable to consider realigning the minimum capital requirements of wholesale DFIs so that they reflect the relatively low level of risk that they assume as “arms-length” lenders that rely on the credit assessment skills of eligible private sector retail banks. xvii The success of the DBN depends on establishing a robust corporate governance framework. Box 1.1 below summarizes the key actions required in implementing the governance framework. Box 1.1: Key actions ensuring strong DBN corporate governance Ownership and Control  Mixed ownership to provide protection from political intervention in the DBN’s day-to-day operations  Control arrangements to ensure that private sector shareholders (equity investors or institutional lenders) exercise control on key management and governance issues  Performance contracts between the ownership entity/ministry and the board members to allow performance to be evaluated against key performance indicators Board and Management  A formal, merit-based, transparent process for appointing independent directors to the board as well as in setting the salaries of board members and DBN management  Majority of the board to be comprised of competitively selected, independent, highly qualified, professional, and experienced directors  Minority shareholders and reputable international investors to be encouraged to participate on the DBN’s board  Reputable international investors to participate in selection and appointment of key executives (including the CEO) in order to minimize the risk of political capture Mandate  Clearly defined to provide financing and partial credit guarantees to eligible financial intermediaries on a market-conforming and fully financially sustainable basis  On-lending to be on a wholesale/second-tier basis through eligible financial intermediaries with the purpose of increasing availability and access to finance to MSMEs and small corporates Operational and Financial Sustainability  Efficient internal audit procedures and audit function monitored by the board and the audit committee  Annual independent external audit based on international standards for listed companies  Disclosure of financial and nonfinancial information according to internationally recognized standards Regulation and Supervision  DFIs to henceforth be regulated and supervised by the CBN  Apply minimum capital requirement to the DBN that is commensurate with the low risk-profile and limited capital investment needs of a wholesale DFI Vigilance in making sure the DBN achieves development impact Constant vigilance by the board and management of DBN will be essential in ensuring implementation of the governance framework outlined above, and that the resources made available to the DBN are used as effectively as possible in addressing the identified market xviii failure in the provision of MSME financing. It will fall on the board of directors and on management to interpret the safeguards built around the DBN, effectively giving them teeth and ensuring that the new bank will successfully fulfill its mission. Private sector participation (and/or that of other nongovernment actors such as international financial institutions or impact investors) will also be crucial to this end. Indeed, with credible and effective implementation and an established track record, there is every reason to believe that the DBN will experience a virtuous cycle, whereby private sector funding will be accompanied by further donor contributions, thereby continuing to ease the DBN’s cost of funding. A crucial link in the governance process will be the accountability framework used in assessing the performance of DBN’s board and management to its shareholders. The DBN should be evaluated according to a set of performance criteria, and its management and board should be held accountable when ex post evaluations indicate underperformance. Experience shows that most reforming state-owned banks use some kind of “performance contract” between the ownership entity/ministry and the DFI/board. This contract should be negotiated with the board (ideally on an annual basis), and should include key performance indicators that allow performance to be clearly understood and evaluated. xix Chapter 1—Why a New DFI in Nigeria Now? Problem Identification and Policy Context I The Nigerian context A critical moment to address Nigeria’s macroeconomic vulnerability The 2015 Nigeria general election opens new opportunities for Nigeria amid social and economic challenges. The 2015 election marked the first time of a democratic transfer of power by a ruling party to an opposition party in Nigeria’s history. Sustainable and inclusive economic development is particularly critical at this juncture in channeling pressures arising from the security situation and an increasing pool of underemployed youth toward development initiatives and improved social welfare. Nigeria recorded robust growth for more than a decade, and its economy has become more diversified with less reliance on oil. Until the recent dip in oil prices, Nigeria was among the fastest-growing countries in the world, with an average economic growth of 6.8 percent a year over the decade 2003–2013 (see Figure 1.1) and its economy accounts for 35 percent of the GDP of Sub-Saharan Africa (IMF Article IV Consultation Nigeria 2014). Nigeria’s recent growth has been led by the non-oil sector (especially domestic consumption), and the service sector contributed over 50 percent of GDP in 2013 (see Figures 1.2 and 1.3). Figure 1.1: Growth rate—BRICS and MINT countries Growth Rates - BRICS & MINT Countries (IMF WEO) 30.00 25.00 20.00 Brazil China 15.00 India Indonesia 10.00 Mexico Nigeria Russia 5.00 South Africa Turkey - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 (5.00) (10.00) Source: IMF World Economic Outlook database. 1 Figure 1.2: GDP by sector 2013 Figure 1.3: Non-oil led growth Services Manufacturing Mining Agriculture 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Pre-rebasing 2010 Post-rebasing 2010 Post-rebasing 2013 Source: IMF Article IV Consultation Nigeria 2014. (Original data sources: National Bureau of Statistics Nigeria.) Nonetheless, oil revenue remains a critical source of fiscal revenues and foreign exchange, and the recent plunge in oil prices demonstrates how exposed the fiscal situation remains to changes in oil prices. Although the oil and gas sector only contributes about 12 percent of GDP, it represents more than 60 percent of total general government revenue (see Figure 1.4). Government revenue generated from non-oil revenue is just 4.5 percent of total non-oil GDP, compared to an average of 10–15 percent of non-oil GDP for other oil producers (IMF Article IV Consultation Nigeria, 2014) (see Figure 1.5). Thus, despite the recent strong growth of the non-oil sectors, budgetary revenues remain highly sensitive to external shocks such as fluctuations in oil prices, which in turn limits the ability of government to provide public goods and services. “Over the medium and longer term, Nigeria’s future prosperity will depend critically on improvements in non-oil growth and non-oil government revenues” (World Bank, Nigeria Economic Report (2014)). Figure 1.4: Government revenue (percent of Figure 1.5: Non-oil tax revenue GDP) 1990–2014 (percent of GDP) Source: IMF Nigeria Selected Issues Paper (2015). The recent oil price plunge provides an opportunity to improve the economic and budgetary structure. Downward pressures on the naira could also provide an opportunity to strengthen the competitiveness of local Nigerian producers, while future positive movements in oil and gas prices may lead to currency appreciation and negatively impact exports of domestic goods and services. 2 A negative effect of the so-called “resource curse” or “Dutch disease” is that governments tend to overspend when oil revenue is high and are forced to cut spending when oil revenues fall.6 The government has set up a sovereign wealth fund that will allow it to build a “cushion” of resources in times when oil prices are high that can be drawn upon when oil prices fall, while also allowing the government to sterilize the impact of liquidity generated by oil revenues when oil prices rise, thus neutralizing the detrimental impact of Dutch disease. Reducing the impact of volatility in oil prices will also be important in sustaining growth in domestic production and thereby reducing the unemployment rate (especially unemployment among the young population). The Nigeria Bureau of Statistics (NBS) official estimate of unemployment almost doubled from 14.8 percent in 2003 to 23.9 percent in 2011. The share of young people in the total population is estimated at about 51 percent according to the Living Standards Survey of 2010, and in 2011 the national youth unemployment rate was 38 percent. Youth unemployment is particularly high in the north, especially in the states of Kaduna (50 percent) and Niger (47 percent). Although the NBS uses a rather unusual definition of unemployment7 that is subject to ongoing methodology review,8 the data do reveal an upward trend in the overall unemployment rate and high youth unemployment (see Figure 1.6). Figure 1.6: Nigeria unemployment rate (official estimates) 2003–2011 Source: NBS Annual Abstracts of Statistics (2011 and 2012). In addition, the majority of jobs created are in the informal sector, highlighting the scarcity of jobs in the more stable and potentially productive sectors. According to National MSME Collaborative Survey 2013,9 MSMEs accounted for 84 percent of Nigeria’s national 6 Literature on the Dutch disease and the pro-cyclicality of government spending include: Gavin and Perotti (1997); Kaminsky, Reinhart, and Vegh (2004); Talvi and Vegh (2005); Alesina and Tabellini (2005); Mendoza and Oviedo (2006). 7 Unemployment estimation adopted by Nigeria authorities was based on 39 hours of work in a week, rather than the International Labor Organization (ILO) definition of one hour of work in a week. 8 Unofficial assessments estimate an unemployment rate likely lower than 10 percent according to the ILO definition, according to NBS and the World Bank Nigeria Economic Report (2014). 9 National Bureau of Statistics and SMEDAN Collaborative Survey (2013). 3 employment.10 According to the NBS Quarterly Job Creation Survey 2014,11 sectors of lower productivity and lower growth, especially the informal sector in which firms are mostly micro or small firms, provided the majority of new jobs (see Figure 1.7). Many Nigerians cannot afford to be completely unemployed and engage in low-productivity and low-paying tasks for survival. Thus, formalizing and strengthening the activities undertaken by the MSME sector12 could significantly contribute toward providing jobs of higher quality. Figure 1.7: Nigeria—quarterly jobs creation by segment (2012 Q3–2014 Q4) New Jobs Crea on for All Sectors Formal sector Informal sector Public ins tu ons 500,000 450,000 400,000 350,000 300,000 250,000 200,000 150,000 100,000 50,000 0 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 2012 2013 2014 Source: NBS (March 2015). Selected Tables from Job Creation and Employment Generation Survey, 4th Quarter 2014. Potential role of financial deepening Evidence and theory support the view that a healthy financial system is important to economic stability and growth. Although inappropriately regulated financial activities and innovations can cause turmoil in financial markets, as was experienced on Western markets and in Nigeria in 2008/2009, provision of access to finance on a sustainable basis continues to be a key constraint in most developing countries. In particular, limited access to finance by MSMEs and rural households increases income volatility and reduces the ability of these households, which are especially exposed to fluctuating incomes, to mitigate exogenous shocks, thereby undermining the ability of the poor households to move out of poverty. Empirical findings confirm a strong causal relationship between a sound financial system and economic growth,13 and that financial system development promotes more inclusive, shared prosperity by providing funding for viable projects where internal funding is not available.14 Efficient and inclusive financial systems are an important catalyst in promoting equitable growth and poverty reduction. 10 http://leadership.ng/business/390893/msme-accounts-75-national-employment-says-aganga. 11 NBS (March 2015). Selected Tables from Job Creation and Employment Generation Survey, 4th Quarter 2014. 12 Reference MSMEs and SMEs are used deliberately throughout this document. While the overall focus is on supporting MSME access to finance, different terminology is used to describe specific issues and/or instruments supporting specific market segments (e.g., SME vs. MSME). 13 An overview is provided by R. Levine (2005), “Finance and Growth: Theory and Evidence” in P. Aghion and S. Durlauf (eds.), Handbook of Economic Growth, Elsevier, Amsterdam. 14 See T. Beck, A. Demirguc-Kunt, and R. Levine (2004), “Finance, Inequality, and Poverty: Cross-Country Evidence,” World Bank Policy Research Paper 3338. 4 Nigeria suffers from a low level of credit provision. Cross-country data show that the level of credit provision to the private sector as a percentage of GDP is low in Nigeria (see Figure 1.8), but that it is also significantly below the level of peer countries at similar levels of per capita income (see Figure 1.9).15 Figure 1.8: Credit to private sector (as % of GDP) and GDP per capita—2013 Source: The World Bank. WDI. The red dot represents Nigeria. Figure 1.9: Private sector credit as a percentage of GDP (2013) South Africa China Brazil Turkey Russian Federa on India Indonesia Mexico Nigeria 0 20 40 60 80 100 120 140 160 180 Source: The World Bank. WDI. Firm-level data also reveal that Nigerian firms face severe credit constraints. According to the Enterprise Survey, bank financing only accounts for 1.6 percent of financing of investments 15 Nigeria is significantly underperforming when compared to peer groups. Closer analysis that controls for Nigeria country-specific factors using the FINSTATS database shows that Nigeria is still performing lower than expected, albeit less dramatically. 5 made by Nigeria firms, far below Nigeria’s peer countries (see Figure 1.10). This further underlines how important it is to strengthen the capacity of Nigeria’s financial system to improve Nigeria’s growth potential. Figure 1.10: Proportion of investment financed by various sources Source: World Bank Enterprise Survey. Latest available survey results for country data: Brazil 2009, China 2012, India 2014, Indonesia 2009, Mexico 2010, Nigeria 2014, Russia Federation 2012, Turkey 2013, South Africa 2007. Among the possible causes of low credit provision to the private sector are warning signs indicating that public sector borrowing may be crowding out private sector borrowing. As shown in Figure 1.11, credit to private sector (as a percentage of GDP) is closely correlated with credit provided by the financial sector (as a percentage of GDP). In Nigeria, credit provided to the private sector in proportion to total credit was below the estimated global average level. A closer look at the breakdown of claims on private sector, public sector, and nonresident banks in Nigeria shows that the proportion of claims on the private sector have been shrinking while claims on the public sector have expanded (see Figure 1.12). Bank credit is the main source of formal financing in Nigeria. Underdeveloped corporate bond and equity markets have resulted in bank credit being the main source of formal financing for Nigerian corporations. At the end of 2013, deposit money bank assets accounted for 30.3 percent of GDP (after the GDP rebasing exercise). The twenty-two deposit money banks dominate the financial system. In addition to dominating the lending market, banks are the main players in the money markets and also act as settlement agents on the capital market. Bank shares account for more than one-third of the market value of listed companies and are among the most actively traded shares. 6 Figure 1.11: Credit to private sector Figure 1.12: Nigeria claims on private correlated with total credit provided by sector, public sector, and nonresident banks financial sector (as % of GDP) Data source: WDI, IMF Article IV Consultation 2013. Following the 2009 banking crisis, the Nigerian authorities took decisive measures to strengthen the banking system. The tenfold increase in the minimum capital requirement for banks in 2005 had led to the consolidation of the banking system (from 89 to 24 banks), and rapid expansion of the banks’ lending. This lending was undertaken partly so as to finance the purchase of the banks’ new share issuance (so-called margin loans) and also in an effort to earn returns for those who had invested in the banks’ significantly expanded capital base. With the stall in economic growth and the collapse in share prices in 2008, both the banks’ margin loans and their new lending contributed to rapid deterioration in the quality of the banks’ assets that resulted in the banking crisis of 2008–2009. In resolving the banking crisis, the authorities established the Asset Management Company of Nigeria (AMCON)—allowing banks to surrender their nonperforming assets to the corporation—strengthened the CBN’s supervisory enforcement and adopted International Financial Reporting Standards (IFRS) for all banks as of end of 2012. These measures resulted in further restructuring and consolidation of the banks (from 24 to 22 banks) and a stronger and more resilient banking system. While the Nigerian banking system has made progress in terms of asset quality and observance of prudential requirements with the support of capital injections made by AMCON and tightened supervision, the banking system remains vulnerable to fluctuations in oil prices. The level of banks’ nonperforming loans (NPLs) declined steadily from the peak of 35.6 percent in September 2010. As of December 2014, the NPL ratio was 2.9 percent. Further, banks’ capital adequacy ratio (CAR) grew from less than 2 percent to more than 17 percent from end of 2010 to end of 201416 (see Figure 1.13). Nonetheless, the quality of bank loan portfolios is vulnerable to the decline in oil prices, due to sizeable direct exposure to the oil sector (24 percent of total loans as of end of June 2014), indirect exposure to the oil sector (lending to the value 16 The regulatory threshold for the CAR is 10 percent, and 15 percent for internationally active banks. In 2014 banks’ return on assets (ROA) was 2.1 percent, while the return on equity (ROE) amounted to 21.2 percent. 7 chains supplying the oil industry), and exposure to loans denominated in foreign currency (21 percent of total loans as of end of 2013) (IMF Article IV Consultation Report Nigeria 2014). Figure 1.13: Banking sector performance indicators Industry Capital Adequacy Ratio Capital Adequacy Ratio (end of 2014) Number of banks by CAR level 14 12 10 8 6 4 2 0 Banks with Banks with Banks with CAR Banks with CAR Banks with CAR CAR>=15% CAR>=10% but <10% but>5% <=5% but >0 <=0 <15% Nonperforming Loan Ratio Returns on assets and equity in 2014 Industry-wide NPL ra os (in %) Returns of banks in Nigeria 2014 40 (in %) 35 30 30 25 25 20 20 15 ROA 15 10 ROE 10 5 5 0 0 Nov '10 Mar '11 Jun '11 Dec '11 Jun '12 Dec '12 Jun '13 Dec '13 Jun '14 Dec '14 Banking industry Large banks Medium banks Small banks Sources: CBN Financial Stability Report 2014 and IMF Article IV Consultation Report Nigeria 2014. A trend toward dollarization of deposits and increasing foreign borrowing increased the banks’ exposure to foreign exchange risks. The authorities view a relatively stable exchange rate as a key plank in maintaining macroeconomic stability. Wanting to avoid any sharper depreciation of the naira/USD that could give rise to inflationary pressures, the authorities remain committed to a continued policy of measured exchange rate depreciation coupled with fiscal prudence; i.e., a combination of macroeconomic policies aimed at enhancing competitiveness and strengthening local production. Nonetheless, dollarization increased from 15 percent of deposits in 2011 to 23 percent by end of 2014. Since 2011, banks have also issued dollar-denominated debt of $3.7 billion to take advantage of the favorable global conditions. As a result, about 20 percent of banks’ total liabilities are denominated in foreign currency, thereby increasing the banks’ vulnerability to foreign exchange risk (IMF Article IV Consultation Report Nigeria 2014).17 17 In 2015–2016 the banks reduced their foreign borrowing due to developments on the foreign exchange market. In preserving a fixed official exchange, the authorities restricted access to foreign exchange, making it difficult for banks to honor their foreign exchange commitments. In addition, the depreciation of the market exchange rate revealed the risks associated with foreign borrowing to the banks’ clients. 8 Interest rates charged by banks to borrowers are relatively high due to a combination of macroeconomic factors and continuing challenges in the operating environment for bank lending. From October 2011 the CBN’s monetary policy rate (MPR)—that sets the benchmark for government borrowing and deposit money bank interest rates—remained stable at 12 percent for a prolonged period. It was increased to 13 percent in November 2014 in response to exchange rate pressures18 (see Figure 1.14). In November 2015 the CBN lowered the MPR to 11 percent and replaced the symmetric plus/minus 2 percent symmetric corridor for money market interest rates with an asymmetric corridor of plus 2 percent and minus 7 percent. After having remained more or less stable at around 11 percent since 2013, the Nigeria treasury bill rate fell to between 5.3 percent (three months) and 8.5 percent (one year) in December 2015. Given continuing pressures on the foreign exchange market, the CBN raised the MPR to 12 percent in March 2016, while in parallel placing increasing reliance on quantitative restrictions on access to foreign exchange. Latest data suggest that the banks’ prime lending rate has remained in the 16 percent to 17 percent range since 2011.19 The question remains as to what extent lower domestic market interest rates will encourage banks to reduce their lending rates, particularly given the policy uncertainty reflected in the wedge between the official exchange rate and black market exchange rate. In addition, as a result of the tightened supervisory requirements introduced subsequent to the 2009 banking crisis and continuing challenges in the overall operating environment, banks have every incentive to charge relatively high spreads on their lending (see Figure 1.15). As regards the operating environment, banks face challenges in several areas. While the coverage of the information provided to Nigeria’s credit bureaus has improved in recent years, the quality and timeliness of the available information remain challenges as do weak property and collateral registration, as well as uncertain and costly foreclosure practices (see further discussion in Section II C below). 18 Information source: http://www.cenbank.org/monetaryPolicy/decisions.asp. 19 Data source: http://www.cenbank.org/rates/mnymktind.asp?year=2015. 9 Figure 1.14: Monthly interest rates in Nigeria money market (2006–2015) Nigeria Money Market Indicators (percent) MPR Treasury Bill Twelve Months Deposit Prime Lending 25 20 15 10 5 0 Jan-06 Apr-06 Jul-06 Jan-07 Apr-07 Jul-07 Oct-07 Jan-08 Apr-08 Jul-08 Oct-08 Jan-09 Apr-09 Jul-09 Oct-09 Jan-10 Apr-10 Jul-10 Jan-11 Apr-11 Jul-11 Oct-11 Jan-12 Apr-12 Jul-12 Jan-13 Apr-13 Jul-13 Jan-14 Apr-14 Jul-14 Jan-15 Oct-06 Oct-10 Oct-12 Oct-13 Oct-14 Data source: CBN. Note: In December 2006, the CBN adopted the MPR, replacing the MDR (minimum discount rate). Figure 1.15: Interest rate spreads in BRICS and MINT countries20 China Indonesia Mexico Russia South Africa Nigeria 20.0 18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Source: The World Bank. WDI. Brazil’s interest spread has been extremely high, despite a recent drop from around 40 percent to around 20 percent. 20 We have thus excluded Brazil data for a graphic purpose. Data for India and Turkey are missing. 10 As an outcome of the factors described above, banks in Nigeria have adopted defensive strategies to manage their credit risks. While banks compete fiercely for the business of large, well-regarded corporate clients, small corporates that operate outside the value chains associated with resource extraction find it challenging to access bank lending on affordable terms. Banks provide loans to large corporates that are predominantly associated with resource extraction industries at interest rates below the prime rate so as to secure both the business of the large corporates as well as the business provided by smaller corporates that deliver directly to the value chains supported by these large corporate clients. Strong competition among banks for the business of the larger corporations leads to a “negative feedback loop” pushing up the borrowing costs of smaller corporates that deliver to “blue chip” large corporations—even those with relatively good credit risk. Thus, when accessing bank financing, smaller corporates rely on the security provided by the “blue chip parent” enterprise in the value chain to which they deliver. Indeed, such an enterprise may become captive or reliant for funding on the bank of the large corporate to which it delivers in the value chain, as finding alternative funding sources with competitive loan terms may be difficult. Bank lending rates for loans to riskier market segments, such as the SME sector, and on loans offered by nonbank financial institutions (i.e., microfinance banks) are much higher than the banks’ effective lending rates to large corporates. In addition, SMEs are exposed to uncertainty as regards unilateral changes in borrowing terms imposed by banks and weak transparency as regards borrowing costs and conditions, which exposes them to ad hoc adjustments in their borrowing terms. Risk aversion as regards SME lending is in part a reflection of weaknesses in the banks ’ lending methodologies. Banks rely heavily on relationship banking and have limited capacity to apply cash flow lending methodologies that leverage information on their clients’ cash flows and make very limited use of small loan scoring systems that would allow them to process a higher volume of loans more efficiently. While SMEs are constrained in being able to meet the banks’ documentation requirements and in having the capacity to prepare bankable projects, banks face high costs in verifying the identity of their clients, in managing their risks effectively, and in providing suitable lending products. The tenors of funding offered by banks are short. Tenors of more than one year for MSMEs are extremely scarce, with longest tenors of about three to five years. In general, the lack of appropriate SME lending methodologies and the banks’ limited understanding of how to assess the potential viability of SMEs and appropriately mitigate the associated risks severely constrain the development of SME finance. II Defining the MSME financing gap A. The MSME financing gap in international context The financing of MSMEs is a global challenge, with a total estimated credit gap of more than $3 trillion. According to the latest available estimates in the IFC Enterprise Finance Gap Database (2011),21 total MSME credit gap is estimated to be about $3.1 trillion to $3.8 trillion (see Figure 21 The IFC Enterprise Finance Gap Database adopts a combination of methods including survey data and assumptions based on experience, whenever data is available. Such data may not be as updated as 2011, and the methodologies applied may not be consistent, thus the robustness of the estimated financing gap is qualified. Nevertheless, the database provides a useful estimate for the range and global comparison of financing gaps. 11 1.16). Sub-Saharan Africa countries alone accounted for $140 billion to $170 billion of the estimated MSME credit gap. Relative indicators show that Sub-Saharan Africa countries suffer from the most severe MSME financing gap. Although high-income OECD countries and East Asia are estimated to face the largest absolute formal MSME credit gap, the size of outstanding credit lines to MSMEs and GDP in these economies are very large compared to those of Africa. Thus, the degree of financing constraints faced by MSMEs across regions is better viewed in relative terms. The SME credit gap in the formal sector in Sub-Saharan African countries amounted to 300 percent to 360 percent of current outstanding SME credit, the highest gap relative to outstanding SME credit among all regions (Figure 1.17). Indeed, globally the informal sector contributed more to the total MSME credit gap (Figure 1.18), and developing countries tend to suffer more from informality in their economic structures. Figure 1.16: Overview of MSME financing gap by region Source: P. Stein, T. Goland, R. Schiff (2010), Two Trillion and Counting: Assessing the Credit Gap for Micro, Small, and Medium-Size Enterprises in the Developing World. 12 Figure 1.17: The financing gap of formal SMEs Source: IFC Enterprise Finance Gap Database (2011). Graph from P. Stein, T. Goland, R. Schiff (2010), Two Trillion and Counting: Assessing the Credit Gap for Micro, Small, and Medium-Size Enterprises in the Developing World. Figure 1.18: Financing gaps in formal and informal SMEs Source: IFC Enterprise Finance Gap Database (2011). Graph from P. Stein, T. Goland, R. Schiff (2010), Two Trillion and Counting: Assessing the Credit Gap for Micro, Small, and Medium-Size Enterprises in the Developing World. 13 Nigeria is among the countries with the most severe MSME credit gaps:  In absolute terms, Nigeria’s MSME financing gap is estimated to be $27.6 billion (IFC Enterprise Finance Gap Database (2011)), accounting for more than 21 percent of the total absolute credit gap in Sub-Saharan Africa (see Figure 1.19).  In relative terms, Nigeria’s estimated credit gap as a percentage of GDP accounts for 16.6 percent of its GDP in 2011, close to the high end of MSME credit gaps relative to GDP among BRICS and MINT22 countries. In 2011, annual credit provided to the private sector in Nigeria only accounted for 12.5 percent of GDP (see Figure 1.20). Figure 1.19: MSME credit gap contribution Figure 1.20: MSME credit gap as a in Sub-Saharan Africa percentage of GDP 2011 in BRICS and MINT countries Credit gap - Sub-Saharan Africa USD Es mated MSME Credit Gap as % of GDP 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 21% Brazil 21.1% Nigeria 16.6% Other Sub-Saharan Africa Turkey 11.8% 10% South Africa India 10.6% Nigeria China 8.1% 69% Indonesia 6.8% Mexico 6.4% South Africa 4.5% Russia 0.5% Source: IFC Enterprise Finance Gap Database (2011), WDI. B. Other indicators of the MSME financing gap The World Bank’s Enterprise Survey uses a representative sample of all firms and also provides evidence of the severity of the MSME financing gap in Nigeria. Firms in BRICS and MINT countries were asked to identify major constraints to their businesses. More than 40 percent of Nigerian firms identified access to finance as a major constraint, the second-highest among all BRICS and MINT countries (only second to Brazil). However, the data also reveal that most large firms in Nigeria do not identify the availability of finance as a constraint. Indeed, among all BRICS and MINT countries, Nigeria stands out as demonstrating the widest gap between the percentage of large firms and the percentage of small and medium firms to identify access to finance as a major constraint (Figure 1.21). This is consistent with the analysis above whereby Nigerian banks have adopted a defensive strategy in managing their credit risks: banks lend predominantly to the value chains servicing “blue chip” enterprises. While there are some exceptions, most banks tend 22 BRICS: an acronym for Brazil, Russia, India, China, and South Africa. This term was created by Jim O ’Neil, the then Goldman Sachs economist. MINT: an acronym for Mexico, Indonesia, Nigeria, and Turkey. This term was created by Fidelity Investments and popularized by Jim O ’Neil. More discussions can be found in O’Neil’s article on Bloomberg Business Who You Calling A BRIC? http://www.bloomberg.com/news/articles/2013-11-12/who-you- calling-a-bric- [Accessed on April 21, 2015]. Although the grouping of countries is still open for debates given the large variety of country specifics in the two groups, such a grouping provides a useful benchmark for Nigeria, in particular in the investment sector. 14 to treat SMEs as “small corporates” with no clear understanding of SME borrowing needs and cash flow profiles. As a result, the percent of small- and medium-sized firms in Nigeria that have a loan or active credit line is among the lowest in BRICS and MINT countries23 (see Figure 1.22). These benchmarking studies support the conclusion that SME financing is a severe constraint in Nigeria. Figure 1.21: Percent of firms surveyed identifying access to finance as a major constraint — BRICS and MINT countries (by country total and firm size) 50 45 40 35 30 25 20 15 10 5 - Russian Brazil - Small Brazil Nigeria - Small Nigeria Russian - Small Mexico - Small South Africa - Small South Africa India Indonesia - Small Indonesia Turkey Turkey - Small China China - Small Brazil - Medium Russian - Large Mexico - Large Brazil - Large India - Medium Mexico Turkey - Large India - Large Indonesia - Large South Africa - Large Nigeria - Large China - Large India - Small Mexico - Medium South Africa - Medium Turkey - Medium Indonesia - Medium Nigeria - Medium China - Medium Russian - Medium Source: World Bank Enterprise Survey. Latest available survey results for country data: Brazil 2009, China 2012, India 2014, Indonesia 2009, Mexico 2010, Nigeria 2014, Russia Federation 2012, Turkey 2013, South Africa 2007. Figure 1.22: Percent of firms with a loan or line of credit (by size category) Small firms Medium firms 70.0 70.0 60.0 60.0 50.0 50.0 40.0 40.0 30.0 30.0 20.0 20.0 10.0 10.0 0.0 0.0 Brazil Turkey Mexico South India Indonesia Nigeria Russia China Brazil Mexico Turkey China South Russia Indonesia India Nigeria Africa Africa Source: World Bank Enterprise Survey. Latest available survey results for country data: Brazil 2009, China 2012, India 2014, Indonesia 2009, Mexico 2010, Nigeria 2014, Russia Federation 2012, Turkey 2013, South Africa 2007. In Nigeria nearly 87 percent of firms surveyed did not apply for a loan in the past year. The majority of these firms did not apply for a loan due to unfavorable terms and conditions (see Table 1.1). Only 13.1 percent of all SMEs applied for a loan, and only 2.6 percent reported having received full access to a requested loan. More than 47 percent of SMEs reported having been discouraged from applying for a loan due to unfavorable terms and conditions. A closer look at the reasons for not applying for loans or credit lines reveals that the major reasons included 23 See discussion of bank business models in Nigeria in Section I B above. 15 complex application procedures (linked to the higher screening barriers imposed by banks and their limited risk appetite), unfavorable interest rates (much higher than those charged to large firms), and high collateral requirements (again linked to banks’ concerns regarding risks associated with SME lending24). Table 1.1: Credit rationing among SMEs in Nigeria Did not apply 86.9%  No need for a loan 39.8%  Did not apply due to terms and conditions 47.1% o Application procedures were complex 13.2% o Interest rates were not favorable 12.1% o Collateral requirements were too high 11.2% o Size of loan and maturity were 1.0% insufficient o Did not think it would be approved 4.5% o Others 5.2% Applied 13.1%  Applied and approved 2.6%  Applied and rejected, partially or fully 2.1%  Applied but didn’t know or report the status 8.4% Source: The World Bank and IFC. Nigeria Enterprise Survey (2014). C. Structural weaknesses in the institutional infrastructure Nigeria suffers from poor institutional infrastructure for financial intermediation , which includes substantial challenges primarily in the areas of entering into and enforcing contracts and collateral agreements, security registration, and legal and judicial processes. A report25 from 2012 links the conclusions from both demand- and supply side surveys (respectively surveying SMEs and financial institutions), and finds that the main obstacle to access to finance for SMEs is the collateral requirements stipulated by financial institutions. Of particular concern to financial institutions in providing SME finance are the lack of a property and collateral registries both for movable and immovable property, the absence of a functioning personal identification system, and the high costs of realizing collateral in the case of default. These issues are of 24 See further discussions on collateral requirements in Section II C below. 25 Gunhild Berg, Michael Fuchs, Leonardo Iacovone, Thomas Jaeggi, Andrew Lovegrove, and Carolina Villegas Sanchez, “Financing Small- and Medium-Sized Enterprises in Nigeria,” July 2012. 16 concern, as they constrain banks in managing their risks effectively. Credit enforcement (e.g., the ability to seize collateral, including leased assets), antifraud mechanisms (e.g., biometric IDs), and the availability of information on potential borrowers provided by credit bureaus are seen by banks as essential to reducing their costs and the risks associated with lending to SMEs. Weaknesses in the institutional infrastructure were identified as the major obstacle to banks’ involvement by the 2012 SME report.26 The legal and regulatory framework affecting banks, macroeconomic factors, SME-specific factors, and the contractual environment were obstacles particularly highlighted in the responses from the institutions (see Figure 1.23). Figure 1.23: Obstacles to banks’ provision of financial services to SMEs 12 Very significant Significant Number of responses (out of 12) 10 8 6 4 2 0 Competition in environment Bank specific of SME lending Macroeconomic SME specific affecting banks Characteristics adequate affecting SMEs demand Contractual Lack of framework framework factors the SME factors market factors Legal Legal Note: Number of institutions citing obstacle as significant or very significant. Source: Gunhild Berg et al., “Financing Small- and Medium-Sized Enterprises in Nigeria,” World Bank, April 2012. The weakness of the institutional infrastructure for bank lending, including the contractual and collateral environment, and credit information infrastructure, is confirmed by Nigeria’s ranking in the World Bank’s Doing Business survey. Nigeria ranked 170th in the World Bank’s Doing Business of 2015, significantly below all of its peer economies. The legal and regulatory framework and credit infrastructure–related indicators (such as “dealing with construction permits,” “registering property,” and “getting credit”) contributed most to Nigeria’s poor performance (see Table 1.2). 