and 2004, the share of credit in banks' total assets declined sharply. ... The largest banks see Basel II as a positive development, as it would lead to a ... higher capital charges; adoption of a formula to smooth the risk curve for SMEs, so as to ... For example, will the regulators have the capacity to validate models and monitor.
International Codes and Standards (C&S) and Development Finance: A Case Study of Brazil
By Ricardo Gottschalk and Cecilia Azevedo Sodré* Institute of Development Studies University of Sussex Brighton, BN1 9RE Tel: 44 1273 606261 Fax: 44 1273 621202
June 2005
•
This study was prepared as part of the project ‘Codes and Standards of International Best Practice: Assessing their Impact on Development Finance’, funded by the UK Department for International Development (DFID). We are thankful to DFID for their financial support and to Laura Bolton for her statistical work. The usual caveats apply.
1
Contents List of Acronyms
3
Executive Summary
4
Introduction
8
Chapter 1. C&S and Development Finance
11
1.1. Why linking C&S to development finance? 1.2. Why are institutions for development finance still important? 1.3. Can C&S address informational problems, thereby helping the banking systems overcome their lending limitations?
11 11 12
Chapter 2. Implementation of C&S, and Basel I and II in Brazil
14
2.1. What steps has Brazil undertaken towards compliance with C&S? 2.2. The Basel Capital Accord – Basel I 2.3. The New Basel Capital Accord – Basel II
14 16 24
Chapter 3. C&S and Reshaping the Financial System in Brazil
35
3.1. General trends and the current structure of the financial system 3.2. The evolution of credit since the adoption of Basel I 3.3. Credit allocation across sectors: what has been the impact on the SMEs and the poor? 3.4. The role of directed credit as a countervailing force
35 40
Chapter 4. New Initiatives in Directed Credit and Other Financing Mechanisms
50
46 49
4.1. Recent initiatives in support of micro-businesses and the poor 51 4.2. Why is micro-credit so important in Brazil? Access to micro-credit by MSEs 52 4.3. The experiences of micro-credit in Brazil 54 Conclusions, Policy Recommendations and Lessons for other Countries
57
Bibliography
60
Annex 1. Table A1. Basel I in Brazil: risk weight for different categories of assets 62 Annex 2. List of interviewees
63
2
List of Acronyms ANBID
Associacao Nacional dos Bancos de Investimento (the National Association of Investment Banks) APLs Arranjos Produtivos Locais BACEN Banco Central do Brasil BANDES Banco de Desenvolvimento do Espírito Santo (Development Bank of Espírito Santo State) BANRISUL Banco do Estado do Rio Grande do Sul BANESTES Banco do Estado do Espírito Santo BASA Banco da Amazonia BB Banco do Brasil BDMG Banco de Desenvolvimento de Minas Gerais (Development Bank of Minas Gerais State) BIS Bank for International Settlements BNB Banco do Nordeste do Brasil BNDES Banco Nacional de Desenvolvimento Economico e Social BOVESPA Bolsa de Valores de Sao Paulo (São Paulo Stock Exchange) BRDE Banco Regional de Desenvolvimento do Extremo Sul C&S Codes and Standards CEF Caixa Economica Federal COPOM Comite de Politica Monetaria do BACEN (Central Bank’s Monetary Policy Committee) CPSS Committee on Payment and Settlement Systems CSFI Centre for the Study of Financial Innovation CVM Comissao de Valores Mobiliarios (the Brazilian Securities Comission) DESENBAHIA Agência de Fomento do Estado da Bahia (Development Agency of Bahia State) FAT Fundo de Amparo ao Trabalhador FEBRABAN Federacao Brasileira de Bancos (Brazil’ Federation of Banks). FSF Financial Stability Forum IBGC Instituto Brasileiro de Governanca Corporativa (Brazilian Institute of Corporate Governance) IAS International Accounting Standards IIF Institute of International Finance IRB Internal Ratings Based Approach MDBs Multilateral Development Banks MSEs Micro and Small Enterprises PCPP Programa de Credito Produtivo Popular PDI Programa de Desenvolvimento Institucional PPA Plano Plurianual (Multiyear Plan) PROER Programa de Estimulo a Reestruracao e ao Fortalecimento do Sistema Financeiro (Programme for the Restructuring and Strengthening of the Financial System) PROES Programa de Incentivo a Reducao do Setor Publico Estadual na Atividade Bancaria (Programme of Incentives for the Reduction of State Level Public Sector in the Banking Activities)
3
PRONAF SCM SDDS SEBRAE SMEs US GAAP
Programa Nacional de Agricultura Familiar (National Programme for Strengthening Family-Based Agriculture) Sociedade de Crédito ao Microempreendedor Special Data Dissemination Standards (IMF) Serviço Brasileiro de Apoio a Micro e Pequena Empresa Small and Medium-sized Enterprises US Generally Accepted Accounting Principles
4
Executive Summary 1) The main purpose of this study is to examine the developmental impact of international Codes and Standards (C&S) as they are applied to the banking system in Brazil. The study is driven by the questions: to what extent has compliance with international C&S affected, or may affect in the future, credit to the SMEs and the poor? Through what mechanisms? What changes (institutional, other) have occurred as a result? If the impacts of adoption of C&S on development finance are negative, what new modalities of development finance, if any, are emerging in Brazil to overcome this problem? 2) To address and find answers to these questions, extensive interviews were conducted in Brazil, with central bank officials (including the top financial regulators), bankers from both public and private banks, and senior financial market consultants. The project also involved literature review, data collection and analysis of key financial indicators. This report is part of a comparative study involving Brazil and India, and the same set of questions was posed for India. 3) The first chapter of the report starts with a discussion on why to explore the link between C&S and development finance, and why institutions for development finance are still important. It argues that C&S are aimed at strengthening domestic financial systems, not at addressing development finance needs. The latter should be addressed through institutional action. But C&S are a package involving a number of general and specific rules for adoption, which are not neutral. They can have unintended negative consequences for development finance, at a time development finance institutions are not being strengthened, but dismantled. Therefore, lack of attention to regulatory implications for development finance may compound the problem. 4) The second chapter looks at what steps Brazil has undertaken towards compliance with international C&S. It focuses on implementation of Basel I in Brazil, and then on how the Brazilian banking authorities intend to apply Basel II in the country. 5) As regards Basel I, the report shows that these rules were adopted in 1994 in the broader context of financial sector reforms, aimed to strengthen Brazil's financial system. The report shows that the reforms have altered dramatically the regulatory and financial system landscapes in Brazil. The view broadly shared in the country – from top regulators to private and public bank representatives – is that they contributed significantly to the improved solidity of the financial system. But the positive assessment is tempered by some sectors of the banking community, particularly in regard to the Basel rules. Bankers from public banks warn that the specific Basel rule on capital requirement is likely to affect public institutions' lending capacity. In the specific case of development banks, some argue that these banks should not be subject to the Basel capital requirements, since their liability structure is based on compulsory savings, not bank deposits. Its implementation has the effect of restricting these banks' capacity to provide financing for developmental projects, which are at the heart of their mission. 6) But how have the implementation of Basel I, and the reforms more broadly, affected the levels of credit in Brazil? The report shows that credit as a proportion of the country's GDP declined gradually between 1994 (when Basel I was adopted) and early this century. Although other factors may also have contributed to credit decline, Basel
5
I seems to have played an important role. An evidence of this is that, between 1994 and 2004, the share of credit in banks' total assets declined sharply. The shift in the banks’ portfolio composition happened because whilst credit to the private sector has risk weight of 100% for capital requirement purposes under Basel I, government bonds have 0% weight – that is, are risk-free. One might argue that the shift in banks' portfolios towards government bonds was due to high interest rates in Brazil. But a similar shift in banks' portfolio took place in India following Basel I adoption in that country, under very different macroeconomic circumstances. 7) The report also shows that Basel I probably have contributed to the decline in the number of banks in Brazil since 1994, and to banking concentration when measured by the share of assets held by the 10 and 20 largest banks. What happened to the distribution of credit in the process? In regional terms, credit has moved away from the poorest regions and towards the wealthiest ones. The distribution of credit across sectors has also changed. Between 1994 and 2004, credit to the public sector fell dramatically – from 15% to 4% of total credit, while to the private sector increased, from 85% to 96%. Within the private sector, credit shares fell for the housing sector (very sharply), and increased for individuals (e.g., consumer credit), the rural sector and other sectors (telecommunications, transport, education, etc). 8) It is possible to cautiously suggest that the increase in the share of credit to the rural sector and to individuals may have benefited small rural producers, and reached less wealthy individuals. At the same time, productive and commercial activities lost financing from the private banks, and this may have harmed mostly the SMEs. A further evidence of the report is that changes in credit patterns were highly influenced by directed credit, which explains why credit to the housing sector fell so sharply, and increased to the rural sector. 9) The report next turns to the New Capital Accord (Basel II). The recently approved Basel II has as one of its main objectives to encourage internationally active banks to adopt risk sensitive models so that credit and other risks can be more accurately measured. Brazil intends to adopt Basel II over a 5-year period – from 2007 to 2011. A gradual approach for the full implementation of Basel II thus is proposed by Brazil's regulators. This approach seems appropriate for a developing country like Brazil where banks probably need time, resources and capacity building to be able to adopt Basel II in full. However, the proposed framework lacks any countervailing mechanisms or instruments to address the following issues the report highlights concerning Basel II: i) to the extent that risk-sensitive models are not universally adopted, but only by the larger banks, it can lead to further banking concentration in Brazil; ii) the use of these models can lead to concentration of banks' portfolio away from SMEs and towards big corporations; iii) being risk sensitive, these models can moreover increase the pro-cyclicality of bank lending. All these issues have clear macroeconomic, systemic and distributive dimensions that are lacking appropriate acknowledgement by Brazil's regulators. 10) The report then presents the views of the banks (and other financial market participants) in Brazil on Basel II. The questions asked include: how are they preparing themselves for the new regulatory framework? Are they considering adopting the risk-sensitive models for credit risk assessment? How do they think Basel II might affect the financial system? What could be the welfare impact? 11) The largest banks see Basel II as a positive development, as it would lead to a strengthened capacity by banks to assess and manage different types of risk and as a
6
result contribute to the solidity of the banking system. They are already taking steps to adopt risk-sensitive models to assess credit risk, and to measure operational risk. The medium-sized banks have expressed interest in adopting risk-sensitive models as well, but they know that would depend on permission being granted by the regulatory authorities, who in Brazil are willing to allow only the largest banks – between 12 and 15 – to use such models. 12) But a number of medium-sized public banks and development banks hold a more cautious position. Whilst acknowledging certain benefits, such as the strengthening of a risk management culture, they point to the operational difficulties in implementing Basel II, the high costs involved especially for the smaller banks, the potential conflict between new supervisory control over managerial practices and the social purpose of certain lending programmes, and the impact of capital requirement for operational risk on the cost and level of credit. Moreover, they point to the fact the new rules may constrain credit to the group of borrowers perceived as of higher risk, which typically are the small businesses. Furthermore, they believe the new rules may also constrain the ability of public banks to play a counter-cyclical role, when needed. 13) The development banks believe they should be given a differentiated treatment, that there is a need to recognise the specific features of development banks, such as their distinct liability structure and their development financing role. The BNDES, the largest of the development banks in Brazil, goes further to say that the bank should not be subjected to the New Accord, partly because its lending operations consist in large measure of passing resources on to other financial institutions, which are the ones that ultimately bear the risk. The banks' views thus are that there is a lack of debate in Brazil on a number of important issues, such as the need for differentiated treatment across the banking system, and the impact of Basel II on the system and on credit provision in particular. 14) The report next describes recent initiatives by Brazil's government to expand credit to micro-business in Brazil, along with other ongoing micro-credit schemes, and asks the question: why are initiatives to provide micro-credit so important for a country like Brazil? The answer is that micro and small enterprises amount to 14 million establishments in Brazil, accounting for over 95% of total enterprises in the country; of which, 9.5 million – or nearly 70% - are in the informal sector. Of these, 7.5 million employ just one person. Of this sub-set, only 472 thousand – less than 5% have access to some form of credit, and less than half of these, have access to bank credit. 15) The report finally suggests a number of proposals being discussed in international policy and academic circles, which could be adopted in Brazil to reduce the potential negative effects of Basel II, regarding banking and portfolio concentration, and procyclicality. These include: adoption of an equalising factor to be applied over the banks adopting the risk sensitive models for credit risk, so as to level up their capital requirements with that of other banks following the simpler credit risk assessment methodology. This could address the inequity issue, and could have the additional benefit of discouraging banks from changing their portfolios away from smaller borrowers, typically the ones deemed as riskier and that therefore require higher capital charges; adoption of a formula to smooth the risk curve for SMEs, so as to address the portfolio concentration risk; the use of models that 'look through the cycle', as opposed to the most utilised models that look at one point of the cycle. The
7
intended objective would be to reduce credit pro-cyclicality. A further measure could be to reward portfolio diversification, also aimed at reduced credit pro-cyclicality. 16) These measures could help address the three issues this study highlights as key ones that are receiving little attention in Brazil. In addition to these, the report raises a number of other issues that constitute a major challenge for Brazil's regulatory authorities. 17) For example, will the regulators have the capacity to validate models and monitor them adequately within the proposed time frame? Is the timetable proposed by Brazil's regulators long enough? Should regulators not need more time to be able to adequately validate and monitor risk assessment systems adopted by banks, especially those that will opt for the most advanced models? Should the proposal for adopting internal models for measuring operational risk not be eliminated, given the sheer complexity of measuring operational risk and the difficulties regulators would face to monitor their use? And in the case it is adopted, could a factor not be equally employed to avoid that some banks end up with lower capital requirements for operational risk than others? Also, it would be important that the regulatory authorities could take account of the fact that the risk management practices should be effective, but not excessively intrusive to the point of inhibiting lending activities and programmes that have a social purpose. 18) The main lessons we can draw for other developing countries are that institutions that support development finance are key and should therefore be preserved, as there is nothing indicating that an entirely market based banking system will serve the financing needs of the small businesses and the poor. This is even more so under Basel I and especially Basel II, as the latter has a clear bias against perceived higher risk borrowers. But many poor countries do not even have development finance institutions, which makes prudence towards the adoption of Basel rules even more necessary for these countries. One should not forget that capital markets in poor countries are under-developed and that the banking sector is still the major source of finance of the economy. Micro-finance in poor countries managed by foreign (and local) NGOs and other organisation has had an important role in providing resources to the small businesses and the poor. However, mainstream finance should also be able to reach these segments, and to finance projects (large or otherwise) that can benefit them indirectly as well. It is thus important that the system is regulated in ways that it can serve the economy and the most needy as well.
8
Introduction Since the East Asian crisis, developing countries have been encouraged to implement codes and standards (C&S) of international best practice in the financial sector.1 The main objective of this initiative has been to strengthen these countries’ financial systems, thereby reducing their vulnerability to shocks and changing circumstances in the global environment. Many developing countries have made strides in enhancing their financial systems through the adoption of international C&S. Of course, initial efforts to improve banking supervision and regulation, which are part of these standards, can be detected since the late 1980s and early 1990s, when developing countries started to undertake financial liberalisation reforms. This study is part of the DFID funded project ‘Codes and Standards of International Best Practice: Assessing their Impact on Development Finance’. The main purpose of the project is to examine the developmental impact of international C&S, as they are applied to the banking systems in developing countries. It was driven by the question: to what extent has compliance with international C&S affected, or may affect in the future, credit to SMEs and the poor? This question is important because efforts have been made to assess the degree of implementation of standards in developing countries, but little has been done to see if and how implementation of standards has affected development finance in these countries. In addressing this question, the study focuses on the implementation of the Basel Core Principles for Effective Banking Supervision, and the Basel Capital Accord. The project is a comparative study involving two countries, Brazil and India. These countries are among the emerging market economies that have succeeded in tapping the international capital markets and whose banks have established international linkages, thereby becoming internationally exposed. Their engagement in international finance has led their banking regulators to undertake measures towards complying with C&S in the banking system. This makes these two countries interesting cases for analysis. The purpose of a comparative study is to reinforce our findings, but most importantly to have additional insights into the likely effects of C&S, made possible by the fact that these countries share common features but most important have a number of specific characteristics. That is, whilst sharing common features, such as the fact both countries had achieved a reasonable degree of institutional development prior to the adoption of C&S, the two countries have important differences in the banking systems, development finance and speed of implementation of C&S. Brazil has a higher degree of banking concentration and foreign ownership. India has a banking system that is mostly state-owned and slower
1
Of importance here are those 12 C&S the Financial Stability Forum (FSF) considers as key for sound financial systems. These are standards in monetary and financial policy transparency, fiscal policy transparency, data dissemination, insolvency, corporate governance, accounting, auditing, payment and settlement, market integrity, banking supervision, securities regulation and insurance supervision – see FSF website http://www.fsforum.org/Standards/KeySTds.html
9
at adopting reforms and meeting international standards. Also, it has a far higher number of financial institutions specialised in rural credit. Based on these country case studies, the project addresses in particular the following questions: 1) Through what mechanisms may the adoption of C&S affect development finance, and what institutional changes may occur as a result? 2) What is the impact of these changes on the provision of finance to SMEs and the poor? 3) If the impacts of adoption of C&S on development finance are negative, what new modalities of development finance, if any, are emerging in these countries to overcome such shortcomings? To address and find answers to these questions, extensive interviews were conducted both in India and Brazil. The interviews were conducted with central bank officials, including top financial regulators based in these institutions; representatives of national associations of banks, bankers from both public and private banks, and senior financial market consultants. The project also involved literature review, and data collection and analysis of key national financial indicators. This report is on Brazil’s case study. The main findings of the study are that Basel I has contributed to a sharp fall in the share of banks’ credit assets in their total assets, and to a declining trend in total credit as a proportion of the country's GDP. This is in itself a worrying outcome, given that in Brazil total credit as a proportion of GDP was already very low. The adoption of Basel I in Brazil has also contributed to a higher degree of banking concentration. This fact can have negative implications for the provision of credit to the SMEs, as the large banks in Brazil have little incentive to cater for this segment of the market. But most worrying among the findings was that, despite international consensus that the New Basel rules (known as Basel II) may restrict credit to the SMEs, scant thought is being given in Brazil on what could be done to mitigate this effect taking place, when the new rules come into effect in early 2007. A negative impact on development finance could happen mainly through further banking concentration and the concentration of banks' credit portfolio away from SMEs and towards big companies. Moreover, little is being done to address a further likely implication, which is the increase in the procyclicality of bank credit. On the positive side, the study has found that the maintenance of a few development banks despite ample banking restructuring towards reducing the number of public banks, and of a long-standing system of directed credit, has helped protect credit levels to productive urban sectors and rural activities. Thus, both development banks and directed credit, which are key parts of the country’s development finance architecture, acted as countervailing forces against the negative effects of the Basel rules on credit to the less favoured.
10
The report is organised in four chapters. The first chapter starts with a discussion on why to explore the link between C&S and development finance, and why institutions for development finance are still important. The second chapter looks at what steps Brazil has undertaken towards compliance with C&S. It then focuses on the discussion about the Basel Capital Accord (Basel I) proposed by the Basel Committee in 1988, and the recently New Basel Capital Accord (Basel II), approved by the Basel Committee in June 2004. As regards Basel I, it explains how the Brazilian government applied it to the country’s banking system, how banks adjusted to it, and what the government did to help them overcome the difficulties they faced in the adjustment process. Concerning Basel II, the study reports how the Brazilian banking authorities intend to apply it to Brazil, and how Brazilian banks are preparing themselves for the New Capital Accord. It discusses possible implications of Basel II for development finance, as well as the views and concerns of both the public and private banking sectors on the matter. The third chapter describes the structure of Brazil’s financial system, and how it together with credit has evolved in the country since 1994, when Basel I was adopted. These trends are analysed with the intent of answering some of the questions raised in chapter 2, such as whether credit declined and if so to whom, and what role Basel I may have had in the process. Following this, the fourth chapter provides a short description of initiatives undertaken by the Brazilian government to expand micro-credit in Brazil, as well as micro-credit programmes initiated by regional development banks in the recent past. Finally, the study provides policy recommendations on ways to reduce the possible negative effects of Basel II on development finance in Brazil, as well as lessons for other developing countries.
11
Chapter 1. Codes and Standards and Development Finance 1.1. Why linking C&S to development finance? C&S have the clear purpose of helping to strengthen domestic financial systems in developing countries. They are not intended to address their development financing needs. The latter should be addressed through institutional action. But C&S are not a fixed set of rules countries should adopt unquestionably. They are a package involving a number of general and specific rules for adoption, and developing country regulators have the discretion to choose those rules – and if necessary adapt them – that are most suited to their countries’ needs and circumstances. This study takes the view that within the C&S package certain rules can be inimical to growth and poverty reduction. They may contribute to the reshaping of a developing country financial system in a way that undermines the ability of the system to provide development finance. Moreover, rules change over time and new rules can sometimes be complex, making it difficult to figure out their possible implications for stability, growth or poverty reduction. Today, developing countries are being encouraged to strengthen their financial systems through compliance with international C&S. In this context, developing country regulators are facing a key challenge: how to adopt these C&S at the national level. Concerning C&S in the banking system, the challenge has become even more difficult, as a new package of international standards – the New Basel Capital Accord (or Basel II), has been recently approved. Although there is an ongoing debate about a number of technical issues involving the adoption of these new standards, little has been discussed about their possible implications for credit provision to development-related projects. This is worrying, because these standards are being implemented at a time when there has been scant support for developing countries to build institutions to support development finance. Where such institutions exist, in most cases efforts have been not to strengthen but to dismantle them. So, lack of attention to regulatory implications for development finance is bound to compound the problem. We therefore believe it is important to inform the debate through generating research that improves our understanding of how the new C&S are likely to affect development finance. 1.2. Why are institutions for development finance still important? Financial markets are characterised by market failures and missing markets. In developing countries in particular there is a lack of certain markets - for example markets for long-term credits. This is partly due to lack of sophisticated instruments, which make it extremely hard for intermediaries to transform short-term liabilities into long-term finance, a crucial ingredient for large development projects. Moreover, in a number of cases private returns differ from social returns. Banks therefore may choose not the
12
project that offers the highest total returns, but the one that the bank itself has the highest return (Stiglitz and Weiss, 1981). Missing markets and externalities thus justify institutional action, to ensure that socially efficient projects are financed. Financial markets suffer in particular from information asymmetry, which impairs the ability of the banking system to assess risk. The result is that credit is rationed (Stiglitz and Weiss, 1981). Because markets are not cleared, the banking system ends up operating in an inefficient way. The system is moreover inefficient in how it allocates resources. Due to information asymmetry, the system becomes biased towards lending to big companies and against small borrowers. 1.3. Can C&S address the informational problems, thereby helping the banking systems overcome their lending limitations? Financial policy transparency and data dissemination are two among the main standards in the financial sector. These are expected to improve financial stability and efficiency. Data dissemination (and therefore information availability) is a key standard in support of financial efficiency. The rationale is that lenders and investors will be able to make better informed allocation decisions. This is a key hypothesis that, if true, would in good measure help tackle the information asymmetry problems raised by Stiglitz and his followers. However, we believe that the information asymmetry problem cannot be entirely overcome. The first reason is that information is costly, especially for banks in developing countries. Second, even if information is made available, still it is not totally exogenous. Borrowers’ behaviour can be affected by the lenders’ actions, for example by the level of interest rates they charge (Stiglitz and Weiss, 1981; Stiglitz, 1993). Given that inefficiencies arise from imperfect information, a case exists for government intervention to improve efficiency in the financial system. Thus, greater transparency and availability of information may improve the stability and efficiency of the financial system, but not entirely. In addition, we ask the question: can certain C&S be inimical to specific development objectives, such as the levels of credit provision, and credit provision to the SMEs and the poor? Going a step further, can they even undermine stability, if not properly implemented? The recently approved New Basel Capital Accord or Basel II has as one of its main objectives to encourage internationally active banks to adopt risk sensitive models so that credit and other risks can be more accurately measured. These risk sensitive models to work properly require the use by banks of a large set of data information on their current and prospective clients. Efforts to obtain information about potential borrowers are seen as a positive aspect. It can reduce lenders’ uncertainty about borrowers’ ability to honour loan contracts, with a consequent positive impact on the cost and possibly availability of funds to projects that are economically sound. We nonetheless believe that there is a limit to the extent to which one can obtain information about borrowers. This is especially true for small borrowers, who in developing countries are in their large majority in the informal sector, which makes
13
extremely difficult for banks to obtain the sort of information they need for risk assessment. More broadly, we raise three objections in relation to Basel II and the models it proposes for adoption: 1) To the extent that risk-sensitive models are not universally adopted, but only by the larger banks, it can lead to banking concentration. 2) The use of these models can lead to concentration of banks’ portfolio away from SMEs and towards big corporations. 3) Being risk sensitive, these models can moreover increase the pro-cyclicality of bank lending. These possible negative implications, which will serve as a guide in our study, have been pointed out by a number of international policy makers and academics, including Borio, Furfine and Lowe (2003) and Griffith-Jones (2003), and in the specific case of bank portfolio concentration away from the SMEs, acknowledged and to some extent addressed by the Basel Committee, in response to pressures from the German government. To summarise the main points made so far, this project takes the view that 1) financial systems do not provide credit to different segments in an efficient way, due to market failures and information deficiencies. As a result, credit to the poor in particular is affected. An increase in information will not solve the problem. There is a need for institutional action. 2) C&S in the financial sector may help ensure stability of the system. But it does not contribute to an increase in credit provision to the poor. That is our main concern. In the specific case of Basel II, which is the focus of our study, the new rules may potentially have a negative effect on development finance, and even on stability.
14
Chapter 2. Implementation of C&S, and Basel I and II 2.1.What steps has Brazil undertaken towards compliance with C&S? The12 standards that the Financial Stability Forum (FSF) 2 considers as key for a sound financial system can be grouped in three main areas: 1) macroeconomic policy and data transparency; 2) institutional and market infrastructure and 3) financial regulation and supervision. As said earlier, the focus of this study is on the Basel Capital Accords, or Basel I and 2, which are standards under the area of financial regulation and supervision. We thus discuss these standards in detail now. Progress Brazil is making in adopting standards under macroeconomic policy and data transparency, and under institutional and market infrastructure, are briefly summarised in Boxes 1 and 2. Standards in Financial regulation and supervision Standards in financial regulation and supervision cover three areas: banking supervision, which are our focus, securities regulation and insurance supervision. Standards in banking supervision are set by the Basel Committee on Banking Supervision, 3 and essentially refer to the Core Principles for Effective Banking Supervision. The Basel Core Principles were developed as a guide to support supervisory authorities from developed and developing countries in their task of strengthening their supervisory regimes and for ultimately contributing to the strengthening of national financial systems all over the world (Basel, 1997). There are 25 Core Principles, which can be grouped under the following headings: preconditions for effective banking supervision; licensing and structure; prudential regulations and requirements; methods of ongoing banking supervision; information and requirements; formal powers of supervisors; and cross-border banking. On the whole, recent assessments of Brazil’s compliance with C&S in the area of banking supervision have been very positive. Areas considered of particular strength have been legal structure, power enforcement, assessment of operational risks of banks, and onsite supervision. An area where weaknesses were identified a few years ago but that since then has improved significantly is supervisor’s capacity to evaluate credit policy of banks and supervise risks. To address this problem but also to strengthen the country’s overall capacity to assess and monitor risks, a number of initiatives have been taken. These include the implementation of a new loan classification system, the central risk information system and the internal control systems. 4
2
The FSF was created soon after the East Asian crisis to identify systemic risk situations and regulatory gaps in the international financial system. 3 The Basel Committee on Banking Supervision was established in 1975 and comprises supervisory authorities of 10 industrial countries – Belgium, Canada, France, Germany, Italy, Japan, Luxembourg, Netherlands, Sweden, Switzerland, United Kingdom and the United States. 4 Resolution 2554, of 29/09/1998.
15
Box 1. Macroeconomic policy and data transparency The standards that fall under the heading of macroeconomic policy and data transparency are: monetary and financial policy transparency, fiscal policy transparency and data dissemination. In Brazil, the monetary area has witnessed in the past few years significant progress towards greater transparency, particularly in the conduct of monetary policy. This includes the adoption of the inflation targeting regime, the disclosure of the minutes of the Central Bank’s Monetary Policy Committee (COPOM) meetings, and the production of inflation reports among other important initiatives. The inflation targeting regime has been seen as transparent in how the monetary policy sets its ultimate objectives and what instruments are used for that purpose. The fiscal area has also benefited from a number of initiatives that have been important to the greater transparency and predictability of the fiscal accounts: the ‘Fiscal Responsibility Law’, the multi-year fiscal framework, and the government recognition of the so-called ‘contingent liabilities’. The Fiscal Responsibility Law, enacted in the year 2000, is a code of conduct for the public administration at all levels. It is aimed at improving fiscal management, accountability and transparency, thereby preventing risks that can affect the balance of the public accounts. The multi-year fiscal framework – the Multiyear Plan (PPA) is also aimed at greater fiscal transparency, but above all fiscal predictability. The latter has since the achievement of price stability become a feasible exercise. Finally, the contingent liabilities refer to past government liabilities that originate in unclear fiscal practices, and in measures taken in the past that implied losses to certain sectors and individuals. The relevance of these liabilities is that, once recognised and acted upon, they can have a large impact on the government’s public debt. Concerning data transparency, a key step has been the country’s adherence to the Special Data Dissemination Standard (SDDS). The SDDS is about disseminating economic and financial data to the markets, and is particularly relevant to countries that have, or are willing to have, access to the international capital markets. Back in 2001, Brazil was already considered by the IMF as having met the SDDS specifications for the coverage, periodicity and timeliness of data. Updated from Gottschalk (2001)
Assessing and monitoring risk is a clear supervisory issue covered by the Basel Core Principles. Moreover, it is addressed again within the New Basel Capital Accord or Basel II, under Pillar 2, on supervisory review. Basel II also has Pillar 1, on minimum capital requirements; and Pillar 3, on market discipline. Of particular interest here are those principles that fall under the heading of prudential regulations and requirements of the Basel Core Principles, which are addressed more thoroughly by the Basel Capital Accords Basel I and Basel II. Specifically, this study focuses on the minimum capital requirements for banks, as established by Basel I, and more recently modified by Basel II, under its Pillar 1. This focus is based on our understanding that regulatory requirements are the ones that have the clearest link to development finance. In what follows, we discuss Basel I and II and their implementation in Brazil.
16
Box 2. Institutional and market infrastructure The standards in the area of institutional and market infrastructure include corporate governance, accounting, and payment and settlement systems. The improvement of corporate governance practices is seen as one among key requisites for the development and deepening of capital markets, and this has been increasingly recognised in Brazil. Various agents in Brazil – the government, the country’s security commission (CVM), the country’s stock exchange (BOVESPA), the Brazilian Institute of Corporate Governance (IBGC) and the National Association of Investment Banks (ANBID) – have taken in the recent past important initiatives in this area. These have included the creation of new codes of conduct for a better corporate governance, and the creation of BOVESPA’s New Market, which is a parallel equity market governed by stricter governance rules. A recent assessment conducted by the Institute of International Finance (IIF) observes that important progress has been made in the area of corporate governance in Brazil. For example, concerning the hotly contested issue of minority shareholder rights, there has been a limiting of non-voting preferred shares in proportion to total shares. However, the IIF assessment points to remaining deficiencies, including the need for a single consistent set of rules, further extending voting rights, higher degree of independence to the board of directors, and improved enforcement (IIF, 2004). Compliance with accounting standards is another key area in which progress has been observed in the recent past. In the assessment of an ex-CVM top regulator and the private sector, Brazil is moving towards complying with most of international accounting standards (IAS). Moreover, a large number of companies are striving for US GAAP adoption. However, supervision and enforcement problems still exist, partly due to lack of human and financial resources the supervisory body – the CVM – faces to carry out its tasks (interview material). Concerning payment and settlement systems, Brazil has developed a new payments system, in place since 2001. Under the previous system, clearing procedures used to take place overnight, with the Central Bank bearing the risk. Under the new system, clearing takes place during the day on a continuous basis, whereby risks have been transferred to the markets, thus forcing the latter towards cautious behaviour. This is an important move towards meeting the core principles for systemically important payments systems set by the Committee on Payment and Settlement Systems (CPSS), which sets the standards of best practice for payment and settlement. Updated from Gottschalk (2001)
2.2. The Basel Capital Accord – Basel I The Basel Capital Accord (Basel I) is an agreed regulatory framework for capital adequacy that the Basel Committee for Banking Regulation and Supervision recommended for implementation in 1988. Its ultimate aim was to improve the soundness and stability of national banking systems and of the international financial system. This was to be achieved through the promotion of international convergence in the rules for setting minimum capital requirements for internationally active banks (Basel, 1998). Of course, it was expected that national regulators would also consider such rules for other banks under their jurisdictions as well. According to this framework, internationally active banks are expected to meet a total capital requirement of at least 8% in relation to their risk-weighted assets. The required capital should have two tiers: core capital and supplementary capital. Core capital is essentially formed by shareholders’ equity and disclosed reserves. Supplementary capital
17
includes undisclosed reserves, revaluation of reserves, general provisions held against unidentified losses and subordinated debt. Assets (and off-balance sheet exposures), in turn, are weighted according to their relative riskiness, with weights ranging from 0% to 100% (applied over the 8% of capital). The framework was initially designed to address credit risk. In the subsequent 10 years, it was amended to include other types of risk, including market risk and concentration risk. The risk weights for different categories of assets are displayed in Table 1. Basel I has been implemented widely across the world – in Latin America alone, all countries claim to be compliant with Basel I (IADB, 2004). As a result, capital adequacy levels have increased significantly worldwide. Over the years, Basel I has been commented upon and criticised on various grounds, these being of mainly two types: the shortcomings of its rules and its implications for borrowers. The first type of criticism included the high level of aggregation of different types of assets for risk-weighing purposes, and focus on credit risk in detriment to other types of risk (a shortcoming gradually overcome as time progressed); the second type of criticism included the fact that efforts to meet capital to risk-weighted assets ratio could result in decline of credit to riskier borrowers, and in particular the fact that the rules discriminated against non-OECD long-term borrowers. The Basel Committee itself observed in their documents the further possible risk that if not implemented homogenously across countries Basel I could lead to competitive inequities. This is because some national regulators could end up imposing stricter capital requirements than others (Basel, 1998). Since then, some of these criticisms have been gradually addressed. For example, in 1996 Basel I was amended to incorporate market risks. Banks were required to measure market risk for capital charge from the end of 1997 onwards, and for that purpose to choose either a standardised approach or an internal risk management model (VAR type of models) for risk measurement (Basel, 1995). In 1999, a new set of weights for capital charge was proposed, which further on evolved into what is now known as Basel II, with a broader range of options on how to measure credit and other types of risk, which includes more risk sensitive measurement techniques. An implication of these proposed techniques is that the discrimination against non-OECD borrowers is reduced. However and as will be discussed further below, whilst addressing some of the initial concerns, Basel II reinforces others such as possible competitive inequities, not only between banks from different jurisdictions but between banks within the same jurisdiction, and the risk of reduced credit to riskier borrowers.