26 See reference in Footnote 19. 17 Table 1.2: Doing Business Index: Nigeria vs. peer economies 2015 Ease of Doing Dealing with Protecting Trading Business Starting a Construction Getting Registering Getting Minority Across Enforcing Resolving Economy Rank Business Permits Electricity Property Credit Investors Paying Taxes Borders Contracts Insolvency Brazil 120 167 174 19 138 89 35 177 123 118 55 China 90 128 179 124 37 71 132 120 98 35 53 India 142 158 184 137 121 36 7 156 126 186 137 Indonesia 114 155 153 78 117 71 43 160 62 172 75 Mexico 39 67 108 116 110 12 62 105 44 57 27 Nigeria 170 129 171 187 185 52 62 179 159 140 131 Russian Federal 62 34 156 143 12 61 100 49 155 14 65 South Africa 43 61 32 158 97 52 17 19 100 46 39 Turkey 55 79 136 34 54 89 13 56 90 38 109 Source: The World Bank Group, Doing Business (2015). Difficulties in contract enforcement and in foreclosing on collateral are a significant constraint to the development of Nigeria’s financial system. According to the 2011 supply side survey, the Nigerian banks were almost unanimous in their views regarding the difficulties associated with registering and enforcing collateral (see Figure 1.24). Figure 1.24: 2011 Survey—Impact of collateral registration and enforcement issues on SME lending Enforcing collateral No Yes Registering collateral 0 2 4 6 8 10 12 Do the following issues affect SME lending? Number of respondents Source: Gunhild Berg et al., “Financing Small- and Medium-Sized Enterprises in Nigeria,” World Bank, April 2012. Nigeria borrowers face extraordinarily high collateral requirements compared to borrowers in peer countries. According to the Nigeria Enterprise Survey (2014), more than 90 percent of loans require collaterals, the highest level among all BRICS and MINT countries, higher than the global average of 77.7 percent and the Sub-Saharan Africa average of 80.7 percent (see Figure 1.25). The average value of collateral was almost 275 percent of the loan amount, also the highest among all BRICS and MINT countries, and substantially higher than the global average of 193 percent and the Sub-Saharan Africa average of 179 percent. Nigerian firms face similarly high requirements for collaterals irrespective of their size, suggesting a fundamental problem in the collateral environment in Nigeria. 18 Figure 1.25: Proportion of loans requiring collateral and value of required collateral Proportion of loans requiring collateral (%) Value of collateral needed for a loan (% of the loan amount) Nigeria Nigeria India India Russian Federa on Mexico Indonesia Turkey Sub-Saharan Africa China All Countries All Countries China South Africa Sub-Saharan Africa Mexico Russian Federa on Turkey South Africa Brazil Brazil 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 0.0 50.0 100.0 150.0 200.0 250.0 300.0 Propor on of loans requiring collateral (%) 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 Br ed ll e Tu Me all ge Af ed l ge M - M all ge Ch ed ll e I n ed l l e Ni - M all ge ian e n de o n on ey um -L m -L m do a do a sia ri a i u m a- m il - m a- m h - M al M a M a -M a I n es i I n es i rg rg rg ss Fed a o ri c i u ico iu in iu az iu di iu il - S m - m m m a- m a Sm m ar ar ar ar ne La La La rk d i ex ed ge ed n n Fe ra ra ey S So rica - S i co - S in - S ria - S -L -L r do az - rk y - di - e ss Fed Af ca Ch ina In ia ge ia ex o Br azil a In Tu rke M exic d Ni iger h fri Ch In Br Ru an Ru n Tu ut A M ia N So uth i ss ut Ru So Value of collateral needed for a loan (% of the loan amount) 350.0 300.0 250.0 200.0 150.0 100.0 50.0 0.0 Br ed l ge h M all ge ge ge Tu Me ll ge M Me all ge ll ge ge a m ia ium in ium ey m ic ium ia m Ch ed l N Me ll Ru M all il m di um l l - M ma a - M ma - M ma a ric i u rk d i u er di u az iu So rica - Sm ey S m - m ia Sm n m ar ar ar ar ar ar ar ar I n ed i Af ed s s ed ex d az - S ic - S di - S ia - S in - S -L -L -L -L -L -L -L -L rk y - er - Af ca In a Br azil ex o ig ia ss n Ch ina a a n o Tu rke di M ic Ru s s i a ut - - - - N er h fri ex o In a Ch il a Br ig ig Tu ut A Ru M N So uth So Source: World Bank Enterprise Survey. Latest available survey results for country data: Brazil 2009, China 2012, India 2014, Indonesia 2009,27 Mexico 2010, Nigeria 2014, Russian Federation 2012, Turkey 2013, South Africa 2007. MSME access to finance is constrained due to weaknesses in the legal and regulatory framework for contract enforcement and foreclosure on collateral. Under the Nigeria MSME 27 Value of collateral data is missing for Indonesia. 19 project that closed in December 2012, the World Bank supported diagnostic work designed to result in reform of the legal and regulatory framework for secured transactions. This review identified the following issues: (i) the current framework is fragmented, with different and conflicting pieces of legislation and serious gaps in certain areas; (ii) the framework is only applicable to companies that are registered in the company registry, therefore excluding the majority of small businesses; (iii) certain types of transactions are in practice transactions that could be secured with movable property, such as discounted invoices, conditional sales contracts, consignments, hire purchase contracts, and long-term leases, but these contractual forms are not encompassed by the legal/regulatory framework; (iv) the current debenture system restricts businesses to accessing credit from only one borrower, thus effectively limiting competition in the market; (v) the existing system fails to establish priority among secured creditors, as only the security interests of corporates are registered. In 2014 the CBN, supported by the IFC, embarked on a project to strengthen the framework for secured transactions that has resulted in the issuance of CBN regulations on registration of security interests in movable assets, the drafting of a bill on secured transactions in movable assets, and the establishment of an electronic centralized registry for movable collateral. While this is still a work in progress, effective implementation will depend on the passage of legislation that strengthens creditor rights. In addition, there is a need to address weak legal enforcement— currently financial institutions cannot rely on timely and reliable court enforcement of contracts. Difficulties in enforcing collateral, as well as long delays in the judicial process, hamper banks’ willingness to lend. Thus, there is a pressing need for a modern, comprehensive, and effective legal framework for insolvency and creditor rights.28 The procedure for taking and registering security is found to be grossly inefficient: high filing fees result in widespread undesirable banking practices. These include reliance on the so-called “equitable charges,”29 and the registration of security by bankers for much diminished values, with a view to increasing the registered amount shortly before seeking foreclosure of the security (a practice referred to as “up-stamping”).30 Altogether, foreclosure, receivership, and litigation are seen as time-consuming and costly due to the weak judicial process. Leasing and asset-based financing remain underdeveloped due to the absence of a legal framework for leasing and a movable asset registry. The advantage of leasing transactions is that they focus less on the balance sheet of the lessee and more on the lessee’s ability to generate 28 This is the conclusion of the “Report on Observance of Standards and Codes on Insolvency and Creditor Rights Systems” (ROSC), World Bank, October 2007. Unfortunately, following the completion of this assessment, little action has been taken on the identified reform needs. 29 “Equitable charge” is a charge that confers a right on the secured party to look to (or appropriate) a particular asset in the event of the debtor’s default, which is enforceable by either power of sale or appointment of a receiver. 30 In practice, parties need not secure the entirety of the borrower ’s obligations in the first instance but may agree on a “notional amount” for stamping purposes and subsequently, where the need arises, “up-stamp” the secured amount to the full obligation. This structure will ensure that the parties only incur the full stamp duty obligation where the need arises. The instrument will only be enforceable in respect to the additional amount from the date of the up- stamping; and charges registered by third parties over the same asset during the intervening period may claim priority. Source: http://www.iflr.com/Article/3127624/Nigeria-Lessons-from-Shell.html. 20 sufficient cash flow to service the lease payment. Even though the leasing market in Nigeria has been expanding, the lack of a specific law clarifying the legal status of leasing is hampering the development of the sector as an alternative to bank financing for SMEs. This problem is reinforced by lack of clarity as to the tax treatment of leases—presently leases are subject to both value-added tax (VAT) and withholding tax. This results in double taxation, as VAT is applied both when the asset is purchased and when it is leased out, making leasing more expensive than other financial products. Microfinance banks are hampered in their ability to fill the financing gap due to regulatory restrictions on their SME lending. Microfinance banks are limited to making only 20 percent of their loans to SMEs (defined as loans above NGN 500,000) and face regulatory uncertainty, for example, relating to proposed maximum daily cash withdrawals and disbursements. The growth of microfinance banks is hampered by tiered minimum capital levels for state, regional, and national licenses, making it onerous to expand the scope of activities. Despite some progress over the recent years, credit infrastructure in Nigeria remains underdeveloped. In 2008 the CBN issued guidelines governing licensing, operations, and regulation of credit bureaus, which facilitated establishment of three private credit bureaus; in 2010 the guidelines were expanded to include a requirement for banks to consult at least two credit bureaus prior to lending; and in November 2013 the guidelines were amended to extend credit reporting to nonbanking credit institutions, and the CBN’s capacity in compliance monitoring was strengthened. To improve data quality, a common data format for credit reporting was developed and adopted by the banks. Nonetheless, the credit bureau industry remains nascent. Key challenges relate to strengthening the quality and timeliness of reported data; ensuring full participation of microfinance banks both as providers of credit information and users of the system; and broadening the scope of credit reporting to select nonfinancial institutions extending credit (e.g., utilities, telecoms, etc.). While credit providers are only gradually gaining awareness of the benefits of credit reporting for their risk management, the existence of the bureaus is resulting in more discipline in the financial system, as borrowers become more aware of the importance of their credit histories. Also banks have started to develop new consumer products based on credit reports. III Addressing the MSME financing gap A. Learning from past policy interventions to address the MSME financing gap in Nigeria Policies designed to support MSME financing in Nigeria are justified. The Nigerian context of a vulnerable macro economy, a shallow financial system with evolving institutional infrastructure, and the imperative of promoting sustainable, balanced economic development justifies policy interventions designed to address the substantial MSME financing gap on a sustainable basis. 21 Yet, when designing new interventions, the lessons from earlier policy interventions need to be borne in mind. Previous attempts to establish DFIs failed to effectively address market gaps, and fell short in facilitating meaningful increases in financial intermediation, or in galvanizing sustainable development in their respective sectors. Previously established, publically owned DFIs in Nigeria were loss-making and thereby had difficulty in fulfilling their development mandate. They encountered difficulties similar to those experienced internationally by public sector DFIs, as further described in chapter 2. Due to their poor financial performance and inability to fulfill their development finance mandate, NERFUND, NBCI, and NIDB were restructured and merged between 1999 and 2001 into a new DFI, the Bank of Industry (see Box 1.2). While this may have resulted in a rationalized institutional structure, evidence from recent years suggests that the Bank of Industry is still run according to a public sector mandate, and as a result the accumulation of losses has continued. Box 1.2: Early DFIs in Nigeria postcolonial times (1) The Nigerian Industrial Development Bank (NIDB) Limited, the first development bank in Nigeria and the processor of BOI, was incorporated in 1964 with an authorized share capital of £2 million. Seventy-five percent of the bank’s equity capital was held by the IFC.31 The primary role of the NIDB was to provide medium- and long-term loans and sometimes equity investments to new and expanding industries. (2) The Nigerian Bank for Commerce and Industry (NBCI) was set up in 1973 to promote the development of small- and medium-size enterprises (SMEs). Until it was subsumed under the Bank of Industry, the principal function of the NBCI had been the provision of long-term investment and equity financing to SMEs. The NBCI had a fully paid share capital of two million ordinary shares at NGN 100 each. The Ministry of Finance held 60 percent of the shares, while the Central Bank of Nigeria (CBN) held the remaining 40 percent.32 (3) Nigeria Agricultural Bank was established with the help of the World Bank in 1973 to provide finance to agriculture.33 (4) The National Economic Reconstruction Fund (NERFUND) was an outreach program that sought to empower the banking sector through the provision of small- and medium- scale industrial loans. NERFUND did not lend directly to beneficiaries: it did so through participating commercial and merchant banks for on-lending to small- and medium-scale enterprises. NERFUND came into existence in January 1989 and commenced operations in September of the same year.34 Existing Nigerian DFIs operate under a public sector mandate and have encountered challenges in achieving their objectives, as highlighted by a recent CBN review. NPL ratios for 31 Website of BOI: http://boinigeria.com/overview/ [Accessed on May 8, 2015]. 32 Nigerianwiki: http://nigerianwiki.com/wiki/Nigerian_bank_for_commerce_and_industries [Accessed on May 8, 2015]. 33 Website of BOA: http://www.bankofagricultureng.com/aboutus/Default.aspx [Accessed on May 8, 2015]. 34 Nigerianwiki: http://nigerianwiki.com/wiki/Nigerian_bank_for_commerce_and_industries [Accessed on May 8, 2015]. 22 the four major development banks in Nigeria ranged from 30 percent to 68 percent in 2010, and almost all of them suffer from losses or barely break even. Benchmarked against the findings of the Global Survey of Development Banks where it is reported that most DFIs make profits or break even, Nigeria’s DFIs show disappointing results. Similar challenges remain to date, as most recently reported by CBN in a report on six35 DFIs that highlights their deteriorating financial performance. This resulted in the decrease of their combined assets of 2.48 percent and of their net loans and advances of 14.73 percent in the six-month period ending December 2015. The capital rating of two DFIs was rated “weak”; one was rated “needs improvement”; and the capital rating for the remaining three was deemed “acceptable.” Moreover, the prudential and soundness analysis of three of the DFIs revealed continued deterioration in their financial performance, owing to inadequate capital, poor asset quality, a continuous stream of operating losses, and weak board oversight. The rather broad mandates of the existing DFIs sanctioned direct provision of credit at the retail level. Their mandates allowed the public sector financial intermediaries to engage in direct investment and commercial risk-taking. On the one hand, the DFIs were constrained in terms of possessing adequate skills to undertake credit decisions and may also have been obliged to make nonviable loans due to their political mandates. On the other hand, the DFIs may in some instances have been making investments and taking credit risks that could otherwise have been undertaken by the private sector (crowding out). This mandate as regards direct lending applied to the Bank of Agriculture (BOA), the Bank of Industry (BOI), the Federal Mortgage Bank, the Nigerian Export-Import Bank (NEXIM), and the Infrastructure Bank (IB). As a result, the Nigerian DFIs were quickly confronted with high levels of NPLs, caused by factors such as lack of appropriate lending methodology and capacity, inadequate incentives to originate and maintain performing loan portfolios and avoid nonperformance, and politically influenced governance structures. As the DFIs struggled to achieve operational and financial sustainability, their ability to fulfill their mandates to any significant scale was constrained. The DFIs’ lack of capital hampered their ability to sustainably stimulate the provision of finance to underserved segments of the economy. According to the government’s Ad Hoc Sub-Committee on Development Finance in Nigeria, which reviewed the DFIs’ performance in 2014, the cumulative losses of the three main DFIs (BOI, BOA, and FMB) in the previous five years had reached approximately NGN 43 billion, eroding their capital to a net negative position despite combined capital injections of approximately NGN 25 billion. The Nigerian authorities also drew on the resources of the CBN to promote funding to constrained segments of the economy, such as the SME sector. The initiatives taken by the CBN were somewhat similar to those taken by central banks in many countries following the 2009 global financial crisis with the purpose of injecting liquidity into the market and stimulating the economy. In Nigeria, however, the central bank has traditionally extended its role beyond preserving the stability of the financial system to promoting finance to specific underserved segments of the economy, including the SME sector. The CBN has fulfilled this extended mandate 35 Bank of Industry (BOI); Federal Mortgage Bank of Nigeria (FMBN); Nigerian Export-Import Bank (NEXIM); Bank of Agriculture (BOA); Infrastructure Bank (formerly Urban Development Bank of Nigeria Plc); and National Economic Reconstruction Fund (NERFUND). 23 by financing subsidized credit schemes, some of which were channeled through DFIs. For example, several of the subsidized credit schemes targeting SMEs were channeled through the Bank of Industry (BOI). The implicit subsidies associated with lending (or on-lending)— calculated as the maximum lending rate minus subsidized interest rate times the credit extended since inception—provided to the BOI were estimated at about NGN 100 billion in 2014 (IMF Article IV 2014 Nigeria Report) (see Table 1.3). Going forward, the risk is that such subsidized CBN schemes may crowd out the opportunities for fully realizing the potential of the DBN, which is required to lend on terms that are market-conform so as to achieve sustainability and scale.36 Table 1.3: Implicit interest subsidies by CBN in 2014 (billion naira) Max Subsidized Amount Interest lending rate outstanding subsidies rate Commercial agriculture credit scheme 25.8 9.0 239.0 40.0 Agriculture credit support scheme 25.8 6.0 0.9 0.2 Refinancing and rediscounting scheme* 25.8 10.0 360.7 56.8 Power and aviation intervention fund* 25.8 7.0 233.6 43.8 Total 140.8 Source: IMF Article IV 2014 Nigeria Report, CBN. (Note: * These items were operated by BOI.) Governance arrangements are crucial to the success or failure of development finance initiatives. The main governance challenges facing DFIs are further discussed in chapter 4. The history of public sector DFIs in Nigeria, a history shared by DFIs in many other countries, reminds us that, despite having mandates focused on addressing financing gaps and correcting market failures, DFIs have difficulty in fulfilling these mandates in an impactful and sustainable way. Despite the imperative of addressing financing gaps and correcting market failures, experience suggests that DFIs often have difficulty using scarce fiscal resources efficiently. In line with the Nigeria Vision 20: 2020, the medium-term implementation strategy, and the World Bank Country Partnership Strategy for Nigeria 2014–2017, the Nigerian government has given priority to MSME development as part of its policy agenda, and has worked actively with the World Bank on finding ways to resolve the MSME finance imperative described in this chapter. The new DBN, with a mandate focusing on providing second-tier finance specifically to the MSME sector, was inaugurated in March 2015, but there is still much work to do to ensure the proper design and operation of the new DFI. Lessons learned from earlier Nigerian experiences—supplemented by lessons garnered from international experience referenced in chapter 2—serve to underline how crucially the success of the new Development Bank of Nigeria (DBN) will depend on effective implementation of the DBN’s specific mandate, strong governance arrangements, and professional risk management capacity. 36 See further discussion in chapter 4 regarding the business model and governance arrangements to be applied to the DBN. 24 B. Complementarity of shorter-term and longer-term solutions It is important to recognize that the weak institutional environment for financial intermediation, calls for both longer- and shorter-term solutions. As noted above in Section II C, weaknesses in the institutional environment relate primarily to challenges in entering into and enforcing contracts; registering and enforcing collateral and security; the availability of reliable credit information; and delays and hurdles in legal and judicial processes. A longer-term solution to these challenges would entail broad, government-led legal and institutional reform programs. For shorter-term solutions, efforts will need to be directed at strengthening the design and efficiency of development finance. This will involve instituting strong governance design and robust safeguards in relation to the establishment of the Development Bank of Nigeria, as further discussed in chapter 4. Embracing both longer- and shorter-term solutions is consistent with the so-called “Public/Private” approach to development finance, as outlined in chapter 2. Both shorter- and longer-term solutions will be important in improving SME access to finance. It is widely recognized that legal and institutional reform programs that contribute to achieving broader societal improvements can take considerable time to implement, as indeed evidenced by experiences in Nigeria. While such legal and institutional reforms are to the benefit of society as a whole, they can be difficult to implement, as entrenched parties benefit from the status quo. Similarly, considerable effort and time are needed for banks to develop skills in serving the financing needs of small enterprises. Thus, a renewed and enhanced effort to implement longer- term solutions needs to be accompanied by efforts that will support provision of SME finance in the shorter term. The activities and possibly also the mandate of the DBN may gradually be adjusted so that its interventions remain on the SME “financing frontier.” This is to ensure that support provided by the DBN supplements the financing being made available by private sector intermediaries, by facilitating access to finance for SME segments that lie beyond the sphere of risk that the private sector is willing to finance. Targeted and flexible administration of the DBN’s mandate will be crucial to achieving results and making sure that the scarce resources at the DBN’s disposal are used to their best advantage. Thus, the shorter- and longer-term solutions need to be implemented hand-in-hand. Either one without the other will significantly reduce and/or delay achievement of the long-term goal of self-sustained, market-based provision of finance to the Nigerian SME sector. Through its efforts in the short term, the DBN also has a role to play in achieving longer- term outcomes. The activities of the DBN can support and strengthen achievement of the broader institutional reforms. The DBN can demonstrate how pilots can be made to work even in the current circumstances through deploying instruments such as partial credit guarantees, or by providing term finance to SMEs where term finance is not otherwise available. Responsibility for undertaking reforms of the broader legal and institutional environment will inevitably require the coordinated and focused efforts of the authorities, and not just the DBN. Even if more concerted efforts are made to strengthen the legal, regulatory, and institutional environment in coming years, the process will take time and only gradually resolve the problem of SME financing in Nigeria. Thus, in pursuing these legal/institutional reforms, the need arises for well- 25 structured funding support to the SME sector that can relieve funding constraints in the shorter term. Institutional reforms, such as strengthening the availability and quality of credit information, could improve MSME access to finance in the shorter term, as they are relatively easy to accomplish, while broader legal reforms of the insolvency and creditor right framework could take longer to put in place. While it is important that the authorities simultaneously embrace both long- and short- term solutions to resolving Nigeria’s SME financing gaps, success is not guaranteed, as design and implementation will be crucial to success. The remainder of this report focuses on how the reform process can be supported by the new Development Bank of Nigeria. Structure of the remaining chapters of this study What role should DFIs play in addressing financing gaps? Based on international experience, what factors determine the success or failure of DFIs? Chapter 2 explains the common approaches that characterize DFIs, and identifies the Public/Private approach as most suited to the Development Bank of Nigeria. Monitoring and evaluation of DFI performance. Chapter 3 reviews efforts to monitor and evaluate (M&E) the performance of DFIs. In referencing the outcomes of existing monitoring and evaluation efforts, guidance is provided as regards establishing an M&E system for the DBN. Governance of the new DFI. Finally, chapter 4 provides guidance on institutional aspects of the new DFI. Factors such as ownership and mandate, selection of the board and management, risk management, regulation and supervision, and transparency and disclosure will prove crucial in making sure that the new institution is successful and in safeguarding implementation against the pitfalls experienced by previous generations of DFIs. 26 Chapter 2—Choosing an Appropriate Approach to the DBN I Defining the role of DFIs A. A limited role in filling financing gaps International experience reveals that DFIs are generally relatively small institutions when measured in relation to the size of the economies where they operate or the assets of local banking systems. Very few DFIs play a role in actually filling financing gaps. Despite growth in the size of the DFI sector in recent years, DFI assets only account for a very small proportion of banking system assets in the vast majority of countries. According to a Global Survey of Development Banks undertaken by the World Bank (2010), the combined total loan profiles of the surveyed development banks increased from $950 billion in 2006 to $1.6 trillion in 2009, and their combined total assets increased from $1.2 trillion in 2006 to $2.1 trillion in 2009, indicating a marked expansion in the nominal size of the DFI sector. The combined total assets to GDP ratio of DFIs in the survey increased from below 10 percent in 2006 to 14 percent in 2009. If five major DFIs37 that are large in size both in absolute and relative terms are excluded, the combined total assets to GDP ratio of DFIs increased from 6 percent to 8 percent between 2006 and 2009 (see Figure 2.1). In addition, in most countries the market share (DFI assets as a percentage of banking system assets) of DFIs was less than 3 percent. For example, in 2009 the Development Bank of South Africa (DBSA) had a market share of only 1.52 percent of total South African banking system assets. Figure 2.1: The relative size and market share of DFIs Rela ve size of na onal DFIs 2006-2009 Combined total assets as % of GDP Combined total assets as % of GDP (excl. major DBs) 16% 14% 12% 10% 8% 6% 4% 2% 0% 2006 2007 2008 2009 Source: The World Bank. WDI and Global Survey of Development Banks (2010). Note: “Major DFIs” refers to BNDES, CDB, KfW, KBN, and Ziraat, which are large in both absolute and relative terms. While most DFIs are relatively small in terms of assets, in some instances DFIs do play a significant role in the domestic economy. BNDES, for example, has become a major player in 37 The five DFIs excluded are BNDES, China Development Bank (CDB), KfW, Kommunalbanken Norway (KBN), and Agriculture Bank of Turkey (Ziraat). 27 the Brazilian economy and banking system. Its total assets relative to GDP grew dramatically from below 10 percent in 2006 to nearly 44 percent in 2013 (see Figure 2.2).38 The assets of BNDES are equivalent to about 80 percent of the assets of deposit-taking banks in Brazil, and the ratio steadily increased from 2009 to 2011. BNDES provided between 13 percent and 18 percent of domestic credit in Brazil during the period 2009 to 2013 (see Figure 2.3). Figure 2.2: Ratio of assets to GDP for selected DFIs (2006–2013) Source: Annual reports for the total assets of DFIs in local currency. WDI for GDP in constant local currency. Figure 2.3: BNDES—a major player in terms of asset size BNDES rela ve size BNDES total assets/total assets of deposit banks in the country Total disbursement as % of domes c credit to private sector 100.0% 80.0% 60.0% 40.0% 20.0% 0.0% 2009 2010 2011 2012 2013 Data source: World Bank FinStats,39 World Bank WDI, annual reports. B. A catalytic role in supporting private sector solutions What then is the desired or optimal size of DFIs? It emerges from this discussion that the vast majority of DFIs are small in terms of assets relative to the size of the financial systems where they operate. Thus, rather than measure their impact in terms of the funding they can muster, the 38 CDB demonstrates a similar pattern, although less dramatic. 39 As yet FinStats does not provide data on total assets of deposit banks for 2012 and 2013. 28 focus needs to be on their catalytic role in supporting pilots or providing funding/risk-sharing arrangements that will encourage the private sector to participate in the financing of activities that would otherwise remain unfunded or underfunded. In this catalytic role, DFIs can contribute to increasing the scope of activities that the private sector is prepared to fund by familiarizing private sector market participants with skills relating, for example, to assessing client cash flow profiles, adopting alternative funding methodologies, designing risk assessment tools, etc., that they would otherwise not be conversant with. As outlined in chapter 1 and further discussed in Section II below, DFIs have a role to play in situations of market failure. For example, the private sector’s risk appetite may be constrained due to gaps in the financial infrastructure. As financial infrastructure reforms take time to implement, DFIs can have a role to play in facilitating credit provision on an interim basis. DFIs can encourage private financial intermediaries to expand the scope of the business and risk profiles that they are willing to fund and thereby contribute toward expanding the frontier of financing possibilities. By adopting this philosophy that is founded in the idea that the public sector has a role to play in supporting the development of private sector–based solutions, DFIs are a vehicle for leveraging private financing and crowding in the private sector (see further discussion of the Public/Private approach in Section II below). The concern that arises when DFIs become large in terms of their assets relative to the private banking sector is that they will inevitably crowd out the private sector and thereby be pursuing a mission that runs counter to their mandate. C. A countercyclical role in “leaning against the wind” Rather than dampening cyclical economic fluctuations, financial systems tend to amplify cyclical fluctuations, pro-cyclically. “Financial developments have reinforced the momentum of underlying economic cycles, and in some cases have led to extreme swings in economic activity and a complete breakdown in the normal linkages between savers and investors” (Borio et al. 2001). One reason given for promoting DFIs is their countercyclical role in times of crisis: they are in a position to time their interventions so as to offset the pro-cyclical tendency inherent in financial systems. Moreover, DFIs can be used as convenient vehicles for channeling funding into markets, especially when the government’s hands are tied due to long budgetary approval processes. The potential role of DFIs as tools of countercyclical policy may help explain increasing interest in the expansion of DFIs since the 2009 global financial crisis. DFIs increased their credit exposures substantially during the recent financial crisis, in some cases dramatically. In nominal terms, DFIs increased their loan portfolios by 36 percent on average in just three years, well above the 10 percent increase in private bank credit (see Figure 2.4 and Table 2.1 below). The annual growth of credit accelerated for all the major DFIs, including BDC, BNDES, CDB, and NAFIN during 2007 to 2009. Individual country data confirm that the expansion of DFI portfolios was particularly concentrated around the time of the global financial crisis (see Figure 2.5 below). Nonetheless, given their relatively small size, the countercyclical role that DFIs are able to play will in most instances be limited. Indeed, it would be ill-advised for publically funded institutions to expand their provision of credit so much so as to replace the fall in credit experienced 29 in an economic downturn, as credit quality generally deteriorates at such times due to cyclical factors. Instead, DFIs can play a useful role in “leaning against the wind” so that credit does not stop flowing to the most credit-worthy borrowers in the economy due to increased (exaggerated) risk aversion during the downturn. A recent study of Latin American countries by Machín (2012) confirms the countercyclical role of public banks and the pro-cyclical role of private banks. Despite the countercyclical role of the public banks, total credit still dropped significantly from previous years during the 2009 financial crisis, reflecting the relatively small size of public banks compared to private banks (see Figure 2.6). Figure 2.4: Loan portfolio of development Table 2.1: Growth rate of credit provided banks during the global financial crisis by development banks (2007–2009) Growth Rate % of DBs Negative growth 16% 0 to 20% 27% 20% to 50% 33% 50% to 100% 15% More than 100% 9% Total 100% Source: Global Survey of Development Banks (2011). 30 Figure 2.5: Annual growth (%) of credit portfolio of selected development banks CDB - China 30 25 20 15 10 5 0 2005 2006 2007 2008 2009 2010 2011 2012 Source: Ferraz (2014) presentation. Original data source: annual reports. Figure 2.6: Average change of credit over same month in the previous year in Latin America40 Credit from Credit from Total credit private banks public banks Source: Machín (2012). Original source: CEPAL, División de Financiación para el Desarrollo sobre datos de Banco Mundial (2012), translated by the authors. Due to the restricted nature of their mandates coupled with fiscal constraints, some DFIs have been known to act pro-cyclically during economic fluctuations. The activities of Khula, 40 Note: Latin American countries include Argentina, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Paraguay, Peru, and Uruguay. 31 a South African development bank, were constrained by a requirement that DFIs be financially self-sustaining on an annual basis. As a result, the trend in the level of support provided turned out to be pro-cyclical (see Box 2.1). Box 2.1: The South African DFI, Khula—An example of pro-cyclicality During 2010, the World Bank undertook a survey of the supply and demand of SME finance in South Africa. A supply-side survey was conducted with eight financial institutions, including the big four banks. The demand-side survey covered 234 SMEs, a panel originally interviewed as part of the 2008 Enterprise Survey. Access to SME finance worsened during the economic downturn. Lending to SMEs in South Africa declined over the downturn; and the banks’ asset quality deteriorated (see Figures 2.7 and 2.8 below). In the perception of the SMEs, financing obstacles became more severe during the economic downturn (see Figure 2.9). This was consistent with the supply-side survey, which showed that loan applications fell (by 23 percent) and loan approval rates declined (from 61 percent to 45 percent), and that the “pricing for risk” (the difference between the best interest rates charged to large enterprises and to small enterprises) increased from 2.5 percent to 3.8 percent. While NPLs for small enterprises remained flat at 4 percent, NPLs for medium-sized enterprises tripled to 5 percent. Khula experienced a sharp drop in new credit indemnities extended during the financial crisis. This happened although Khula had a specific mandate for sponsoring the provision of credit to the SME sector by providing risk-sharing using partial credit guarantees. There was a sharp drop in both volume and value of credit guarantees extended by Khula to SMEs during and after the financial crisis (see Figure 2.10), suggesting that Khula played a pro-cyclical rather than countercyclical role. Figure 2.7: Banks’ exposure to SME Figure 2.8: All credit impairment 260 40 R bn R bn 240 35 220 30 200 25 180 20 160 15 140 10 120 5 100 - Mar-09 Mar-10 Jun-08 Sep-08 Jun-09 Sep-09 Dec-08 Dec-09 2001 2002 2003 2004 2005 2006 2007 2008 2009 SME Retail SME Corporate Source: SARB returns (BA120, DI200, BA200). 32 Figure 2.9: Demand side—Firms’ perception of finance as obstacle for business Source: World Bank Survey South Africa (2010). Figure 2.10: New credit indemnities provided by Khula 2006–2010 Source: Fuchs et al. (2011). Financing Small and Medium Enterprises in the Republic of South Africa. The World Bank, Africa Finance and Private Sector Development Department. Where DFIs do play a countercyclical role, they may have difficulty managing this role. During economic downturns, DFIs would be required to expand their lending quite rapidly, if they were to play a more important countercyclical role. While it takes time for the impact of credit expansion to feed through into asset quality, raising credit provision is often associated with a loose credit policy, weaker selection standards, and as a result, more rapid accumulation of nonperforming loans. The development bank in South Africa, DBSA, faced just such a trade- off in playing the role of countercyclical lender and maintaining financial performance: “Although the DBSA has sought to play a counter-cyclical role, it has not been immune to the impact of the economic climate, particularly given its dependence on the financial market as a source of financing.” (DBSA annual report (2013)). DBSA’s asset quality deteriorated significantly post crisis with a higher NPL ratio of 7.3 percent in 2013 (compared to 4.9 percent 33 in the previous year), and it reported a net loss of R 826 million in 2013, largely due to loan impairment and loss write-offs of R 1.6 billion (see Figure 2.11). Figure 2.11: Challenges encountered by DBSA as a countercyclical lender DBSA Deteriora ng performance post crisis 10.0% 8.0% 6.0% 4.0% ROE 2.0% NPL 0.0% -2.0% 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 -4.0% -6.0% Source: DBSA annual reports. If DFIs are to be used more proactively for countercyclical purposes, exit can also pose severe challenges. As opposed to the longer-term developmental mandate usually associated with DFIs, countercyclical activities are short-term in nature. Once the macroeconomic situation stabilizes, the lending activities—and as a consequence the capital base—of DFIs should be reduced appropriately. Both politically and administratively it is often much easier to increase the capital base of DFIs in times of emergency, than to reduce the capital base when the economic situation improves. DFI management will tend to have strong incentives to maintain a larger balance sheet, and government’s incentives to reduce the capital base may not be that strong, especially if the DFI is performing well commercially. Indeed, DFIs are likely to deploy the newly available capital to expand their business lines and to take on risks that stretch their management capacity.41 For example, when DBSA management reported loan impairments and write-offs of R 1.6 billion and a net loss of R 826 million in 2013, they emphasized that they were “of the view that…the recent deterioration is concentrated mainly in the non-public sector investment book in South Africa, which is more susceptible to changes in economic conditions” (DBSA annual report (2013)). In deviating from the bank’s traditional focus on financing public infrastructure projects, DBSA had been challenged due to competition from the private sector and capacity constraints in risk management and investment evaluation. DBSA’s losses led to the reconsideration of the scope of the bank’s activities. The so-called “sleeping beauty syndrome”42 relates to the countercyclical role of DFIs. The idea is that DFIs can be “awakened” when cyclical downturns occur, and assume a “sleeping” role when the economy recovers. On the one hand, “Once a financial crisis hits, it is too late for governments to create institutional capacity to provide fallback credit support. The institutions therefore need to already be in existence, with a clear operating mandate, experienced professional staff, and the financial capacity to respond to the financial needs and ramp-up their operations 41 Gutierrez (2011). 42 The idea of the sleeping beauty syndrome is borrowed from the discussions of export credit agencies after the 1997 Southeast Asia crisis (Stephens 1999). 34 when the private market fails.”43 The “sleeping beauty syndrome” thus provides an argument for maintaining the readiness of DFIs throughout economic cycles, instead of creating them only in times of crisis.44 On the other hand, if DFIs “are to have and maintain the technical expertise and experience that is essential to operating in current market conditions, and if they are to be regarded as serious and value-adding players by others involved in structuring and financing projects, there is a challenge for them to disengage or withdraw and later be brought back to an active and viable operation.”45 To some extent the revival of development banks in recent years or “the awakening of the sleeping beauty” has come about as a natural response to the global financial crisis (see Table 2.2). Table 2.2: Examples of recent DFI initiatives in advanced economies Time Country Event 2008 Japan Japan Finance Corporation (JFC 2008) was established as the outcome of a merger of existing DFIs, to fund other banks (using fiscal resources) during crises and natural disasters with a mandate to foster national priorities (sustainability and innovation). 2012 United Green Investment Bank, a new DFI, has a mandate to foster environmental and energy efficiency investments. Kingdom 2012–2013 United British Business Bank, a new DFI, was established to support SME finance. Kingdom The Korea Finance Corporation merged with Korea Development Bank 2015 to improve the efficiency of development finance. Subsidiaries with Korea activities not strictly developmental in nature were privatized. Source: Ferraz (2014), updated by the authors. In conclusion, DFIs need to be wary of the potential risks to their asset quality associated with deviation from their developmental mandate. In controlling the risks associated with becoming conduits for countercyclical actions, DFIs need to have in place robust governance, strong technical capacity, and risk management systems. DFIs also need to carefully consider and be confident they can effectively implement an exit strategy. Governance, risk management, and professional capacity are key factors in ensuring the financial sustainability of DFIs as considered further in chapter 4. 43 Conference Board of Canada, January 2010, as cited in Lessons from the Recession and Financial Crisis; Lesson 2—Public Sector Financial Institutions Prove Their Worth. Available at: http://www.shcp.gob.mx/ApartadosHaciendaParaTodos/banca_desarrollo/pdf/presentacion_theodore_homa.pdf. 44 Gutierrez et al. (2011). 45 D. Smallridge and F. de Olloqui (2011), IDB. 35 II Core features of alternative approaches to DFIs A. A conceptual framework: Evolving views46 The conceptual framework as regards the role of DFIs—how they are designed and how they operate—has evolved over time. Fundamentally there are three approaches to addressing the role of DFIs: (1) The Public Sector approach: According to this approach, problems of access to finance result from widespread market failures that can only be overcome by public sector intervention. Thus, the government has to assume a more hands-on involvement in mobilizing and allocating financial resources. This view was popular in post-WWII and postcolonial times, and until the late 1980s. (2) The Laissez-Faire approach: According to this approach, even though there may be market failures, the costs associated with direct government intervention are likely to exceed those of the market failures themselves. Thus, instead of intervening directly to fill funding gaps, government efforts are better deployed in improving the functioning of the market by strengthening the legal, institutional, and enforcement environment. While this view gained popularity in the early 1990s, it has been questioned widely in recent years. The evolution of the Public Sector and the Laissez-Faire approaches are summarized in Table 2.3. (3) The Public/Private approach: This approach emerged as the middle ground and has become more accepted following the global financial crisis and subsequent economic downturn: more active government involvement gained ground as a response to the crisis. While recognizing the role of market fundamentals in resource allocation, this approach places emphasis on the catalytic role of government interventions in addressing market failures in well-tailored, targeted, and nontraditional ways. While the authorities should promote the development of markets, they should not seek to replace them. Rather, the authorities should address financing gaps by correcting specific market failures. 46 This section draws on the discussions in de la Torre et al. (2007): Innovative Experiences in Access to Finance: Market Friendly Roles for the Visible Hand? 36 Table 2.3: Evolution of the Public Sector and Laissez-Faire approaches Public Sector approach (before about 1989)47 Laissez-Faire approach (about 1989–2005) Theoretical o Keynesian economics advocating the role of government interventions o The neoclassical school emphasizing the importance of efficient markets and foundation (emerged in 1930s and popularized during World War II, and the postwar rational expectations, and placing more importance on the role of markets in economic expansion) resource allocation o Development of economic theories on market failures: imperfect o The Washington Consensus placed emphasis on stabilization, privatization, and information, incomplete markets, and insufficient competition (in the liberalization, as well as “best practices” in institution building. 1970s) Political o Huge post-WWII financing needs for reconstruction and to support o Following the fall of the Berlin Wall and the collapse of the Soviet Union, economy expanding demand both in the West and in postcolonial times former socialist countries made bold leaps toward establishing market economies backdrop o Import substitution policies48 of the 1950s and 1960s through privatization and liberalization. o Newly established independent developing countries faced challenges in o Failures of the interventionist approach and inward-looking import substitution weak domestic financial systems, largely a legacy from the colonial policies: times.49 o Disappointing performance of developing countries due to failed structural reforms50 in the 1970s and 1980s led to economic stagnation, severe swings in cyclical growth, and inflation.51 o Distorted price and resource allocation, rent seeking and corruption, large public deficits, and excessive foreign debts Impact on o Public banks grew rapidly: by the 1970s, the state owned on average 40 o A wave of market-oriented reforms (such as large-scale privatization, DFI practices percent of the assets of the largest banks in developed countries and about deregulation, and trade liberalization). 65 percent in developing countries. o A trend toward downscaling DFIs: a large number of DFIs were closed or o Consistent with the market failure rationale, public banks tended to focus merged with commercial banks. Some DFIs changed strategy and adopted on areas where private markets were missing, such as long-term finance, commercial operational codes and emphasized commercial viability. lending to SMEs, housing finance, and agricultural credit. o Focus on creating strong legal and institutional framework for financial markets o Public banks became key vehicles in government efforts to support social designed to strengthen property rights, collateral registration, and foreclosure and developmental goals through selective allocation of (often subsidized) processes, improving credit bureaus to facilitate information exchange and credit. reduce transaction and screening costs Impacts on o Nigeria set up several development banks after its independence in 1960 o Although Nigeria did privatize the Infrastructure Bank Plc (previously known as Nigerian DFIs so as to finance the demands of economic development: the Nigerian Urban Development Bank of Nigeria Plc) in 2006, most of Nigeria ’s DFIs Industrial Development Bank Limited (NIDB), the Nigerian Bank for continued to be run with an interventionist mind-set. Commerce and Industry (NBCI), Nigeria Agriculture Bank, and the o The Nigeria authorities initiated efforts to undertake Washington Consensus– National Economic Reconstruction Fund (NERFUND) style reforms to improve the financial market infrastructure.52 47 There is hardly a clear cut between the periods when different views were prevailing. Prescriptions relating to Latin American countries by Williamson in 1989 (see Williamson 2004) gained popularity among policymakers and international institutions, and marked the beginning of the Washington Consensus era. From around 2005 the World Bank undertook a systematic retrospective review of the results of reforms guided by the Washington Consensus in 1990s (Zagha and Nankani 2005). 48 Import substitution policies sought to reduce a country’s import dependency through expanded local production, thereby generating demand for local capacity and financing. 49 In postcolonial Africa, most commercial banks were foreign-owned and provided financing mainly to foreign-owned trading companies, mines, and plantations. Indigenous companies found it hard to obtain external financing from the formal financial system. 50 Interventions include trade restrictions, state ownership of firms, financial repression, price controls and foreign exchange rationing. 51 For example, in the 1980s Latin American countries were dragged into an economic slump by a severe debt crisis: the “lost decade.” 52 As outlined in chapter 1, experience suggests that such reforms typically take a concerted effort over a longer period of time. As in many countries in Sub-Saharan Africa, Nigeria’s financial system is still constrained by its weak institutional infrastructure, a problem that is unlikely to be fully addressed in the near term. 37 B. Defining features of the Public/Private approach The emerging Public/Private approach is more a trend toward rebalancing the Public Sector and Laissez-faire approaches than an attempt to define a new set of policy prescriptions. The Public/Private approach is an emerging view, which balances the Public Sector approach and the Washington Consensus–styled Laissez-Faire approach. Although this approach builds on an acceptance of the fundamental role of markets in resource allocation and the importance of policy initiatives that catalyze the private sector, it also recognizes there still is a role for the government in addressing financing gaps—that is, going beyond the government’s role in institution building. To better understand this new approach, the following compares this evolving approach with the two more-established approaches: (1) The Public/Private approach versus the Laissez-Faire approach: While the Public/Private approach recognizes the fundamental role of markets in resource allocation and a good institution environment, it differs from the Laissez-Faire view in the following aspects:  According to the Public/Private approach, institution building may not be enough. This case is made by Rodrik (2002) who argues that “[t]he record suggests that an adequate growth program needs to be anchored in two strategies: an investment strategy designed to kick-start growth, and an institution building strategy.” Similarly, Zagha (2004) argues in reviewing the experience of the 1990s, it’s seen that “selective government interventions can contribute to growth when they address market failures, when and where they are carried out effectively, and are subject to institutional checks.”53  The Public/Private approach builds on tailor-made policy advice and interventions rather than a generic, rigid blueprint. This distinguishes the Public/Private approach from the Laissez-Faire approach, which advocates applying “best practices” to developing the legal and institutional environment. Under the Public/Private approach, the specific country context does matter.  The Public/Private approach emphasizes the importance of short-term interventions, while the Laissez-Faire approach relies on longer-term solutions and focuses on building the fundamental legal/institutional framework. (2) The Public/Private approach versus the Public Sector approach: While the Public/Private approach recognizes that there is a role to play for government, this role is not to replace the market but to focus on providing support to the private sector, so that it becomes conversant with and adopts instruments tailored to filling the identified market gaps. Rather than replace or substitute private sector funding provision, public sector intervention is seen as catalytic, thereby encouraging private sector participation in addressing market failures. Thus, the Public/Private approach leverages tailored, 53 See de la Torre et al. (2007) for further discussion. 38 targeted, and restricted interventions and emphasizes the role of innovative instrument design. These different approaches to the role of DFIs are associated with different design features as regards: (1) ownership and funding structure, (2) mandate and strategy, and (3) instruments and clients. These design features are briefly summarized below:  Ownership and funding structure: (1) The Public Sector approach is usually associated with state ownership or a situation where the state is the controlling shareholder. The board executives of public sector DFIs are political appointees. Funding is usually provided through government budget transfers and/or subsidized borrowing. This may result in greater political influence by the government in the daily operations of the DFIs, known as the “superdirector syndrome.”54 (2) The Laissez-Faire approach involves no public ownership and no public funding. This approach typically encourages downscaling and privatization of DFIs, and an operational mandate that is primarily based on commercial principles. Under the Laissez-Faire approach, DFIs need to rely on private capital markets and internal funds to finance their activities, thereby raising the cost of services provided so that they are equivalent to commercial terms. (3) The Public/Private approach reflects the recent trend of moving away from 100 percent state ownership to increasing use of a mixed public-private capital structure. Under this approach, board members are selected according to transparent, professional criteria. The funding structure is also more diversified with internal funds and financing from capital markets being combined with funding provided by one-off government capitalization and/or a diversified range of donors or impact investors. While this may allow DFIs to provide services at a small discount compared to commercial terms, the focus of this approach is primarily on catalytic instrument design.  Mandate and strategy: (1) Under the Public Sector approach, the policy mandate of DFIs tends to focus on filling financing gaps and replacing private financial institutions. The assumption is that the public sector makes better choices than market players. Private market signals are suspended, and projects are selected based not on their financial returns but on decision-making criteria administered by the public sector.55 When DFIs 54 As defined by T. Home in his presentation “Managing Public Banks: Case for a Holding Structure.” More details can be found in Part B of this section. 55 Here the public sector is defined to include agents of the public sector. In the Nigerian case, this includes schemes administered by the CBN. 39 play a countercyclical role, as discussed earlier in this chapter, timely exit is a concern. Similarly, it may be difficult for DFIs to “reawaken” their capability when the economic cycle weakens, and after having previously retreated from certain markets, known as the “sleeping beauty syndrome.”56 (2) Taking the Laissez-Faire approach, commercially run DFIs will focus more on financial and operational performance and not focus on market failures or market gaps. While the activities of institutions run according to these principles are likely largely to remain within existing market structures, this may be at the cost of being able to effecting their developmental mandates. Were DFIs to achieve commercial success according to market principles, they would most likely come to compete with and crowd out private capital. (3) Adopting a Public/Private approach, DFIs will seek to leverage the private sector’s capacity in funding, risk assessment, etc., while balancing the discipline that this private sector involvement implies in fulfilling their development mandate sustainably. The role of DFIs under this approach is to focus on identifying and resolving specific market failures. DFIs address such market failures with instruments that encourage learning, adoption, and supplementary funding by private intermediaries. Thus, the primary aim of DFIs under the Public/Private approach is to ensure that their funding (“start capital”) is leveraged. The aim is to ensure that scalability is reached, while also maintaining financial sustainability. This can only be achieved if: (a) the instruments and funding terms provided by DFIs are market-conform, i.e., they are consistent with and encourage private sector financing on similar/parallel terms; and (b) the DFI devotes attention to remaining on the frontier of financial service provision—seeking to test market segments and/or instruments that the private sector has as yet been unwilling to adopt or is not conversant with. With the aim of encouraging private sector participation, financial viability, and scalability under the Public/Private approach, it is advisable to restrict the mandate of DFIs to performing the role of second-tier, wholesale intermediation. Advantages associated with establishing DFIs as wholesale rather than retail intermediaries are that: (a) second-tier DFIs will be perceived as supporting retail-level institutions rather than competing with or looking to replace them; (b) second-tier DFIs will need to rely on private retail intermediaries as regards expertise in managing client relationships and the assessment of credit risks. 56 The idea of the sleeping beauty syndrome is borrowed from the discussions of export credit agencies after the 1997 Southeast Asia crisis (Stephens 1999). More discussions can be found in Part B of this section. 40  Instruments and clients: (1) DFIs with a Public Sector approach tend to choose grants and subsidized lending, including directed credit programs,57 thereby providing balance sheet funding on terms not otherwise available on the market to clients exposed to financing gaps. (2) DFIs with a Laissez-Faire approach operate on commercial terms and use client selection criteria similar to those of commercial banks. (3) Under the Public/Private approach, DFIs target specific market failures using tailored instruments. Simultaneously they seek to encourage private sector participation so that they can focus on new markets and clients as soon as the private sector becomes conversant with and is willing to service those targeted markets or market segments. Table 2.4: Comparing core features of the Public Sector approach with the Public/Private approach Public Sector approach T approach Public/Private Ownership o State ownership o Mixed government/nongovernmental ownership and Funding o Board and management o Board and management selected according to appointed by government transparent, professional criteria o Government funding that o Initial capital funded by government/donors, includes fiscal supplemented with wholesale market funding transfers/subsidies Mandate and o Social policy mandate o Strategy tailored to addressing specific market Strategy supplemented with failures/financing gaps countercyclical interventions o Emphasis on market conformity so as to encourage o Projects selected according to learning, adoption, and supplementary funding public sector’s assessment of from the private sector needs o DFIs assessed on whether they achieve: o Projects characterized by high  financial sustainability risk/low return activities and  meaningful scale poor viability  crowding-in of private sector funding 57 Directed credit programs provide financial resources to a specific sector of the population for a specific purpose on nonmarket terms. 41 Instruments o Direct lending provided on the o Second-tier lending relying on private sector to and Clients balance sheets of DFIs establish/maintain client relationships and assess o Targeting underserved client credit risks groups o Focus on catalytic instrument design (providing o Subsidized credits and financing otherwise unavailable on the market) and guarantees targeting financing gaps (excluded client groups) o Targeted client groups adjusted as private sector becomes conversant and takes them on o Market-conform pricing—leveraging public/donor- funded initial capital (with zero required return) to provide limited discount to market rates The Public/Private approach encourages private sector intermediaries to engage in the provision of finance in areas where the private sector lacks capacity and/or which the private sector finds too risky by providing longer-term finance that is otherwise unavailable in the market, or by sharing the risk assumed by private financial intermediaries. The challenge faced by the Nigeria authorities is to move from a purely public sector approach to DFIs to an approach that is focused on ensuring that DFIs leverage the involvement of the private sector. With this transition in mind, Table 2.4 summarizes the core features of the Public Sector and Public/Private approaches as outlined above. III International experience relating to alternative DFI approaches A. Issues confronted by the Public Sector approach The Public Sector approach has generally not been successful: indeed, most empirical studies suggest that public banks tend to do more harm than good.58 Several studies find that greater government participation in bank ownership is associated with lower levels of financial development, less credit to the private sector, wider intermediation spreads, greater credit concentration, slower economic growth, and recurrent drain on fiscal resources.59 Ownership and funding State ownership and funding assistance—where the government plays an influential role in the daily operations of DFIs—give rise to the so-called “superdirector syndrome.”60 Political intervention in DFIs is common, partly due to their social policy mandate, but also the government’s incentives for establishing DFIs may reflect the desire to avoid the budgetary approval process for specific expenditures and/or the possibility of 58 de la Torre et al. (2007) provides a good summary of literature and empirical results. 59 See Barth, Caprio, and Levine (2001); Caprio and Honohan (2001); IADB (2004); and La Porta, Lopez- de-Silanes, and Shleifer (2002). 60 As defined by T. Home in his presentation “Managing Public Banks: Case for a Holding Structure.” 42 using DFI patronage as a source of rent seeking. If, instead of using the budget to finance projects, the government uses budget funding to inject seed capital into a politically controlled DFI, this provides the government with the opportunity to leverage the DFI’s balance sheet. As a result, the government gains access to funding (and political patronage) equal to a multiple of the seed capital injected. As many governments have come to realize, there are potentially severe economic costs associated with such an approach. If the funded projects do not generate enough cash flows—and quite typically after multiple capital injections, the government decides to close the DFI—it will have to write off an amount equivalent to several times the initial seed capital. The move toward privatization of state financial institutions in developing countries in the past few decades has largely come about in recognition of the consequences of the poor performance of these institutions, characterized by the unsustainable nonperformance of loans and a continuous process of recapitalization to overcome mounting losses.61 While a robust governance framework provides some assurance against such losses, they are unlikely to be effective while DFIs remain in public hands. For example, the appointment of an independent, accountable, and professionally qualified board of directors can support DFI management to avoid investment in projects with poor risk- adjusted returns. Similarly hiring a professional and suitably qualified DFI management team is important in managing investment risks. Nonetheless, even if all members of the board of directors are independent and the management is professional, experience suggests they will not be fully protected from political interference. Additional measures are needed to ensure the DFI directors’ independence from political interference.62 This applies particularly in more fragile institutional environments that characterize developing and emerging markets. Several countries took measures designed to address these governance concerns with whistle-blower and other safeguard policies, but doubts remain as to their effectiveness in more fragile institutional environments. For example, the safeguards policy in the case of the Business Development Bank of Canada (BDC), known as the Referral Policy states that whenever a member of parliament, senator, or fellow board member exerts undue pressure on a BDC employee, the board of directors is notified. The person who makes the referral is notified in writing that client confidentiality supersedes all third-party involvement, and that BDC retains sole authority for its decisions. In the case of the National Bank of Chile (BancoEstado), a law bans BancoEstado from lending money to public institutions. In addition, the members of the supervisory board, who are typically politically appointed, are not allowed to participate in deliberations relating to credit decisions. While these initiatives are positive, questions remain as to their sufficiency in more fragile institutional environments where such safeguards can all too 61 Gutierrez et al. (2011). 62 See Gutierrez et al. (2011). More discussions on corporate governance is to be found in chapter 4. 43 easily be overturned by political influence. As described further in Section C of this chapter and in chapter 4, a pivotal feature of the Public/Private approach is the requirement that government ownership be supplemented with ownership participation by outside nongovernmental parties. Mandate and strategy The activities funded by DFIs with a public sector mandate tend to be associated with high risks and low returns, thereby threatening the DFIs’ commercial viability. Public sector development banks are often associated with large losses and the need for recurrent recapitalizations. There are many examples. In the case of Turkey, the cost of recapitalizing the two largest public banks (Ziraat Bank and Halk Bank) in 2001 amounted to 15.5 percent of GDP.63 In 1999 the Mexican government had to recapitalize Banrural, a development bank providing financing to the rural sector, with about US $1.1 billion, even after having significantly downscaled its operations in previous years.64 A more recent case is the substantial loss of DBSA in 2012–2013 due to large loan impairments and write-offs (discussed in Section I C of this chapter). These international experiences (as well as those of the Nigerian DFIs referenced in chapter 1) confirm that publically funded DFIs that have a mandate to lend directly to the private sector are not effective because the capacity of the public sector to manage financial risks is highly constrained. The skills required to assess and manage financial risks and the discipline associated with well-functioning financial infrastructure (i.e., foreclosure on collateral and ultimately insolvency procedures) are simply absent. Instruments and clients The experience with directed credit programs typically used by Public Sector DFIs has more often than not been unsuccessful:65  Directed credit programs often fail to reach their intended beneficiaries. Within priority sectors, the tendency is for larger and more influential borrowers to find favor. Efforts to direct credit to particular sectors or borrowers have often failed: lenders misclassify loans to provide credit to other sectors, and borrowers divert credit to other uses. As a result, funding provided through directed credit programs often ends up in the hands of politically well-connected firms and is not used to correct market failures.  The fiscal cost of subsidies to directed credit programs are often substantial. For example, at the time when such programs were most in vogue in 1987 in Brazil, their cost was estimated to amount to between 7 percent to 8 percent of GDP. In Korea, the subsidy provided by directed credit was approximately 1 percent of GDP during the 1980s (Booth et al. 2001). Given scarce fiscal revenues and the resource 63 Fouad et al. (2005). 64 Brizzi (2001). 65 World Bank, 1989, 2005a. 44 mismanagement associated with the poor resource allocation decisions taken by publically managed DFIs, the resource wastage implied by directed credit programs can be a severe drag on economic development.  Directed credit programs tend to set unfortunate precedents. Once established, directed credit programs create a strong constituency of beneficiaries, making it difficult for governments to reduce their support to such programs, regardless of how inefficient or costly they may have become. Directed credit programs, even if channeled through participating private banks, may leave little responsibility for managing credit risk with private banks, resulting in weak credit assessment and monitoring.66 Also, the availability of extensive refinance schemes at low interest rates compromises the banks’ incentives to mobilize resources of their own, thereby encouraging a downward spiral whereby provision of credit by the public sector crowds out private sector funding, resulting in less intermediation by the private sector. B. Challenges facing the Laissez-Faire approach According to the Laissez-Faire approach, the authorities are encouraged to withdraw from any form of direct intervention and to rely on Washington Consensus–style reforms of the legal and institutional infrastructure to create a conducive market- enabling environment. Empirical evidence suggests that reforms to the enabling environment do have a positive impact on financial development. For instance, Djankov, McLiesh, and Shleifer (2006) find that improvements in creditor rights and the introduction of credit bureaus are associated with increases in credit to the private sector. Similarly, de la Torre, Gozzi, and Schmukler (2005) find that capital market–related reforms tend to be followed by significant increases in stock market capitalization, trading, and capital raising. Nonetheless, despite the progress made in strengthening the enabling environment, the overall outcomes have not lived up to the expectations of the Washington Consensus. Box 2.2 provides insight into some of the challenges faced by the Washington Consensus. Box 2.2: Why the Washington Consensus failed67 The disappointing outcomes of Washington Consensus–type reforms appear to be attributable to a combination of insufficient reform implementation and impatience. In some cases, key reforms were not even initiated, while other reforms were implemented in an incomplete or inconsistent fashion. In many cases, only laws were approved, but they were not effectively implemented or adequately enforced. Moreover, policymakers were too impatient, often expecting results to materialize sooner than could be expected. Building strong institutions 66 Examples of such directed credit programs are the schemes managed by the Central Bank of Nigeria, where ultimate responsibility for loan approval resides with the central bank rather than the participating private banks. 67 See de la Torre et al. (2007) and Birdsall et al. (2010) for more discussion. 