18
Table 1. Risk Weights for Different Types of Assets as defined by Basel I Weight 0%
Loans to/Investment in: • OECD central governments • Central governments that borrow in the national currency • Borrowers with OECD central governments’ collaterals or guarantees From 0% to 50% (at the discretion of national • Domestic public sector entities outside the regulators) central government; borrowers with such entities’ collaterals or guarantees 20% • Multilateral development banks (MDBs); borrowers with MDBs collaterals or guarantees • OECD banks and securities firms; borrowers with OECD banks and securities firms’ collaterals or guarantees • Non-OECD banks with maturity of up to one year; borrowers with non-OECD banks’ collaterals or guarantees, with maturity up to one year. • Non-domestic OECD public sector entities outside the central government; borrowers with such entities’ collaterals or guarantees 50% • Mortgage borrowers who inhabit the residential property or rent it. 100% • Private sector • Non-OECD Banks with maturity of over one year. • Non-OECD central governments (unless they borrow in the national currency). • Real estate • Capital instruments issued by other banks. Source: Basel (1998).
The Adoption of Basel I in Brazil Brazil adhered to Basel I in September 1994 through the Resolution 2099 of Brazil's Central Bank. 5 The resolution established that to appropriately address credit risk, Brazil's financial institutions had to meet a minimum level of capital of 8% in relation to risk-weighted assets. Later in 1997 the 8% limit was raised to 11%, 6 thus higher than the 8% recommended by the Basel Committee, a decision justified on the grounds that Brazil’s financial institutions were subject to higher macroeconomic volatility and shocks than financial institutions based in the rich countries. In addition, the risk weights assigned to different categories of assets were slightly adapted – see Table A.1, annex 1. The Central Bank established in addition capital requirements for market and other types of risks, following advice from the original Basel Capital Accord document and subsequent ones that addressed these other types of risks in greater detail. 5
Resolution 2099, of 17/09/1994. Specifically, in June 1997 the minimum capital level was raised to 10% (Resolution 2399/1997) and later in November of the same year, to 11% (Circular no. 2784/1997).
6
19
The reason given for the adoption of Basel I in 1994 was that Brazil's domestic banks were becoming increasingly exposed to the international financial markets, and therefore there was a need to adjust regulation to this new reality. But, in retrospect, it can be said that Basel I in Brazil was part of a bigger package of banking reforms undertaken in the second half of the 1990s and early this century. Moreover, the Basel rules were implemented just after the adoption of Brazil’s Real Stabilisation Plan, in July 1994, which opened a phase of price stability in the country for the first time after many years. We thus next examine the impact of Basel I in Brazil taking a broader picture, to include analysis of the banking reform package, which interacted with the changes in the regulatory framework for the banking system. But we first set the context that triggered that banking reforms. Inflation declined sharply in the second half of 1994 in response to the stabilisation measures adopted in July of the same year. In response, banks started changing their lending practices towards credit expansion to the private sector, to compensate for the end of gains derived from inflation-related sources of revenues. This process was interrupted shortly after, when the government responded to the Tequila crisis by adopting restrictive monetary and credit policies in the first quarter of 1995. These measures put a number of private banks in a very fragile financial position, as the quality of their loan portfolio started to deteriorate rapidly with the recession that ensued the adjustment. This, together with the similarly fragile positions of the public banks, which had a portfolio of bad loans and were under-capitalised, led the government to take initiatives to clean and strengthen Brazil's financial system. The Basel I and other regulatory measures that have been adopted since September 1994 should be seen in this broader context. That is, as part of efforts to improve the solidity of the financial system, to be pursued through both regulatory and financial restructuring measures. Two major instruments were set up to undertake a major cleaning and restructuring of the banking system: the Programme for the Restructuring and Strengthening of the Financial System (PROER), and the Programme of Incentives for the Reduction of State Level Public Sector in the Banking Activities (PROES). 7 The PROER provided credit lines and fiscal incentives to support organisational restructuring that resulted in mergers and acquisitions among private banks. The PROES provided credit lines so that public banks at the State Levels could be closed, privatised or transformed into public development agencies. 8,9
7
The PROER or Programa de Estimulo a Reestruracao e ao Fortalecimento do Sistema Financeiro, was set up in November 1995; and the PROES or Programa de Incentivo a Reducao do Setor Publico Estadual na Atividade Bancaria, in September 1996. Both PROER and PROES were modified on a number of occasions thereafter. 8 Andrezo and Lima (2002) provide a descriptive summary of these programmes, and how they evolved over time. 9 The PROES also had a clear political purpose. It was motivated by the diagnosis that public banks constituted a source of monetary and fiscal profligacy. This was explained by the existence of political pressures from State Governments for their controlled banks to lend to them (despite the existence of legal restrictions for banks to lend to their controllers), and their subsequent inability to service the debt, which was in the end re-financed by the Central Government to protect the overall stability of the financial system.
20
By the end of 1997, 48 banks of a total of 270 private banks (retail, investment) were restructured and ended up being incorporated by other financial institutions, or transferred to the Federal Government under the PROER (Andrezo and Lima, 2002). The restructuring process continued until into this century, with new rounds of mergers and acquisitions. During the period, large retail banks with national coverage, such as Nacional, Bamerindus, Excel-Economico and Real were acquired by Unibanco, HSBC, Bilbao Vizcaya and ABN Amro Bank, respectively. Under the PROES, nearly all banks at the State level were either closed, privatised, had their control transferred to the Central Government, or were transformed into government agencies. The Central Government provided through the PROES up to 100% of the resources to support the restructuring, but only up to 50% in those few cases in which banks remained under the State Government control, as was the case of the Banrisul and Nossa Caixa (see below). The restructuring process led to a higher degree of banking concentration and foreign ownership. Total assets held by the 10 largest banks went up from 52% in 1999 to 62% in 2001 and 66% in mid-2004 (see Table 6 in the next section). The share of the foreign banks in the system's total assets increased from 7.5% in 1994 to 23% in 1999 and then 22.5% in 2001 (see Table 4, also in the next section). The costs to finance the restructuring process were of sizeable magnitude. By the end of 1999, the financial assistance provided by the PROER to six large private banks summed up to R$ 17 billion, while the support provided by the PROES to the restructuring of public banks reached R$ 82,8 billion (or around 9% of the country’s GDP at that time). 10 Furthermore, Brazil also had the public federal banks, two of which are among the largest financial institutions in the country, which similarly to many private banks and public banks at the State level, were in poor financial shape. In response to that, these banks were also re-capitalised to be able to comply with the new capital requirements, although they were not privatised. The Banco do Brasil – the largest bank in Brazil - was recapitalised in 1996, as part of a broader restructuring process that led to new management – and again in 2001, when only by then was the bank able to meet Basel I minimum capital requirements. Today, the bank has its capital adequacy levels at 15.6%. The Caixa Economica, a public bank that is also very large and that operates in the housing, sewage and infrastructure segments, went through a similar restructuring process to that adopted for the Banco do Brasil. But the bank's pathway towards compliance was even steeper, as the insolvency levels of its portfolio and its capital and provisioning deficiencies were of considerable magnitude at the early stage of the process. The bank was finally brought in line with Basel I capital requirements in 2001. 10
Figures provided by BACEN and Diario Oficial do Estado de Sao Paulo; cited by Andrezo and Lima (2002). To know exactly how much the values in the domestic currency (R$) are in US dollars, one would have to have information on the chronogram of the resources’ disbursements. But just to give a rough idea of the scale of the resources in U$ dollars, if 75% of these are converted using the 1998 exchange rate at 1.16 (which was not much different from the preceding years) and 25% using the 1999 exchange rate at 1.82, the values for R$ 17 billion and R$ 82.8 billion are US$ 13.3 billion and US$ 64.9 billion, respectively.
21
Among the development banks, the Banco do Nordeste do Brasil (BNB) was also recapitalised in 2001, which helped the bank increase its capital ratio to 16,2% in the same year, and to 22,6% in 2003. In 2002, the Banco da Amazonia (Basa) was in turn recapitalised, reaching the capital ratio of 52% in mid-2004. The two major banks at the State levels that remained under control of the State Governments despite the PROES – Banrisul and Nossa Caixa, Nosso Banco - experienced a similar adjustment process. Banrisul (controlled by the State of Rio Grande do Sul) and Nossa Caixa, Nosso Banco (controlled by the Sao Paulo State) were cleaned and recapitalised, with 50% of the resources coming from the State Governments, and 50% from the Central Government. As a result, Banrisul reached a capital ratio of 19% in 2002 (although that declined to 11% thereafter as a result of rapid credit expansion) and Nossa Caixa a ratio of 25%. The private banks (national and foreign) in turn, also adjusted their capital levels to meet the Basel I requirements. This, together with the recapitalisation of those parts of the system that were in poor shape at the beginning of the process, led to a gradual increase in the system's capital level as a whole, reaching the level of 18% in December 2003. All these developments - the country's compliance with Basel I, the adoption of supervisory measures (the new loan classification system, the central credit risk information system 11 , the internal control system 12 ) and the broader restructuring process – have altered dramatically the regulatory and financial system landscapes in Brazil. Each of these elements was part of an agenda of reforms that undoubtedly contributed significantly to the improved solidity of the financial system. That is a view broadly shared in Brazil, from top regulators to private and public bank representatives, and financial consultants. 13 The positive assessment is tempered by some sectors of the banking community in Brazil, particularly in regard to the Basel rules. Whilst recognising concrete benefits stemming from Basel I, such as the development of a credit risk assessment culture, some bankers from the public sector warn that the specific Basel rule on capital requirement is likely to affect public institutions' lending capacity. In the specific case of development banks, some argue that these banks should not be subject to the Basel capital requirements, since their liability structure is based on compulsory savings, not bank deposits, and that compliance with Basel is therefore not only unnecessary but counter-productive. Its implementation has the (admittedly unintended) effect of restricting these banks' capacity 11
The Central Credit Risk Information system is managed by the Central Bank of Brazil, and compiles information on credit extended to clients corresponding to a value of R$5000 or more. The information has to be provided by all banks and non-bank financial institutions. These institutions also have to provide the level of risk category the credit falls under (from AA to H – see Resolution 2682), in addition to the kinds of guarantees attached to the credit they provide to the client. 12 The internal control system is based on the Basel's document 'Framework for International Control Systems in Banking organisations, and is aimed at changing banks' management structures and strengthen internal and risk control systems to improve clarity of functions, rules and procedures, transparency and risk control for various types of risk, particularly liquidity risk (see Central Bank resolutions 2554 of 24/09/1998, and 2804 of 21/12/2000). 13 Private banks see all these changes as important, including those that affected them most directly, such as the required modifications in their management structures. To these banks, efforts to adopt internal control systems had the positive effect of making their management practices, particularly their risk control systems, converge towards those observed among foreign banks based in the country (interview material).
22
to support financing for developmental projects, which are at the heart of their mission. For this reason, in their view the homogeneous treatment given across private and public banks that differ in their liability structures and purposes should be reviewed. But how have the implementation of Basel I, and the reforms more broadly, affected the levels of credit in the country? Figure 1 in the next section shows that total credit as a proportion of the country’s GDP declined gradually between 1994 and early this century. However, the majority view held in Brazil is that Basel I did not affect the total level of the country’s credit, or even credit to the rural sectors, the micro-enterprises or SMEs. The reason provided is that historically credit to the private sector has been very limited in Brazil – at around 25% and that other factors explain better why this is so. These include high levels of public financing requirements, which makes government bonds the most important asset held by banks; the country's macroeconomic conditions characterised by high levels of interest rates and spreads (which did not come down with the entry of foreign banks in the system), high insolvency levels among private borrowers, the existence of government taxes (what Brazilians call the ' fiscal wedge'), high levels of deposit rates with the Central Bank, and bankruptcy law and a judiciary system that are tilted towards the interests of the debtors. This study contests this view. Although it accepts that a number of factors explain why credit has historically been so low in Brazil, the fact is that it has declined even further since 1994, and Basel I seems to have played an important role in it. As shown further below, an evidence of this is that the share of credit in banks’ total assets declined sharply between 1994 and 2004. This evidence is consistent with recent simulations run by Barrel and Gottschalk (2005) using a macro-econometric model for Brazil, which shows that an increase in banks’ minimum capital requirements brings about a fall in banks’ credit to the private sector, and an increase in the levels of government bonds held by banks. The shift in portfolio assets composition happens because whilst credit to the private sector has risk weight of 100% for capital requirement purposes under Basel rules, government bonds have a 0% weight – that is, these assets are risk-free. One might argue that the shift in banks’ portfolio towards government bonds was due to high interest rates. But a similar shift in banks’ portfolio took place in India following the adoption of Basel I in that country, under very different macroeconomic circumstances. There is thus strong evidence suggesting that credit was indeed affected by the Basel capital rules. If credit was not so more affected, this was due to the fact that efforts towards meeting the minimum capital requirements took place mainly through banks’ recapitalisation, as indicated earlier, thereby reducing the need for credit cuts or cuts in the riskier types of credit. As regards credit allocation, the discussion further below suggests that it is difficult to judge what happened to credit to SMEs or the poor, due to lack of information. But for rural credit information is available. It shows that rural credit patterns are strongly determined by directed credit schemes, which has historically been a key government policy instrument in support of the rural sector. This policy instrument clearly has countervailed any negative impacts Basel I may have had on rural credit.
23
Looking forward, it should be seen whether with the various constraints on credit being gradually relaxed in the future, Basel capital requirement will or not constitute a major binding constraint for credit expansion in Brazil. It should be recalled in this regard that although the current Basel requirement levels are high, they are likely to decline very rapidly as banks' portfolio composition moves away from government bonds that have 0% risk weight towards credit to the private sector, whose risk weight is well above 0% and therefore requires capital allocation (in addition to 'ex-ante' provisions for assets of doubtful value – see Box 3 below).
Box 3. The development of own systems of credit risk assessment by Brazil's public banks The Central Bank has established the need for banks to allocate loss provisioning for credits; in that, the percentage of provisioning should be crescent according to the risk category assigned to the credit, from AA to H (it starts 0.5% for category A and moves upwards gradually to end at 100% for the category H – see Resolution 2682 of 1999). To comply with the Central Bank guidelines in this area, the main Brazilian financial institutions have developed their own credit risk assessment systems. Following the Central Bank's Resolution No 2682, these banks categorise credit from AA to H. The federal public banks, for example, categorise loan-related risks from AA to C for projects in general, big and small rural producers, and more recently for microcredit as well, reflecting the fact the latter is gaining growing importance in Brazil. But some banks place loans to small rural producers with high levels of insolvency under the extreme risk category H. In what follows, we provide descriptive information on individual development banks’ experiences with new credit risk systems, and the challenges they are facing in the process. The Banco da Amazonia (BASA) started to design systems of credit risk assessment in 1997. In that year, the Bank established the Department of Credit Risk. In 1998, it conducted the first assessment of its risk system. Until 2000, the Bank still did not have a network with data information from all the Bank's agencies across the country. In the assessment of the Bank's representatives, the Banco Basa is still outdated in relation other federal public banks regarding its technology. The bank’s system of credit risk assessment has three platforms: development, homologation and thesis, and production. Under this framework, the assessment process involving a rural producer, who constitutes a major part of the bank’s portfolio, starts with a description of the economic group to which the producer belongs, type of property, monthly gross income, type of activity. Through this system, information about the net worth is gathered. For example, for those that request credit lines of R$ 600,00, and do not possess documents that provide evidence of their worth or income, a risk analysis is undertaken on the basis of the amount of resources the producer holds, like animal stock, machines and equipments. The bank also has been developing a mode of analysis that includes interviews and visits to the potential borrower’s property, assessment of his/her entrepreneurial capacity, credit experience, type of business (family, otherwise), length of business, annual income, and whether he/she is in litigious land area where conflicts take place. In general, the microcredit (annual income up to R$ 6,000.00) falls under the A to C risk categories. The Banco do Nordeste do Brasil (BNB) also created a department charged with credit risk analysis, which has been gradually refining its risk models. Until 1998, the bank extended credit without a risk analysis, a procedure that had a negative effect on the bank’s credit portfolio. With the creation of the risk analysis department (along with an area of internal control, auditing and an auditing committee), the bank’s representatives believe the bank has nowadays greater capacity to manage and control public resources, especially those coming from federal financial institutions, as a result of reduced political interference in the process. The risk assessment system has been refined to include visits by a credit department personnel to credit borrowers, as well as their assessment and punctuation according to credit scoring and behaviour scoring.