45 takes time, and the agenda in removing institutional impediments may remain unfinished and legal constraints to access to finance can still be large. Another view is that the Washington Consensus reform agenda was not wrong but incomplete, leaving out essential areas for reform actions. The implication of this view is that one needs to take into account the broader social context of the reform agenda. External shocks, macroeconomic disruptions, and social unrest can easily derail efforts to strengthen the legal/institutional environment. To be successful legal/institutional reforms depend on a conducive environment: (1) instead of temporary factors that spur economic growth, such as natural resource discoveries or a new free trade agreement, the basis for economic growth needs to be sustained and broadly based; and (2) domestic capacity needs to be developed to manage external shocks, such as terms of trade declines or reversals in capital flows, to avoid temporary growth disruptions.68 According to a third view, the Washington Consensus is fundamentally flawed due to: (1) Unrealistic expectations as regards the shorter-term impact of institutional reforms: According to the Washington Consensus, the role of the government is to focus on developing the enabling environment for market-based financing, leaving capital mobilization to the private sector. However, given the shallowness of the financial system, it is unlikely that the private sector will be able to fulfill this role in the shorter term; this leaves a role for DFIs—to fill financing gaps in the short term. Over the medium to longer term, the exit of DFIs will depend on the outcomes of efforts to strengthen the legal and institutional infrastructure. The Washington Consensus places altogether too little emphasis on shorter-term dilemmas, and too much reliance on the longer-term impact of financial market liberalization. (2) The fallacy of “one size fits all”: The traditional criticism of the Washington Consensus is that one size most certainly does not fit all—the framework does not address the local political economy.69 Although global best practices provide experience and benchmarks, these practices will need to be customized to the country-specific context. In addition, the Washington Consensus is based on a rather naïve view of the local legal/institutional reform process: even though reforms are for the greater good of society, undertaking them is likely to meet strong local resistance from those who benefit from the status quo. Empirical evidence as to the outcomes of the Laissez-Faire approach also suggests that reliance on market forces is insufficient. While placing overall greater reliance on market forces as suggested by this approach may be sound, weaknesses in incentives, intransigence to change, and market failures can delay or thwart the achievement of development outcomes. Additionally, evidence suggests that instituting market-based reforms may not be required to achieve improved outcomes in the shorter term.  Economic growth: Cross-country comparisons have been unable to establish a strong causal link between particular institutional design features and economic growth. 68 Rodrik (2006). 69 Hausmann, Rodrik, and Velasco (2008), among others, advocate a case-by-case approach. 46 Although we know that stronger institutions make investors feel more secure and may promote growth, broad goals such as improving the investment climate can be achieved using divergent approaches, and the literature does not provide guidance as to which institutional design characteristics are best suited to achieve the desired improvement in the investment climate. In the mid-1990s, China was able to attract sizeable investment by the private sector, although the economy was still predominantly state- owned (Rodrik 2006). While intense efforts were made by Latin American countries to implement the Washington Consensus framework, outcomes in the form of increased economic growth were disappointing. GDP growth averaged only 3 percent during the 1990s, and per capita income in Latin America has continued to diverge from levels in developed countries, such as the United States (Birdsall et al. 2010). On the other hand, some East Asian countries that operated outside of the Washington Consensus framework significantly narrowed their per capita GDP gap with the United States, experiencing economic growth at twice the rate of Latin America (World Bank, 2004).  Access to finance: Despite intense reform efforts, access to finance did not seem to increase significantly in most developing countries in the period following the adoption of the Washington Consensus in the early 1990s. While many countries experienced strong growth in deposits, this did not translate into an increase of similar magnitude in credit to the private sector, as most of the additional loanable funds were absorbed by larger holdings of public sector debt (Hanson 2003). As discussed in chapter 1, substantial financing gaps, especially in developing countries, persist and continue to be the focus of attention of policymakers.  Financial markets: Similarly, the performance of domestic securities markets in many emerging economies has been disappointing (World Bank 2004a). Although some countries experienced growth of their domestic securities markets, this growth was in most cases significantly slower than the growth of securities markets in industrialized nations. Improving the financial performance of DFIs may strengthen their market- conformity and be consistent with a Laissez-Faire approach, but in focusing on financial performance, DFIs may deviate from their mandate of addressing financing gaps—referred to as the “Sisyphus syndrome” (de la Torre 2002). One reaction to the track record of poor financial and operational performance of DFIs is to call for radical reforms designed to strengthen the regulation and oversight as applied to DFIs and their governance, and to insist on adoption of commercial practices. However, in as much as DFIs come to mimic the activities of private sector banks, their focus will move away from achieving their developmental mandate, and the challenge then becomes to ensure that they still add value and achieve additionality. DFIs that adopt commercial practices consistent with the Laissez-Faire approach will come to compete with the private sector, and may well be crowding out private commercial banks both in their investment and fundraising activities: the “crowding- 47 out syndrome.” The focus on profitable activities may lead DFIs to direct competition with the private sector, so-called investment crowding-out. With explicit or implicit government support, DFIs tend to have easier access to investment opportunities than the private sector. The intention may well be that DFIs should exit when the market takes off, but such intentions are difficult to implement in practice. In addition, DFI funding activities may also contribute to crowding out private sector borrowing due to explicit or implicit sovereign guarantees on the bonds issued by DFIs and their advantage in benefiting from higher credit ratings than those of domestic private financial institutions. The case of the China Development Bank (CDB) showcases the dilemmas associated with a Laissez-Faire approach. Despite being a wholly state-owned bank, CDB is one of the most commercially successful state development banks, having adopted a market- oriented approach. However, CDB’s commercial success and its competitive advantage compared to commercial banks have become controversial. Recently, the CDB veered from pursing the path toward full commercialization and reemphasized its role in development finance (see Box 2.3). Box 2.3: The case of CDB: The trade-off between development finance and commercialization As early as 2003 CDB was among the best-ranked banks in China in terms of its low NPL ratio. CDB was first among the state banks to introduce commercial standards to its operations. Competing with “commercial” banks In many respects, CDB is little different from China’s big four state-owned commercial banks. CDB and the commercial banks may compete for the same projects, the only difference being that CDB can provide longer maturities and larger loans. CDB finances itself by issuing bonds that are sold to the commercial banks, and by lending to key infrastructure sectors such as power, road construction, railways, petrochemicals, and telecommunications. As early as 2003, CDB provided two-thirds of the funding made available to these sectors. (Source: Downs 2011, Sanderson and Forsythe 2012.) Switching back to development finance The stalling of the commercialization process that took place in China in the wake of the global financial crisis indicates that the State Council of China may have decided that CDB should shift focus so as to continue to function as a policy bank. CDB shelved plans to make an initial public offering and introduce foreign investors—reforms that would make CDB answerable to actors other than the central authorities. In 2009, a CDB official told the Financial Times that CDB would continue with “business as usual”—supporting government projects and policies, with the international expansion of Chinese firms ranking as one of CDB’s most important objectives. Chen Yuan, the then-chairman of CDB, emphasized CDB’s emergence as the main financier of China’s “going out” strategy, including provision of funding to cross-border energy deals, as a key reason for retaining CDB as a policy bank. (Source: Downs (2011), Inside China, Inc.: China Development Bank’s Cross-Border Energy Deals.) 48 C. The defining features of the Public/Private approach Ownership and funding The Public/Private approach supports the view that, particularly in the context of developing and emerging markets, shared ownership will considerably enhance the governance of DFIs. In principle, 100 percent state ownership is not ruled out, but where institutional settings are weak and potentially subject to political interference, the Public/Private approach recognizes the importance of inviting or ceding parts of the ownership of DFIs to nongovernmental beneficiaries or stakeholders. Given that the primary role of such third-party investors is to strengthen DFI governance, it is likely that this role, at least initially, is best fulfilled by so-called “impact investors.” The intention of impact investors is to generate a measurable, beneficial social impact alongside a financial return.70 The approach of impact investors and their experience in introducing best-practice governance, including robust performance measurement practices and merit-based reward systems that are linked to performance indicators, will be pivotal in ensuring successful implementation of the DFI mandates. Ceding ownership to nongovernmental investors is important, as doubts remain as to the effectiveness and sufficiency of the introduction of whistle-blower and other governance safeguards, particularly in countries with more fragile institutional environments (as noted in Section III A above). Hence, the imperative of addressing potential governance weaknesses through shared ownership arrangements. The ownership structure can vary on a case-by-case basis and can include foreign donor capital, which as suggested here is at least initially preferable, as well as domestic and foreign private companies and individuals. While the recent trend in ownership of DFIs does show some evidence of the trend of moving away from 100 percent state ownership toward increasing use of a mixed public-private capital structure (see Figure 2.12), such data need to be interpreted with care, as in some instances the non-state shareholders may be state-owned enterprises or state-run pension funds ultimately controlled by the state. Altogether, despite the widely acknowledged advantages spelled out above, providing “voice” to private parties in running DFIs that are overwhelming capitalized by the government is still an experimental rather than a widely implemented approach. 70 Impact investments of the kind envisaged here could be undertaken by the private sector –focused donor agencies, such as the International Finance Corporation (part of the World Bank Group), the Commonwealth Development Corporation (the UK government agency focused on investment in private sector development), or Kreditanstalt für Wiederaufbau (the German government agency with a mandate to take ownership stakes in private companies in developing countries). 49 Figure 2.12: State ownership of DFIs 5% 21% State owns 100% State owns 50% to 99% State owns less than 50% 74% Source: The World Bank, Global Survey of Development Banks (2010). To finance their operations, Public/Private DFIs rely on financial markets to leverage initial capital provided by the government and donors. While Public/Private DFIs endeavor to reduce their reliance on fiscal transfers, their primary source of funding is envisaged to be from wholesale financial markets through the issuance of debt instruments. Reliance on deposits from the general public is not a preferred option. Maturity mismatch makes DFIs (and banks) vulnerable to interest rate volatility, thus potentially becoming a severe problem particularly for DFIs, as most DFIs focus on providing medium- to long- term financing. Additionally, the interface with depositors involves investment in deposit mobilization, an activity that requires significant resources and is not core to achievement of the development mandate pursued by DFIs. While more than half of DFIs in the World Bank’s Global Survey of Development Banks (2010) still take deposits from the general public, more recently established DFIs are less likely to accept deposits from the public (see Figure 2.13). Even among DFIs that do accept such deposits from the public, the proportion of total DFI funding coming from this source has fallen. Although they are fully government-owned institutions, CDB and BNDES provide examples of commercially viable DFIs that rely predominantly on capital markets and internal funds to finance their operations. CDB provides an example of a DFI that relies on the bond market to finance its operations. In 2013, CDB issued RMB 1,240 billion in debt, while its internal funding (net profit) amounted to only RMB 80 billion.71 CDB is the second-largest bond issuer in China’s RMB debt market, second only to China’s Treasury and Municipals, and accounting for a market share of almost 20 percent (see Box 2.4 below). 71 CDB’s net profits were largely allocated to appropriation to surplus reserve, appropriation to general reserve, and dividends. 50 Figure 2.13: Public deposits—No longer a preferred funding source Source: The World Bank, Global Survey of Development Banks (2010). Box 2.4: CDB and China’s RMB debt market Source: CDB annual report 2013. Original data sources: www.chinabond.com.cn, www.shclearing.com, www.chinaclear.cn. 51 BNDES relies more on its internal funding to finance its operations. Returns earned on the operations of BNDES comprise the main source of resources for its disbursements and investment budgets. In 2013, returns on operations accounted for almost 78 percent of BNDES’s annual funding, followed by funds raised on the capital market that accounted for 11.5 percent of total funding. Figure 2.14: BNDES funding sources BNDES 8.30% 2.80% 11.50% FAT Return on opera ons Market Treasury 77.40% Source: BNDES annual report 2013. Note: FAT represents transfers of 40 percent of the proceeds from Workers’ Assistance Fund as required by the constitution. In the case of Nigeria, a diversified ownership and funding structure will be critical to the sustainability of the DBN. As discussed in chapter 1, given Nigeria’s fiscal pressures, the DBN is not likely to be able to rely on budgetary resources to provide sustainable funding, nor would it be advisable for it to do so. The involvement of impact investors as shareholders will strengthen the DBN’s governance as will other efforts in this regard, such as the appointment of a management board chosen according to criteria regarding professional qualifications (see further discussion in chapter 4). Such measures will also serve to strengthen confidence in the practices being pursued by the DBN and thereby ease its access to capital market financing. Indeed, access to domestic capital markets will be essential if the DBN is to be able to source sizeable, sustainable financing at reasonable cost. Mandate and strategy According to the Public/Private approach, DFIs should achieve both additionality in addressing specific financing gaps and financial viability. Financial viability is important as the basis for ensuring sustainability and in order to allow Public/Private DFIs to accumulate capital,72 providing the foundation for future growth and allowing DFIs to source funding on financial markets and thereby reach scale. Key indicators of financial 72 Under the Public/Private approach, DFIs are required to live up to prudential requirements of the central bank that include accumulating capital relating to the size of the risks they assume. See further discussion in chapter 4. 52 viability include: (1) maintaining a sufficient capital cushion against risk as measured by the level of capital adequacy ratio (CAR); (2) sustaining asset quality and preventing the accumulation of nonperforming loans as measured by the NPL ratio; and (3) achieving an adequate return on invested capital, as measured by the return to equity (ROE) and return to assets (ROA). Although these indicators are similar to those applied by prudential regulators to commercial banks, due to their social/developmental mandates, DFIs are more likely to focus on the sustainability of their activities than achieving high returns.73 The mandates of Public/Private DFIs can be designed to strengthen their financial viability and support their catalytic role by leaving responsibility for assessment of credit risk with private sector intermediaries and restricting the mandate of DFIs to wholesale, second-tier institutions. DFIs operating as Public/Private institutions will need to adhere to disclosure and transparency standards equivalent to those required of private financial institutions. Very few DFIs comply with disclosure standards similar to publicly traded commercial banks—in most cases, transparency and disclosure policies are simply not applied. Though a number of DFIs issue annual reports, disclosures in such annual reports are often lacking or rather opaque. Enhanced transparency standards could strengthen accountability and would bring pressure to bear on DFIs to achieve financial viability.74 CDB is audited by a Big Four audit firm75 and is among the few DFIs to adopt high disclosure standards in its annual reports. In addition, very few DFIs disclose assessments of the impact of their activities by, for example, publishing their project evaluation results, as is done by KfW (see further discussion in chapter 3). It will be important that the DBN adopts fully commercial standards of financial accounting, reporting, and disclosure; maintains full compliance with CBN supervisory standards; and issues audited financial statements at least annually. Instruments and clients76 The mandates of Public/Private DFIs are directed at addressing specific market failures and crowding-in private capital, thereby effectively leveraging limited public fiscal resources and DFI capacity. This has implications for their choice of instruments and the clients that they target. Certain instruments have more Public Sector features (such as grants or subsidized loans),77 while others have more Laissez-Faire features (such as lending products provided on commercial terms). The Public/Private approach relies on a combination of instruments to be deployed at the “risk frontier” (Beck and de la Torre 2006). According to this approach, the focus is on a market-inducing role, avoiding 73 See also discussion of the suggested prudential framework for DFIs discussed in chapter 4. 74 Chapter 3 discusses issues relating to the measurement of DFI performance in more depth. 75 The four largest international professional services networks in audit, assurance, tax as well as other services are Deloitte, Ernst & Young (EY), KPMG, and PwC. 76 This section draws on examples cited in de la Torre et al. (2007) unless otherwise specified. 77 As discussed earlier in this chapter, the outcomes supported by such public sector–type instruments have been disappointing largely due to the distortions and disincentives that accompany nonmarket-conform pricing. 53 situations where DFI interventions could be replacing the market. Examples of instruments in line with this approach are outlined below. An example of a Public/Private subsidy: Subsidies deplete fiscal resources and can create (price) distortions and moral hazard.78 A program called SIEBAN (Sistema de Estímulos a la Banca) provided by FIRA, a Mexican development agency, was designed to address these concerns in the following ways: (1) The subsidy was designed to last three years and decrease over time. The temporary nature of the subsidy was intended to mirror the falling cost of acquiring credit information about bank borrowers. Once borrowers have established credit histories, screening costs for financial institutions would be significantly reduced. The need for the subsidy would fall in line with the reduction of the distortion (information asymmetry) that the subsidy was designed to correct. (2) The moral hazard associated with SIEBAN was limited because it only provided a small subsidy during the initial period, and private financial intermediaries still bore the costs associated with eventual default, leaving the incentive to manage credit risk with the private sector. (3) Any enterprise receiving the subsidy was required to register with the credit bureau so as to ensure that the subsidy contributes to addressing the targeted market failure. An example of Public/Private credit guarantees: In developing countries, credit guarantee systems suffer from constant threats to their sustainability due to high default levels and/or poor risk management. FOGAPE (Fondo de Garantía para Pequeños Empresarios), a Chilean state fund designed to provide partial credit guarantees on loans issued by commercial banks to small firms, is considered to a success story in terms of fostering market activity while minimizing the problems that have characterized other guarantee schemes. FOGAPE works on a commercial basis with banks, where banks select loans and FOGAPE then checks to see if they meet the partial credit guarantee’s eligibility criteria. Banks have maintained high screening and monitoring standards because (a) they share in the risks of default—FOGAPE credit guarantees only cover 70 to 80 percent of loans, and (b) FOGAPE tests the banks’ risk appetite by regularly auctioning guarantees among participating banks. As a result, default rates on loans guaranteed by FOGAPE have been relatively low. An example of a Public/Private approach to structured finance: In providing working capital to shrimp producers, while also addressing potential principle-agent problems, FIRA has adopted an innovative practice in structured finance. FIRA outsources the client- screening work to a large shrimp distributor, Ocean Garden. Ocean Garden has information advantages in undertaking the screening of producers, and a strong incentive to select producers who can fulfill their obligations. This arises because Ocean Garden also provides shrimp producers with guarantees to cover any initial credit losses. By adopting this approach, FIRA successfully includes the private sector as a risk-sharing partner, leveraging its expertise in managing risk exposure. 78 Moral hazard refers to the case where one party (the borrower) tends to take more risks, because the borrower relies on another party (the DFI) to assume the risks. 54 NAFIN, a Mexican development bank, has adopted an innovative Public/Private approach to enhancing the supply of working capital. NAFIN has established an online market for factoring and reverse factoring services to SMEs. Factoring is a type of financing in which firms sell their accounts receivables at a discount to a financial firm (the factor) and receive immediate cash. Reverse factoring is a type of financing where a financial firm (the factor) purchases accounts receivables issued by certain larger reputable firms. This reduces information problems, as the factor only needs to assess the credit- worthiness of a select number of larger buyers. Factoring and reverse factoring are asset sales, not loans, and thus they help firms as they are not dependent on updated secured lending laws, or well-functioning registries and foreclosure mechanisms. NAFIN provides an online platform for reverse factoring, enabling SMEs that operate within supply chains (delivering their products to larger firms) to borrow against their receivables. Suppliers participating in the program are required to be screened by buyers (large credit-worthy firms) and need to be invited to join the chain, thereby effectively outsourcing supplier screening to the buyers. All transactions are carried out on an electronic platform, which reduces transaction costs and improves efficiency and security. NAFIN is responsible for the development, production, and marketing costs related to the electronic platform. 55 Chapter 3—Developing and Implementing a Robust Monitoring and Evaluation (M&E) Framework I Motivation for establishing an M&E framework for the DBN A results-oriented approach to monitoring and evaluation will not only help ensure that results are measured and can be communicated, but will also contribute to focusing the attention of the DBN’s management on effectiveness and efficiency. Understanding and defining the criteria of success at an early stage is crucial to being able to measure results. Moreover, being able to provide information on monitoring and assessment of results is a critical part of the feedback cycle: such information will be critical in identifying design gaps and implementation weaknesses, and thus improving the DBN’s effectiveness and efficiency. Despite the important role of robust DFI M&E frameworks, the monitoring and evaluation of DFIs is still an evolving science. In shaping the M&E process of the DBN, it would be relevant and desirable to draw on the M&E methodologies deployed as well as the outcomes achieved by other DFIs, given that most countries have established DFIs over past decades. However, as discussed further in this chapter, the available materials are not as strong as one might expect. There are several reasons for this. Firstly, while DFIs may have had limited capacity and faced technical challenges relating to developing appropriate M&E methodologies, the more likely explanation is that insufficient attention was paid to M&E by DFI management. The foundation for any M&E system is solid and comprehensive data, so it is crucial to establish routines from the outset that ensure the collection of relevant data. Secondly, the public sector mind-set that was pervasive in earlier generations of DFIs paid little heed to establishing an environment of transparency and accountability. In effect the seeds of the failure associated with the Public Sector approach described in chapter 2 were sown from the outset. If DFIs are allowed to function without appropriately designed M&E systems, they will be unable to provide the M&E evidence required to assess their performance. In the absence of any background information, due diligence by the DFI’s board and management becomes very difficult, and designing and taking corrective action to strengthen DFI efficiency and effectiveness becomes as good as impossible. II Selection and application of performance and impact indicators Due to methodological and measurement difficulties encountered in attributing impact on outcomes, most monitoring and evaluation frameworks focus predominantly on measured outputs.  Outputs are intermediate results directly produced by the DFI’s operations and can encompass the following elements: 56 1) Process: Process relates to how the DFI, the instruments at its disposal, or the project it manages operate. Process outputs are largely a reflection of the governance and implementation mechanisms deployed by the DFI, and do not relate to the financial and social outputs. Process outputs are typically qualitative and focus on whether a DFI follows preset governance practices, or on whether a development finance project follows its project cycle targets in implementation. 2) Sustainability: Sustainability measures the direct financial output of the DFI as reflected by financial indicators, such as the DFI’s capital adequacy ratio, returns (such as on equity, or internal rate of return in the case of project evaluation), and asset quality (measures such as loan nonperformance, the NPL ratio). Such measures can be reflective of the financial performance of the DFI, or particular instruments or projects. 3) Outreach: Outreach measures evaluate the DFI’s role and impact in addressing financing gaps, as discussed in chapter 1. These measures are mostly based on descriptive statistics, such as total assets, total annual credit and/or financial commitment, or total clients reached, combined with benchmark analysis using cross-country comparisons or time series comparisons. 4) Policy role: The policy role of DFIs (and the particular instruments or projects deployed by them) relates to whether they operate consistently within their mandates. Assessments of the policy role of DFIs, and whether they effectively address the specific financing gaps defined by their mandates, typically combine quantitative and qualitative analysis.  Outcomes refer to the DFI’s development impact in terms of achieving social and economic results, often referred to as “additionality” in the development literature. Additionality M&E indicators can be as specific as the enhancement of returns of firms that have received investment financing from DFIs, or as general as how the policy environment is impacted by a DFI. Due to the complexity of measuring additionality, studies using the additionality concept are sparse and focus mostly on specific instruments and projects, given that it is technically challenging to evaluate additionality of a DFI as an institution.79 As described in the following sections of this chapter, the usage of performance and impact indicators varies according to which M&E methodology is deployed in assessing DFI outputs and outcomes. While outputs are measured across all methodologies, measures of outcomes (additionalities) can typically only be measured at instrument or project level. 79 See Spratt and Collins (2012) for a discussion of four types of additionality (financial, design, policy, and demonstration) applicable to DFIs. 57 A. Use of scoring matrices At the institutional level, scoring matrices provide a feasible, flexible framework for DFI M&E. Scoring matrices list evaluation criteria, provide a score to measure the DFI under each criterion, and assign weights to the scores to generate an overall score for the DFI. An advantage with scoring matrices is that they provide room to integrate various analytical methods, both qualitative and quantitative, under each criterion of the evaluation. While it would be desirable to place more emphasis on measuring results, due to both the complexity of interpreting DFI mandates into a purely quantitative assessment framework and difficulties associated with implementation of more comprehensive quantitative evaluation methodologies, scoring matrices provide a suitable platform for integrating both available data and qualitative assessments in a combined approach. Hitherto institutional evaluation frameworks for DFIs using scoring matrices have predominantly focused on qualitative analysis. Due to the rather narrow focus of current institutional analysis on qualitative analysis, the reliability of current evaluation outcomes has been called into question, and greater efforts are being undertaken to incorporate quantitative analysis into prevailing scoring matrices. The score card adopted by the African Association of DFIs (AADFI) to conduct peer review evaluations provides a useful benchmark (see Table 3.1). The evaluation elements are grouped into process outputs under “governance standards” and “operational standards,” accounting for 60 percent of the total score. “Financial prudential standards” are viability outputs, accounting for the remaining 40 percent of the score. The methodology provides a comprehensive set of indicators relating to performance, but falls short in the evaluation of the outcomes of the DFI activities, such as their outreach or success in addressing financing gaps. Table 3.1: African Association of DFIs: Peer review evaluation matrix Rating category/weight Key evaluation elements Governance standards  Sufficient independence from government  Management independence and incentives (40 percent)  Operating in accord with commercial principles  Accounting and auditing  Management information systems and procedures  Other governance Financial prudential standards  Capital adequacy  Profitability and efficiency (40 percent)  Asset quality  Asset diversity and safety  Liquidity  Funding Operational standards  Risk management practices  Lending policies 58 (20 percent)  Loan appraisal policy and procedures  Supervision and collection policies  Operation strategy policies Source: Yuma Morisho (2014). Using scoring matrices has the following advantages: (1) Comprehensiveness and compatibility: Use of a scoring matrix can serve as a framework for integrating several assessment methodologies.80 (2) Simplicity: The methodology and the final score are easy to understand. (3) Comparability: Scores generated provide a consistent benchmark for cross-country comparison: the AADFI publishes a comparison table of African DFIs participating in its peer review process according to the evaluation matrix in Table 3.1 above. (4) Stability: Scores are generally relatively stable. This may be helpful in stabilizing DFI funding costs when accessing capital markets. Caveats associated with scoring matrices include: (1) Lack of objectivity and precision: Scoring matrices to a considerable extent rely on qualitative assessments. In administering a scoring approach, there is need to encourage use of quantitative analysis so as to strengthen the assessment of institutional effectiveness. (2) Slow responsiveness to changes in the policy environment and the DFI’s performance: The rather broad categories used in published scoring matrices may result in slow adjustment in response to changes in the policy environment and/or the DFI’s performance. On the other hand, DFI management would be in a position to follow the various components of each score and react more immediately as appropriate. The so-called “DFI health diagnostic tool” developed by the Inter-American Development Bank (IDB), incorporates more results and additionality indicators as regards outreach, financial sustainability, and policy roles (see Table 3.2 and Smallridge and de Olloqui).81 In addition to the commonly used process output indicators, similar to the indicators used by the AADFI, this scoring matrix seeks to include measures of development impact where available. The health diagnostic tool reflects the increasing emphasis on results and additionality of DFIs and has been applied in evaluating several Latin America and Caribbean programs. Table 3.2 summarizes results for the Bank of Foreign Trade of 80 KfW adopts a scale system that integrates other evaluation methods; see further discussion in the section on evaluation of projects and instruments. 81 Smallridge and de Olloqui (2011). 59 Colombia (Bancóldex), a state-owned commercial bank that operates as Colombia’s entrepreneurial development and export-import bank. Largely due to the unavailability of more rigorous additionality studies, this health diagnostic tool is still more qualitative than quantitative, and thereby more output-focused than results-focused. In presenting their results, the authors point to the weaknesses of IDB’s methodology in assessing the development impact of Bancóldex (see items bolded by the authors in Table 3.2). Table 3.2: IDB—Summary of health diagnostic tool for DFIs Parameters Dimensions Bancóldex review results Public policy and a. The government has a clear Bancóldex reflects the vision of its supervisory corporate strategy for developing and ministry (Ministry of Trade, Industry, and Tourism) governance promoting a target area, and in its planning, and uses the ministry’s overall this strategy has been a key economic targets (listed below) to develop its input in drafting the DFI’s strategic plan and objectives. vision, strategic plans, and budget. b. The DFI’s mission statement The law and bylaws clearly define the mandate of and resources are clearly Bancóldex. defined by law, act, or company bylaws. c. The DFI has a solid code of Bancóldex has a Code of Corporate Governance corporate governance. embedded in its regulatory framework, and is one of the top performers in the country according to a scoring system. d. The DFI has clearly defined Bancóldex is very clear in defining what an SME is the market gap and plays a and how budget should be split over various complementary role to private segments, products, and tenors. sources of capital. Development e. The DFI has clearly defined Bancóldex has clearly defined development impact development objectives. objectives at various levels. First, the government has introduced a Governability Management System (SIGOB). This system allows the government and every citizen to monitor the quality of the implementation of social and economic policies that are part of the country’s development plan. As part of the SIGOB framework, Bancóldex has objectives in the following domains (among others) for the period covering 2006 to 2010:  Total disbursements  Volume and number of disbursements to microenterprises  Volume and number of disbursements to SMEs  Number of entrepreneurs reached for training. In its annual report, Bancóldex is tracking a complementary set of development parameters, each of which has the potential to cover specific gaps. The most important parameters are:  Number of nonbanking correspondents (through which loans are sold)  Increase in regional presence (geographic gaps)  Introduction of new products (product gap)  Promotion of industrial renewal (industry gap) 60  Improving levels of capitalization of domestic companies. f. The DFI has clearly defined None of the DFIs interviewed (including Bancóldex) development criteria fully meets this criterion. This is, in part, a incorporated in its consequence of their tier-two operational models, lending/investment policies wherein the responsibility for credit allocation has and processes. been left in the hands of their commercial bank intermediaries. For example, Bancóldex leaves credit decisions to the financial intermediaries, explicitly stating that it does not analyze the underlying credit applications, but only reviews whether the application is in line with the formal requirement of the program g. The DFI regularly monitors its In general, impact evaluation studies are not development impact, and the carried out. Bancóldex is finalizing two studies to lessons learned are integrated measure the impact of some of its major into subsequent strategic programs. Bancóldex is also measuring the level of plans. compliance with its development KPIs and development objectives at various levels. h. An independent evaluation None of the DFIs interviewed (including unit carries out a review of the Bancóldex) presently has an independent unit or DFI’s development impact. department to review the development impact or return on its activities at any level (whether individual loan, program, business unit, or PDB as a whole). Financial and i. The DFI has a comprehensive Bancóldex has set up a very efficient multilayered operational marketing strategy that is distribution model capable of penetrating almost all performance consistent with its mandate. geographic areas and reaching all targeted segments and subsegments. j. The DFI has an independent Bancóldex has established an independent risk risk management function in function that reports directly to the president of the place that covers all types of board of directors. risk (credit, market and liquidity, and operational risk). k. The DFI has a clearly defined Bancóldex clearly understands a healthy risk risk strategy, supported by a management framework is essential for meeting its risk- and development- objectives. This is seen at various levels. adjusted financial reporting system, and a capital management framework. l. The DFI is cost-efficient and Barring some specific exceptions that should be productive and remains addressed by its budget, Bancóldex is precluded financially sustainable. from operating at a loss. Moreover, Bancóldex has clearly indicated that the efficiency of the organization, next to financial sustainability, is one of its strategic “internal action blocks.” The quality of financial results is such that financial sustainability is protected, which is demonstrated by the solid ratings assigned by the various credit rating agencies. However, Bancóldex has not defined any specific cost targets. Source: Smallridge and de Olloqui (2011). 61 Similar to evaluations at the institutional level, scoring matrices can also be used in instrument and project evaluations. Combining quantitative and qualitative criteria is somewhat easier when applied at the instrument or project level, as data at this level is generally more readily available. Such evidence can be derived from enterprise surveys that may provide information as to how development finance helps firms. Questions in the surveys could explore firms’ use of funding provided by DFIs; alternative financing sources available to firms; reasons for preferring DFI products; recipient firms’ level of satisfaction with the products; and the terms and conditions on which funding is provided. KfW evaluates its projects on a scoring matrix that integrates other evaluation methods (see Box 3.1). The scoring system serves as a framework, and as part of this framework, other methods (such as the cost–benefit analysis) are employed for the evaluation of specific areas. The KfW framework illustrates how quantitative analysis, such as cost–benefit analysis and appropriate quantitative indicators, can be incorporated into the scoring matrix assessment methodology. Box 3.1: The use of scoring matrices by KfW KfW evaluates its projects using a scoring matrix according to the following criteria: relevance, effectiveness, efficiency, overarching developmental impact, and sustainability. The overall rating on a six-point scale is the weighted average of scores on each of the criteria. Projects are considered developmentally “successful” only if they achieve a satisfactory score on project objective (“effectiveness”), impact on the overall objective (“overarching developmental impact”), and sustainability. According to KfW’s 12th Evaluation Report on Projects and Programs in Developing Countries 2011–2012 (KfW 2013), the “success rate” by number of projects is estimated to be 80.3 percent for sample projects undertaken from 2010 to 2012. Criterion Meaning Content (and method)82 Relevance—Is The extent to which the  Assess the extent to which the project focuses on KfW doing the objectives of a an important development problem (development right thing? development intervention priority), and whether there is a plausible causal are consistent with link between the project and its development beneficiaries’ objectives (validity of the results chain). requirements, country  Assess the extent to which the intervention is needs, global priorities, aligned with (sector) policies and strategies of the and partners’ and donors’ partner country (national plans, poverty reduction policies strategy) and partner institutions, as well as with the goals and guidelines and international standards. Effectiveness— The extent to which the  Assess the actual results of a project in terms of its Is KfW development intervention’s direct benefits. The intended results are reflected achieving the objectives are achieved, in the project or program objectives. To be able to objectives of taking into account their evaluate effectiveness, the project objectives, the relative importance starting from the appraisal phase, have to be development supported by concrete indicators in order to intervention? measure performance. 82 Bolded text highlights areas where the KfW scoring matrix incorporates more rigorous quantitative analysis. 62  For example: Supply of fifty liters per day of drinking water to each of 50,000 inhabitants year- round; 98 percent of water samples meet WHO standards. Acceptable limits must be established for anticipated negative side effects during project appraisal.  Unexpected (positive or negative) effects are included in the evaluation of effectiveness in the same way as the intended results. Efficiency— A measure of how  Assess whether the goods and services (outputs) Are results economically generated by the project were produced at an achieved in a resources/inputs (funds, appropriate cost (production efficiency). cost-effective expertise, time, etc.) are  More importantly, assess allocation efficiency, i.e., manner? converted to results the appropriate relation between the funds spent and the outcomes/impact achieved. Evaluating allocation efficiency requires a comparison of alternative options that achieve similar results. Here, cost–benefit analyses provide important indicators.83 Impact—Does Positioned above project  Impact cannot always be measured precisely but the objectives are overarching has to be estimated and made plausible based on development development goals, i.e., the circumstantial evidence. intervention goals that ultimately help achieve justified supporting the overarching activities on development goals? policy grounds. In the case of a water supply project, for example, the main issue is not how much water the target group consumes (direct benefit), but rather the improvements to the group’s health status resulting from the improved water supply, through reduced health risks from waterborne diseases. Sustainability— Sustainability is one of the  Risks that might affect the sustainability of the Are outcomes more ambiguous terms in development intervention are evaluated based on long-lasting? the international the likelihood that they will materialize. development debate. The  While the first four criteria pertain to the actual sustainability criterion is state of affairs at the time of an evaluation, met when the project assessing sustainability rests on expectations implementer or target regarding the future course of an intervention, and groups are able—once thus depends particularly on estimating the external financial, prospects and risks that will influence its future organizational, or technical impact. support has ended—to continue the project activities independently 83 For an example of a cost–benefit analysis as applied at the project level to the Canadian Small Business Financing Program, see Seens (2015). 63 and generate positive results for an appropriate period. Source: KfW website. KfW (2013). B. Use of summary quantitative performance indicators Rather than trying to develop more comprehensive measures of additionality, another approach to strengthening the institutional evaluation of DFIs is to develop summary indicators that lend themselves to easier measurement. With a view to strengthening the measurement of the performance of DFIs and their achievement of development objectives, Francisco et al. (2008) provide suggestions for such summary indicators. The suggested methodology measures a subsidy dependence index (SDI) and an output index (OI) of DFIs. SDI is defined as the annual subsidy received by the DFI as the proportion of the average annual yield earned by the outstanding loan portfolio of the DFI. This ratio measures to what extent the DFI relies on subsidies. The output index (OI) is defined as the weighted sum of loans extended to different income groups as the proportion of the sum of loans extended. The output index is designed to reflect whether the DFI lends comparatively more to lower-income groups, indicating the level to which the DFI fulfills its social objectives. The combination of the two indices may provide enhanced guidance as to whether DFIs are financially sustainable and whether they meet their social objectives. The subsidy dependence indicator approach was applied to two pilot case studies , Banadesa (the Honduran state agricultural development bank) and Banrural (the Guatemalan state agricultural development bank). The SDI estimate for Banadesa was high, suggesting the need to focus on reducing its subsidy dependence. The OI estimate indicated that Banadesa was lending comparatively less to low-income individuals, below the objective set by the government. While such indicators can be measured and compared across DFIs, their focus is rather partial compared to the more comprehensive institutional evaluation frameworks outlined above. Questions also arise as to the relevance of measuring the effectiveness of the impact of the subsidy element in DFI operations, when the Public/Private approach supported by this study places less emphasis on public subsidies and more emphasis on instruments designed to address specific financing gaps using market-conform funding. C. Use of statistical methods Statistical methods provide a more rigorous basis for evaluating DFI activities, although these are applied at the instrument level.84 In evaluating the impact of DFI activities, a counterfactual scenario is created that identifies what would have happened to the same firms in the absence of the DFI’s intervention. Thus, outcomes for two groups of firms are constructed: one group of firms benefiting from the DFI intervention (the treatment group), and the other group of firms that do not benefit from the DFI intervention (the control group) but are not systematically different from firms in the treatment group. 84 This section draws on Hurrell, A. and Aisa, M. G. (2015). 64 In order not to require random trials, quasi-experimental methods are used in constructing the counterfactual control group. The “difference in differences” (DID) methodology compares the change in the situation of participants and nonparticipants over time, assuming participants and nonparticipants would follow the same trend in the absence of the intervention. Box 3.2 provides further discussion of the DID methodology and an example of its implementation.85 Largely due to the data required to undertake the statistical evaluation of the additionality of DFI instruments, such evaluations are rather rare. As explained earlier in this chapter, application of more stringent evaluation methodologies depends on farsighted political commitment, particularly in regard to assembling relevant data. Table 3.3 summarizes the outcomes of several available quantitative evaluations of instruments funded by DFIs. The following are a few takeaways from these evaluations: (1) Despite the data requirements and technical challenges, additionality evaluation of DFI instruments can be done in a rigorous way. The studies are heavily reliant on data inputs both in undertaking the evaluations and in establishing the robustness of the results. A key challenge encountered in undertaking DID analysis is to construct a control group that has the same characteristics as the treated group with the absence of the DFI intervention, so that there are to be no significant differences (even in the unobserved characteristics) between the treated group and the control group. (2) Several lessons can be learned from these studies as regards the design of DFI instruments. There is evidence that DFI interventions in the form of credit lines and partial credit guarantees do improve firms’ access to finance and reduce financing costs, and certain studies show a positive impact on employment. Nonetheless, issues such as moral hazard are flagged to highlight the importance of taking care in design of recipient selection criteria and in implementation of DFI interventions.86 For example, there is evidence that partial credit guarantees issued by DFIs encourage firms to take excessive risks and thereby become more likely to go bankrupt. This arises due to the comfort provided by the DFI intervention, whereby the borrower no longer has to provide personal guarantees (Lelarge et al. 2010). This may also result in a situation where borrowers are less likely to repay their guaranteed credits (Cowan et al. 2012). One response to these findings, now quite commonly adopted by DFIs issuing partial credit guarantees, is to reduce moral hazard by prohibiting participating banks from informing their borrowers as to whether their loans are counterguaranteed or not. 85 This experimental method is often known as randomized controlled trials (RCTs). The control group is constructed by random selection (and is referred to as experimental) so as to control for any systematic differences. Establishing such randomly selected control groups may not be feasible or ethically acceptable in contexts such as DFI instrument evaluation where DFI activities have to be intentionally withheld from the selected firms. 86 Here moral hazard refers to situations where one party enjoys protection from risks and therefore has a higher risk appetite, as the other party bears the risk and will incur the cost should it materialize. 65 Box 3.2: An illustration of a quasi-experimental method The “difference in differences” method (DID) has been widely applied in policy research.87 DID is a statistical technique that attempts to mimic a scientific approach, “natural experiments,” and requires data to be assembled for two groups at two or more different time periods. Data such as the profitability or capital expenditure of MSMEs are assembled for two periods: (1) the first time period when both treated and control groups have not yet benefited from the DFI funding; (2) the second time period when the treated group has been exposed to the funding provided by the DFI. The impact of the DFI’s intervention can be estimated by calculating the difference in the changes of the indicators as between the treated and untreated groups from time period 1 to time period 2 (thus the term “difference in differences”). While it is not a requirement of the DID approach that the two groups have the same characteristics, DID does require that other than the treatment or intervention (in our case, the MSME finance provided by DFIs), the two groups should follow similar patterns of changes over time. Machado et al. (2011) provides an example of the application of DID in measuring additionality of an intervention by BNDES known as the “BNDES Card.”88 Data were analyzed for (i) MSMEs that used the card (the treated group: 7,765 firms) and (ii) MSMEs that were issued with the card, but did not use it (the control group: 14,807 firms). The study analyzed the impact on the variation of the number of employees using the DID method. If there had been no BNDES financing (through use of the card), the average results for both treated and control groups would have evolved on parallel trajectories. The evolution of the control group was used as the counterfactual for the analysis. The impacts were examined by firm size and period of time (“short term” referring to 2007–2008 and “medium term” referring to 2007–2009) (see Figure 3.1). For both periods, the study finds a positive impact on the average number of employees of the firms that received credit lines through the BNDES card. Figure 3.1: Estimated impact of BNDES Card on employment—Short term and medium term difference-in-differences calculations Average number of employees Medium- Average number of employees Short-term term impacts: impacts: 9.6% 7.9% Didn’t use card Used card Counterfactual Didn’t use card Used card Counterfactual Source: Machado et al. (2011), based on data from BNDES and Brazil Ministry of Labor and Employment (MTE/Rais). 87 Early applications of DID in policy study include Ashenfelter (1978) and Ashenfelter and Card (1985). 88 A revolving credit granted to MSMEs for the purchase of goods, industrial inputs, and services. 66 Table 3.3: Instrument evaluation studies Instrument and Study additionality under Methodology and data Key findings evaluation Credit lines  In the short term, the study finds an average estimated increase  Instrument: “BNDES of 7.9 percent in the number of employees. Card” (a revolving credit  In the medium term, the study finds an average estimated  Methodology: DID granted to MSMEs for the positive increase of 9.6 percent in the number of employees.  Data: MSME firms who used the card (the Machado purchase of goods,  The impact decreases as firm size increases. When breaking treated group: 7,765 firms), and MSME firms et al. industrial inputs, and down the impact by firm size, impact is statistically significant who were issued the card but did not use it (the (2011) services) only for micro and small firms. In the short term, the impact control group: 14,807 firms)  Additionality: Impact on on employment by micro firms was 10.2 percent, whereas for  Data time period: 2007–2008; 2007–2009 firm employment (number small firms the impact was 4.3 percent. In the medium term, of employees) the impact was 13 percent for micro firms, and 4.7 percent for small firms.  The study suggests that loans supported by BNDES reduce the firms’ financing costs significantly.  Methodology: Cross-sectional analysis, DID  Instrument: BNDES loans  BNDES support does not seem to affect firm-level operational (fixed effects) Lazzarini  Additionality: Impact on performance and investment decisions, and the subsidized  Data: Firm-level panel data of 286 publicly et al. firm financial loans do not seem to improve firms’ profitability, market listed companies (which may present a (2011) performance and valuation, or investment. selection bias of this study) investment decisions  In addition, the study finds that BNDES appears to be  Data time period: 2002–2009 generally selecting firms with capacity to repay their loans, as regular commercial banks would do.  Instrument: World Bank’s Turkey Credit Line  Methodology: DID (fixed effects regression)  Sales: sustained 21 percent sales growth Projects—four Export  Data: seventy-two firms (out of a sample size  Employment: sustained 14 percent employment growth from World Finance Intermediation of 168) project participants, fifty control group investment loans Bank Loan (EFIL) projects firms  Exports: sustained 28 percent export growth (2011)  Additionality: Impact on  Data time period: 2005–2008  Investment: 79 percent higher investment in the year the loan financial and operational  Also made reference to firms’ self-assessment was received data of firms (sales, of the program to support qualitative analysis employment, number of 67 workers, exports, investment)  Instrument: World Bank’s Turkey Credit Line  Methodology: DID (fixed effect regression) Projects—two Access to  Sales: no significant effect found  Data: 166 firms that received a loan through Finance for SME projects  Employment: sustained 15 percent employment growth from the program between 2007 and 2009  Additionality: Financial investment loans  Data time period: 2006–2008 and operational data of  Investment: 210 percent higher investment in the year the loan  Also made reference to firms’ self-assessment firms (sales, employment, was received of the program to support qualitative analysis number of workers, investment)  In general, SME support has been effective in improving  Methodology: DID (fixed effects) combined  Instrument: Interventions employment. with propensity score matching (PSM)91 provided by major SME  Credit support has the most significant positive impacts on techniques to construct a control group, and support programs to employment and wages. Credit is estimated to have created other robust tests manufacturing businesses three jobs per establishment on average, representing a 13  Data: Firm-level data.92 Credit (treated: 6,919 in Brazil, including credit, percent increase in the number of workers. Credit support also firms; control: 9,855–29,429 firms); Pires et al. agglomeration,90 export, has significant positive impacts on exports (in value) and agglomeration (treated: 695 firms; control: (2014)89 innovation, and consulting patents and trademarks. 1,547–29,429 firms); export (treated: 964 support  Agglomeration, export, and consulting also have statistically firms; control: 2,212–29,429 firms);  Additionality: Impacts on significant impacts on employment. Consulting is estimated to innovation (treated: 103 firms; control: 287– employment, wages, have created nearly four jobs per establishment on average, 29,429 firms); consulting (treated: 194 firms; exports value range, and representing a 16 percent increase in the number of workers. control: 421–29,429 firms) patents and trademarks  Export intervention has a significantly positive impact on  Data time period: 2003–2012 exports. Innovation also has a positive impact on exports. 89 There are other studies evaluating the same instrument, such as: Larraín, F. and Quiroz, J. (2006). Evaluación de la Adicionalidad del Fondo de Garantías de Pequeñas Empresas, manuscript, BancoEstado, Santiago, Chile; and Benavente, J. M., Galetovic, A., and Sanhueza, R. (2006). FOGAPE: An economic analysis. 90 A general support provided by SEBRAE to reinforce and build up cooperation and governance among the public and private sectors. 91 Propensity score matching (PSM) constructs the counterfactual (control group) by matching individuals/households/firms receiving the intervention treatment, with those with (preintervention) characteristics as statistically similar as possible but who are not receiving treatment, based on the probability that the individual/household/firm is covered by the intervention (propensity score). PSM is an intensive approach that requires individual models for subgroups of firms. It is data-heavy in requiring a large sample of data units. 92 Three control groups are constructed, thus the sample size for control groups vary. 68  Agglomeration does not have a robust impact other than a small positive impact on employment. Partial credit guarantees (PCGs)  The program significantly impacted the development of newly  Methodology: DID and ordinary least squares created firms (OLS). A unique aspect of the design of this  Firms targeted by the program are found to systematically study is utilizing a quasi-natural experiment of  Instrument: Partial credit raise more external finance, pay lower interest expenses, and a policy shock. In 1995, the government guarantees (PCGs) enjoy higher growth rates than other similar firms. decided to increase the loan guarantee scheme, provided by SOFARIS’s not only by further increasing the budget  On the industry level, the program is found to be mostly Creation Fund to new effective on the intensive margin. While the availability of Lelarge et allocated to SOFARIS (and thereby increasing ventures mostly through loan guarantees allows newly created firms to become larger, al. (2010) the amount of subsidized loans in already medium- to long-term it does not trigger an increase in the overall number of firms eligible industries) but also by providing loans created. eligibility to additional industries.  Additionality: : Impact on  A potential concern is that loan guarantees cause firms to  Data: Firm- and loan-level data sourced from credit available to SMEs become more likely to go bankrupt. While banks can in the SOFARIS Information System (with a and repayments general ask entrepreneurs for personal guarantees, they cannot total number of observations of 207,214 for the study) do so if the loan they provide is guaranteed by SOFARIS, which provides entrepreneurs with incentives to take more  Data time period: 1989–2000 risks.  Instrument: Partial credit guarantees (PCGs) on  Credit guarantees increase the amount of credit: one additional SME credit availability  Methodology: DID, regression discontinuity dollar of insurance increases the total credit for SME by $0.65. Cowan et and repayment in Chile, design (RDD)  Borrowers are 2 percent less likely to repay insured loans, al. (2012) FOGAPE  Data: Loan and PCGs data by operations which could be explained by a decrease in the collection effort  Additionality: Impact on  Data time period: 2003–2006 by the bank. credit available to SMEs and repayments 69 III Designing an M&E framework for the DBN While different evaluation methods have their strengths and weaknesses, as noted in previous sections of this chapter, scoring matrices provide a useful framework for assembling both qualitative and quantitative information about DFI performance. By applying higher scores to quantitative elements, the evaluator can compensate for the subjectivity of qualitative scores, while still preserving the scope of the overall assessment methodology. Similarly, the scoring methodology also provides the evaluator the opportunity to place greater emphasis on core aspects of the DFI’s operations, thereby tilting the evaluation outcomes toward those aspects of the assessment that are regarded as more important to fulfilling the DBN’s mandate. Given that there are qualitative elements to the evaluation process, it will be important that the evaluation is undertaken by an adequately staffed independent unit within the DBN. In a similar way to the autonomy usually assigned to an enterprise’s internal audit committee, the evaluation unit should be established as an independent unit referring directly to the chairman of DBN’s board. While the evaluation unit does not need to be an external party, it should be independent of the DBN’s management and operational staff. The evaluation unit should report directly to the board of directors and operate independently of the management and operation units. It should have a separate budget, personnel, and reporting lines. Hiring reputable external consultants may be a way to introduce technical expertise in specific areas and even help achieve a more objective assessment, but it is unlikely that external consultants will be able to replace the first-hand understanding of the evaluation context provided by an internal evaluation unit. Two examples of internal DFI evaluation units are the World Bank Group’s Independent Evaluation Group (IEG), and KfW’s evaluation unit. IEG follows the principle of “independence within”: while still part of the World Bank Group, it is independent of the management and operation units. Key factors that help IEG meet the independence criteria are summarized in Box 3.3. Box 3.3: Independence within the World Bank’s Independent Evaluation Group (IEG) IEG is an independent evaluation function within the institutional structure of the World Bank. While the IEG faces internal pressures to its independence, an external review noted that engagement with management did not undermine independence: “on the contrary, such interaction should be increased to ensure the usefulness of evaluation products.” IEG Independence Criteria Criterion Factors that help meet criterion Organizational IEG reports directly to the World Bank board through the Committee on independence Development Effectiveness (CODE) and is thus organizationally independent from management and operational staff whose activities are being evaluated. IEG’s scope of responsibility extends, without restriction, to all the determinants of the World Bank Group’s operational results. Behavioral IEG’s work program and budget are endorsed by CODE and approved by the independence board. 70 IEG’s budget is separate, and World Bank management does not have authority over IEG’s budget or its use. IEG’s reports are transmitted to the World Bank’s board through the Director General for evaluation, without any clearance from management. Although management is given the opportunity to review and comment on draft reports, IEG decides whether or not, and how, to address such comments. Avoidance of IEG’s staff does not evaluate activities that they were previously responsible for conflict of interest or were involved in. The head of IEG and its Director General are not eligible for employment in other positions in the World Bank Group or for consulting assignments. Protection from IEG’s three-year rolling consolidated work program and budget are prepared outside influence independently of management for endorsement by CODE and approval by the board. Source: Boehmer et al. (2016). The evaluation unit of KfW has a similar feature of “independence within.” The Financial Cooperation Evaluation Unit of KfW concentrates on independent ex post project assessments while remaining within the operational structures of KfW. Box 3.4 provides more details about the operational independence of KfW’s evaluation unit. Box 3.4: The operational independence of KfW’s evaluation unit Independence As an administrative entity, the evaluation unit reports directly to KfW’s executive board. The evaluation unit works independently of operational country departments, which are responsible for appraising projects and supervising their preparation and implementation. Completed projects and programs have been evaluated for their effectiveness since the 1960s. Earlier this task fell to the operational departments. Since 2000 an independent evaluation unit has had responsibility for assessing project development effectiveness, drawing on its own staff and commissioning independent experts. These experts may be employees from KfW operations or external specialists, but are never individuals who themselves were involved with the evaluated projects. Evaluation methodology Initially all completed projects were evaluated, and results as summarized by a success rate were published. Given the growth in the number of projects since 2007, the success rate has been estimated by sector using a representative random sample stratified of about 50 percent of completed projects. Moving from the study of a full project inventory to a sample allows for more extended analysis of particular interventions and themes. Beyond individual projects, the evaluation unit undertakes evaluations in selected thematic areas in order to explore the context of particular results, specific sectorial questions or the appropriateness of certain ideas and development frameworks. More rigorous statistical methods can be applied where appropriate. 71 A database with results from over 2,000 project evaluations provides the basis for cross-cutting analysis on specific topics. Disclosure KfW’s transparency portal (http://transparenz.kfw-entwicklungsbank.de/en) provides up-to-date information on the source, use, and impact of development projects by country, sector, and project. It also provides summaries of all of KfW’s evaluation reports. Source: KfW (2013). The M&E unit should be adequately staffed and be assured access to any information about the DBN that it may require in undertaking its evaluations. Indeed, the success of the M&E process very much depends on adequate resourcing of the evaluation unit. The board’s endorsement of an evaluation methodology should be sought as part of the inception of DBN’s activities. It is to be expected that the evaluation process will need to be modified as the DBN’s activities are rolled out and changed, and also as experience in using particular evaluation processes reveals weaknesses or opportunities for improvement that might be adopted and implemented. Clearly, as with all its activities, the DBN will need to pay close attention to the cost-effectiveness of the evaluation process itself.93 Table 3.4 assembles some core components and suggested indicators of an M&E framework suitable to the DBN. The focus of the suggested M&E framework is on monitoring inputs, processes, products, and outputs. The majority of the suggested indicators are quantitative, although there are also qualitative indicators (such as on corporate governance, internal cost control systems). Table 3.4: Suggested key elements of an M&E framework for the DBN Evaluation items Key indicators Process/inputs Corporate governance review Review of internal control systems Operational efficiency measures: costs per employee, costs per loans extended Products Measure of compliance of activities with intentions of DBN lending programs Financial sustainability indicators: capital adequacy ratio (CAR), nonperforming loan ratio (NPL), provisions over NPLs, return on equity (ROE) Subsidy dependence index (SDI) Outputs Percentage of firms supported by DBN programs Share of total MSME credit supported by DBN programs Share of long-term finance supported by DBN programs The output index (OI): weighted sum of loans extended to different income groups as the proportion of the sum of loans extended 93 While quantitative processes are encouraged, a balanced approach is important, as quantitative assessments will require that the DBN assemble specific data of undertaken surveys on a regular basis, which can themselves be costly processes. 72 Outcomes Growth in productivity and employment of firms supported compared with economy-wide trends Full support by DBN’s shareholders and DBN leadership (the board of directors and management) will be critical to undertaking effective M&E. In giving credibility to the M&E process, it will be important that DBN’s shareholders (especially the principle shareholder, the Federal Ministry of Finance) commits to using the outcomes of the M&E process as a key indicator in evaluating the performance of DBN’s board and management. As in many other countries, existing Nigerian DFIs have not defined a rigorous analytical approach to M&E, or assembled the data required to perform systematic analysis or make public the outcomes of such analysis. It will be essential that DBN’s shareholders and board pay particular attention to (i) defining the institution’s approach to M&E at the time of its inception, and (ii) setting up a rigorous results framework that details specific data collection requirements, and specifies the frequency of M&E reporting and how the outcomes of the evaluation will be disclosed. The DBN would be well-advised to set up mechanisms for data collection prior to rolling out its product offerings. In addition to requiring periodic disclosure of basic financial and operating data from participating financial institutions (PFIs), the DBN should periodically collect financial and operating data from financial intermediaries and firms that participate in DBN programs, making the submission of such data a requirement of the participating financial intermediaries. 73 Chapter 4—What Makes DFIs Work or Fail? I DFI governance arrangements A. Introduction Too little attention has hitherto been paid to the governance arrangements, related safeguards and design of development finance initiatives. Until now very little has been done in Nigeria to support the establishment of an effective development finance framework. As discussed in chapter 1, while various development finance initiatives were largely motivated by the need to compensate for weaknesses in the institutional/financial framework, neither the DFIs nor the CBN’s schemes have been able to address the market gaps in a sustainable way. Their impact was often limited and associated with sizeable fiscal or quasi-fiscal costs. The question posed by this chapter is—given the decision to establish a new DFI, how should it be governed to minimize risks and maximize chances of success? This chapter focuses on defining the institutional arrangements that will allow the DBN to fulfill its role more effectively, in particular identifying the most appropriate governance structure for the DBN. This chapter reviews the following core elements of the governance landscape:  DFI ownership and control;  appointment and responsibilities of the board of directors and management;  defining and achieving the mandate of the DBN;  achieving financial sustainability;  disclosure and accountability; and  regulation and supervision. In each of these areas, the recommendations are based on the review of international guidelines and experience as well as previous experience in Nigeria. The final section of the chapter pulls together lessons learned as regards ensuring a new approach to development finance in Nigeria, such that the DBN can achieve the additionality envisaged by the Public/Private approach to development finance. Box 4.1 below summarizes the key governance principles discussed in this chapter. Box: 4.1 Summary of key governance principles To ensure sustainable impact and scale, it will be important that the Development Bank of Nigeria is grounded in internationally recognized, good corporate governance principles. Observance of these principles includes the following arrangements/criteria (see respective subsection below): A. A mixed public and private ownership structure B. Board and management to be composed of independent and highly qualified individuals selected through a merit-based, transparent process C. A mandate to support the financial services needs of Nigerian MSMEs and for its products to be designed in a manner that: 74 a. effectively addresses market gaps, and complements and leverages private sector funding; b. pricing fully reflects costs and credit risk; and c. ensures full financial sustainability D. Wholesale-only, second-tier business model coupled with an effective risk management system to ensure maximum financial sustainability and minimize risks E. Regulations and governance arrangements to provide for adequate transparency and disclosure requirements, including standards for accounting and auditing—overall, DFI standards should be comparable to those of publicly listed firms F. Prudential regulation and supervision by Central Bank including (i) enforcement of requirements essentially similar to those applied to commercial banks (but taking into account the significantly lower risks taken by the DFI); (ii) strong technical oversight by MOF; and (iii) eligibility requirements for participating financial institutions. Brief introduction to corporate governance (CG) Corporate governance (CG) refers to the mechanisms, processes, and relations by which corporations and other institutions are controlled and directed. Indeed, CG involves a set of relationships between a company’s management, its board, its shareholders (owners), and other stakeholders. According to the OECD and the G20,94 CG is a key element in improving economic efficiency and growth as well as enhancing investors’ (as well as citizens’) trust and confidence in companies and markets. CG also provides the structure through which the objectives of a firm or institution are set, and the means by which those objectives are attained and performance is monitored. A good CG framework not only provides incentives for the board and management to pursue key objectives that are in the interest of the institution and its shareholders, but also facilitates effective monitoring of the institution’s performance and operations. The presence of an effective CG system, both within a company but also across an economy as a whole, helps to provide confidence in the institutional framework that is necessary for the proper functioning of a market economy, reducing the cost of capital and encouraging institutions to use resources more efficiently—thereby underpinning economic efficiency, sustainable growth, financial stability, and social development. This is as true to privately owned institutions as it is to DFIs. According to the World Bank95 and the recently released G20/OECD Principles on CG, a robust corporate governance framework is one that focuses on: (i) the strategic guidance of the 94 OECD, Principles of Corporate Governance, available at: http://www.oecd.org/daf/ca/corporategovernanceprinciples/31557724.pdf (accessed on May 11, 2015); and the G20/OECD Principles of Corporate Governance, released in September 2015, available at: http://www.oecd.org/corporate/principles-corporate-governance.htm (accessed on September 26, 2015). 