24
The development agency Desenbahia (from the Bahia State) has developed internal methodologies for assessing credit risk for all sizes of enterprises (from micro to big ones). It also adopted a more advanced risk assessment method for projects above R$ 200,000.00) It uses simulation system to analyse the behaviour of a firm’s cash-flow in the future. The system is used for bigger projects with the purpose of assessing their viability. The Banco de Desenvolvimento de Minas Gerais (BDMG) has changed its credit risk analysis methodology between 1999 and 2001, in response to the Central Bank’s resolution No 2682. Nowadays, the method consists of aggregating long-term financing and practising surveillance of borrowing companies every 6 months. The risk management process is considered relevant by the bank’s representatives. However, they remark that it is difficult to reconcile lending practices to high risk level projects that, notwithstanding, have a clear social purpose. Under the adopted method, the bank has to assess the project’s impact on the company’s net worth, and on the basis of that, take the decision of whether to lend or not. However, this takes away from the bank flexibility to decide whether it might want to undertake certain risks and bear the resulting loss, having in view its mission to support social projects. If the bank decides to grant a company better risk category than other financial institutions, the Central Bank will require from the bank explanations about its risk assessment method, and will monitor it henceforth. Source: interview material.
2.3.The New Basel Capital Accord – Basel II The main purpose of the New Basel Capital Accord approved in June 2004 is to further strengthen the soundness and stability of the international banking system, through encouraging banks to improve their risk management practices. To the extent that various internationally active banks have been adopting internal models to assess different types of risks, the new accord's intent is to align the rules that determine capital allocation with what has been already practised in the markets. The new framework has three mutually reinforcing pillars: 1. The minimum capital requirement, 2. The supervisory review and 3. Market discipline. Pillars 2 and 3 relate closely to the Basel Committee's Core Principles for Effective Banking Supervision (BCP), but in this new context in which new risk management systems are encouraged for adoption, emphasis is put on supervising the quality of banks' new systems for risk assessment, and on disclosure of information on risk management practices and on different types of risk exposures, along with disclosure of other types of information, such as banks’ financial performance and financial position (Basel, 2004). But the main novelty and challenges for banks and regulators world-wide concern the new rules under Pillar 1 for capital requirements. The minimum capital adequacy level at 8% recommended by Basel I is maintained, but three different approaches are suggested for determining the risk for different types of assets: the standardised approach, the foundation internal risk based (IRB) approach and the advanced IRB approach. Under the standardised approach, different risk levels can be assigned to different categories of assets, and the approach allows for external rating agencies to determine risk levels. The basic and advanced IRB approaches differ from the standardised approach in that they require the use of internal modelling techniques to measure risk. The difference between the latter two approaches is that under the foundation IRB approach banks can use their own models to determine default risk, but the parameters for loss given default is furnished by the regulatory authorities. In the case of the advanced IRB approach, banks are allowed to determine through their modelling techniques and data base both default risk and the loss given default. 25
In addition, the new accord requires the allocation of capital for operational risk (in addition to credit, market risks, international exposure and other risks), and proposes three methods for measuring this type of risk: the basic indicator method, the standard indicator method and the advanced measurement method. The new framework has been designed for adoption by the G-10, and the Basel committee expects that these countries will be ready to implement the framework by the beginning of 2007. At the same time, the Basel Committee recognises that many non-G10 countries world-wide may wish to adapt the new framework to their own national realities and circumstances, and to have their own timetable for adopting the new rules. 14 The Committee goes further to say that national regulators should aim to ensure the regulatory systems in their countries meet certain pre-conditions before attempting to implement the new framework in its entirety. They specifically recommend a sequencing approach, in which national regulators should aim for strengthening the country's regulatory infrastructure through the implementation of the Pillars 2 and 3, which deal with supervisory systems and market discipline, as just mentioned; only when these Pillars are firmly in place, should they focus on Pillar 1. This suggested approach reflects a main concern that many countries face limited resource capacity (human, financial) to implement Basel II, and that efforts to adopt the Pillar 1 may have the undesirable effect of diverting resources needed to ensure a satisfactory level of compliance with the BCP, many elements of which are embodied in the Pillars 2 and 3. The new framework was approved after several rounds of consultations and debates that involved numerous stakeholders such as financial market participants, senior policy makers (national, international), national regulators and academics. In this consultative process, in addition to capacity limitation, a number of other issues were raised in relation to the proposed framework. For the purpose of this study, we highlight the following: 1) Inequity leading to banking concentration; 2) loan portfolio concentration; 3) and pro-cyclicality. 1) Inequity leading to banking concentration The inequity issue had been raised before by the Basel Committee when Basel I was created. Their concern was that if Basel I did not ensure a minimum degree of homogeneity of rules across different jurisdictions, this could grant competitive advantage to internationally active banks based in certain jurisdictions against banks based in others. The point was that if rules were applied differently across different jurisdictions, some banks would end up facing higher capital requirements than others. There was the further risk that these differences could be magnified by specific tax, accounting and other rules across jurisdictions.
14
In this regard, the Basel Committee set up a Working Group in 2003 composed of representatives from mainly non-G-10 countries, including Brazil, to assess Basel II and provide recommendations on how supervisors might want to promote changes towards the new framework (Basel, 2004).
26
Basel II provides a menu of options for calculating capital charges, and this can also cause the inequity problem mentioned above, of banks working with different levels of capital requirements. But in this case that would happen not only among banks across countries but also within countries. For example, in a same jurisdiction there could be banks adopting the IRB approach along with others adopting the standardised approach. But those banks adopting the IRB approach would be at advantage in relation to the others. This is because the IRB approach is likely to result in lower levels of capital requirements. The larger and more sophisticated banks are more likely to adopt the IRB approach and therefore to benefit from it, in detriment to smaller banks, which are more likely to adopt the standardised approach. This type of inequity could, in turn, lead to banking concentration favouring the bigger banks, and in the case of many developing countries, it could favour the foreign banks. 2) Loan portfolio concentration The use of risk measurement techniques to determine the amount of capital to be allocated for different types of assets is likely to result in both more expensive and rationed credit to borrowers perceived as of higher risk, and more and cheaper credit to borrowers perceived as of lower risk. For reasons such as information asymmetry, small borrowers such as SMEs are likely to be judged as of higher risk than the larger ones, such as large companies. This can cause a concentration in banks’ loan portfolio away from small borrowers and towards the larger companies. Moreover, portfolio concentration implies that risk is being concentrated thereby making financial institutions more vulnerable to shocks and unexpected changing circumstances. This goes against the intended objective of regulatory measures, which is to reduce risks and vulnerabilities to which banks are normally exposed. 3) Pro-cyclicality The use of risk-sensitive models under the IRB approach is bound to result in these models detecting an increase in the probability of default during economic downturns. As a consequence, the assets of a portfolio will be downgraded – what is called migration – which in turn will lead to higher capital charges. Recent empirical evidence supports the claim that the use of the IRB approach to measure risk may have the effect of a higher variation in the capital charge over the business cycle, as compared to the use of Basel I type of rules for measuring risk (see Goodhart and Segoviano, 2005). This in itself may lead to both increased cost and reduced quantity of credit. Furthermore, the fact that it is harder to raise capital during economic downturns may reinforce the tendency in credit reduction, ultimately leading to a credit crunch and a deepening of the economic downturn, with further impacts on banks' portfolios. A reason why the measured risk by these models tends to be so much time-variant is that even when they are forward-looking, their time horizons often are limited to one year (see Borio et al, 2003). These models therefore result in assigning borrowers ratings in light of their current (or over a limited time-horizon) status. That is what is called the ‘point-in-time’ approach. But if models could instead look ‘through-the-cycle’, so as to reduce or eliminate variations in the ratings caused by changing conditions during the cycle, then their pro-cyclicality effects could be avoided or at least significantly reduced.
27
An additional problem is that Basel II by encouraging different banks to use similar models – VAR models – could exacerbate pro-cyclicality even further, as banks would behave and react in similar ways to the same events (Persaud, 2000; Danielsson et al., 2001). In times of financial crises, their effects could be magnified throughout the system. 15 The potential problems of inequity (i.e. banking concentration) and portfolio concentration show that regulatory measures are not neutral, that they can have an important impact on competitive and equity issues. Moreover, they can exacerbate procyclicality of bank credit and thereby contribute to larger swings in the business cycle. The latter problem in particular should be a concern for regulators, as it also has a bearing on the stability of the financial system. Indeed, accentuated macroeconomic volatility is a major factor underlying banking crises, due to sharp variations in key prices, such as exchange and interest rates, and therefore in banks’ balance sheets. In what follows we will provide an assessment of to what extent Brazil fulfils the preconditions the Basel Committee has indicated, and whether the country’s banking regulators, and more broadly financial market participants and academics, are showing concern, and indicating solutions, to the problems we have just highlighted. Does Brazil fulfil the pre-conditions as recommended by the Basel Committee? We have seen above that Brazil’s compliance with C&S in the area of banking supervision is in good shape, including supervisory and monitoring capacity. Moreover, the creation of the new loan classification system, the central risk information system and the internal control system are steps that respond to the specific concerns of Pillars 2 and 3 with regard to supervisory processes and market discipline. More broadly, the financial system has been strengthened as a result of the restructuring process. Today the system is financially solid and meets comfortably the capital adequacy levels as established by the Central Bank. All these factors combined – strengthened supervision and financial position of the banking system – provide a firm platform for Brazil’s banking regulators to take steps towards implementing the Pillar 1 on capital requirements of the new capital accord. In line with this broadly positive assessment, Brazil’s regulators have established the procedures for implementing Basel II, with particular attention to Pillar 1. 16 At the same time, in recognition that it is important to adapt the new framework to Brazil’s specific conditions, a phased approach, consisting of five steps, has been proposed. That is, the regulatory authorities established a chronogram that covers a period of 7 years – from 2005 to 2011 – for the full implementation of the New Accord. In addition, they established that those banks with significant weight in the domestic financial system and with international exposure will be permitted to adopt the foundation IRB approach (and 15
On this point, the Financial Times editorial of 22nd February 2005 also calls attention to a survey of financial opinion conducted by the Centre for the Study of Financial Innovation (CSFI) in 54 countries, which shows that business respondents differentiate between ‘banking risk’, which is the focus of Basel II, and the safety of financial institutions, which can be threatened by systemic risk, which could increase as a result of the use of similar risk assessment models by banks. 16 See Brazil Central Bank’s Communication No. 12.746 (08/12.2004).
28
the advanced IRB approach at the end of the transitional process as well), while the remaining banks will have to adopt the standardised approach. The standardised approach will not draw on external ratings for determining credit risk. It will consist of an upgrading of the current approach, with the incorporation of risk mitigation instruments. The same rules will be applied to national and foreign banks. Table 2 provides detailed information on the chronogram proposed for Basel II in Brazil. Table 2. Basel II in Brazil – Announced Chronogram for Implementation Period Until end of 2005
Measures/Action Review of capital requirements for credit risk under the standardised approach; new capital requirements for those market risks still not covered by current rules; impact studies regarding operational risk Eligibility criteria for adoption of the IRB approach for Until end of 2007 credit risk and internal models for market risk assessment; capital requirement for operational risk. Validation of models for assessing market risk; 2008-2009 chronogram for validating the use of the foundation IRB approach; initial validation of the IRB approach and criteria for the adoption of internal models for operational risk. Validation of the advanced IRB approach for credit 2009-2010 risk and chronogram for the advanced approach for operational risk. Validation of internal models for operational risk. 2010-2011 Source: Brazil Central Bank’s Communication No 12.746 of 8/12/2004.
It can be seen from Table 2 that a chronogram has been established for the adoption of internal models for market and operational risks, in addition to credit risk. The adoption of the phased approach for the transition from the current to the new framework (whose time span goes considerably beyond the deadline established by the Basel Committee for the G-10), the limitation of the IRB approach to the larger and internationally exposed banks, and the non-use of external ratings under the standardised approach, are factors that together reveal an intention by regulators to be cautious. This in all probability reflects a view that Brazil’s banks and the regulators themselves still need a considerable amount of time to become ready for the IRB approach. The proposed approach is broadly consistent with what Brazil’s top regulators had indicated during our interviews on what they would do. First, that only the larger banks – in the regulators’ words, between 12 and 15 – would be permitted to adopt the IRB approach at some stage, thus following the US approach to Basel II. Second, that the standardised approach to be adopted by the majority of banks would be indigenised to suit better Brazil’s needs. Third, that the basic indicator method whereby capital charge for operation risk should be calculated as a percentage (e.g. of 15%) over the banks’ gross revenues would be adopted - although as we can see the proposed approach indicates that this will be the case only during the first years, as a timetable exists for the adoption of internal models for measuring operational risk. Finally, all banks, public and private, and regardless of their liability structure or mission, would be subject to Basel II rules.
29
During the interview phase, Brazil’s regulators stressed the point that addressing operational risk adequately was a key challenge facing Brazil’s banks. Their assessment was that this particular type of risk is of enormous complexity, involving both operational and conceptual challenges, such as how to define operational risk, and what types of risk to measure, and how to measure them. There is an uncertainty as to whether banks have the capacity to adopt models capable of measuring different types of operational risk. The longer time-span allowed for the adoption of internal models for operational risk reflects this concern. But given the costs involved in developing models and the conceptual problems that will never be entirely overcome, one begs the question of whether the regulators should not have been even more cautious by discarding the possibility of the use of internal models for measuring operational risk in the foreseeable future. More broadly, the relatively long timetable of the proposed framework for Brazil reflects awareness concerning operational and resource challenges for implementing the new framework. According to a Brazilian academic (and financial market participants as well), this problem concerns not only banks, but also regulators, the question being: will Brazilian regulators be prepared to validate and supervise the use of sophisticated risk assessment models? Will there be human resources available in sufficient numbers to do the job? Will these resources not be attracted to the the private sector? Although regulators also recognise the existence of a variety of limitations and have responded with that through putting forward a gradual approach, the fact that the IRB approach will be introduced at some stage shows that the framework is fundamentally accepted in its principles and aims. This point was made clear during the interview phase, when Brazil’s regulators expressed the view that Basel II signified steps in the right direction, in that it would reinforce the country’s prudential rules for the financial system. The phased approach that Brazil’s regulators have proposed for adoption looks appropriate for a developing country where banks probably need time, resources and capacity building to be able to adopt Basel II in full. However, the proposed framework lacks any countervailing mechanisms or instruments to address any of the issues outlined above – of banking concentration, portfolio concentration away from SMEs and increased pro-cyclicality. All these issues have clear macroeconomic and systemic dimensions that are lacking appropriate acknowledgement by Brazil’s regulators. This begs the question of whether Brazil’s regulators, in not addressing the three issues just mentioned, are not excessively focused on micro-prudential risks (e.g. the risk facing individual banks), but not paying sufficient attention to macro risks, such as shocks or large swings in the business cycle, which are common to the whole banking system (Borio et al., 2003). Macro risks can be exacerbated by a banking system that has concentrated portfolios and that uses risk models that accentuate credit pro-cyclicality. Brazil’s regulators did not seem very concerned about these issues when asked during the interview phase of this research. To illustrate the point, a senior regulator expressed the view that Basel II should not lead to further concentration (though if it did, he believed there would be no problem with that), and that the current number of banking institutions
30
is of good size to stimulate competitive behaviour and to gradually incorporate large segments of the population still outside the reach of the banking system. The question therefore is: does that mean that Brazil's regulators are not concerned or mindful of macro risks? A possible explanation for their lack of concern might be that the demands from Basel are so numerous and complex that they have concentrated their efforts to address the technical and operational aspects of the framework. Moreover, the broader and perhaps subtler implications of Basel seem to have been overlooked due to a lack of debate in the country on these issues which, if existed, would have drawn their attention to them. This point was made by bank representatives during the interviews. It is worth to turn then to the views of the Banks (and other financial market participants) in Brazil on Basel II. A large number of financial market participants were interviewed (both private and public). 17 The questions asked included: how were they preparing themselves for the new framework? Were they considering adopting the IRB approach? What challenges would they face in the process? How did they think Basel II could affect the financial system? What would be the welfare impact? Views of the Private Sector (and academics) and what steps they are taking The private sector sees the adoption of Basel II rules in Brazil as a positive development. It would lead to a strengthened capacity by banks to assess and manage different types of risk and as a result contribute to the solidity of the banking system. The largest banks are already taking steps to adopt the internal models to assess credit risk, and to measure operational risk. The three largest private banks – Bradesco, Itaú and Unibanco – have already established Directorships of Risk Management, charged with the developments of both credit and operational risk management models, and specifically in regard to credit risk, their expectation is to be able to adopt the advanced IRB approach. They expect that their credit risk models will be up and running until the end of 2006. One large bank informed that they perceive the build up of a data base for measuring operational risk as the major challenge facing the bank at present. Indeed, as with this large bank, Brazilian banks more generally perceive as a major challenge how best to address operational risk - specifically how to quantify this kind of risk, and the necessary capital required to protect against it. Although a good deal of losses arising from operational risk is clearly identified by banks – losses associated with labour and civil litigation and frauds, a whole universe of unidentifiable losses still exists, making their measurement very difficult. Like the larger banks, the medium-sized banks have expressed interest in adopting internal models for both credit and operational risk, but accepted that whether they will be able to use these models will depend on permission being granted by the regulatory and supervisory authorities, whose position on the issue was still not clear to them at the time the interviews were being conducted. 17
Most interviews were conducted in July and August 2004.