95 The World Bank, “Template for Country Assessment of Corporate Governance ,” Revision 5.0, July 2007. The template—based on the OECD CG principles and used as a diagnostic tool for assessing the CG framework of a particular country—assesses, inter alia, the basis for an effective CG framework, rights of shareholders and their equitable treatment, the role of stakeholders, disclosure and transparency, responsibilities of the board, and key ownership functions. 75 company, (ii) the effective monitoring of management by the board, and (iii) the board’s accountability to the company and the shareholders. Governance in the DFI sector Recently published international guidelines place emphasis on applying similar governance principles to DFIs as to private commercial enterprises, consistent with the Public/Private approach to DFIs. According to the OECD Guidelines (2005)96 and the more recently published World Bank Toolkit,97 the main principles to be followed by state-owned firms and DFIs are the following:  Ensuring an effective legal and regulatory framework for state-owned enterprises To avoid market distortions, the legal and regulatory framework for SOEs should ensure a level playing field in markets where SOEs and private sector companies compete. Such a framework implies clear separation among the state’s ownership function, simplified operational practices for SOEs, uniform application of general laws and regulations to all enterprises including SOEs, and no privileged access to SOEs for factors of production including finance.  The state acting as an owner The state should act as an informed and active owner and establish a clear and consistent ownership policy, ensuring that the governance of SOEs is carried out in a transparent and accountable manner, with the necessary degree of professionalism and effectiveness. For example, the state should have no involvement in the day-to-day management of SOEs; and should let SOE boards exercise their responsibilities and respect their independence.  Equitable treatment of shareholders The state and SOEs should recognize the rights of all shareholders and ensure their equitable treatment and equal access to corporate information. For example, SOEs should be highly transparent with all shareholders, develop an active policy of communication and consultation with all shareholders, and protect the rights of minority shareholders. 96 OECD (2005) Guidelines on Corporate Governance of State-Owned Enterprises, available at http://www.oecd.org/daf/ca/34803211.pdf, accessed on May 14, 2015. 97 World Bank (2014). Corporate Governance of State-Owned Enterprises: A Toolkit. DOI: 10.1596/978-1-4648- 0222-5. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 IGO. Available at https://openknowledge.worldbank.org/bitstream/handle/10986/20390/9781464802225.pdf?sequence=1, accessed on December 1, 2015. 76  Relations with stakeholders The state ownership policy should fully recognize the SOEs’ responsibilities toward stakeholders and request that they report on their relations with them. For example, large SOEs, and SOEs pursuing important public policy objectives, should report on stakeholder relations.  Transparency and disclosure SOEs should observe high standards of transparency, such as developing consistent and aggregate reporting, and publishing an annual independent external audit based on international standards.  Responsibilities of SOE boards SOE boards should have the necessary authority, competencies, and objectivity to carry out their function of strategic guidance and monitoring of management. They should act with integrity and be held accountable for their actions. For example, SOE boards should be assigned a clear mandate, be responsibility for the company’s performance, and be fully accountable to the owners. SOE boards should be constituted so that they can exercise objective, independent judgment. In the DFI sector, three overarching objectives98 for which boards and executive managers need to be held accountable are: 1. achieving public policy objectives; 2. ensuring the institution’s long-run financial sustainability, i.e., that the DFI continues as a going concern, without need for extraordinary financial support from the state; and 3. achieving certain financial performance objectives defined in terms of criteria, such as return on equity and dividend payment targets. While these governance principles are important, they also imply trade-offs. A major challenge for DFIs is to find a balance between the government’s responsibility for actively exercising its ownership functions (such as the nomination and election of the board) and refraining from imposing undue political interference on the management of the company. Another important challenge for the authorities relates to finding the right balance between the core objectives above, as it is conceivable that tensions could arise between achieving public policy objectives and ensuring the DFI’s financial sustainability. How best should management, the board, and the shareholders manage these conflicting goals? 98 The objectives of DFIs are discussed in greater detail in chapter 3, where they are reflected in the proposed M&E framework for the DBN. 77 Many of the problems commonly recognized to afflict DFIs can be associated with, if not directly attributed to, weaknesses in corporate governance.99 Identified examples of such weaknesses include:  Government officials acting in the capacity of shareholders directly intervening in day-to-day operational decisions, but also government in all its forms (even without a formal role) asking DFIs for special favors (e.g., to whom to lend, on what terms to lend, and when to forgive indebtedness): political intervention or political capture;  Executives acting almost autonomously (without clear reporting lines), pursuing unintended objectives: mission creep; taking decisions contrary to commercial and/or financial management principles thus eroding the institution’s self-sustainability;  Board members lacking the necessary experience, skills, and capacities to effectively and properly exercise their functions according to the institution’s objectives; and  Lack of accurate and complete reporting (on financial and nonfinancial matters alike), giving rise to inadequate decision-making by those who rely on reporting, and thereby misleading shareholders, investors, legislatures, and society in general. In this context and so as to guard against such pitfalls, it is important to bear in mind that governance arrangements need to undergo periodic, independent evaluations, especially regarding the workings of the board of directors (as addressed below). B. Ownership Most governance problems originate from ownership and control being exercised by different parties. Indeed, those who own an institution (the shareholders) are seldom the same people who run it on a day-to-day basis (the board and management). This is especially true of large, publicly listed or state-owned enterprises, because owners are usually representatives of state agencies whose interests may well differ from those of the board and/or management. Ownership entails providing a certain set of rights to the owners of a company—such as the right to obtain accurate, material, and timely information; participate and vote in board meetings; and elect and remove board members. The manner in which board powers, independence, and objectivity (as regards judgment on corporate affairs) might be underpinned depends on the ownership structure of the company. A dominant shareholder has considerable powers to appoint the board and the management. However, the board still has a fiduciary responsibility to the company and to all shareholders including the minority ones. In addition, institutional interests of the owner (e.g., MOF) and the DFI may differ. For example, the ownership entity may want the DFI to carry out certain public policy goals without necessarily wanting to directly bear the costs for them. Indeed, this is one kind of issue that arises from the “state as owner” problem (referred to above): Does the state as owner have the 99 David H. Scott, “Strengthening the Governance and Performance of State -Owned Financial Institutions,” Policy Research Working Paper, The World Bank, August 2007. 78 capability and will to exercise its ownership rights professionally, and play its role as shareholder in guiding the DFI to achieve its objectives? Will the state accept a limited, hands-off role that gives autonomy to the DFI to pursue its objectives? The advantages of mixed public/private DFI ownership DFI ownership can be divided into three types: full government ownership, mixed public and private ownership, and full private sector ownership:  There are many examples of fully government-owned DFIs, by far the most frequently adopted ownership structure. Examples include: BNDES (Brazil’s Banco Nacional de Desenvolvimento Economico e Social), DBSA (South Africa’s Development Bank of Southern Africa), NAFIN (Mexico’s Nacional Financiera), KfW (Germany’s Kreditanstalt für Wiederaufbau), CDC (the United Kingdom’s CDC Group Plc), Sweden’s SwedFund, Norfund (Norwegian Investment Fund for Developing Countries), IDC (South Africa’s Industrial Development Corporation), and OPIC (the United States’ Overseas Private Investment Corporation).  While there are examples of DFIs that have been able to attract private sector ownership, these are largely DFIs established in advanced market economy circumstances.100 As pointed out in chapter 2, in developing and emerging markets, given the fragile domestic institutional infrastructure, private sector investors’ willingness to invest in DFI equity is limited. Indeed, the primary role of third-party investors would be to strengthen DFI governance, and it is likely that this role, at least initially, will be filled by so-called “impact investors.” The intention of such impact investors is to generate a measurable, beneficial social impact alongside a financial return.101  Finally, examples of completely privately owned DFIs are rarer and include: SIFEM (Swiss Investment Fund for Emerging Markets) and TKSB (Turkey’s Türkiye Sınai Kalkınma Bankası). Multilateral and regional DFIs generally have multiple shareholders from various countries.102, 103 100 Examples of DFIs with a mixed public and private ownership structure include: Proparco (France’s Société de Promotion et de Participation pour la Coopération Economique), FMO (the Netherlands’ Nederlandse Financierings Maatschappij voor Ontwikkelingslanden n.v.), COFIDES (Spain’s Compania Española de Financiación del Desarrollo S.A.), and SIMEST (Italy’s Società Italiana per le Imprese all’Estero SpA). 101 As already noted in chapter 2, impact investments of the kind envisaged here could be undertaken by the private sector–focused donor agencies, such as the International Finance Corporation (part of the World Bank Group), the Commonwealth Development Corporation (the UK government agency focused on investment in private sector development), or Kreditanstalt für Wiederaufbau (the German government agency with a mandate to take ownership stakes in private companies in developing countries). 102 For example, the multilateral DFIs—PIDG, IFC, EBRD, EIB, AFDB, ADB are all owned by their shareholders or member governments. Therefore, national governments tend to own or be shareholders in several DFIs at the same time. DFID, for instance, is a shareholder in CDC, PIDG, EIB, EBRD, IFC, AFDB, and ADB. 103 On the categorization of DFIs proposed here, see “Comparing DFIs—Literature review.” Kingombe, Massa, and te Velde, January 2011, Overseas Development Institute, available at: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/67635/comparing-DFIs.pdf, accessed on March 16, 2015. 79 Although the majority of DFIs are government-owned, the mixed public–private ownership structure provides a number of safeguards and is emerging as good practice. The advantage of mixed ownership is that it reduces the risk of governance failures because it provides protection in the form of safeguards from political intervention in the DFI’s day-to-day operations while maintaining the DFI’s focus on public policy goals. In the case of investment by so-called “impact investors”—who are themselves “soft investors” with broader socioeconomic objectives and/or multilateral development institutions—DFIs in developing and emerging markets stand to benefit considerably from the experience and technical expertise that such investors have to offer, both as regards governance and management practices, but also in designing and implementing development programs. Indeed, the presence of such impact investors is likely to be more beneficial to the governance of DFIs than private investors in the traditional sense, each with much smaller stakes and thereby less influence than institutional impact investors. The board and management selection processes are key mechanisms by which mixed ownership reduces the risk of governance failures. These selection processes provide assurance that a DFI’s board members and management are selected (and removed) based on competitive criteria and processes that favor independent, professional, and highly experienced persons. In this context, mixed ownership can also ensure some degree of whistle-blowing by the nongovernment stakeholders, should problems arise.104 For example, where third-party impact investors assume minority ownership stakes in the DBN, their role on the board of the DBN may primarily be to act as a deterrent—namely, in assuring that agreed standards set for the operation of the DBN are fully observed, thereby preventing any unwanted interference that might set aside or compromise the principles for board and management selection as defined in the DBN’s Memorandum and Articles of Association.105 In addition, mixed-owned banks tend to operate under private commercial law and are not subjected to public sector administration procedures. This allows mixed-owned banks to offer competitive salaries to attract qualified professionals (outside of the civil servant cadre) and to execute contracts under commercial practices. By having both public and private sector representatives on the board, mixed ownership structures also help DFIs align themselves with the interests of a broader set of stakeholders. Due to enhanced governance, mixed ownership may provide additional benefits to DFIs in facilitating access to private sources of funding, as has been argued in the case of Proparco.106 If the DBN is to achieve scale and impact, it will need to be able to access capital market funding, and in order to do so, it may initially need to rely on a government guarantee. However, in subsequent funding rounds, it is envisaged that the presence of reputable impact investors as minority owners will work toward reducing the DBN’s financing costs. It is difficult to establish, with any degree of certainty, whether there is an “optimal” mix of public–private ownership.107 But, as outlined below, even a small degree of private ownership 104 See discussion around the “superdirector syndrome” in chapter 2. 105 These principles refer to competitive and transparent selection among candidates based on their professional qualifications. 106 According to Kingombe, Massa, and te Velde, op. cit. 107 Among DFIs on Western markets with a mixed ownership structure, the size of the private sector shareholding varies, for example, from 15 percent in Proparco to 49 percent in FMO. The mixed ownership structures of the cited 80 will have the preventative impact described above, providing a platform for intervention before governance risks materialize, thereby minimizing the risk of governance challenges, such as the ones faced by BNDES and DBSA discussed in the case studies below. The choice of the state entity responsible for executing the state’s ownership rights can have an important impact on governance: whether the state’s ownership stakes are administered by a specific department within a ministry, an autonomous agency, or another government body or entity. The OECD Guidelines on Corporate Governance of SOEs 108 suggest the state should exercise its ownership functions through a centralized ownership entity, or effectively coordinated entities, which should act independently and in accordance with a publicly disclosed ownership policy.109 The guidelines also suggest the strict separation of the state’s ownership and regulatory functions in order to minimize the risk of conflict of interest. Control arrangements In addition to direct private ownership, “control arrangements” can also be put in place as a way to ensure that private sector shareholders (equity investors or institutional lenders) exercise control on key management and governance issues. These mechanisms are designed to ensure that the DFI is properly run; fulfills its performance, financial, and public policy objectives; maintains financial sustainability; and is as far as possible insulated from political interference. Such arrangements can be set up to the advantage of outside parties, even if those outside parties do not have the equivalent stake in the institution that would warrant such a level of control. That is to say, as long as there is agreement among all stakeholders, arrangements such as an “investor rights agreement” can stipulate a certain level of nongovernment control over specific strategic decisions. Given the advantages of mixed public–private ownership, the DBN would benefit from having private sector shareholders110 holding enough equity to provide them with effective control over sensitive and strategic issues concerning the institution. Alternatively, if such level of private sector shareholding cannot be attained, “control arrangements” could be put in place, whereby private sector entities (shareholders and/or lenders) would have an effective say on predefined strategic issues. These issues could include but need not be limited to appointment of directors and the observance of specific criteria in the selection of board members. public–private DFIs are as follows. Proparco: AFD (59 percent); French Financial Institute (26 percent); IFIs (11 percent); French companies (3 percent); funds and ethical foundations (1 percent). FMO: Dutch government (51 percent); Dutch banks (42 percent); private companies, trade unions, and individuals (7 percent). COFIDES: Spanish Foreign Trade Institute, Official Credit Institute, and ENISA (61 percent); BBVA; Santander; Sabadell. SIMEST: Ministry for Economic Development (76 percent); IMI; UniCredito; Banco Popolare di Vicenza; Banca Popolare di Sondrio; Banca Intesa. 108 Available at http://www.oecd.org/daf/ca/34803211.pdf, accessed on May 14, 2015. 109 A document that “defines the overall objectives of state ownership, the state ’s role in the corporate governance of SOEs, and how it will implement its ownership policy” (Guideline II.A). 110 Here, as mentioned above, the private sector includes impact investors. 81 Consistent with this approach the Memorandum and Articles of Association (MemArt) for DBN provides for a list of “reserved matters,” which require approval by 75 percent of “total voting power of members”:111  any change to this Memorandum and Articles of Association;  any change to the nature and purpose of the business of the company, its lending policy and/or its general strategy;  any amalgamation, merger, consolidation, reconstitution, restructuring, or similar transaction that results in a change in control of the company;  any termination of operations, or liquidation, of the company;  any alteration of share capital or share repurchase; and  any declaration of dividends. Case studies: DFI ownership Recent evidence suggests that BNDES’s project selection process, financing structure, and governance arrangements have rendered it less effective in achieving its development goals.112 Previously BNDES enjoyed a good reputation as a sound, lean, and effective DFI that provided a substantial portion of total credit in Brazil. Recent weaknesses can to a significant degree be attributed to the government’s 100 percent ownership of BNDES: the absence of nongovernmental shareholders discourages involvement by independent and experienced directors. A somewhat similar assessment can be also made concerning DBSA. Long regarded as a well- managed (and financially sound) institution with high levels of capacity, it also experienced challenges following investment in projects that fell outside its core mandate. Similar to BNDES, had there been some form of nongovernment ownership or control in DBSA, the bank’s board would have been better able to steer the institutions away from risky investments. The South African National Treasury requested an organizational review of DBSA in 2012– 2013, which aimed to “refocus” the bank on its core activities and a governance challenge that arose in the absence of safeguards that could effectively keep the bank from straying outside its mandate. 111 To some extent, discussion of the prerogative of voting members as regards reserved matters becomes mute in as much as on-lenders to the DBN (the World Bank and the African Development Bank) will be providing much of the initial funding for the DBN’s operations. Thus, donors (whether providing equity and loans) will at least initially have considerable influence over the so-called reserved matters. 112 BNDES first fell into disrepute when economists and some politicians started to question the issue of its selection of “national champions” following the 2008–2009 financial crisis, during which it was much praised for its countercyclical role during President Lula ’s second term. It ran into criticism once again recently, following the 2014– 2015 Petrobras corruption scandals. See: http://www.ft.com/cms/s/0/c510368e-968e-11e4-922f-00144feabdc0.html#axzz3a9QEjJqk; http://www.telegraph.co.uk/finance/newsbysector/energy/oilandgas/11292004/How-allegations-of-corruption-at- Latin-Americas-biggest-company-came-to-threaten-Brazils-future.html; and http://www.conectas.org/en/actions/business-and-human-rights/news/25287-bndes-and-human-rights. 82 On the other hand, NAFIN and BancoEstado, fully state-owned banks from Mexico and Chile respectively, provide examples of well-run state-owned banks. One could argue that it is not the type of owner (public or private) but the decision-making process by the owner that makes a difference in how those institutions are run. Ownership and control arrangements to be applied to the DBN Experience with already-existing DFIs in Nigeria (as summarized in chapter 1) provides ample evidence of the need for instituting strong governance procedures. Procedures regarding the process of selection of board members and management and the process of assessing performance, including application of sanctions, where performance falls short, will be crucial. Most SOEs use some kind of “performance contract” between the ownership entity/ministry and the DFI/board. This contract should be negotiated with the board (ideally on an annual basis) and should include key performance indicators that allow performance to be clearly understood and evaluated. As regards implementation of performance contracts, the involvement of nongovernmental owners will encourage compliance with such performance contracting procedures, even if only as a deterrent, as nongovernmental owners are represented on the board and are privy to the decision-making process around performance measurement and assessment. Thus, nongovernmental stakeholders have a crucial role to play in averting dilution of the DBN’s mandate that might result from political interference or “mission creep” on the part of management. C. Board of directors and management Board and management Arrangements as regards the responsibilities of the board of directors113 and management are central to the governance of any institution. Arrangements around the board are important because they involve the establishment of structures and processes with appropriate checks and balances to enable directors to discharge their legal responsibilities and oversee compliance with legislation. Arrangements around management are also important in that they provide for effective monitoring of management by the board, which includes the selection, motivation, and replacement of the management team, while at the same time ensuring management is allowed full operational autonomy to achieve the DFI’s defined objectives. The board is the highest policymaking body of an institution, second only to the Annual General Meeting of Shareholders (AGM). It is the board that exercises strategic oversight and planning; reviews and guides risk policy, annual budgets, and business plans; sets performance objectives; monitors implementation and corporate performance; selects (sets compensation 113 The reference here is to the board responsible for governing the institution and overseeing the executive management. In countries with a unitary board structure, this is the board of directors that normally will include executive and nonexecutive directors. In countries with a two-tier board structure, the reference is to the supervisory board (in contrast to the management board, which is normally composed entirely of executive directors/managers). 83 levels, monitors, and when necessary replaces) key executives; exercises objective and independent judgment on institutional affairs; manages conflicts of interest; ensures integrity of the institution’s accounting and financial reporting systems; and oversees the process of disclosure. In exercising these functions, the board should act in good faith, with due diligence and care, and in the best interests of the institution. While the board decides on strategic matters, management is in charge of running the institution on a day-to-day basis, fulfilling the policies, strategies, and business plans set by the board. Management needs to fulfill these functions in a dynamic, efficient, and professional fashion. Management has to be able to make business decisions rapidly in a fast-moving environment where roles and responsibilities need to be clearly demarcated and well understood. In the DFI sector, responsibilities of the board and management can be broadly divided into three main categories, which reflect the DFI’s objectives as set out in the M&E framework proposed in chapter 3: (i) achieving public policy objectives, (ii) ensuring long-term financial sustainability, and (iii) achieving financial performance objectives. The board appointment process is one of the key factors to ensure the DFI fulfills its objectives. As noted in the study of the governance of state-owned financial institutions cited above:114 “A key ownership task is to appoint the Board of directors. An appropriately constituted, qualified and empowered Board of directors is an essential component of good corporate governance. The Board appointments process is thus a key governance matter. Too often DFI Board member appointments are seen as a form of political patronage, with the result that the Board does not collectively possess the necessary skills and experience to do its job. A better approach is for the shareholder representative to use a structured and transparent Board appointment process that adheres to explicit policies and procedures and seeks to ensure the ability of the Board to exercise its responsibilities in an independent manner, including the use of competence and experience requirements consistent with the strategy and business of the DFI.” The study cites cases in a number of countries where steps have been taken in this direction. In some countries, legal and regulatory texts have been modified to mandate use of skill and experience criteria by the shareholder representative or other parties with appointment powers. In some instances, this involves a requirement for specialized board committees, including an independent board nominations committee to assess the skill requirements of the board and the capacities of existing board members, and to make recommendations to the shareholder David H. Scott, “Strengthening the Governance and Performance of State-Owned Financial Institutions,” Policy 114 Research Working Paper, The World Bank, August 2007. 84 representative. In other cases, specialized units have been created within government to coordinate and advise on the appointment process. The Memorandum and Articles of Association (MemArt) of the DBN provides for the creation of, at the minimum, “the finance, audit committee, remunerations committee, nominations committee and risk management committee,” with “audit and remunerations committees…each be[ing] solely constituted by Independent Directors.” Regarding the appointment of directors, the MemArt stipulates that the board will be constituted by a majority of independent directors who shall be jointly appointed by all the shareholders, and that each director of the company shall fulfill certain minimum qualification criteria for appointment to the board, which will be agreed among the shareholders. The MemArt also stipulates the process of the selection of directors to be undertaken by the nominations committee. CEO appointment process There are usually shortcomings associated with the direct appointment of the CEO by the shareholders, chief among them the inability of the board to hold the CEO accountable. It is thus advisable that the board have a direct role in the appointment and potential dismissal of the CEO. In Norway, Finland, and New Zealand, for example, SOE boards are granted authority to recruit, hire, dismiss, and set the compensation of the CEO. An alternative arrangement (when for political reasons the board cannot have express authority to appoint/dismiss the CEO) is for the board at least to take the lead in the CEO recruitment process. In this model, the board, or a board committee, would lead the search for CEO candidates. This approach has been used in certain instances in Canada, where it is viewed by chairpersons, boards, and CEOs as the most effective arrangement among several alternatives in supporting good corporate governance and a strong accountability relationship between the CEO and the board.115 In exercising its authority to appoint or participate in the appointment of the CEO, the board should adopt clear rules and procedures. Selection criteria should be explicit, and based on professional qualifications and experience. The board might well utilize the services of a professional executive search firm to support its efforts. The MemArt for the DBN provides the board with authority to competitively select and appoint the CEO and other senior executives of the DBN. Case studies: Selection of DFI board and management The boards of DBSA, NAFIN, BNDES, and KfW116 have a majority of government appointed/delegated members. Although DBSA’s board is composed of a majority of nonexecutive, independent directors (drawn from the private sector), they are appointed by the government of South Africa in its capacity as shareholder. Such a board structure provides an extra 115 David H. Scott, “Strengthening the Governance…” op. cit. 116 As it pertains to KfW’s supervisory board. 85 layer of safeguard against governance issues, provided of course that the board members do regard themselves as independent of the government. NAFIN’s board is required by law to be composed of independent members, although the definition of “independent” is rather vaguely crafted in the law. In practice, board members were earlier nominated on the basis of political criteria. In the period of 2000–2006, NAFIN started a process of modernization in order to achieve financial sustainability and improve its corporate governance, which included changes to the legal framework bringing greater transparency to the selection of management and independent directors in the bank’s governing bodies. BNDES’s board is comprised of ten government appointed members,117 with one representative from among BNDES’s employees, plus the CEO (who acts as the board’s vice- chair). Board decisions have been subject to criticism by economists and business people in Brazil due to lack of transparency and clarity, and weak rationale. The board does not directly approve the bank’s operational decisions but rather defines the operational decision-making autonomy of management. KfW’s case is different from the others above: although government appointed, supervisory board members hold limited decision-making powers. In reality, business decisions are taken by the executive board,118 which is composed of independent, highly qualified banking professionals. As a result, the executive board’s credit and investment decisions are seen as protected from outside interference. This independence is also further strengthened by the wholesale-only domestic business model adopted by KfW.119 Other DFIs—such as CDC, PIDG, FMO, IDC, and OPIC—have a majority of nongovernment, independent board members. In the case of CDC, although its sole shareholder is the UK government, all board members are nonexecutive, independent directors who are appointed based on their private sector investment expertise. Table 4.1 below summarizes the ownership structures and board representation of a number of DFIs. A key challenge in board composition is to ensure that the board collectively has the mix of skills, experience, and capacity needed to oversee the business of the DFI in an efficient, professional way. As seen in Table 4.1, KfW, DBSA, PIDG, FMO, and OPIC all require that board members have proven expertise and professional experience in the areas relevant to the DFI’s operations. Two other questions that arise with respect to board composition are whether public servants should serve on the board of a DFI, and whether the CEO should be prohibited from also serving as chairperson of the board. IDC and DBSA do not allow for the chairperson to be the same as the CEO, and in the case of DBSA the chairperson cannot be a government official. 117 These members are appointed by the President of Brazil, and are usually government ministers. 