31
The interest among large and medium-sized banks in preparing themselves for Basel II was captured by a research undertaken by the Febraban. 18 The results indicated that at the time the research was being conducted, 19% of the 163 banking institutions in Brazil were already adopting risk management procedures as part of their preparations towards Basel II; 45% were evaluating carefully what steps to take, and 36% claimed they would not adopt any measures to comply with the new rules (Gazeta Mercantil, 2004). The banks that claimed not to be taking any significant steps in preparation to Basel II probably were the smaller ones. This position looks coherent with the view shared that these banks would have to adopt a modified standardised approach, whose specific rules would be determined by the Central Bank regulators, as indeed has been the case. But whether that suits these banks and the banking system as a whole is not an uncontested issue in Brazil. Some large banks have expressed the opinion that the smaller banks should also make efforts to adopt the internal risk assessment models, so that they avoid being downgraded by rating agencies, which could hinder their ability to tap international capital markets. But a more serious concern is whether the adoption of different models by banks could lead to banking concentration, as argued earlier. Would that be likely to happen as a result of the use of the standardised approach by some banks but not others? A recent study by Carneiro et al. (2004) based on simulations for Brazil, shows that the use of the IRB approach by banks would, for the majority of banks, imply a reduction in capital requirements between 0% and 40%. For a few banks, the needs would be reduced even more, by up to 82%. This indicates that the risk envisaged here, of a few banks gaining substantial competitive advantage through use of the IRB approach, is very real. A market insider believes further banking concentration as a result of Basel II is possible, although the majority view is that in Brazil banking concentration is a phenomenon driven by market factors and economies of scale, and that in this context regulatory measures would not be an important factor. To illustrate the point, interviewers observed that the pull out of foreign banks from the Brazilian market was the result of their lack of ability to compete with Brazilian banks. The latter have shown to hold competitive advantages in relation to the foreign ones, due to their scale, reach and better knowledge of the market and the profile of their customer base, which could be used as an effective tool for analysis and provision of credit. Yet, even if these factors are in fact important ones, this does not preclude regulation having the potential role of reinforcing concentration trends. Still in relation to this issue, an academic expressed the view that the banking system has already reached a high level of concentration, and that therefore there exists only little room left for further banking consolidation. The existing small medium-size and small banks have either their own established niche or are family-owned, in the latter case having been effective in servicing the purpose of supporting the family business.
18
Febraban stands for Federacao Brasileira de Bancos (Brazil’ Federation of Banks).
32
Views of the Public Banks and what steps they are taking A good number of Brazilian banks still remain public despite the banking restructuring and banking privatisation in the past few years. These are mainly federal banks with large retail base and development banks, although a few state-level banks still exist (see section above). All of them are taking steps to be prepared for the new capital rules. Like the private banks, the largest retail public banks hold a favourable view of Basel II, and believe that these rules should be adopted in Brazil in its entirety to the benefit of the financial system. The new instruments of risk and managerial controls and increased transparency are seen as positive developments, among other reasons because they can contribute to reduced political influence on lending decisions and thereby to greater efficiency. But a number of medium-sized and development banks hold a more cautious position. Whilst acknowledging certain benefits, such as the strengthening of a risk management culture, they point to the operational difficulties in implementing Basel II, the high costs involved especially for the smaller banks, the potential conflict between new supervisory control on managerial practices and the social purpose of certain lending programmes, and the impact of capital requirement for operational risks on the cost and level of credit. Moreover, they acknowledge the fact that the new rules may constrain credit to the group of borrowers perceived as of higher risk, which typically are the small businesses. There is therefore a much higher degree of heterogeneity of views among public banks than among private banks, and within the former group this reflects a divide along the lines of size and nature of the banking activity. The largest retail bank in Brazil – Banco do Brasil (BB) – takes a favourable view about Basel II, and is seen as leading the process in developing internal risk assessment models both for credit and operational risk. As regards credit risk, they claim to be at a considerably advanced stage in developing a model and believe that this will be ready for use by 2007. They are also investing considerable amount of resources in developing a VAR model to measure operational risk, and in preparing the data base, which will have a 5-year period coverage by 2007. Other banks are also taking steps to be prepared for Basel II. The public banks at the state level are improving the managerial practices of their credit portfolios, through upgrading their credit risk assessment models and pursuing modelling design for operational risk. They intend to adopt internal risk models in the future and are hoping to have them fully developed and tested by 2007. Other banks have only recently created risk departments and therefore are running behind other banks in developing risk assessment models, particularly for operational risk, and in building the required data base, admittedly a task of great complexity, especially for medium-sized and small banks, which face high fixed costs in relation to the scale of their operations. The public federal banks other than the BB – Caixa Economica, Banco da Amazonia (Basa), Banco do Nordeste do Brasil (BNB), are also attempting to improve their risk assessment systems, and intend to adopt internal models for credit and operational risk assessment. Given their limited internal capacity, they are working on these areas with
33
the assistance of external consultancies. But even with external help, they point to the difficulties they face in taking these steps. A particular difficulty relates to how to map and quantify operational risk, and especially how to disentangle operational risks from other types of risk, including credit risk. Some institutions also fear the risk of investing in the development of internal models for risk assessment, but not having them validated by the Central Bank. Both groups of banks – federal and state-level ones – share a number of concerns. In addition to their difficulties in developing and putting in practice new risk assessment models, and the costs that these tasks involve, especially for the smaller banks, they raise a number of other points that need to be addressed. First, they believe the use of internal risk models will imply less capital requirement, and that if they end up not adopting these models they will find themselves at disadvantage in relation to those banks adopting them, as it will imply allocating higher levels of capital and therefore higher costs. Furthermore, there would be the additional negative impact on their ability to raise external borrowing in the external markets, a view that coincides with that of some private banks, as noted earlier. Second, banks are worried that the capital requirement for operational risk, by increasing the banks' total capital requirement, will lead to higher costs, which are likely to be reflected in more expensive credit. The larger banks might be permitted to adopt the standardised model at some point, which means measuring risk by type of business and thus requiring less capital. But the smaller banks will have little alternative but to adopt the basic indicator method (i.e., capital required corresponding to 15% of banks' gross revenues) and therefore will face higher capital requirements, both in absolute terms and relative to other banks adopting a more advanced method. There are therefore two problems arising from the need to allocate capital for operational risk: 1. overall higher level of capital requirements with banks facing higher costs as a result, 19 and 2. the competitive effect affecting negatively those banks adopting the simplest approach. Third, some of these banks (especially the retail ones at the state level) believe they have a relatively homogeneous portfolio of clients to which credit extension is in many cases consigned, which reduces the credit risk they face. Their current risk controls may not be among the most sophisticated ones, but are deemed as sufficient in light of their customer profile. However, to the extent they attempt to expand their client base to include clients with different and riskier profiles, they fear that the new risk control systems will inhibit this process from taking off. That is, the system will delimit the sorts of products offered by the bank and therefore affect its business activities. There would thus be a heightened conflict between different areas of the bank. This indicates that elements of Pillar 2, such as stricter supervisory controls and monitoring, are likely to restrain credit expansion policies. (In relation to Pillar 3, banks have pointed out that there is a need to clarify better what sort of information needs to be disclosed, and within that, to clearly separate strategic information and information that can be made available to the markets. The 19
It has been noted that banks adopting the IRB approach for credit risk could end up requiring less capital for this type of risk, thus offsetting the added capital for operational risk. But banks adopting the standardised approach would not be able to generate this balancing effect – see IADB (2005, chapter 16).
34
underlying concern is that excessive information disclosure might be harmful to banks and the system as a whole.) Fourth, public banks have a social mission. In line with that, many of their lending programmes derive from Federal and State level social policies. But the New Basel rules are likely to exacerbate the tension between profit maximising and social objectives, as the latter should be expected to involve activities deemed as of higher risk. As it is put in a IADB report, ‘[p]ressures for profitability may induce public bank managers to deviate from their social mandate and mimic private banks in their credit allocation criteria’ (IADB, 2004, p. 144, footnote 8, based on De La Torre, 2002). Moreover, the new rules may also constrain the ability of public banks to play a counter-cyclical role, when needed. 20 A final point that relates closely to the previous one is that development banks, such as the BNDES, Basa, BNB and BDMG, believe they should be given a differentiated treatment. They recognise that the recent restructuring process involving cleaning and recapitalisation provided public banks with conditions to compete with private banks on an equal basis, but they firmly believe there is a need to recognise the specific features of development banks, such as their distinct liability structure and their development financing role. Accordingly, it would be important to make the C&S related rules more flexible to this group of banks. That could include a lower capital adequacy requirement, whose minimum level in Brazil is higher at 11% compared with the 8% determined by the Basel Committee for the G-10. The BNDES goes further to propose that the bank should not be subject to the New Accord, partly due to its liability structure based on compulsory savings, partly because its lending operations consist in large measure of passing resources on to other financial institutions (banks and development agencies) which are the ones that ultimately bear the risk. The banks' views are that there is a lack of debate in the country on a number of important issues, such as the need for differentiated treatment across the banking system, and the impact of Basel II on the system and on credit provision in particular. There is a feeling that Brazil's regulators have missed the opportunity to raise these issues more forcefully in international fora and with the Basel Committee. There is a debate of some of these issues domestically within the Febraban, as well as at national and international fora, but this has been limited. It is therefore felt that more needs to be done. Thus, the views between the private sector and public banks on the potential benefits but especially costs of Basel II diverge fairly significantly. This divergence reflects their differences in terms of size, capacity to adopt more advanced risk assessment approaches, and their nature and purpose. But a particular concern that emerges very strongly and that reflects public banks' social concerns is that credit can be affected by Basel II rules through a variety of mechanisms. Unfortunately, this aspect has received very little attention so far.
20
This point has been made mainly by academics. Moreover, an IADB study presents evidence that public banks in Latin America are less pro-cyclical than private banks in extending credit (IADB, 2004, p. 23 and chapter 11).
35
Chapter 3. C&S and Trends in the Financial System in Brazil This chapter of the report discusses in more detail some of the points raised or explored only to a limited extent in the second chapter: what impact has Basel I had on the structure of the financial system? Has it affected the levels of credit, and if so, to what sectors? To what extent have institutional factors helped counter-balance the negative effects of Basel I? The next section starts with the description of recent trends in the structure of the financial system, followed by an analysis of the evolution of credit in Brazil since the early 1990s. 3.1. General trends and the current structure of the financial system The financial system in Brazil has undergone major changes in the past two decades. Between 1988 and 1994 – that is, a period of just 6 years – the number of banks increased from 106 to 246. Since the mid-1990s, however, a steady decline in the number of banks has taken place, reaching the total of 164 banks in 2003 (see Table 3). Thus, in a space of just 15 years or so, two marked trends were observed: an upward one between 1988 and 1994, and a declining one between 1995 and 2003. The increase in the number of banks in the late 1980s and early 1990s can be attributed to the government’s decision undertaken in 1988 to allow financial non-banking institutions to become banks, and to permit commercial and investment banks to become universal banks. 21 The number of banks, which jumped from 106 to 179 between 1988 and 1989, and continued to increase steadily in the early 1990s, was a phenomenon that can in addition be explained by an environment of high inflation, which permitted banks to expand in size and numbers through a business strategy that combined low loan levels with high profits derived from the high inflation-related revenues. But the upward trend was reversed from 1995 onwards, when the number of banks started to exhibit a steady decline. As seen earlier, the reversal in the trend has been the result of the banking restructuring process, driven by a government aiming to strengthen the banking sector through the recapitalisation, mergers and acquisitions, privatisation and the entry of foreign banks. An underlying component of this process was the government's adoption of prudential regulation determining minimum capital requirements for banks (Basel I).
21
See Resolution 1.524 of 21.09.1988, from the Central Bank of Brazil. Also, as Troster (2004) observes, Brazil's 1988 Constitution reduced barriers to entry.
36
Table 3. Number of banks in Brazil 1988-2003 Selected years Year 1988 1989 1990 1992 1994 1995 1998 2001 2003 106 179 216 234 246 242 204 182 164 Total 32 22 15 14 Public banks 172 123 95 88 Private national banks1 Foreign Banks 2 38 59 72 62 Source: Central Bank of Brazil. 1 Includes national banks with foreign participation. 2 Includes foreign banks’ branches in Brazil.
The restructuring process undertaken from 1995 onwards changed the ownership structure of the banking system, with the number of public banks declining from 32 to 14 between 1995 and 2003, and of private national banks from 172 to 88. During the same period, the number of foreign banks increased from 38 to 62 (see Table 3). 22 Table 4. Participation of banks in the banking system’s total assets -1995-2004 % 1995 1996 1997 1998 1999 2000 2001
2002
2003
2004
Public banks 52.2 50.9 50.1 45.3 43.0 36.6 32.0 34.7 37.2 34.4 Private National banks1 38.9 38.3 36.8 35.3 33.1 35.2 37.2 36.9 40.8 41.7 Foreign banks2 8.7 10.5 12.8 18.4 23.2 27.4 29.9 27.4 20.7 22.4 Source: Central Bank of Brazil. 1 Includes national banks with foreign participation. 2 Includes foreign banks’ branches in Brazil.
The reduction in the number of public banks was reflected in a substantial fall in the share of their assets in the banking system's total assets, from 52% in 1995 to 34% in 2004. Despite the decline in their numbers, the percentage share of private national banks in the total assets of the banking system went up from 39% to 42% during the same period. Among foreign banks, their assets' share went up from 8.7% to 22% (see Table 4). 23 The sharp decline in the proportion of assets held by public banks reflected mainly a steep fall in the share of assets of the state-level banks, from 21.9% in 1995 to 5.2% in 2004. Of course, this was the result of the fact that nearly all such banks were either closed or privatised during the period, as discussed earlier. At the same time, the federal public bank Banco do Brasil, witnessed an increase in its percentage share from 13.9% to 17.4%, while the other large federal public bank, Caixa Economica, had its share declined from 16.4% to 11.5% (see Table 5).
22
For an analysis of the growing foreign ownership of Brazil’s banking system, see Carvalho et al. (without date). 23 In truth, the share of assets held by private banks went first down to 33% in 1999, and then up to 42% in 2004, while of foreign banks went up to 30% in 2001, and down to 22% in 2004. These inflexions were due to purchases of foreign banks by domestic ones, as the former started leaving the country due to fierce competion.
37
Table 5. Participation of different categories of public banks in the banking system’s total assets -1995-2004 % 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Public state-level banks 21.9 21.9 19.1 11.4 10.2 5.6 4.3 5.9 5.8 5.2 Banco do Brasil 13.9 12.5 14.4 17.4 15.8 15.6 16.8 17.1 18.4 17.4 Caixa Economica Federal 16.4 16.5 16.6 17.0 17.1 15.4 11.0 11.7 13.0 11.5 Source: Central Bank of Brazil.
The reduction in the total number of banks also led to a higher degree of banking concentration in the system. Between 1999 and mid-2004, the percentage of assets held by the largest 10 banks in the total assets of the financial system went up from 52% to 66%; when the largest 20 banks are considered, the percentage of assets went up from 62% to 77% (see Table 6). Table 6. The largest banks in Brazil 2000-2004 1999
2000
2001
% of total assets1 2002
2003
Jun/2004
Largest 10
52.0
59.8
62.1
65.5
67.6
66.1
Largest 20
61.7
74.6
75.3
77.2
76.9
76.8
Largest 50
Nd
83.8
85.1
84.7
82.9
83.0
Source: Central Bank of Brazil. 1 Excludes BNDES. Years 2003 and 2004 also exclude Volkswagen, BRDE, GM and CNH Capital. Excluding these latter banksfor the years up to 2002 alter the results only marginally. For the years 2001 and 2002,the 50th largest bank was assumed to hold the same value of assets as the 49th largest bank.
The trend in banking concentration is less clear when measured by credit operations. In national terms, the trend points to the opposite direction. For the largest 10 and 20 banks, their share in the total credit operations in the financial system declined slightly between 1999 and mid-2004. For the largest 50 banks, it declined more pronouncedly between 2000 and mid-2004 (see Table 7). These findings are similar to those obtained by Troster (2004) - concentration of assets and de-concentration of credit between 1999 and 2003, and between 1994 and 2003 as well, using the Herfindahl index to measure banking concentration. Table 7. The largest banks in Brazil 2000-2004 1999
2000
2001
% of total credit operations1
2002
2003
Jun/2004
Largest 10
67.4
65.9
61.3
63.2
65.1
66.0
Largest 20
77.8
78.5
75.4
75.1
73.6
74.6
Largest 50
Nd
85.6
84.2
82.2
77.3
78.7
1
Source: Central Bank of Brazil. Excludes BNDES. Years 2003 and 2004 also exclude Volkswagen, BRDE, GM and CNH Capital. Excluding these latter banks for the years up to 2002 alter the results only marginally. For the years 2000-2002, the 50th largest bank was assumed to hold the same value of assets as the 49th largest bank.
38
The process of concentration in the past 10 years or so follows one of strong deconcentration, between 1980 and 1993 (see Rodriguez de Paula, 1998). In regional terms, a clear trend in credit concentration can be observed. For the north and north-east regions, which are the smallest regions in terms of bank deposits and credit operations, and the poorest ones in terms of income per capita levels, the decline in credit operations between 1997 and 2003 was from 1.9% to 1.4% and from 13.6% to 6.2%, respectively (see Table 8). For the Centre-West region, the decline was from 12.3% to 8.7%. By contrast, the south-east and south regions, the wealthiest ones in income per capita terms, witnessed an increase in credit operations in the period 1997-2003 from 59.4% to 70.9% and from 12.7% to 12.8%, respectively. Interestingly, regional concentration has not happened when measured by level of deposits, which increased for the north and centre-west regions, and declined only slightly for the north-east region (see also Table 8). Table 8. Participation in total banking deposits and credit operations, by regions 1997-2003 % Total Regions
1997 Deposit
1999
2001
Credit Deposit Credit operation operation
Deposit
2003
Credit Deposit operation
Credit operation
North
1.2
1.9
1.5
1.5
1.4
1.2
1.3
1.4
Northeast
7.6
13.6
7.2
9.0
7.1
5.8
6.1
6.2
Southeast
71.3
59.4
69.5
64.2
67.2
72.2
65.9
70.9
South
10.0
12.7
10.6
12.6
10.2
11.2
10.0
12.8
9.9
12.3
11.2
12.8
14.2
9.6
16.7
8.7
Centre-West
Source: Central Bank of Brazil (http://www.bcb.gov.br)
Thus, the restructuring of the banking system from 1995 onwards led to 1. a significant reduction in the number of banks operating in the country, 2. a reduction in the public sector participation in the banking system (when measured by share of assets in total assets), 3. an increase in the participation of private domestic and especially foreign banks; and 4. a higher level of banking concentration, both in terms of asset holdings by the largest 10 and 20 banks, and in terms of regional credit distribution. As regards the latter, the concentration of credit away from the poorest regions and towards the richest ones was quite dramatic. The lack of concentration of credit by the largest banks might be explained by the sharp reduction in credit by the Caixa Economica Federal (which has figured among the largest 10 banks over the years), from 18.9% in 2000 to 6.2% in mid2004. It is likely that, within private banks, credit concentration was the case. Where does the financial system stand today? In June 2004, the total assets of the financial system in Brazil were equivalent to 87% of the country’s GDP, a proportion that has not varied much in the last few years, but that was much higher than in 1995, when total assets reached a low of 48% of the GDP (see 39
Table 9). This means that, since the period of strong adjustment in the financial system in the early period of the Real plan, the banking system has expanded quite strongly. Table 9. Total Assets of the Financial System as a proportion of the GDP % Year 116.7 1989 148.8 1993 71.1 1994 47.6 1995 44.6 1997 88.6 2001 91.5 2002 85.6 2003 87.4 20041 Source: Central Bank of Brazil, and de Paula (1998), Table 5, for the 1989-1997 period. 1 June 2004.
Of the total assets held by the banking system, 33.7% were total credit operations. The other main components of total assets were cash and inter-bank loans at 14%, bonds and stocks at 26%, and other assets at 22% (see Table 10). Table 10. Components of total Assets % Total
Cash and inter-bank loans Bonds and stocks Credit operations Permanent assets Other assets
Jun 2002
Dec 2002
Jun 2003
Dec 2003
Jun 2004
9.8
13.8
11.4
14.7
13.6
30.3
27.1
27.7
27.5
26.4
30.7
30.1
31.2
33.9
33.7
6.4
5.3
5.1
4.9
4.8
22.8 23.7 24.6 18.9 Source: Financial Stability Report, Central Bank of Brazil, various issues.
21.5
According to Soares (2002), before the implementation of the Real Plan in 1994, the proportion of credit in the banks' total asset was constant at around 44%, declining sharply to 33% in 1999. His explanation for this decline is the adoption of Basel I in 1994, which encouraged banks to move away from credits and towards acquiring government bonds, whose risk weight is 0% thereby making it easier to meet the minimum capital requirements determined by the new regulatory framework (see below for a further discussion of this point). The decline in the share of credit in banks’ total assets, in turn, can help explain why concentration trends measured in assets and credits diverged, as reported earlier. Thus, since the adoption of the Real plan, one can observe a rapid expansion of the banking system's total assets as a proportion of the GDP, but within that a fall in the participation of credit in total assets. That takes us to the central question: what has happened with total credit as a proportion of the country's GDP since the Real plan?
40
3.2. The evolution of credit in Brazil since the adoption of Basel I and before As of November 2003, total credit in Brazil as a proportion of GDP was at 26%. This level is fairly low when compared with other countries around the world. But how does that compare with the long-term levels of credit in the country, and how has it evolved over time? Tracing this information back to 1990, one can see a similar level of credit. But between these two points in time, it is also possible to detect an upward increase in credit between 1990 and 1994, from 24% to 36% of the country's total GDP, and since then a gradual declining trend down to 26%. Figure 1. Credit in Brazil 1990-2004 as a proportion of total GDP %
40
35
30
% GDP
25
20
15
10
5
0 1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Source: Central Bank of Brazil.
The decline in credit from 1994-1995 onwards reverses an upward trend until then. The reversal coincides with the adoption of the Real stabilisation plan, and is opposed to what analysts expected at the time. The expectations were that, with stabilisation, banks would lose a major revenue source derived from high inflation, and would thus expand credit as an alternative. Indeed, that is what happened in the first few months of the adoption of the Real plan. However and as described earlier, this process was aborted by the drastic monetary tightening adopted by the government in response of the Tequila crisis. Since then, the economy has been hit by a number of shocks - the Asian crisis, the Russian crisis, Brazil's devaluation of early 1999, and so on. In this context, credit really has never regained a path of sustained recovery. Given that credit declined between 1994-95 and 2004, the next question is: was this decline uniform across the banking system?
41
Turning to the composition of credit between the private and the public sectors, one will see that the share of credit by the public banks, which were already on a downward trend in the early 1990s, continued to decline, and quite sharply, from 62% in 1995 to 31% in 2004. Of course, a good deal of this decline simply reflects the fact many public banks were either closed or privatised during the period. At the same time, the share of credit by the private banks increased steadily, from 38% to 67% during the period. Figure 2: Public and private banks' share in total credit by the banking system 1990-2004 % 80.00
70.00
% total banking system
60.00
50.00
40.00
30.00
20.00
10.00
n91 ec -9 1 Ju n92 D ec -9 2 Ju n93 D ec -9 3 Ju n94 D ec -9 4 Ju n95 D ec -9 5 Ju n96 D ec -9 6 Ju n97 D ec -9 7 Ju n98 D ec -9 8 Ju n99 D ec -9 9 Ju n00 D ec -0 0 Ju n01 D ec -0 1 Ju n02 D ec -0 2 Ju n03 D ec -0 3 Ju n04 D ec -0 4 D
Ju
D
ec -9 0
0.00
Public Banks
Private Banks
Source: Cosif, Central Bank of Brazil
In the context of declining credit levels in proportion of the GDP, credit granted by public banks clearly declined, and quite sharply. But what happened to credit granted by the private banks?
42
Figure 3: Credit by Private and Public Banks as % GDP 1990-2004 25.0
20.0
% GDP
15.0
10.0
5.0
n91 ec -9 1 Ju n92 D ec -9 2 Ju n93 D ec -9 3 Ju n94 D ec -9 4 Ju n95 D ec -9 5 Ju n96 D ec -9 6 Ju n97 D ec -9 7 Ju n98 D ec -9 8 Ju n99 D ec -9 9 Ju n00 D ec -0 0 Ju n01 D ec -0 1 Ju n02 D ec -0 2 Ju n03 D ec -0 3 Ju n04 D ec -0 4 D
Ju
D
ec -9 0
0.0
Public Banks
Private Banks
Source: Cosif, Central Bank of Brazil
If we assume that the proportion of credit by the banking system remained more or less constant in relation to credit by the financial system as a whole, then we can cross information from Figures 1 and 2, and obtain information of credit by private and public banks as a proportion of total GDP. By doing so, we can see from Figure 3 that credit by private banks, which was on a steep increase until 1994, declined between 1994 and 1996 (from 14.6% to 12.8%), and then increased gradually until 2001,reaching nearly 20%, to fall slightly again until 2004, when it was at 17.3%. Overall, it went up between 1994 and 2004, although not very much. As seen earlier, the post-1994 was a very turbulent one in Brazil, not least because of the many shocks the economy was subject to. A key feature in this period has been the high real interest rates in a context of historically low inflation. This factor, together with a number of other constraints (see previous section) probably inhibited the expansion of credit in Brazil, which was expected to take place as a consequence of stabilisation. Moreover, it was a period of banking restructuring, in which the participation of the public banks in the banking system was drastically reduced. As a result, credit from this segment alone was drastically reduced. These factors altogether seem sufficiently important to explain why credit did not expand from 1994 onwards. But, as Soares (2002) argues, not only did credit not expand, a pre1994 expansion phase was interrupted in the post-1994 period. In his analysis, credit remained stable in real terms between 1994 and 1999, which is consistent with our data, which shows a decline, when measured as a proportion to a growing GDP. In the face of 43
these developments, what it seems is that a key factor in explaining why credit did not continue to expand as expected is the adoption of Basel I in September 1994, which was in itself very strict and turned even stricter in the subsequent years with additional regulatory measures adopted by the Central Bank of Brazil. Basel I was stricter than what was recommended by the Basel Committee, as initially applied and later amended between 1994 and 1999. First, a minimum absolute level of capital (which was higher than the prevailing ones) had to be observed by banks to be able to operate. Second, the minimum capital requirements to risk-weighted assets were set at 11% rather than 8%. Third, swap operations and market risk had to be included in the calculus of minimum capital requirement as well. Fourth, the risk weight for tax credit, initially set at 20%, went up to 300% (Soares, 2002). A good way to see if Basel I had a major impact on credit in Brazil is through looking at the banks’ credit to total assets ratio. Between 1994 and 1999 this ratio fell from 44% to 33% (Soares, 2002), which was a critical period of banks’ adjustment to the Basel rules. Moreover, looking at the broader period 1994-2004, Table 9 shows that total assets of the banking system increased from 71.1% to 87.4% of the country’s total GDP, a huge increase that contrasts drastically with the decline in credit for the period, from 36% to 26%. This means that the credit to total assets ratio fell from 50.6% to 29.7% (and from 73.5% if the year 1995 is taken as a base). Data on the Basel index of capital ratios are not available in aggregate terms for the 1990s. But for three of the largest five banks in Brazil – Banco do Brasil, Caixa Economica Federal and Unibanco – data are available both for 1995 – the first year of Basel in Brazil, and 2004. These are displayed in Table 11. The Table shows that an inverse relationship exists between the Basel index and the credit-total assets ratio for all the three banks. Table 11. The Basel Index and the credit-total assets ratio – selected banks % Banco do Brasil Basel Index Dec 1995 7.9 Sep 2004 15.6 Sources: Brazil’s Central Bank.
Credit to assets ratio 42.4 31.9
Caixa Economica Federal Basel Credit to Index assets ratio 9.8 48.8 18.4 18.7
Unibanco Basel Index 17.3 15.4
Credit to assets ratio 30.0 34.9
High real interest rates from 1995 onwards may partly explain the change in banks’ asset composition away from credit, and towards federal government bonds. But clearly, efforts to comply with Basel I has also certainly been an important factor in explaining the change in banks’ portfolio composition, as it induced banks to acquire risk-free government bonds, for which no capital is necessary for meeting the minimum capital requirements. The biggest effort of adjustment by banks to the Basel rules occurred mainly in the second half of the 1990s. For the largest five banks, the Basel index went from 9.8 in December 1995 to 12.03 in 1999, and then 13.6 in 2001.
44
For the whole banking system, the index, available from December 2001 onwards, showed further increases between then and December 2003, from 16.4 to 19.0 – see Figure 4, which displays the trend in the Basel index for the Brazilian banking system for this latter period, on a quarterly basis. But, unlike the previous period, the credit to total assets ratio for the banking system exhibited stability, remaining at around 30% over the 2001-2003 period –see Figure 5. Figure 4. The Basel Index in Brazil – Dec 2001-Dec 2003 25.0
Basel Index
20.0
15.0
10.0
5.0
0.0 Dez
Mar 2001
Jun
Set
Dez
2002
Mar
Jun
Set 2003
Years
Source: Central Bank of Brazil.
45
Dez
Figure 5. Credit to Assets Ratio for the Brazil’s Banking System 35
30
Credit-Assets Ratio
25
20
15
10
5
0 Dec- Jan- Feb- Mar- Apr- May- Jun- Jul- Aug- Sep- Oct- Nov- Dec- Jan- Feb- Mar- Apr- May- Jun- Jul- Aug- Sep- Oct- Nov- Dec01 02 02 02 02 02 02 02 02 02 02 02 02 03 03 03 03 03 03 03 03 03 03 03 03
Source: Central Bank of Brazil.
But were trends across banks uniform –that is, with portfolio composition remaining constant as capital requirements continued to go up, as the overall data suggests? To see if trends across banks were uniform or not regarding their portfolio composition over the later period, we tested for the correlation between the Basel index and the credit-total assets ratio for the 50 largest banks in 2001 for the Dec-2001-Sep 2004 period, for which data are available on a quarterly basis for individual banks Changes did occur, and the direction of change across banks was rather mixed, which conforms with the aggregate pattern of credit-total asset ratio stability, as trends in opposite directions probably have cancelled each other. But can we draw a coherent story by grouping the individual banks in broad banking categories, and thus looking at trends across these different categories? The answer is clearly positive. Among public banks, the correlation is positive for some, negative for others. Apparently, this was the case because some public banks succeeded in raising their capital requirements through government re-capitalisation. Among foreign banks, a mixed picture is also found; but most importantly, for the majority of private domestic banks, a negative correlation is found, which indicates that for this category of banks, further portfolio adjustments took place in response to their efforts to further increase their capital ratios. It would be interesting to extend the exercise to the totality of private banks, to see if a negative correlation indeed dominates. Moreover, further research involving a multi-variable econometric exercise would be desirable, to control for the effects of other factors affecting the banks’ portfolio composition. 46
3.3. Credit allocation across sectors: what has been the impact on the SMEs and the poor? In the context of overall credit decline between 1995 and 2004, how was total credit distributed across sectors? Which sectors increased their share in total credit, and which sectors have lost access to credit? Tables 12, 13 and 14 show respectively how total, private and public credit is distributed across different sectors of activities, and how distribution shares have evolved since 1994, when financial reforms started and Basel I was adopted. Table 12. Total Loans From the Financial System – Distribution by Sectors1 as percent % of total Public Sector
Private Sector Industry
Housing
Rural
Commercial Individuals Other Activities Services
Total
1994
15.2
22.4
21.2
9.4
11.4
8.3
12.1
84.8
1995
14.9
23.9
19.8
9.6
13.1
6.5
12.1
85.1
1996
17.9
23.9
19.0
7.5
11.2
8.9
11.6
82.1
1997
9.5
26.0
19.4
8.6
11.2
13.1
12.2
90.5
1998
7.6
26.2
19.6
9.0
9.2
12.6
15.7
92.4
1999
6.1
29.3
18.5
8.9
10.0
13.9
13.3
93.9
2000
3.9
26.8
17.5
8.5
9.9
19.0
14.4
96.1
2001
3.0
29.7
7.2
7.9
10.9
23.3
18.0
97.0
2002
3.6
30.6
6.4
9.2
10.5
21.6
18.2
96.4
2003
3.7
28.5
6.1
10.9
10.5
23.0
17.3
96.3
20042
4.1
26.1
5.4
11.0
11.2
25.2
16.9
95.9
Source: Central Bank of Brazil. 1 December - balance end of period. 2 October 2004.
47
Table 13. Loans From the Private Financial System – Distribution by Sectors1 as percent % of total Public Sector
Private Sector Industry
Housing
Rural
Commercial Individuals Other Activities Services
Total
1994
2.0
29.4
12.1
2.9
18.9
13.9
20.8
98.0
1995
3.0
33.4
10.2
2.7
21.9
9.8
19.1
97.0
1996
2.6
34.7
8.8
3.8
19.3
14.5
16.3
97.4
1997
1.8
33.1
7.1
4.5
17.5
20.5
15.6
98.2
1998
1.2
33.2
7.4
4.3
15.1
21.9
16.9
98.8
1999
0.9
34.0
6.4
3.8
15.6
21.0
18.3
99.1
2000
1.0
29.5
5.6
4.5
14.3
25.7
19.4
99.0
2001
0.9
28.9
4.3
4.8
13.2
27.7
20.2
99.1
2002
1.1
28.7
3.6
6.1
13.4
27.0
20.1
98.9
2003
0.8
27.6
3.4
6.7
13.6
29.8
18.1
99.2
20042
1.2
25.9
2.8
7.4
14.3
31.8
16.7
98.8
Source: Central Bank of Brazil. 1 December - balance end of period. 2 October 2004.