118 KfW, like many DFIs, has a two-tier board system comprised of supervisory and management boards that make decisions within an agreed investment policy. The composition of the supervisory board varies. For example, in FMO and CDC, the supervisory board does not include direct representatives of government ministries, whereas the ministry is represented in KfW. 119 Wholesale-only for domestic lending, carried out by intermediary institutions and not by KfW. 86 Selection of the DBN’s board and management It is critical that the DBN institute and maintain transparent appointment processes for board members and key executives to ensure strong governance. Going forward, it will be important to ensure the independence, professionalism, and experience of the DBN’s board members and key executive appointments. Implementing a selection process that requires candidates to live up to demonstrated professional and technical qualifications will be important in this regard. Table 4.1: Criteria applied to the selection of DFI boards Ownership structure Board representation DBSA Fully gov’t owned Fifteen directors:  Thirteen nonexecutive + CEO and CFO  Twelve independent (majority)  All appointed by gov’t  Limited and staggered terms  Performance-based reappointments  Chair (can’t be a gov’t official) ≠ CEO (nominated by board but appointed by gov’t) KfW Fully gov’t owned Two-tier board system:  Supervisory: thirty-seven government appointed members  Executive: independent, highly qualified banking professionals BNDES Fully gov’t owned Ten government appointed members NAFIN Fully gov’t owned Directors nominated based on government-defined criteria (although law requires “independent” directors): five “civil servants” (minister and deputy minister of finance, minister of economy, minister of energy, and central bank governor), and three private sector directors (who have to be presidents of Mexican business associations) CDC Fully gov’t owned No gov’t representation on board (all nonexecutive directors from the private sector) PIDG Fully gov’t owned Directors are infrastructure and finance professionals with deep understanding of operations FMO Mixed public (51 Two-tiered board: percent) and private (49  Supervisory: seven independent members with specific percent) expertise in relevant areas  Management: three statutory members appointed by supervisory board IDC Fully gov’t owned One exec and thirteen nonexec members (all appointed by the gov’t based on their “ability and experience in business or administration, and their suitability otherwise for appointment as directors”); Chair ≠ CEO OPIC Fully gov’t owned Fifteen members (eight from private sector; seven from gov’t) 87 The DBN’s MemArt institutes a transparent and merit-based process for the selection of the board and management, including provisions for:  a formal, merit-based, transparent process for appointing independent directors to the board as well as performance-based assessment and salaries for DBN management;  a majority of the board to be comprised of independent, highly qualified, professional, and experienced directors who are competitively selected; and  involving all shareholders in the process of selection and appointment of key executives (including the CEO). D. Mandate DFIs benefit from having clearly and tightly defined mandates, tailored to address local conditions and to fill specific financing gaps. Among the factors that help identify the “best scope” of a DFI’s mandate are macroeconomic conditions, the strength of local financial markets and supporting institutions, and the ability of the government to regulate, coordinate, and monitor activities of the DFI sector. DFI mandates need to be tight enough to prevent “mission creep” and flexible enough to give room to adjust the path of the DFI should its mission become less relevant. It is important to have checks and balances in place to avoid “mission creep” while also creating governance arrangements that encourage the DFI to adjust its mandate as it (partly) becomes obsolete. The latter is particularly important to the Public/Private approach to DFIs (as defined in chapter 2)— to make sure that DFIs preserve their “additionality”120 by continually seeking to push the frontier of the production possibilities of the financial system. Given that development challenges are dynamic, the mandates of DFIs should be monitored and reviewed on a regular basis to gauge whether they remain valid and appropriately defined. In order to make mandates enforceable and binding, it is advisable that they are stipulated in either legislation or regulation. Case studies: DFI mandates NAFIN and KfW have focused their operations on the SME sector. In addition to its lending activities, KfW undertakes equity investments in SMEs and start-ups via investments in equity funds. SMEs, together with green finance, are KfW’s favored “policy niches.” DBSA’s mandate provides a good example of the desirable breadth and depth of a DFI mandate. DBSA’s mandate is clearly articulated in the DBSA Act.121 The mandate is further specified in the “Shareholder Compact,” which documents the government’s policy and strategic 120 “Additionality” is, broadly speaking, the extent to which something happens as a result of an intervention that would not have occurred in the absence of such intervention. It will be further addressed in Section II below. 121 The government, through Act no. 13 of 1997, mandated the DBSA to: “(i) Promote economic development and growth, human resources development and institutional capacity in the region; (ii) Support sustainable development projects and programs in the region; and (iii) Focus on infrastructure and leverage the private sector. ” Available at http://www.dbsa.org/EN/About-Us/Pages/About-Us.aspx, accessed on May 14, 2015. 88 objectives, as well as key performance measures and indicators to be attained by DBSA; and in the “Corporate Plan,” which elaborates upon agreed objectives, performance indicators, and quantitative targets.122 The DBSA case is also a good example of the refocusing of an institutional mandate undertaken in response to deviation from the institution’s original mandate. This refocusing of DBSA’s mandate in 2012–2013 exposed trade-offs between mandate specificity and self- sustainability. Before its restructuring, DBSA provided support for initiatives that sometimes were outside or in extension of its mandate. Subsequent to its restructuring DBSA’s funding activities have become more focused.123 Global trends point to the advisability of having a tight mandate to prevent “mission creep” into costly and outside-of-scope ventures. Experience also suggests the need for regular review of the mandate of development banks to avoid situations where DFIs “crowd out” the private sector—thereby avoiding situations where DFIs provide financing in areas where there is relative ease of access to commercial bank funding, or where markets have deepened so that provision of “DFI facilitation” is no longer needed. The comparative advantage of DFIs established according to the Public/Private approach is to provide funding where the private sector is reluctant to venture, building on the DFI’s better understanding of higher-risk markets and in-depth knowledge of the clients in these markets. The DFI provides facilitation to the private sector, thereby encouraging private financial intermediaries to assume credit risks judiciously where they would be unwilling to engage absent the DFI’s intervention. Defining the DBN’s mandate The Memorandum and Articles of Association provides the DBN with quite a narrow development finance mandate. The DBN’s main objective is “[t]o alleviate financing constraints faced by MSMEs and small corporates in Nigeria through providing financing, partial credit guarantees and technical assistance to eligible financial intermediaries on a market-conforming and fully financially sustainable basis.” Furthermore, the DBN is mandated only to on-lend its funds on a wholesale/second-tier basis through “eligible financial intermediaries” in order to fulfill its purpose of “increas[ing] availability and access to finance” to “MSMEs and small corporates in Nigeria.” One aspect of the stated mandate that could be reconsidered relates to provision of technical assistance to financial institutions. This is due to a potential conflict of interest (i.e., the expectation that financing would be provided to institutions that benefited from technical assistance without full adherence to eligibility criteria) and because of the financial implications of such an undertaking. While DBN could facilitate knowledge exchange and dissemination to 122 In practice, the strategic objectives and performance measures and targets generally are proposed by the CEO, the executive management, and the board, and agreed with the Minister as shareholder representative. They form the basis of reporting by the board on performance against specific financial and nonfinancial objectives in the annual report and in other reports. 123 DBSA’s mandate is to provide: (i) project development advisory services, (ii) project finance, (iii) accelerating infrastructure delivery, and (iv) project implementation; while its funding activities address two priorities: (i) support to projects that address backlogs and provide municipal services to poor communities, and (ii) support of infrastructure projects that promote economic growth (in the transportation, telecommunications, and energy sectors). 89 participating financing institutions, it would be preferable that such support be externally funded by professionally qualified external parties (and not included on DBN’s balance sheet). E. Sustainability International experience shows that operationally and financially sustainable DFIs perform better in terms of outreach and developmental impact than their peers. Clearly, a loss-making DFI will not be able to count on its own resources to provide funding for the support it provides, and will thus constantly have to fall back on the government for funding support, rendering it more susceptible to political capture. DFIs experience governance challenges, with respect to managing the trade-offs inherent in conflicting objectives. Like most SOEs, DFIs have the difficult task of reconciling financial performance (operating under commercial criteria) and development objectives. Thus, fulfilling policy mandates can conflict with financial performance objectives and jeopardize long-run financial sustainability.124 From a governance perspective, two crucial aspects to the DFI business model can ensure sustainability: the first is to adopt a business model for lending/funding125 that substantially reduces the DFI’s credit risk by sharing it with the private sector while maintaining the DFI’s focus on expanding credit supply to otherwise excluded market segments; the second is to put in place risk management policies that minimize the risk of nonperforming loans (NPLs), defaults, and bankruptcies. These two aspects are discussed below. Lending model (wholesale v. retail) Cross-country evidence reveals that DFIs perform better when DFI lending takes place through participating financial institutions (PFIs), avoiding direct lending at the retail level—that means using a wholesale/second-tier model. Second tier–only DFIs report lower NPL ratios than retail-only DFIs according to the Global Survey of Development Banks. Indeed, all second tier–only DFIs fall within the “less than 5 percent” bracket, while only 27 percent of retail-only development banks are in this bracket.126 Confining DFIs to second-tier operations contributes to improved performance and leverages the expertise of participating private sector financial institutions (PFIs) in managing credit risk. This contributes both to more cost-effective utilization of scarce public funds and to reducing the risk that DFIs expand their activities into areas, such as managing credit risk, where they would be duplicating rather than supplementing the capacity of private sector intermediaries. Wholesale DFIs rely on credit assessment undertaken according to commercial practices by the private sector and avoid the high costs associated with establishing a network of 124 See discussion on sustainability in chapter 3, and mandate and strategy in chapter 2. 125 For funding models, see discussion in chapter 2 (“Ownership and Funding”). 126 Global Survey of Development Banks conducted by the World Bank. Luna-Martínez, J. and Leonardo Vicente, C. (2012). 90 retail outlets. They leverage the network of branches, ATMs, etc., established by participating private sector financial intermediaries. Case studies: Lending model KfW is exclusively wholesale on its domestic market. It may lend directly (retail) in international development assistance projects, which combine wholesaling of funds via intermediaries for target areas (e.g., SMEs) with direct credit and investments. KfW has been run with a profit,127 and its mission and scope of activities are considered to have evolved along with the economy.128 The intermediary banks that lend with KfW’s funding do so according to criteria set out by KfW, and have access to refinancing their loans at the appropriate KfW window. As a result, KfW has little exposure to direct credit risk on its domestic operations in Germany. KfW uses its quasi-sovereign credit rating to be able to fund priority sectors both at longer maturities and lower cost. This could carry the risk of “crowding out” private investors, but criteria applied by KfW are explicitly designed to reduce such risks: “To prevent crowding out of private investments and to ensure a prudent use of internal and public resources, access to KfW’s promotional programmes is dependent on certain conditions. Within a given programme, applicants have to verify their eligibility (innovative nature of investment projects, certain type of start-up, investments in energy efficiency) and are subject to ex-ante and/or ex-post audits. Furthermore, KfW finances only part of the local investment costs and is thereby a co-lender or a co-investor alongside its on-lending partner.” 129 BNDES offers both retail (direct130) and wholesale (via intermediary banks131) transactions, but all schemes supporting access to finance for SMEs are wholesale. This mixed business model does not add as much value as it could, and may increasingly be “crowding out” the private market.132, 133 According to a World Bank study comparing BNDES to Germany’s KfW, this mixed credit model may be responsible for BNDES’s showing poorer performance on credit quality than KfW. The study bases this conclusion on the fact that BNDES itself bears significant credit and 127 Its consolidated profit was EUR 1.5 billion in 2014 and EUR 1.2 billion in 2013. Source: KfW ’s 2014 Financial Report, available at https://www.kfw.de/PDF/Download-Center/Finanzpublikationen/PDF-Dokumente-Berichte- etc./3_Finanzberichte/Finanzbericht_2014_D-2.pdf. 128 According to one paper, “the ability to rapidly adapt to changing economic conditions rests on KfW ’s business model and its unique position within Germany’s financial system—thus allowing financing programs to be funded via the capital market, with a decreasing share of budget sources in the balance sheet of KfW .” (http://www.ltic.org/IMG/pdf/A_PUBLIER_Etude_KFW_Promoting_LT_May13.pdf) 129 Available at: http://www.ltic.org/IMG/pdf/A_PUBLIER_Etude_KFW_Promoting_LT_May13.pdf (accessed on March 12, 2015). 130 For transactions above BRL 20 million. 131 The so-called “BNDES Automatico.” 132 Lazzarini, Musacchio, Bandeira de Mello, and Marcon, “What do development banks do? Evidence from Brazil, 2002–2009,” Harvard Business School, working paper, 12-047, December 2011. Authors argue inter alia that BNDES provides more capital to firms with political connections (measured as campaign donations to politicians who won an election). 133 Aldo Musacchio and Sergio Lazzarini, Reinventing State Capitalism: Leviathan in Business, Brazil and Beyond . Sao Paulo: Portfolio-Penguin, 2015. 91 investment risks from loans and investments made directly; whereas KfW—using primarily the wholesale-only model—bears little direct credit risk because (i) its counterparty is the financial institution receiving refinancing (whose risk levels are generally lower than those of the final borrowers) and (ii) domestic financial institutions typically diversify their credit risks among the final borrowers. In line with the above views, BNDES is advised to “modernize” its business model, and to become a second-tier,134 wholesale-only institution.135 Initially the business model of FIRA (Mexico) consisted of government intervention in the provision of directed credit at lower-than-market rates and subsidized credit guarantees. This kind of intervention was characterized by high administrative costs, widespread strategic default induced by debt-forgiveness programs, and failure to reach the intended clientele, resulting in significant fiscal drains. Since 2003, FIRA has gradually migrated from direct or first-tier credit to second-tier schemes136 and has shown better financial results.137 FIRA has also moved to using credit guarantees to encourage private sector funding and thereby ensure more efficient use of its capital. Further, as FIRA has moved from first-tier to second-tier lending, a larger number and broader range of financial intermediaries have been attracted as participants in the system (e.g., banks and nonbank financial institutions operating with FIRA).138 Similar to FIRA, NAFIN has since the mid-1990s increasingly moved away from direct lending (retail), becoming mostly a second-tier (wholesale) financial institution. NAFIN’s second-tier lending model is based on granting credit at longer maturities and lower costs than typically available in the Mexican market, passing on the cost advantage resulting from its ability to raise capital at sovereign interest rate levels. Under this strategy, NAFIN promotes SME lending by providing credit to PFIs at below-market interest rates. This source of funding has become less attractive for private sector financial intermediaries following reduction of the spread between sovereign interest rates and interbank rates, but still continues today. This change in demand for its products provided NAFIN with the incentive to create an innovative program (Cadenas Productivas—Productive Chains) that facilitates reverse factoring services to SMEs139 and 134 Second-tier loans are the “repasses” from BNDES to banks, which then on-lend the funds to their clients, subject to BNDES guidelines. Client credit risk is borne by the financial intermediary. 135 These new “directions” would likely require a major reshaping of its governance practices, which would necessarily include: (i) clearer criteria and justifications for BNDES involvement ( “matching operations with motives,” if possible, justifying them based on “quantified cost–benefit analyses”); and (ii) a further boosting of transparency (not only making projects and disbursements publicly available, as it already does, but including sharing the rationale and assessment methodologies for state involvement and debating them as widely as possible). 136 Between 2003 and 2007, second-tier financing increased from 17 percent to 33 percent of Mexico’s development banks’ total portfolio. Source: Financial System Report June 2010, Banco de México. November 2010. Available at http://www.banxico.org.mx/publicaciones-y-discursos/publicaciones/informes- periodicos/reporte-sf/indexpage.html, accessed on June 24, 2015. 137 Its net income has been growing steadily. It was 355 million pesos in 2012; 546 million pesos in 2013; and 928 million pesos in 2014. 138 Source: “Mexico: Financial Sector Assessment Program Update—Technical Note—Strategic Issues in Development Bank Reform.” International Monetary Fund, May 15, 2007. 139 Further discussion of innovative programs undertaken by DFIs can be found in Section III below. 92 significantly increased demand for NAFIN’s funding while also removing reliance on interest rate subsidies. Preserving DBN’s financial sustainability The benefits of the wholesale-only lending model lie in limiting political interference in allocating the DBN’s credit resources, leveraging the underwriting skills of private financial institutions, and reducing the risk of competing with and crowding out the private sector. Nonetheless, as indicated by the NAFIN case study above, this model does require that the DBN is able to provide products that are attractive to the private sector and is innovative as regards the instruments and products it deploys. For example, in times of ample liquidity the on-lending provided by the DBN may not be attractive to the banks, as the interest rates on the wholesale loans will be similar to the rates on alternative sources of wholesale funding. Nonetheless, DBN funding may well still be attractive, as it is of longer tenure than banks are able to provide, and thereby contributes toward reducing the debt service costs experienced by bank borrowers. F. Risk management In addition to adopting a wholesale-only model, another effective tool to strengthening financial sustainability is to put in place strong risk management systems and policies. This will allow the board and management to manage the DFI’s exposure to risk and ensure successful implementation of a DFI’s mandate and strategy. Such risk management systems are designed to identify, measure, monitor, and control the risks faced by DFIs, as well as to determine that they hold adequate capital against those risks. Risk management is a critical component of the overall accountability framework and ultimately an essential determinant of performance.140 It is also an important governance tool with a direct impact on DFI sustainability and financial performance. In principle DFIs face financial risks similar to other financial institutions such as credit risk, market risk, liquidity risk, and operational risk.141 However, in practice one of the advantages of establishing DFIs as second-tier institutions is that their exposure to these risks is substantially reduced or simplified. Lending through participating financial intermediaries rather than directly 140 Pietro Calice, “African DFIs: Unlocking the potential.” African Development Bank Group. Working paper series, no. 174, May 2013, available at: http://www.afdb.org/fileadmin/uploads/afdb/Documents/Publications/Working%20Paper%20174%20- %20African%20Development%20Finance%20Institutions-%20Unlocking%20the%20Potential.pdf, accessed on March 16, 2015. 141 Credit risk refers to the risk of sustaining losses resulting from a decline in the value of assets (including off- balance-sheet items) due to deterioration in the financial condition of the borrower. Market risk—which can be broadly classified into interest rate risk, exchange risk, and stock market risk—describes the risk of loss from fluctuations in the value of assets or liabilities (including off-balance-sheet items) owing to changes in interest rates, exchange rates, stock markets, and various other markets. Liquidity risk is the risk of a mismatch occurring in the periods when funds are used and raised, causing unexpected differences in the flow of funds (cash flow risk). This situation makes securing funds difficult and creates situations in which interest rates on borrowed funds are substantially higher than usual rates. Operational risk can be defined as risks from operational processes, inappropriate actions by executives or regular employees, or those arising from risks of failure due to extrinsic events (portion excluded in calculating the capital adequacy ratio) as well as from the DFI itself (portion included in calculating the capital adequacy ratio). 93 to end-borrowers significantly reduces exposure to credit risk; and relying on wholesale market funding mitigates liquidity and market risks, as second-tier DFIs do not need to establish the infrastructure (branch networks, ATMs, etc.) required to attract retail deposits. Boards of financial intermediaries normally establish specialized risk management committees, which are generally responsible for ensuring the adequacy of the institution’s risk management policies and the management’s adherence to such policies, including specific risk limits. A sound risk management framework includes four major components: (i) risk identification, (ii) risk measurement, (iii) risk mitigation, and (iv) risk monitoring and reporting. DFIs need robust risk management systems, not only so as to ensure their safety and soundness but also for compliance purposes.142 A DFI’s risk management function should be linked to its mandate and governance structure. The board has a central role to play in setting risk appetite across different activities/products and risk management policies as part of its business strategy. Finally, an internal audit department should be established under the direct supervision of the CEO and independent from other operating departments. Its mandate is to conduct inspections to ensure the appropriateness and effectiveness of internal controls—including the DFI’s overall operational compliance, risk management, and performance evaluations—as well as to recommend improvements. Case studies: Risk management When DBSA experienced losses in 2009–2010, it responded by strengthening the management of its strategic, operational, and business risks, enhancing staff capacity, and reviewing the principles of its engagement. The core risk management focus areas identified by DBSA were enterprise-wide risk management (which relates to people, processes and systems, regulatory compliance, legal risk, and business continuity); credit and investment risk management (including development impact and investment risk); and financial risk management (incorporating market and capital management risks). DBSA uses a quantitative assessment of expected loss in setting risk-adjusted pricing.143 KfW’s Audit Committee, which oversees internal and external audit, also has oversight over risk management and internal controls. KfW is subject to inspection by the Federal Audit Office. Even though KfW is not subject to the German Banking Act (Kreditwesengesetz, KWG), it does apply the relevant norms, particularly the minimum requirements for risk management (Pillar II of Basel II) and the German solvency regulation (Core of Pillar I). KfW today already voluntarily applies key rules and standards of the KWG, such as the Minimum Requirements for Risk Management (“MaRisk”144), and complies with the solvency ratios. KfW’s lending reached 142 See discussion on the application of conventional banking standards to DFIs under “Regulation and supervision” below. 143 Probability of Default (PD), which determines the likelihood that the client will not be able to meet its debt repayments based on creditworthiness; Exposure at Default (EAD), which calculates the size of exposure and thus potential loss at the point of default; and Loss Given Default (LGD), which estimates the portion of exposure that is expected not to be recovered at the event of default. 144 Mindestanforderungen an das Risikomanagement. 94 EUR 74 billion at 2014 year end, while its tier-one capital was EUR 20 billion, providing a capital coverage ratio of 14 percent.145 Instituting the DBN’s risk management practices By focusing exclusively on second-tier activities and market-conform pricing, the DBN’s mandate is quite different from previous development finance initiatives in Nigeria. As noted above, the DBN’s MemArt states that its purpose is “to increase availability and access to finance on a wholesale basis only through eligible financial intermediaries” and “[t]o alleviate financing constraints faced by MSMEs and small corporates in Nigeria through providing financing, partial credit guarantees and technical assistance to eligible financial intermediaries on a market- conforming and fully financially sustainable basis.” Risk management is a critical component of the overall accountability framework and ultimately an essential determinant of performance. It is also an important governance tool with a direct impact on the sustainability. Risk management should be an integral part of the performance contract established between the shareholders and the board, used in assessing the performance of DBN’s board. G. Disclosure and transparency State-owned enterprises should observe high standards of transparency146 according to the World Bank’s toolkit for Corporate Governance of State-Owned Enterprises and the OECD Guidelines on Corporate Governance of SOEs,147 which include the following aspects:  efficient internal audit procedures and audit function monitored by the board and the audit committee;  annual independent external audit based on international standards;  same high-quality accounting and auditing standards as listed companies;  disclosure of financial and nonfinancial information according to high-quality internationally recognized standards (IFRS148);  disclosure of material information on key matters (e.g., financial and operating results, remuneration policies, related party transactions, governance structures, and governance policies). Examples of such information include: o clear statement of the institution’s objectives and their fulfillment; o the ownership and voting structure of the institution; o any material risk factors and measures taken to manage such risks; 145 Its tier-one ratio fell from 20.6 percent in 2013 due to implementation of new EU capital requirements (“CRR/CRD IV”) as well as adoption of IFRS. Source: KfW Annual Report, available at https://www.kfw.de/KfW- Group/Service/Download-Center/Financial-Publications-%28D-EN%29/Gesch%C3%A4ftsbericht/, accessed on June 24, 2015. 146 See also discussion on disclosure in chapter 2 (“Public/Private Approach”) and chapter 3 (“M&E”). 147 See references in Section A of this chapter (footnotes 2 and 3). 148 International Financial Reporting Standards (IFRS), issued by the International Accounting Standards Board. 95 o any financial assistance, including guarantees, received from the state and commitments made on behalf of the institution; o any material transactions with related entities (such as the equity investment of one SOE in another SOE, which might be a source of potential abuse). The board is responsible for ensuring publication of an annual report, which is the principal vehicle for most of the disclosures mentioned above. The annual report should also include full financial statements, with a detailed balance sheet, income statement, supporting notes and schedules, and a report on internal controls. Case studies: Disclosure and transparency The DBSA Act149 requires the board to publish an annual report “on the financial status and the achievement of the objectives” of DBSA. It also requires DBSA to be audited annually by persons registered under the Public Accountants’ and Auditors’ Act. In addition, DBSA is subject to the accounting and disclosure requirements of the Companies Act. The Public Finance Management Act requires the board to submit an annual report and audited financial statements based on generally accepted accounting practices, to be presented before Parliament. DBSA is required to adhere to South Africa’s Generally Accepted Accounting Practices, but DBSA has gone further and has adopted the IFRS. DBSA’s actual disclosure practices are extensive, including an annual report incorporating the audited Annual Financial Statements and the Director’s Report; an accompanying Activities Report describing the development impact of the bank’s activities during the year; an Information Manual issued in accordance with the Promotion of Access to Information Act; and six-monthly results incorporating a strategic overview. All reports, as well as extensive additional information, are available on DBSA’s website.150 In addition, DBSA submits financial and operational information to the National Treasury on a quarterly basis. Many DFIs do not have explicit transparency and disclosure policies in relation to financial and nonfinancial information.151 It is suggested that making available basic data in the following areas would increase transparency and benefit the DFI sector and its direct beneficiaries:152 1. the extent to which DFIs receive technical assistance; 2. the interface between DFIs and official development assistance (ODA153); 149 Act no. 13 of 1997. 150 www.dbsa.org. 151 Kingombe, Massa, and te Velde, op. cit. Available at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/67635/comparing-DFIs.pdf, accessed on March 16, 2015. 152 te Velde and Warner (2007), “The Use of Subsidies by Development Finance Institutions in the Infrastructure Sector.” Working Paper 283. London: ODI. 153 More on ODA can be found on the OECD website: http://www.oecd.org/dac/stats/officialdevelopmentassistancedefinitionandcoverage.htm. 96 3. the terms and conditions of operations;154 and 4. the overall size and sector allocation of DFI activities. DFI annual reports reveal very little on transparency and disclosure policies.155 The DFIs referenced by Kingombe et al. are mostly from European countries, but include several multilateral institutions as well (e.g., IFC, EBRD, AFDB, ADB, etc.). CDC discloses as much as publicly listed companies do. OPIC has announced new transparency measures in order to make more information available to the public about the projects it supports, and to encourage a new level of public involvement in the institution’s development. FMO follows the Global Reporting Initiative’s (GRI) sustainability reporting guidelines156 and recently subscribed to the Dutch transparency benchmark. Lastly, IFC has an Access to Information Policy (AIP) in place,157 designed to make it possible for interested parties to understand better and engage in informed discussions about IFC’s business activities and the overall development impact of its activities. Overall, there seems to be a difference between domestic and multilateral DFIs when it comes to disclosure and transparency policies. For example, while IFC and EBRD have disclosure and transparency policies, DEG/KfW refers to the German Banking Secrecy Law, which it is legally bound to uphold,158 although it is also obliged to observe IFRS rules on transparency and disclosure.159 Applying disclosure and transparency standards to the DBN Instituting a public accountability framework that records performance data and financial outcomes is essential to effective oversight and transparency. The DNB’s MemArt requires keeping of accounting and other records and distribution of balance sheets (as required by law) as well as the recording of minutes regarding the board workings, but is otherwise silent on the more extensive transparency and disclosure requirements outlined above. This could be an area where the presence of donors, both as funders and third-party investors in the DBN, will have a role to play in making sure the DBN lives up to the principles as outlined in the OECD’s principles and the World Bank’s toolkit. 154 Many DFIs do not reveal this information due to possibly warranted commercial confidentiality concerns. In May 2015, President Dilma Rousseff of Brazil vetoed legislation that would have put an end to BNDES operationsal secrecy, citing bank and business secrecy concerns and the competitiveness of Brazilian firms in the global market for goods and services. 155 See Kingombe, Massa, and te Velde, op. cit. Available at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/67635/comparing-DFIs.pdf, accessed on March 16, 2015. 156 Available at https://www.globalreporting.org/information/sustainability-reporting/Pages/gri-standards.aspx. 157 Available at http://www.ifc.org/wps/wcm/connect/98d8ae004997936f9b7bffb2b4b33c15/IFCPolicyDisclosureInformation.pdf? MOD=AJPERES. 158 Kingombe, Massa, and te Velde, op. cit. 159 For a discussion on the commercial disclosure approach by the Chinese Development Bank, and the complete disclosure of evaluation results by KfW, see chapter 3. 97 H. Regulation and supervision Independent and effective regulation and supervision of DFIs is a basic condition for sound governance and for ensuring good performance and financial sustainability. Studies of comparator countries indicate that, while technical oversight may stay with the Ministry of Finance, prudential regulation and supervision is best exercised by the Central Bank with a view to enforcing essentially the same regulatory framework for DFIs as is applied to commercial banks. The aim is to safeguard solvency/liquidity and assure financial sustainability. While the regulatory framework applied by the Central Bank should follow the same principles as applied to commercial banks,160 adjustments are appropriate to take into account the DFI’s risk profile. For example, the risks assumed by a wholesale-only DFI are reduced on both sides of its balance sheet, as it neither takes on credit risk directly nor solicits deposits from the public. The regulatory framework should include licensing and prudential requirements as regards capital adequacy, loan provisioning, and recognition of loan losses, in addition to the strong transparency and accountability standards outlined in the previous section. Other more- specific standards include oversight regarding prudent lending criteria; efficient use of resources; professional risk management; pricing of products; liquidity planning; internal/external audits; and finally a lean organization structure. Further, only banks and other financial institutions living up to a full set of eligibility requirements, including the CBN’s prudential requirements, should be eligible to becoming participating financial institutions (PFIs) that receive financing from the new DFI.161 DFIs are unique financial institutions in that they perform functions which are not identical to those of banks, while being systemically part of a country’s financial system, thus requiring prudential safeguards on liquidity and solvency. This “uniqueness” calls for a nuanced approach to their regulation and supervision: on the one hand, they need tailored oversight as regards their development profiles as set by the government—oversight which is normally best exercised by the Ministry of Finance—while, on the other hand, they should observe 160 Observance of the similar commercial principles applicable to private sector banks also constrains DFIs in “crowding out” the private sector. In order to remain compliant with capital adequacy and other Central Bank prudential requirements, DFIs are required to accumulate capital, thus limiting their ability to pass on the advantages associated with their cheap cost of funding to their borrowers. Passing on such subsidies to the borrowers without paying heed to the sustainability of the DFI ’s capital base is one of factors that cause DFIs to “crowd out” private sector banks. 161 The following should be the minimum eligibility criteria for the so-called participating financial institutions or PFIs: (i) the PFI must be duly licensed and at least two years in operation; (ii) PFI’s owners and board of directors should be fit and proper; (iii) the PFI must have qualified and experienced management, adequate organization, and institutional capacity for its specific risk profile; (iv) the PFI must be in good standing with its supervisory authority and remain in compliance at all times; (v) the PFI must have well-defined policies and written procedures for management of all types of financial risks (liquidity, credit, currency, interest rate and market risk, as well as risks associated with balance sheet and income statement structures); (vi) the PFI must maintain capital adequacy as prescribed by prudential regulations; (vii) the PFI must have adequate liquidity; (viii) the PFI must have positive profitability and an acceptable risk profile; (ix) the PFI must have adequate portfolio quality; (x) the PFI must have adequate internal audits and controls for its specific risk profile; (xi) the PFI must have adequate management information systems; and (xii) the PFI must demonstrate commitment to serving the MSME sector and have in place satisfactory MSME loan approval processes and risk management procedures. 