It can be seen from Table 12 that the share of total credit from the financial system to the public sector fell dramatically between 1994 and 2004 – from 15% to 4%, while the share of credit flowing to the private sector increased from 85% to 96% over the same period. The decline in the share of credit to the public sector, in the context of overall decline in credit as a percentage of the GDP, can be explained mainly by the reduction of public banks, which were the main lenders to the public sector, which constrained the ability of governments, especially at the sub-national levels, to borrow from the financial system. This happened along with new fiscal rules constraining the state-level governments capacity to borrow. The latter is evidenced by the fact that lending to the public sector by private banks also declined during the period – see Table 14. The decline is also reflecting the fact that a good deal of public enterprises was privatised. The increase in the share of credit to the private sector benefited mostly individuals (e.g., consumer credit, others) and other services (telecommunications, transport, education and culture, press, informatics) with their shares in total credit going up from 8% to 25 and 12% to 17%, respectively. Other sectors whose shares in total credit increased were the industrial (which may be a statistical effect due to privatisation) and rural sectors. Credit to housing, which refers mainly to mortgage lending, fell sharply. These trends are broadly similar between private and public credit, with the big difference being that whilst the share of private credit to industry and other services declined, the share of public credit to these activities went up. But the latter may just be
48
reflecting a statistical effect, as big state-owned companies that fall under these categories were privatised. It is difficult to say whether the SMEs and the poor benefited or not from these trends. But it is possible to cautiously suggest that the increase in the share of credit to the rural sector and to individuals may have benefited small rural producers, and reached less wealthy individuals. At the same time, productive and commercial activities lost financing from the private banks, and this may have harmed mostly the SMEs. So, what could be said is that the redistribution of credit across sectors seems to have been in the form of a slight shift from productive and commercial activities to consumer credit. Thus, a preliminary hypothesis is that, whilst Basel I together with broader financial sector reforms have affected the level of total credit negatively through requiring banks to raise capital to meet the Basel rules for capital requirements, the redistribution of credit across sectors may not have discriminated against the less favoured households, though it may have affected the SMEs negatively. Table 14. Loans From the Public Financial System – Distribution by Sectors1 as percent % of total Public Sector
Private Sector Industry
Housing
Rural
Commercial Individuals Other Activities Services
Total
1994
25.3
17.0
28.2
14.4
5.6
4.0
5.5
74.7
1995
24.6
16.3
27.6
15.1
6.1
3.9
6.4
75.4
1996
30.1
15.3
27.2
10.4
4.8
4.5
7.7
69.9
1997
16.7
19.4
31.0
12.5
5.3
6.1
9.0
83.3
1998
12.8
20.6
29.4
12.8
4.5
5.1
14.8
87.2
1999
11.1
24.7
30.3
13.9
4.5
7.1
8.5
88.9
2000
7.7
23.3
32.8
13.6
4.3
10.3
8.0
92.3
2001
6.9
31.2
12.6
13.6
6.7
15.0
14.1
93.1
2002
7.6
33.7
10.8
14.2
5.8
12.8
15.1
92.4
2003
7.8
29.6
10.1
17.1
6.0
13.1
16.3
92.2
20042
8.8
26.6
9.5
16.6
6.5
14.9
17.2
91.2
Source: Central Bank of Brazil. 1 December - balance end of period. 2 October 2004.
49
3.4. The role of directed credit as a countervailing force Whilst it is difficult to assert how new credit distribution patterns affected the SMEs and the poor, one clear fact is that the changing patterns closely follow what happened to directed credit in Brazil. This sort of credit accounts for a large part of total credit. In the 2000-2004 period, it varied between 36% and 44%; if directed credit to the housing sector is excluded, it stayed around 30%. Table 15. Directed Credit as a Proportion of Total Credit – 2000-2004 % Housing
1
BNDES
Rural
Other
Total
Total minus Housing
Dec-2000
15.7
8.5
17.8
2.0
44.0
28.3
Dec-2001
6.4
7.9
19.7
1.6
35.6
29.2
Dec-2002
5.7
9.2
22.4
0.8
38.1
32.3
Dec-2003
5.6
11.0
22.2
1.0
39.8
34.2
Oct-2004
5.1
11.0
20.0
0.8
36.8
31.7
Source: Central Bank of Brazil.1Includes both direct resources and resources passed to other banks.
As can be seen from Table 15, directed credit explains why total credit to the Housing sector fell so dramatically between 2000 and 2004, and why rural credit expanded. Indeed, during the period, it contributed to nearly 90% of the fall in the Housing sector credit, and 100% of the increase in rural credit. Also, the BNDES, which figures as the largest development bank in Brazil, accounting for about 20% of total directed credit in the country 24 , may well explain why credit from the public sector to the industrial sector expanded between 2000 and 2004, thus largely offsetting the decline in credit to the sector by private banks (see Tables 12-14). 25 For the purpose of this study whose focus is on SMEs and the poor, it is important to notice that the BNDEs traditionally lends to large projects and companies, not the SMEs (although resources it lends through other banks and development agencies may reach the SMEs). So, where credit may have benefited the SMEs and/or the poor, this was due to directed credit. For example, there is no directed credit in Brazil to the SMEs, which therefore may have lost (apart from directed credit to micro-businesses implemented in 2003, which may have reached small business, but mainly through credit to individuals – see below); where the poor may have gained, for example small rural producers through the expansion of rural credit, this was due to increases in directed rural credit.
24
This includes both direct resources provided by the BNDEs and those resources the bank distributes via other banks. Direct resources account for 48% of total resources managed by the bank in October 2004. 25 Credit by the BNDES accounted for 12.2% of total credit in Brazil in December 2004. If to that we add credit by the other four major development banks – BNB, BASA, BRDE and BDMG – the proportion of credit provided by Brazil's development banks in total credit goes up to 13.8%.
50
But has the increase in the share of rural credit been accompanied by more credit to small producers? It is hard to gauge this, due to lack of available data. But information on credit by size of credit, which bears some correlation with size of the rural producer, suggests that, at least for the 2001-2003, no discernible change took place. This is true both for total and free rural credit – see Table 16. The only noticeable change is a relative decline in credit to small producers through PRONAF (National Programme for Strengthening Family-Based Agriculture) – from 15% of total rural credit in 1999 to 10% in 2003. Table 16. Total and Free Rural Credit, by Size of Loans1 Credit as a percent % of total credit I
II
III
IV
2001
40.9
21.3
7
30.8
2002
35.8
25.15
8.9
30.1
2003
42.7
15.8
11.2
30.3
Free credit as a percent % of total credit 2001
2.6
0.8
0.3
0.3
2002
2.6
1.1
0.4
1.1
2003
3.1
0.7
0.4
1.1
Source: Central Bank of Brazil.1 Category I, II, III and IV correspond respectively to the ranges 0 to 40.000 Reais; 40.000 to 150.000 Reais; 150.000 to 300.000 Reais; and above 300.000 Reais; for the year 2003, the value 40.000 is increased to 60.000.
The main message is that directed credit in Brazil is a powerful factor in determining credit patterns in the country, and may have had an important countervailing role to credit decline as a result of financial reforms and Basel I. Moreover, although the credit share by public banks (including development banks) has been drastically reduced, they still seemed to have a crucial role in credit provision for productive urban sectors and rural activities. So, what it seems is that financial reforms and Basel I did not have a major impact on credit allocation in ways that harmed the poor or the SMEs – at least not in in a major way – due the maintenance of two key institutional factors that have historically strongly featured in Brazil's financial system: large public and development banks, and directed credit. Whilst the presence of public banks in Brazil has been downscaled, with no plans to change it, directed credit remains as seen by the government as an important instrument for credit promotion and allocation towards the less favoured segments. In what follows, we will briefly discuss recent initiatives that have been undertaken in Brazil in this area.
51
Chapter 4. New Initiatives in Directed Credit and other Financing Mechanisms This chapter looks at the recent initiatives undertaken by the Brazilian government in support of micro-businesses and the poor in Brazil. It then asks the questions: why is it so important to provide micro-credit in Brazil? What has been the access to micro-credit by the country’s micro and small enterprises (MSEs). Finally, it looks at ongoing experiences of micro-credit conducted by the country’s development banks and agencies. 4.1. Recent government initiatives in support of micro-business and the poor The Brazilian government takes the view that credit expansion to the less favoured segments of the population, and in support of productive activities, can play a key role in helping promote social inclusion and sustainable growth. 26 Three vectors are identified for the expansion of credit in Brazil: the micro-credit, the credit cooperatives and the provision of banking services in non-banking commercial establishments (the so-called correspondentes bancarios), such as post offices, lottery houses and supermarket chains. The micro-credit is targeted at individuals and micro-business; the co-operatives to benefit its associates, mainly in the rural sector; and the correspondente bancario is aimed at providing financial services in regions and locations that do not have access to such services. A number of initiatives have already been undertaken in these areas. The government created the National Plan of Micro-credit in June 2003, whereby 2% of banks' deposits have to be directed to this type of credit. 27 The aim is to provide bank loans to 6 million of micro-entrepreneurs who at present do not have access to banking services. In addition, the consigned credit has been created as well.28 It essentially means a credit provision for workers in the formal sector, with charges being incurred directly on their payrolls. Since its creation, credit through this latter instrument has reached R$ 13.5 billion, accounting for 37% of total credit to individuals, with interest rates much lower than other credit modalities, due to the low risk lenders face. As regards provision of banking services through the correspondentes bancarios, it is not guaranteed that the provision of such services will at some point include bank loans, or that such loans will be for productive purposes. To address these shortcomings, a number of initiatives to promote lending to productive activities are under analysis in the National Congress. 29 26
The position of the Brazilian government in relation to credit and the role it can play in fostering growth and social inclusion were articulated in two recent documents: Ministry of Finance (2003) 'Economic Policy and Structural Reforms [Politica Economica e Reformas Estruturais], April, and Ministry of Finance (2004) 'Microeconomic Reforms and Long-term Growth' [Reformas Microeconomicas e Crescimento de Longo Prazo]. 27 Law No. 10.735/2003 and CMN Resolution No. 3.109/2003. 28 Law No. 10.280/2003. 29 These include: MP no. 226.2004, which concerns the implementation of the National Programme of Micro-credit Oriented to Productive Activities [Programa Nacional de Microcredito Produtivo Orientado],
52
Initiatives regarding the Correspondente Bancario include the use for the provision of banking services of lottery houses by the public bank Caixa Economica Federal; the right of using post offices throughout the country, gained by the private bank Bradesco through a tender; and the agreement between the supermarket chain Pao de Acucar and the private bank Itau to use the supermarkets' many superstores and other establishments for the provision of banking services as well. The outreach of such initiatives can be illustrated by the case of Caixa Economica Federal, which has a present 13 thousand points spread throughout the country, with 9 thousand of which based on lottery houses, and 3.7 thousand on commercial establishments of different kinds. Between August 2003 and January 2004, 1.1 million bank accounts were opened through this scheme, with 85% of account holders having monthly income of less than R$ 500 (Carvalho and Abramovay, 2004). The sorts of financial services provided by the bank through the lottery houses include payments of bills, pension payments and FGTS (a compulsory fund that workers accumulate while working and that can be withdrawn under certain circumstances, such as unemployment, and acquisition of their own house). The expectations are that these services will be extended to include credit cards and financing. 4.2. Why are initiatives to provide micro-credit so important for Brazil? Access to microcredit by MSEs Initiatives as those described above are important, particularly the ones geared towards micro-businesses, given the latter's limited access to banking credit. Research conducted in 1999 and 2004 by Sebrae – Servico Brasileiro de Apoio a Micro e Pequena Empresa, which is an entity that provides services in support of Micro and Small Enterprises (MSEs), shows that only 37% of micro-business and 55% of small business have ever had access to bank credit (Sebrae, 2004). It also shows that only 6% of such enterprises drew on bank credit when setting up their business (Bede, 2004). The research covers an universe of 450 enterprises of the formal sector, based in the State of Sao Paulo. 30 But access to bank lending has been gradually improving in the past few years. Access by MSEs to credit from public and private banks went up from 3% and 5% respectively in 2000 to 12% and 10% in 2004. 31 It is crucial that credit to MSEs is expanded. The universe of MSEs in Brazil amounts to about 14 million of establishments in industry, commerce and services, of which 9.5 million – or nearly 70% - is in the informal sector. 32 with funding from the Fundo de Amparo ao Trabalhador (FAT), and resources from the 2% bank deposits for micro-credit; complementary MP no. 210/2004, which concerns the provision of simplified tax, social protection and labour treatment for small entrepreneurs with revenues up to R$ 36 thousand; the use of mediation as an alternative mechanism for resolving conflicts between the lending and borrowing parties; and improvement of tax legislation for Micro and Small Enterprises (MSEs). 30
Sebrae defines micro-enterprises as those activities with annual gross revenues up to R$ 244 thousand and with up 19 employees in the industrial sector and 9 in commerce and services. Small enterprises have an annual gross revenue of up to R$ 1.2 million, with between 20 and 99 employees in the industrial sector and between 10 and 49 employees in commerce and services. The universe of micro and small enterprises accounts for 95% of industrial enterprises in Brazil; 98% of commercial establishments, and 99% of establishments operating in services (Sodre, 2000). 31 In 2004, the main form of financing was through supplier credit, used by 66% of MSEs. 32 Although less numerous, the formal MSEs account for 20% of Brazil's GDP and employ about 56% of the total labour in the formal sector (Revista Rumos, mar/abril 2004).
53
If household agricultural establishments are added, the total figure goes up to 16 million (Revista Rumos, March/April 2004). The figures on access to credit just provided refer to the formal MSEs. For the informal MSEs, access to bank credit is even more limited. Moreover, the figures provided are based on access to bank lending by formal MSEs from the wealthiest state in Brazil. For other states, the numbers should be even lower. A main reason for their limited access to banking credit is lack of real guarantees. The result is that banks end up lending not to business but to individuals (Carvalho and Abramovay, 2004). This suggests the numbers above probably underestimate access to bank lending by MSEs, as part of it takes place through lending to the MSEs owners, as individuals. The majority of the MSEs – 53% of the total - identified a decline in interest rates as the main factor for making access to bank credit easier; other 29% pointed to reduction in paperwork as a main factor to facilitate access to credit (Bede, 2004, drawing on Sebrae research). In face of these facts, a clear policy recommendation that emerges is the desirability to incorporate the MSEs into the formal sector. This would bring two benefits. One would be to improve information availability on such business activities, and the second would be to make it easier for these to provide real guarantees. On the other hand, given that still a large portion of the business activities will remain outside of the formal sector, it shows that there is a clear limit to which information availability can be improved, so that banks can increase lending. This also shows that one recent government initiative – of encouraging positive credit information on borrowers, may benefit only those borrowers who have already had access to the banking system. The reality in Brazil is that the financial system does not have a sufficiently large range of products to support the MSEs, most of which are represented by self-employed workers, who, in the words of Carvalho and Abramovay, account for 'the single largest segment of the poor in the country'. Of the 9.5 million of informal MSEs in Brazil, 7.5 million employ just one person. Of this sub-set, only 472 thousand – less than 5% - have access to some form of credit, and less than half of these, have access to bank credit (Carvalho and Abramovay, 2004). 33 In the view of Carvalho and Abramovay (2004), lack of access to credit by the MSEs reflects adverse selection and asymmetry of information. The role of these factors in barring access to credit is magnified in a country characterised by high income and educational inequalities. The authors believe that support towards the small business (through enabling them to have access to bank financing) is among the most important policy actions to combat poverty in Brazil. Lack of banking products to support micro-business activities is particularly the case among the urban based micro-business activities. In the rural sector, a few programmes exist to support small producers, the most prominent of which is the PRONAF, managed by the Banco do Brasil, as mentioned above. In what follows, we briefly report how 33
Figures on the self-employed are drawn from the IBGE research 'Economia Informal Urbana', on nonrural business activities undertaken by self-employed workers, conducted in 1997.
54
micro-credit has been practised in Brazil, as well as discuss two micro-credit schemes that have proved successful in the recent past. 4.3. The ongoing experiences of Micro-Credit in Brazil Micro-credit has been dominated by NGOs in Brazil. Even the BNDES, the largest Brazil's development bank, has been involved with micro-credit programmes mainly indirectly, through transferring resources to NGOs. That is, until 2002 the BNDEs focus, in terms of micro-credit operations, had been on supporting micro-finance institutions set up by NGOs and associations of credit for micro-entrepreneurs [Sociedades de Credito ao Microempreendedor (SCMs)]. In 2003, the bank tried to strengthen support to networks for the entrepreneurship, and at present it emphasises regional and local development through supporting the so-called local productive clusters – arranjos produtivos locais (APLs). 34 According to data from the BNDES, there is a total of 121 entities in Brazil dealing with micro-credit. These are public agencies, NGOs and the SMCs. Apparently reflecting a sub-set of the universe of entities linked to micro-credit identified by the BNDEs, data from Ibam show that micro-credit institutions in Brazil grew from just 2 in 1984 to 56 in 2002. Only recently has micro-credit become part of commercial banks' portfolio of products, as a result of a number of initiatives such as the above mentioned micro-credit programme launched by the Federal Government, and the creation of the Banco Popular, which is a Banco do Brasil's subsidiary to provide banking services to the poorer segments of the population. But the most carefully designed micro-credit schemes are those created by regional development banks and the newly created development agencies at the state level. Two of the most prominent ones - CrediAmigo and Nossocredito – are described below in Box 4.