98 basic principles as regards sound financial and risk management as outlined above. Such financial oversight is best conducted by the Central Bank as the institution responsible for prudential regulation and supervision of the financial system. Nonetheless, it is important that the prudential framework administered by the Central Bank be adapted to the risk profile of the particular DFI, given that DFIs are constrained by their mandates and take considerably less risk than private banks when functioning as wholesale institutions. Lastly, it is advisable that DFIs should be subject to the same general regulations which apply to private banks such as those contained in procurement, contract, and labor laws. Case studies: Regulation and supervision The primary regulator of BNDES is the Central Bank. BNDES complies with commercial bank regulations and is subject to oversight by the Brazilian Minister of Development, Industry, and Foreign Trade (from the perspective of industrial policy and socioeconomic development issues). By and large there is agreement that BNDES’s observance of the Central Bank prudential standards has not hampered its ability to fulfill its developmental mandate: the volume of disbursements in the period of 1998–2010 grew steadily year on year, reaching record highs, and the bank’s outreach increased significantly, especially after 2007.162 One criticism163 directed at the recent application of tighter prudential standards to BNDES is that these standards may have encouraged BNDES to become more risk-averse, with an increased focus on investment-grade borrowers.164 However, both the Central Bank and BNDES emphasize that BNDES must be supervised exactly like any other commercial bank to ensure a level playing field. In South Africa, regulation and supervision of the DBSA rests with the government as opposed to the banking regulator. While DBSA is ultimately subject to review by Parliament, as the only DFI under the Ministry of Finance, it is in the unique position of referring to one entity (National Treasury) that fulfills three roles: financial oversight (as for all other SOEs), policy- setting, and executive authority. In practice, the Treasury is the regulator and appoints members of DBSA board. Over time differences within the Treasury about the role of the DBSA have led to shifting requirements and emphasis. A positive feature, however, is that the Treasury has provided DBSA with full operational autonomy. Although DBSA is not subject to regulation or supervision by the South African Reserve Bank (SARB), thus not required to comply with the Banks Act or prudential regulations, it is subject to regulatory oversight by the Financial Services Board (FSB)165 (which supervises nonbank financial institutions) and the Financial Intelligence 162 Rogerio Sobreira and Norberto Martins, “Os Acordos de Basileia e bancos de desenvolvimento no Brasil: uma avaliação do BNDES e do BNB”—Rio de Janeiro 45(2):349–76. 163 See Rogerio Sobreira and Norberto Martins op. cit. 164 According to the above authors (Sobreira and Martins, op. cit.), “the tendency towards increasing the share of investment grade borrowers in the bank’s portfolio seems to be linked to the need to keep its risk-weighted assets under control so as to avoid substantive expansions in the required regulatory capital, even though that did not seem to be contrary to the functions assigned to [BNDES] ” (free translation). Investment-grade operations grew from 58 percent of total operations in 2001 to 90 percent after 2007. 165 The FSB oversees the nonbanking financial services industry, which includes retirement funds, short-term and long-term insurance, companies, funeral insurance, schemes, collective investment schemes (unit trusts and stock market), and financial advisors and brokers, including the Johannesburg Stock Exchange. Its oversight criteria are fair treatment of consumers of financial services and products, financial soundness of FIs, systemic stability of financial 99 Centre (for anti-money laundering matters). It is also mandated to “seek” to adhere to international corporate best practices. In this regard, it monitors and manages its capital adequacy within certain regulatory constraints on leverage as well as a self-imposed capital adequacy framework approved by the board. While the current arrangements may work for DBSA, they are undoubtedly a second-best solution. As a banking institution, it would be more appropriate that DBSA be under the supervision of SARB. Not only would this provide enhanced credibility, it would also be consistent with policies being developed by the South African government under the “twin peaks” reform program that envisages the creation of a prudential regulator housed in SARB with FSB being transformed into a dedicated market conduct regulator.166 In Germany, KfW has recently moved from being supervised by the MOF to being regulated and supervised by the German Federal Financial Supervisory Authority (“BaFin”). This followed a legislative change in 2013 to allow certain provisions of commercial banking regulations (mainly the German Banking Act) to be applied to KfW by analogy. The move is seen as reinforcement of previous initiatives designed to modernize and professionalize KfW. The recent regulatory change resulted in technical/prudential supervision of compliance with the bank supervisory laws and regulations being exercised by BaFin in cooperation with the German Central Bank (Deutsche Bundesbank). This has no bearing on the legal supervision, which continues to be exercised by the MOF167 in cooperation with the Federal Ministry for Economics and Technology (“BMWi”). Regulation and supervision of the DBN International experience suggests that it is important that DFIs be regulated and supervised by the banking regulator, which in the case of Nigeria is the CBN. In 2015, the CBN developed Regulatory and Supervisory Guidelines for Guidelines Framework for Development Finance Institutions, bringing development financial institutions under its full regulatory and supervisory remit. The CBN’s guidelines outline the supervisory principles that it intends to adopt and apply for all Nigerian DFIs (see summary in Box 4.2 below). The supervisory standards to be applied to DFIs (including the DBN) are very similar to those applied to commercial banks. As mentioned above in relation to international practices, in implementing these standards it will be important that due care is taken to take into consideration the wholesale nature of the business of DBN. According to its mandate the DBN’s risk exposure will be considerably lower than that of commercial banks and other existing DFIs: on the asset side, the DBN will not undertake any direct lending and will only be performing second-tier, services industries, and integrity of financial markets and institutions. Source: http://www.fsb.co.za/aboutUs/Pages/mission.aspx. 166 The prudential regulator’s objective will be to maintain and enhance the safety and soundness of regulated financial institutions; while the market regulator will be tasked with protecting consumers of financial services and promoting confidence in the South African financial system. 167 In this capacity, the Federal Ministry of Finance is empowered to adopt all measures necessary to ensure that KfW operates its business activities in accordance with its bylaws and other rules and regulations. 100 wholesale functions; and on the liability side, the risk exposure of the DBN will be considerably less than that of commercial banks because it will not be soliciting deposits from the public. The framework proposed by the CBN distinguishes between wholesale and retail DFIs, but applies higher minimum capital requirements to wholesale DFIs (i.e., the DBN). Wholesale DFIs provide funds to participating financial institutions (PFIs) for on-lending to final borrowers, while retail DFIs lend directly to end-borrowers. The minimum capital requirements to be applied to the DBN has been set at NGN 100 billion payable over a maximum period of four years, out of which a minimum of NGN 20 billion is required as a condition for receiving approval of the license in principle. This minimum capital requirement is ten times the NGN 10 billion minimum capital requirement applied to retail DFIs, and four times the minimum capital of the NGN 25 billion applied to commercial banks. Given the lower level of risk assumed by the DBN as a wholesale DFI, it would seem advisable to consider adjusting the minimum capital requirement applicable to the DBN so it better reflects the relatively low level of credit risk that it is envisaged to assume as an “arm’s-length” lender that relies on the credit risk management skills of prequalified participating private sector financial institutions. 101 Box 4.2: Key provisions of CBN’s set of draft regulatory guidelines for DFIs  DFIs’ objectives are: o to fund MSMEs for economic development; o to foster growth in sustainable businesses; o to create jobs; and o to reduce poverty and improve quality of lives.  CBN powers and duties in relation to DFIs are: o to license the DFIs; o to determine the DFIs’ minimum capital requirements; o to approve the appointment of board and top management of DFIs; and o to regulate and supervise the DFIs’ business operation.  Corporate governance requirements for board and senior management should include: o qualifications for board members and senior management; o functions of the board and senior management; o duties of directors and senior management.  DFIs’ sources of funds should consist of any one or more of the following: o paid-up share capital and reserves (equity); o preference shares; o long-term loans from international financial institutions (IFIs) subject to prior CBN approval of the draft Memorandum of Understanding (MOU); o debentures and/or bonds; o loans from national and supra-national governments and other bodies; o funds from development partners; o gifts, grants, and donations.  DFIs’ rendition of statutory returns should include: o periodic returns; o domestic report; o audit opinion on going-concern status; o publication of audit accounts; and o report on internal control.  DFIs’ prudential requirements should include rules on: o capital adequacy (“A DFI shall maintain at all times a minimum ratio of Tier I capital to total assets (Tier I leverage ratio) of not less than 5 percent”); o liquidity requirements; o credit extension and collateral policies; o management of market risk; and o investment policies and statutory reserves.  Other regulatory approvals include appointment and duties of external audits; branch expansion, relocation, and closure; and changes in ownership structure.  Rules on examination, reporting, and off-site surveillance by CBN.  Rules on administrative sanctions and actions. 102 II. Making sure DFIs achieve development impact As seen in chapter 3, the focus of the Public/Private approach to development finance institutions is on expanding the scope of the “financing frontier” and providing finance to businesses that would not otherwise receive financing from banks. Were DFIs not providing such services, then their efforts would not be “additional” and they would in practice be “crowding out” private sector banks using taxpayers’ money.168 Establishing a clear “yes or no” answer on whether a DFI has created development additionality is extremely difficult due to the lack of a counterfactual. Referring to DFI- supported infrastructure projects, the study cited above169 found very few cases of clear crowding out, where crowding out is defined as DFIs investing in the place of private financiers and thus prejudicing the development of a healthy private sector market for infrastructure financing. However, many cases were found of projects which the evaluators believed could have gone ahead without DFI involvement. Oftentimes a project would probably have gone ahead without DFI involvement (sometimes referred to as “non-additionality”), but it was not entirely clear that the DFI had crowded out other private investors (i.e., invested in their place when they’d wished to make the investment themselves). In the case of the DBN, additionality will predominantly be achieved through improving financial access conditions—for example, by providing longer maturities and thereby reducing the cash flow burden for the MSME borrower or by strengthening the availability of finance due to risk-sharing arrangements. Again here, while ensuring additionality is central to the Public/Private approach, monitoring its achievement will be difficult. Even though every effort should be made to monitor the additionality of the DBN’s activities, and safeguards are applied that encourage PFIs to use the DBN’s funding and risk-sharing instruments to expand the frontier of financing opportunities, it will be difficult to ascertain whether banks will be adding new borrowers due to the support provided by the DBN or whether the DBN’s support is provided to borrowers whom the banks would anyway have been willing to service. Also the tendency will be that participating financial intermediaries and MSMEs will continue to want to take advantage of the facilitation provided by the DBN on conducive terms, even after the private sector would be prepared to provide financing and there is no longer need for the DBN’s support. These considerations highlight the importance of the governance of development banks. Strong commitment to implementing the mandate of the DBN will be crucial in ensuring that the new development bank does not use its instruments—with the longer maturities and risk-sharing 168 Stephen Spratt and Lily Collins, “DFIs and Infrastructure: A Systematic Review of Evidence for Development Additionality,” Private Infrastructure Development Group, January 2012. This paper establishes four different forms of “additionality” created by DFIs: (1) financial, (2) design, (3) policy, and (4) demonstration. Available at https://www.ids.ac.uk/files/dmfile/2012IDS-PIDGSystematicReviewFinal.pdf, accessed on March 16, 2015. 169 Stephen Spratt and Lily Collins, op cit. 103 facilities otherwise not available on the market—to crowd out private financing.170 Rather, the DBN should strive to catalyze private financing in areas where it is slow to come forward. Thus, while it has been established that Nigeria suffers from an SME financing gap, constant vigilance by the board and management of DBN will be essential in ensuring that the resources made available to the DBN are used as effectively as possible in addressing the identified market failure. While outside the mandate of the DBN, it will be important that both financial intermediaries and MSMEs receive technical assistance. As outlined in chapter 1, constraints to MSME financing permeate the demand side as much as the supply side in Nigeria. With very few exceptions, banks in Nigeria are not well-acquainted with the operations of MSMEs or their financing needs, and one component of the World Bank’s efforts in supporting the establishment of the DBN has been to set aside funding to be administered by the project implementation unit to the advantage of capacity-building among banks and microfinance banks. Also as seen from the demand side; that is, from the perspective of Nigerian banks, there is a dearth of “bankable” projects. Thus, donor-funded programs that support MSMEs in developing business plans and provide training to business managers will be important in expanding the scope of businesses potentially eligible for finance by banks or microfinance banks, thereby supporting the DBN as a “first-mover.” While it would be inappropriate for the DBN to both undertake such activities and provide funding (i.e., to facilitate preparation of projects that it intends to fund or support through credit guarantees), donor funding for this purpose is already available 171 and is more likely to become available, once the DBN is fully operational and has established a credible track record. Given a credible implementation track record, there is every reason to believe that the Public/Private approach adopted by the DBN will give rise to a virtuous cycle of increased private sector funding supported by further donor contributions to ease the DBN ’s cost of funding. It is likely that if the DBN adopts the governance arrangements outlined in this chapter and establishes a credible implementation track record, this will both stimulate greater involvement in MSME financing by private financial intermediaries—as intended under the Public/Private approach—and encourage further donor contributions to the DBN in the form of equity investment and on-lending. This will support the penetration of the DBN’s activities by delaying the time at which the DBN is obliged to access the capital market for funding. Achieving additionality in Nigeria For the DBN to be effective and sustainable, it is critical that it adheres to and maintains its founding principles. The safeguards that have been established around the DBN, which are to a large extent included in the DBN’s Memorandum and Articles of Association, live up to the recommendations outlined in previous sections of this chapter. They are designed to ensure that the DBN’s activities will create additionality; that it will act as a catalyst for private sector investments; and that it pilots projects on the “financing frontier” where the private sector is 170 This is also a point emphasized by the European Commission in its discussion of potential crowding out caused by DFIs. Green Paper on Long-Term Financing of the European Economy. March 25, 2013. 171 A Business Innovation and Growth platform is being launched in early 2016 under the World Bank ’s Growth & Employment Project (GEM) to provide training programs and consulting services to about 15,000 MSMEs. 104 reluctant to engage. It will fall on the board of directors and on management—selected and appointed according to the processes and criteria discussed above—to interpret these safeguards, effectively give them teeth, and ensure that the DBN is successful in fulfilling its mission. Private sector participation, as seen throughout this study, will also be crucial to this end. An enhanced and sustained focus on governance will be important in ensuring the effectiveness of development finance interventions. The role of governance extends to choosing instruments and modalities of intervention that effectively promote the mandate of the DBN. In taking into consideration the lessons learned from chapter 2, it will be important that the design of DFI interventions complements and promotes financial intermediation provided by the private sector, rather than replacing it, but also takes into account factors such as sustainability and transparency. Finally, as outlined in chapter 3, a crucial link in the governance process will be the accountability framework to be established in measuring the performance of the DBN. The DBN should be evaluated according to a set of performance criteria, and its management and board should be held accountable when ex-post-evaluations indicate underperformance. The performance criteria, to which DBN’s board of directors and management will be held responsible, should be aligned with the business strategy and mandate of the DBN and include high-level indicators reflecting development impact, sustainability/efficiency, and organizational capacity. The governance issues around performance management include holding the board of the DBN accountable for setting and achieving its goals. Experience shows that most reforming state-owned enterprises use some kind of “performance contract” between the ownership entity/ministry and the DFI/board. This contract should be negotiated with the board (ideally on an annual basis) and should include key performance indicators that allow performance to be clearly understood and evaluated. 105 Bibliography Acemoglu, D., Johnson, S., and Robinson, J. (2001). The Colonial Origins of Comparative Development: An Empirical Investigation. American Economic Review, 91(5), pp.1369–1401. Ashenfelter, O. (1978). Estimating the Effect of Training Programs on Earnings. Review of Economics and Statistics, 60(1), 47–57. Ashenfelter, O., and Card, D. (1985). Using the Longitudinal Structure of Earnings to Estimate the Effect of Training Programs. Review of Economics and Statistics, 67(4), 648–660. Banco de México. (2010). Financial System Report. Available at: http://www.banxico.org.mx/publicaciones-y-discursos/publicaciones/informes- periodicos/reporte-sf/indexpage.html [Accessed 24 June 2015]. Bator, F. (1958). The Anatomy of Market Failure. The Quarterly Journal of Economics, 72(3), p. 351. BDC Strategy & Planning, GY Consulting (2009). International Benchmark Study on Development Institutions. Available at: http://s3.amazonaws.com/zanran_storage/www.bdc.ca/ContentPages/111679891.pdf [Accessed 13 February 2015]. Beck, T., Demirguc-Kunt, A., and Martinez Peria, M. S. (2005). Reaching Out: Access to and Use of Banking Services across Countries, mimeo, World Bank. Beck, T., and de La Torre, A. (2007). “The basic analytics of access to financial services.” Financial Markets, Institutions & Instruments, 16(2), 79–117. Berg G., and Fuchs, M. (2013). Bank Financing of SMEs in Five Sub-Saharan African Countries: The Role of Competition, Innovation, and the Government. World Bank Policy Research Working Paper, 2013 (6563). Berg, G., Fuchs, M., Iacovone, L., Jaeggi, T., Lovegrove, A., and Villegas Sanchez, C. (2012). Financing Small- and Medium-Sized Enterprises in Nigeria. The World Bank, Africa Finance and Private Sector Development Department. Birdsall, N., de la Torre, A., and Caicedo, F. (2010). The Washington Consensus: Assessing a Damaged Brand. Washington, DC: The World Bank. Boehmer, Hans-Martin. “The Importance of Evaluation in Development Policy,” in The Future of Evaluation: Global Trends, New Challenges, Shared Perspectives, edited by Wolfgang Meyer and Reinhard Stockmann. 2016. Borio, C., Furfine, C., and Lowe, P. (2001). Procyclicality of the Financial System and Financial Stability: Issues and Policy Options. BIS Papers chapters, 1, 1–57. 106 Calice, P. (2013). African DFIs: Unlocking the Potential. African Development Bank Group. Working paper series, no. 174. Available at: http://www.afdb.org/fileadmin/uploads/afdb/Documents/Publications/Working%20Paper%2017 4%20-%20African%20Development%20Finance%20Institutions- %20Unlocking%20the%20Potential.pdf [Accessed 16 March 2015]. Conference Board of Canada (2010). Lessons from the Recession and Financial Crisis. Lesson 2: Public Sector Financial Institutions Prove their Worth. Cowan, K., Drexler, A., and Yanez, A. (2012). The Effect of Credit Guarantees on Credit Availability and Delinquency Rates. Datta-Chaudhuri, M. (1990). “Market Failure and Government Failure.” Journal of Economic Perspectives, 4(3), pp. 25–39. DBSA (2010). The Evolution of the DBSA 2010. [online] Issuu. Available at: http://issuu.com/developmentbankofsouthernafric/docs/the_evolution_of_the_dbsa_2010/3?e=0 [Accessed 16 January 2015]. de la Torre, A. (2002). Reform Development Banks: A Framework. Public Sector Banks and Privatization, World Bank Workshop. Washington, DC: The World Bank. de la Torre, A., Gozzi, J., and Schmukler, S. (2007). Innovative Experiences in Access to Finance. Washington, DC: The World Bank. de la Torre, A., and Schmukler, S. (2007). Emerging Capital Markets and Globalization. Stanford University Press and World Bank. Washington, DC. Denzer-Speck, D., and Lob, H. (2013). “Promoting Long-Term Investments under Changing Regulatory Framework Conditions. The Case of Germany’s KfW Bankengruppe.” Available at: http://www.ltic.org/IMG/pdf/A_PUBLIER_Etude_KFW_Promoting_LT_May13.pdf [Accessed 12 March 2015]. Development Committee (2015). “From Billions to Trillions: Transforming Development Finance. Post-2015 Financing for Development: Multilateral Development Finance,” prepared jointly by the African Development Bank, the Asian Development Bank, the European Bank for Reconstruction and Development, the European Investment Bank, the Inter-American Development Bank, the International Monetary Fund, and the World Bank Group for the April 18, 2015 Development Committee meeting. Available at: http://siteresources.worldbank.org/DEVCOMMINT/Documentation/23659446/DC2015- 0002%28E%29FinancingforDevelopment.pdf [Accessed at April 14, 2015]. Diamond, W. (1957). Development Banks. [Washington]: Economic Development Institute, International Bank for Reconstruction and Development. 107 Downs, E. (2011). Inside China, Inc.: China Development Bank’s Cross-Border Energy Deals. John L. Thornton China Center at Brookings, Monograph Series, Number 3. Available at: http://www.brookings.edu/~/media/research/files/papers/2011/3/21-china-energy- downs/0321_china_energy_downs.pdf [Accessed 25 February 2015]. Etim, K., and Okuribido, D. (2012). Nigeria: Lessons from Shell. International Financial Law Review (IFLR). Available at: http://www.iflr.com/Article/3127624/Nigeria-Lessons-from-Shell.html [Accessed 12 March 2015]. European Commission (2013). Green Paper, Long-Term Financing of the European Economy /* COM/2013/0150 final */. Available at: http://eur-lex.europa.eu/legal- content/EN/TXT/?uri=CELEX:52013DC0150 [Accessed 21 June 2015]. Ferraz, J. (2012). Central Banks, Financial Systems and Economic Development: The Experience of Development Banks. Presentation at 2012 Money and Banking Conference in Buenos Aires, October 1, 2012. Ferraz, J. (2014). Financing Development: The Strategic Role of Development Banks (and the Case of BNDES). Presentation at IPD/JICA Task Force on Industrial Policy and Transformation Meeting in Jordan, June 5–6, 2014. Ferraz, J., Alem A., and Madeira, R. (2013). Sketching Profiles of Development Banks. Presentation at Seminar Global Economic Governance in Madrid, March 11, 2013. Francisco, M., Mascaro, Y., Mendoza, J. C., and Yaron, J. (2008). Measuring the Performance and Achievement of Social Objectives of Development Finance Institutions. World Bank Policy Research Working Paper 4506. Fuchs, M., Iacovone, L., Jaeggi, T., Napier, M., Pearson, R., Pellegrini, G., and Sanchez, G. V. (2011). Financing Small and Medium Enterprises in the Republic of South Africa. The World Bank, Africa Finance and Private Sector Development Department. Gutierrez, E., Rudolph, H., Homa, T., and Beneit, E. B. (2011). Development Banks: Role and Mechanisms to Increase Their Efficiency. Policy Research Working Paper Series, No 5729, World Bank, Washington, DC. Hausmann, R., Klinger, B., and Wagner, R. (2008). Doing Growth Diagnostics in Practice: A ‘Mindbook.’ Center for International Development working paper, (17). Hausmann, R., Rodrik, D., and Velasco, A. (2008). Growth Diagnostics. The Washington Consensus Reconsidered: Towards a New Global Governance. Oxford University Press: New York, 324–355. 108 Homa, H. (2011). Managing Public Banks: Case for a Holding Structure. BDC International Consulting. Presentation available at: http://www.shcp.gob.mx/ApartadosHaciendaParaTodos/banca_desarrollo/pdf/presentacion_theo dore_homa.pdf [Accessed April 8, 2015]. Independent Evaluation Group (IEG) World Bank Group (2014). The Big Business of Small Enterprises Evaluation of the World Bank Group Experience with Targeted Support to Small and Medium-Size Enterprises, 2006–12. [online] Available at: http://ieg.worldbank.org/Data/reports/sme_eval1_0.pdf [Accessed 9 February 2015]. IFC (2012). Access to Information Policy. Available at: http://www.ifc.org/wps/wcm/connect/98d8ae004997936f9b7bffb2b4b33c15/IFCPolicyDisclosur eInformation.pdf?MOD=AJPERES [Accessed 12 March 2015]. IFC (2013). Closing the Credit Gap for Formal and Informal Micro, Small, and Medium Enterprises. Available at: http://www.ifc.org/wps/wcm/connect/4d6e6400416896c09494b79e78015671/Closing+the+Cred it+Gap+Report-FinalLatest.pdf?MOD=AJPERES [Accessed 21 January 2015]. International Monetary Fund (2007). Mexico: Financial Sector Assessment Program Update— Technical Note—Strategic Issues in Development Bank Reform. Isern, J., Agbakoba, A., Flaming, M., Mantilla, J., Pellegrini, G., and Tarazi, M. (2009). Access to Finance in Nigeria: Microfinance, Branchless Banking, and SME Finance. CGAP and World Bank. Kaminsky, G. L., Reinhart, C. M., and Végh, C. A. (2005). When It Rains, It Pours: Procyclical Capital Flows and Macroeconomic Policies. In NBER Macroeconomics Annual 2004, Volume 19 (pp. 11–82). MIT Press. Kaminsky, G. L., and Schmukler, S. (2003). Short-Run Pain, Long-Run Gain: The Effects of Financial Liberalization (No. w9787). National Bureau of Economic Research. KfW (2014). Financial Report. Available at: https://www.kfw.de/PDF/Download- Center/Finanzpublikationen/PDF-Dokumente-Berichte- etc./3_Finanzberichte/Finanzbericht_2014_D-2.pdf [Accessed 24 June 2015]. KfW. Annual Reports. Available at https://www.kfw.de/KfW-Group/Service/Download- Center/Financial-Publications-%28D-EN%29/Gesch%C3%A4ftsbericht/ [Accessed 24 June 2015]. Kingombe, C., Massa, I., and te Velde D. W. Comparing DFIs—Literature Review. January 2011. Overseas Development Institute. Available at: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/67635/comparing- DFIs.pdf [Accessed on 16 March 2015]. 109 Lazzarini, S., Musacchio, A., Bandeira-de-Mello, R., and Marcon, R. (2015). What Do State- Owned Development Banks Do? Evidence from BNDES, 2002–09. World Development, 66, pp. 237–253. Lelarge, C., Sraer, D., and Thesmar, D. (2010). “Credit Constraint and Entrepreneurship: Evidence from a French Loan Guarantee Program,” in International Differences in Entrepreneurship, edited by J. Lerner and A. Schoar. University of Chicago Press. Lemma, A. (2015). Development Impact of DFIs—What Are Their Impacts and How Are They Measured? Overseas Development Institute. Levine, R. (2005). “Finance and Growth: Theory and Evidence,” in Handbook of Economic Growth, edited by P. Aghion and S. Durlauf. Elsevier: Amsterdam. Lieberman, M. B., and Montgomery, D. B. (1990). To Pioneer or Follow? Strategy of Entry Order. Graduate School of Business, Stanford University. Love, I., Martinez Peria, M. S., and Singh, S. (2013) Collateral Registries for Movable Assets: Does Their Introduction Spur Firms’ Access to Bank Finance? World Bank Policy Research Working Paper 6477, June. Luna-Martínez, J., and Leonardo Vicente, C. (2012). Global Survey of Development Banks. Policy Research Working Paper. World Bank, Financial and Private Sector Development, Financial Systems Practice. Machado, L., Parreiras, M., and Peçanha, V. (2011). Avaliação de impacto do uso do Cartão BNDES sobre o emprego nas empresas de menor porte. Revista do BNDES, 36, 6. Available at: http://www.bndes.gov.br/SiteBNDES/export/sites/default/bndes_pt/Galerias/Arquivos/conhecim ento/revista/Rev3601.pdf. Machín, A. (2012). Countercyclical Strategy of the Development Banks. MPRA Paper No. 42632, posted 22. November 2012 02:58 UTC. Online at http://mpra.ub.uni- muenchen.de/42632/. McKinsey Global Institute, McKinsey Infrastructure Practice, McKinsey & Company (2013). Infrastructure Productivity: How to Save $1 Trillion a Year. [online] McKinsey & Company. Available at: http://www.mckinsey.com/insights/engineering_construction/infrastructure_productivity [Accessed 22 January 2015]. Musacchio, A., and Lazzarini, S. (2015). Reinventing State Capitalism: Leviathan in Business, Brazil and Beyond. Sao Paulo: Portfolio-Penguin, 2015. Newlyn, W., and Rowan, D. (1954). Money and Banking in British Colonial Africa. A Study of the Monetary and Banking Systems of Eight British African Territories. pp. xiv. 301. Clarendon Press: Oxford. 110 Organisation for Economic Co-operation and Development (OECD) (2004). Principles of Corporate Governance. Available at: http://www.oecd.org/daf/ca/corporategovernanceprinciples/31557724.pdf. [Accessed 11 May 2015]. Organisation for Economic Co-operation and Development (OECD). (2005). Guidelines on Corporate Governance of SOEs. Available at: http://www.oecd.org/daf/ca/34803211.pdf [Accessed 14 May 2015]. Ottaviano, G. I. P., and Sousa F. L. (2007). The Effect of BNDES Loans on the Productivity of Brazilian Manufacturing Firms. Working paper. Park, S. G. (2012): Central Banks Quasi-Fiscal Policies and Inflation. IMF working paper. Available at: https://www.imf.org/external/pubs/ft/wp/2012/wp1214.pdf. Pierrakis, Y., and Westlake, S. (2009). Reshaping the UK Economy—The Role of Public Investment in Financing Growth. Research report. National Endowment for Science, Technology and the Arts (NESTA). Piketty, T. (2014). Capital in the 21st Century. Cambridge: Harvard University. Pires, J. C. L., Lodato, S., Cravo, T., and Vellani, S. (2014). A Comparative Analysis of IDB Approaches Supporting SMEs: Assessing Results in the Brazilian Manufacturing Sector. Inter- American Development Bank. Available at: http://publications.iadb.org/handle/11319/6683?locale-attribute=en. Rodrik, D. (2002). After Neo-Liberalism, What? The Economic Times, November 19. Rodrik, D. (2006). Goodbye Washington Consensus, Hello Washington Confusion? A Review of The World Bank’s Economic Growth in the 1990s: Learning from a Decade of Reform. Journal of Economic Literature, 44(4), 973–987. Sanderson, H., and Forsythe, M. (2012). China’s Superbank: Debt, Oil and Influence—How China Development Bank Is Rewriting the Rules of Finance. John Wiley & Sons. Scott, D. (2007). Strengthening the Governance and Performance of State-Owned Financial Institutions. Policy Research Working Paper, The World Bank. Seens, D. (2015). Cost–Benefit Analysis of the Canada Small Business Financing Program. Small Business Branch, Research and Analysis Directorate, Industry Canada. Serra, N., and Stiglitz, J. (2008). The Washington Consensus Reconsidered. Oxford: Oxford University Press. 111 Smallridge, D., and de Olloqui, F. (2011). A Health Diagnostic Tool for Public Development Banks. Inter-American Development Bank, Institution Capacity and Finance Sector, Technical Notes No. IDB-TN-225. Available at: http://services.iadb.org/wmsfiles/products/Publications/35558811.pdf [Accessed 28 February 2015]. Sobreira, R., and Martins, N. (2011). Os Acordos de Basiléia e bancos de desenvolvimento no Brasil: uma avaliação do BNDES e do BNB. Revista de Administração Pública (RAP)—Rio de Janeiro 45(2): 349–76. Spratt, S., and Collins, L. R. (2012). Development Finance Institutions and Infrastructure: A Systematic Review of Evidence for Development Additionality. Private Infrastructure Development Group (PIDG), Institute of Development Studies and Engineers against Poverty. Available at: https://www.ids.ac.uk/files/dmfile/2012IDS-PIDGSystematicReviewFinal.pdf [Accessed 16 March 2015]. Stein, P., Goland, T., and Schiff, R. (2010). Two Trillion and Counting: Assessing the Credit Gap for Micro, Small, and Medium-Size Enterprises in the Developing World. International Finance Corporation and McKinsey & Company. Available at: http://www.ifc.org/wps/wcm/connect/3d5d09804a2d54f08c1a8f8969adcc27/Two+trillion+and+c ounting.pdf?MOD=AJPERES [Accessed 21 January 2015]. Stephens, M. (1999). The Changing Role of Export Credit Agencies. IMF Research Paper. Stiglitz, J. E., and Weiss, A. (1981). Credit Rationing in Markets with Imperfect Information. The American Economic Review, 393–410. te Velde, D. W., and Warner, M. (2007). The Use of Subsidies by Development Finance Institutions in the Infrastructure Sector. Working Paper 283. London: ODI. van Zwieten, K. (2014). Debt Resolution and Business Exit, The World Bank Viewpoint Series Number 343, July 2014. Venugopal, S., Srivastava, A., and Polycarp, C. (2013). Survey of Public Financing Institutions’ Use of Instruments. [online] World Resources Institute. Available at: http://www.wri.org/sites/default/files/pdf/survey_of_public_financing_institutions_use_of_instru ments.pdf [Accessed 28 January 2015]. World Bank (2004). Inequality in Latin America: Breaking with History? World Bank, Latin American and the Caribbean Region, Flagship Report. Washington, DC. World Bank (2007). Template for Country Assessment of Corporate Governance. Revision 5.0. 112 World Bank (November 2009). Making Finance Work for Nigeria. [online] Available at: http://siteresources.worldbank.org/INTAFRSUMAFTPS/Resources/Making_Finance_Work_for _Nigeria.pdf [Accessed 20 January 2015]. World Bank (2009). Research—Knowledge in Development Note: Global Inequalities. [online] Available at: http://econ.worldbank.org/WBSITE/EXTERNAL/EXTDEC/EXTRESEARCH/0,,contentMDK:2 2451005~pagePK:64165401~piPK:64165026~theSitePK:469382,00.html [Accessed 30 January 2015]. World Bank (2011a). Financing of SMEs in South Africa: Results of a Survey of SMEs and Financial Institutions. Africa Region presentation. World Bank (2011b). Turkey Improving Conditions for SME Growth Finance and Innovation. Report No. 54961-TR. World Bank (2012). Global Financial Development Report 2013: Rethinking the Role of the State in Finance. Washington, DC: World Bank. Doi:10.1596/978-0-8213-9503-5. License: Creative Commons Attribution CC BY 3.0. Available at: https://openknowledge.worldbank.org/bitstream/handle/10986/11848/Global%20Financial%20D evelopment%20Report%202013.pdf?sequence=1 [Accessed 20 March 2015]. World Bank (2014). Nigeria Economic Report, no. 2. Washington, DC: World Bank Group. Available at: http://documents.worldbank.org/curated/en/2014/07/19883231/nigeria-economic- report-no-2 [Accessed at 15 January 2015]. Yuma Morisho, P. (2014). Consultant Report on the 4th Peer Review with the AADFI Prudential Standards, Guidelines and Rating System (PSGRS) in 2014. Available at: http://www.adfi- ci.org/downloads/ [Accessed 4 March 2015]. Zagha, R. (2004). Results from the Exercise on the Lessons of the 1990s, in Searching for Alternatives—Beyond the Washington Consensus. International Policy Dialogue: Berlin. Zagha, R., and Nankani, G. (2005). Economic Growth in the 1990s: Learning from a Decade of Reform. Washington, DC: World Bank. [online] Available at: http://www1.worldbank.org/prem/lessons1990s/ [Accessed 8 January 2015]. 113