34
To support these various activities, the BNDEs has specific micro-credit schemes, such as the BNDES Solidario and the BNDES Trabalhador. In 1996, it created the Programa de Credito Produtivo Popular (PCPP), which exists already in 17 States and 492 cities across the country; and in launched in collaboration with the IADB the Institutional Development Programme [Programa de Desenvolvimento Institucional (PDI)], aimed at raising NGO and SCMs' capacitiy in microfinance operations.
55
Box 4. Micro-credit schemes in Brazil – The CrediAmigo and Nossocredito 1) The CrediAmigo The CreditAmigo is a micro-credit scheme created by the BNB – the regional development bank based in Northeastern Brazil – in 1998. The scheme was designed to reach micro-entrepreneurs form the Northeastern region of Brazil, plus the northern part of the States of Minas Gerais and Espirito Santo. The scheme has three credit lines: (i) working capital for Solidary Groups, consisting of loans to groups comprising between 3 and 10 micro-entrepreneurs; (ii) working capital for individual entrepreneurs; and (iii) credit for fixed investment. In the first case, credit is provided in a gradual fashion, starting with values between R$ 500 and R$ 2000, but with the possibility of going up in subsequent operations to R$ 8.000 for each borrower. Reimbursements can be on a weekly, fortnight or monthly basis, with no grace period and with maturity between one and six months. Interest rates are proportional to maturity and frequency of payments, being at 2% a month for values up to R$ 1.000. Credit for fixed investment has a ceiling of R$ 3.000, with monthly reimbursements, without grace, and maturity between 1 and 18 months, with interest rates of 2.5% a month. The typical borrowers are mainly owners of micro-business in the informal sector of the economy. Examples include sale men in commerce; owners of shoe-making and garment firms; beauty saloons, car repairs, carpentry; and service providers. Commercial activities account to 91% of the resources provided by the scheme; industry accounts for 4%; and services for 5%. Fifty percent of the total borrowers are either illiterate or have primary school incomplete. Only 2% have a university degree. Of the universe of borrowers, 62% have household income of less than R$ 1,000; and 36% less than R$ 600. The scheme covers 1.172 towns across the North-eastern Brazil and the northern most part of Minas Gerais and Espirito Santo. The scheme has at present 166 agencies plus 44 reception points; it has been making major strides mainly in the interior areas due to the higher levels socialisation and solidarity among borrowers in these areas, but also due to lack of alternative credit sources. In December 2003, the programme had 138.5 thousand borrowers, and it was predicted 2004 would end with 194 thousand borrowers. The active portfolio of the scheme increased from 8.2 million in 1998 to 85.5 million in 2003, and was projected to reach 122.5 million in 2004. 2) Nossocredito Nossocredito is also a micro-credit scheme created by the Governor State of Espirito Santo in 2003. The programme started in 4 municipalities and is expected to end 2004 covering 13 towns. The scheme is aimed at supporting micro and small entrepreneurs of the State of Espirito Santo. There are 209 thousand of them across the state, of which only 6% have ever had access to credit. Between December 2003 and July 2004, the programme had 247 credit operations, of which 205 were to the informal sector of the economy. The programme provides credit varying between R$ 200 and R$ 5,000; for cooperatives and associations, the total value can reach R$ 25,000. Interest rates are 1% a month, with maturity for working capital up to 6 months, and for fixed investment, up to 12 months. The scheme is operated through credit agents, who are nominated by the council towns, and who are charged with searching for potential borrowers, assisting them on how to use the resources, and follow the implementation of the projects. The scheme is conducted in collaboration with the council towns, through Bandes (a development agency of Espirito Santo State) and Banestes, a state-level public bank. The State government has created a fund called Fundapsocial (Fundo de Desenvolvimento das Atividades Portuarias) to be managed by the Bandes, aiming to provide R$ 3.5 million a year to the programme (Conde, 2004).
What can be grasped from these schemes is that they aim at targeting not just the poor but the poor who have their own business, and to reach mainly the informal sector of the economy, in those localities with little access to banking services and banking credit. 56
These initiatives are encouraging, but the scale of resources they involve, although growing, are still very little as a proportion of total credit in Brazil. More broadly, the micro-credit initiatives by the government are welcome, including efforts to encourage banks to have outposts in remote localities and in the peripheries of the large cities. However, provision of banking services in itself does not constitute credit extension. One should wonder whether banks can accumulate knowledge and build a long-term relationship with customers, that can then evolve into a lender-borrower relationship, as usually happens with bank agencies located in small towns. The point is whether a credit culture and practices can be easily developed through the correspondente bancario form of operating, or conventional banking presence is still required for credit operations to take place. Particularly in relation to the national micro-credit programme, which aims at directing 2% of total deposits to the poor, scepticism was expressed by our interviewees whether private banks would really use it as a starting point to penetrate the SMEs market, or whether they would simply deposit the resources with the Central Bank. The question thus still remains of how to make mainstream lending more widespread across income groups, and how to ensure the new regulatory framework for capital adequacy does not work as a limiting force for the expansion of credit in Brazil, especially to the SMEs and the poor.
57
5. Conclusions and Policy recommendations The Brazilian government is taking a number of initiatives to provide banking services to larger segments of the country's population, and credit to micro-business. These initiatives are welcome in light of the reduced levels of credit in Brazil, and to counteract possible negative effects on credit expansion of the new capital accord. But we hold the view that the new regulatory framework for the banking system should be better aligned with the governments’ policy aims. Our assessment is that this is not the case at present. The New Basel rules, as Brazil’s regulators intend to apply in the country, may have at least three effects that can affect credit to the SMEs and the poor negatively: further banking concentration, banking portfolio concentration away from the SMEs, and increased credit pro-cyclicality. Our research shows that these possible effects are not part of the concern of Brazilian regulators. The banking community, in its turn, is too busy in its efforts to be prepared for the new rules when these come into effect in early 2007, and therefore is not addressing these issues either. But it is important to do so. Looking at the current capital rules (or Basel I) as adopted in Brazil, it was possible to see that these rules have contributed to the banking concentration observed in the late 1990s and early this century, to a sharp decline in the share of credit in banks’ portfolio of assets – and related to that, to a decline in credit as a proportion of total GDP in the ten years since the mid-1990s. The India study undertaken under the same project shows that, under different economic circumstances, there too credit share in banks’ total assets has declined as a result of Basel I; it moreover shows that credit to small enterprises declined sharply relative to total credit. The effects of Basel I both in Brazil and India thus clearly demonstrate that changes in the regulatory framework for banks can have important effects on the structure of the banking system and on credit patterns. It is therefore important to avoid a repeat of the negative consequences that often accompany the introduction of new banking rules. Particularly at a time international efforts are being made to reduce poverty worldwide, it is important to raise awareness – and encourage the debate on the possible negative implications of the new capital rules – or Basel II – which will come into effect in early 2007, for the SMEs and the poor. The debate could help create a consensus around measures that could be implemented to address the shortcomings of the new rules; in particular, measures that can help remove or at least reduce the potential bias of such rules against credit, especially to the neediest segments. Since 1999-2000, when discussions on the initial proposals for a new capital rules started, a number of ideas have emerged on how to mitigate the possible negative impacts of the new rules on credit patterns and pro-cyclicality. For example, a menu of options exist on how to reduce the pro-cyclicality of credit, which developing countries in particular should worry about, due to the fact that economic volatility in these countries are higher than in the OECD countries. Of course, there are some technical challenges associated with each proposal, which require careful examination. Nonetheless, the options exist and
58
are feasible. What really seems to be missing is lack of political initiative, which can be partly explained by a lack of debate on these issues, as this study on Brazil suggests. In what follows, we refer to a few possible measures for adoption put forward and discussed in international academic and policy circles, which could address the issues raised in this study. •
To address inequity arising from the use of the IRB approach by the large banks and the standardised approach by the remaining banks, an equalising factor could be applied over the banks adopting the IRB approach, so as to level up their capital requirements. That would address inequity issues, and could have the additional benefit of discouraging banks from changing their portfolios away from smaller borrowers, typically the ones deemed as riskier. That would be moreover consistent with the Basel Committee's primary intention to address relative rather than absolute risk.
•
To address portfolio concentration, in addition to the application of a factor as proposed above, regulators could work on a formula to smooth the risk curve for SMEs, as the Basel Committee has done in the past between the Consultative Papers CP2 and CP3. As said earlier, this could be done in a number of ways, but would be to the country’s regulators to choose which method might be the most appropriate one; the decision could be based on technical studies to assess the impacts of alternative measures on credit to the SMEs.
•
To deal with pro-cyclicality, the smoothing of the risk curve proposed above would be beneficial for that purpose. In addition, other counter-cyclical measures could be adopted, such as to encourage the use by banks of different models (Persaud, 2000), and the use of models that ‘look through the cycle’, as opposed to the most utilised models that look at one point of the cycle. A further measure would be to reward portfolio diversification. The reason for the latter is that, in addition to reducing risk for a given level of return (which is why diversified portfolios are desirable in the first place), portfolio diversification could contribute to reduced credit pro-cyclicality (IADB, 2004; Griffith-Jones, Spratt and Segoviano, 2004). This is because a negative event would affect only that part of a bank's portfolio that share similar characteristics and therefore is vulnerable to the same types of shocks, not the entire portfolio.
The measures proposed thus far address the three issues this paper has highlighted as key ones that are receiving little attention in Brazil. In addition to these, a number of other issues that have been raised in this report constitute a major challenge for Brazil’s regulatory authorities. For example, will the regulators have the capacity to validate models and monitor them adequately within the proposed time frame? Is the timetable proposed by Brazil's regulators long enough? Should regulators not need more time to be able to adequately validate and monitor risk assessment systems adopted by banks, especially those that will opt for the most advanced models? Should the proposal for adopting internal models for measuring operational risk not be eliminated, given the sheer complexity of measuring operational risk and the difficulties regulators would face to monitor their use? And in the
59
case it is adopted, could a factor not be equally employed to avoid that some banks end up with lower capital requirements for operational risk than others? Also, it would be important that the regulatory authorities could take account of the fact that the risk management practices should be effective, but not excessively intrusive to the point of inhibiting lending activities and programmes that have a social purpose. This study on Brazil thus shows that implementation of Basel rules poses a number of challenges to national banking regulators, and to the country at large. But what lessons can we learn from the Brazilian study for other developing countries? The Brazil study shows that the Basel rules are not neutral, and this should be borne in mind when a country is considering adopting such rules. In this regard, it is important that measures being considered for adoption are carefully examined, and that their implications for development finance are identified and properly addressed. But the Brazil study also shows that, had the country not had its development finance architecture in place, the impacts of Basel I on development finance would have been far bigger. For example, although Basel I did affect credit in Brazil, there is no evidence that the credit to the SMEs, to rural producers or to the urban poor was negatively affected, at least in a major way. A main reason for this outcome is that credit patterns during the period under Basel I have been influenced by directed credit policy, which in a number of cases were intended to protect the less favoured segments. From this, the lesson we can draw for other developing countries is that institutions that support development finance are key and should therefore be preserved, as there is nothing indicating that an entirely market based banking system will serve the financing needs of the small businesses and the poor. This is even more so under Basel I and especially Basel II, as the latter has a clear bias against perceived higher risk borrowers, which usually are the small businesses and the poor. But many poor countries do not even have development finance institutions, in a number of cases because they have reformed their banking systems and in the process dismantled such institutions. The lack of such institutions makes prudence towards the adoption of Basel rules even more necessary for these countries. One should not forget that capital markets in poor countries are still very small and that the banking sector is still the major source of finance to the economy. Of course, one may contend that micro-finance in poor countries managed by foreign NGOs and other organisations have had an important role in providing resources to the small businesses and the poor. However, mainstream finance should also be able to reach these segments, and to finance projects (large or otherwise) that can benefit them indirectly as well. It is thus important that the system is regulated in ways that it can serve both the economy and the most needy as well.
60
Bibliography Barrel, R. and Gottschalk, S. (2005) 'The Impacts of Capital Adequacy Requirements on Emerging Markets: The Cases of Brazil and Mexico', National Institute of Economic and Social Research (NIESR), June, London, unpublished. Basel (2004) 'Implementation of Basel II: Practical Considerations', Basel Committee on Banking Supervision, July. Basel (1997) ‘Core Principles for Effective Banking Supervision’, Basel Committee on Banking Supervision’, Basel, September. Basel (1998) ‘International Convergence of Capital Measurement and Capital Standards (July 1988, updated to April 1998)’, Basel Committee on Banking Supervision, Basel, April. Basel (1995) ‘Planned Supplement to the Capital Accord to Incorporate Market Risks’, Consultative proposal by the Basel Committee on Banking Supervision, April. Bedê, M.A. (2004) ‘Gargalos no financiamento dos pequenos negócios no Brasil’. In: Sistema Financeiro e as Micro e Pequenas Empresas:Diagnósticos e Perspectivas, Brasília, Sebrae. Borio, C., Furfine, C. and Lowe, P. (2003) ‘Procyclicality of the financial system and financial stability: issues and policy options’, BIS Papers No 1. Carneiro, F, Vivan, G., e Krause, K. (2004) 'Novo Acordo da Basiléia: Estudo de Caso para o Contexto Brasileiro', Resenha BM& F, no 63, São Paulo. Carvalho, C.E. e Abramovay, R. (2004) ‘O Difícil e Custoso Acesso ao Sistema Financeiro’. In: Sistema Financeiro e as Micro e Pequenas Empresas: Diagnósticos e Perspectivas, Brasília, Sebrae. Carvalho, C. E., Studart, R. and Alves Jr., A. J. (without date) ‘Desnacionalizacao do Setor Bancario e Financiamento das Empresas: A Experiencia Brasileira Recente’, Estudo Cepal-IPEA. Conde, W. (2004) ‘A vez dos pequenos negócios no Espírito Santo’, Revista Rumos, ano 28, n. 216, jul/ago. Danielsson, J., Embrechts, P., Goodhart, C., Keating, C., Muennich, F., Renault, O. and S hin, H.S. (2001) ‘An academic response to Basel II’, Special Paper 130, London: LSE Financial Markets Group. Goodhart, C.A.E. (2004) ‘Some Reflections on Financial Stability’, Lecture presented at the Central Bank of the Czech Republic, Prague, November.
61
Goodhart, C. and Segoviano, M. (2005) 'Basel and Procyclicality: A Comparison of the Standardised and IRB Approaches to an Improved Credit Risk Method', LSE, London, unpublished. Gottschalk, R. (2001) ‘A Brazilian Perspective on Reform of the International Financial Architecture’, report prepared for DFID, July. Griffith-Jones, S. (2003) ‘How to Prevent the New Basel Capital Accord Harming Developing Countries’, paper presented at the IMF-World Bank Annual Meetings at Dubai, September. Griffith-Jones, S., Spratt and Segoviano, M. (2004) ‘CAD 3 and Developing Countries: The Potential Impact of Diversification Effects on International Lending Patterns and Pro-cyclicality’, IDS, unpublished. IADB (2005) 'Unlocking Credit: The Quest for Deep and Stable Banking Lending', Economic and Social Progress in Latin America and The Caribbean', 2005 Report, Banco Inter-Americano de Desenvolvimento, 281 paginas, Washington DC. IIF (2004) ‘Corporate Governance in Brazil: An Investor Perspective’, Task Force Report, June.1 Persaud, A. (2000) ‘The disturbing interaction between the madness of crowds and the risk management of banks’, paper presented at the Commonwealth Secretariat conference on ‘Developing Countries and Global Financial Architecture’, London, 22-23, June. Revista Rumos (2004), ano 28, n. 214, março/abril. Rodriguez de Paula, L. F. (1998) ‘Tamanho, Dimensao e Concentracao do Sistema Bancário no Contexto de Alta e Baixa Inflacao no Brasil’, Revista Nova Economia, V. 8, No 1, pp. 87-116, Jul-Dec. Sebrae (2004) ‘Sistema Financeiro e as Micro e Pequenas Empresas:Diagnósticos e Perspectivas’, Brasília. Stiglitz, J. (1993) ‘The Role of the State in Financial Markets’, Proceedings of the World Bank Annual Conference on Development Economics, pp. 19-52. Stiglitz, J. and Weiss, A. (1981) ‘Credit Rationing in Markets with Imperfect Information’, The American Economic Review, Vol. 71, No. 3, June, pp. 393-410. Soares, R.P. (2002) ‘Evolução do Crédito de 1994 a 1999: Uma Explicação’, Planejamento e Políticas Públicas, n. 25, Jun/Dez., pp. 43-87. Troster, R. L. (2004) ‘Concentracao Bancária’, Febraban, unpublished.
62
Annex 1. Table A.1. Basel I in Brazil: risk weights for different categories of assets Weight Loans to/Investment in: 0%
• • • • • • •
Brazil Central Government’s bonds Foreign currencies deposited with the Central Bank Compulsory deposits with the Central Bank 20% Bank deposits in other banks Gold Deposits and credits in foreign currencies Tax related credits (then raised to 300% in August 1999, through Circular no. 2916). 50% • Government bonds outside the Central Government • Inter-bank deposits with own resources • Foreign currencies abroad • Mortgages 100% • Private bonds with own resources • Investments in variable income assets • Investments in commodities • Operations linked to stock exchanges and future markets • Exchange operations • Diverse credits Source: Annex IV, Resolution No. 2.099 of 17/08/1994. Available at: www.bcb.gov.br
63
64
