Francisco Carlos Fernandes Paulista School of Politics, Economics and Business Federal University of São Paulo (UNIFESP), Brazil.

Transcription

1 European Journal of Scientific Research ISSN X / X Vol. 148 No 1 December, 2017, pp europeanjournalofscientificresearch.com The Subprime Crisis and Reflections in Credit Risk Management of Brazilian Banks: A Comparative Study in the Largest Public and Private Banks in Brazil Júlio César da Silva Department of Business Administration Regional University of Blumenau (FURB), Blumenau, Brazil profjuliosilva72@gmail.com Silvio Aparecido Teixeira Accounting Departament Pontifical Catholic University of Paraná (PUCPR) State University of Londrina (UEL), Londrina, Brazil Rodrigo Moreira Casagrande Postgraduate Program in Governance and Sustainability Higher Institute of Administration and Economics (ISAE), Curitiba, Brazil Francisco Carlos Fernandes Paulista School of Politics, Economics and Business Federal University of São Paulo (UNIFESP), Brazil Abstract The objective of this study was to compare the credit risk management of the main Brazilian public and private banks, before and after the 2008 global crisis, with the purpose of answering the research question: What behavior in credit risk management presented by public and private banks, between 2006 and 2011? The research is characterized descriptive, documentary and with quantitative treatment of the data. The choice of sample was intentional for the comparative evaluation between the two largest public banks and the two largest private national banks with commercial profile at the time of the crisis: BancodoBrasil, CaixaEconômica Federal, Itaú and Bradesco. The data were collected from the financial statements and the explanatory notes, from 2006 to 2011, for the collection of data on the credit operations carried out and the risk at the level recorded in the Balance Sheet and in the Income Statement of the Financial Years and the credit rating collected in explanatory notes. The analyzes showed different reactions to credit risk management, preand post-crisis of 2008, as the public banks analyzed had higher exposure to risk in the Medium and High portfolio than the private banks. Keywords: Credit Risk Management; World Subprime crisis; Brazilian banks 1. Introduction The credit risk is the oldest existing form of risk in the financial market and is the result of a financial transaction between a supplier of funds (credit donor) and a user (the borrower). The simple act of lending a sum to someone embeds the likelihood of this not be received, which characterizes

2 The Subprime Crisis and Reflections in Credit Risk Management of Brazilian Banks: A Comparative Study in the Largest Public and Private Banks in Brazil 131 uncertainty about the return. In essence, the credit risk is the possibility of a counterparty not honoring the commitment made in a credit granting (Figueiredo, 2001). Assess the credit risk is to examine the borrower agent probability of it not meet its obligations as they fall due. The credit granting systems, scoring or judgmental, are based on the premise that the future, at least in the short term will be very similar to the recent past. There is no way to know in advance whether a company will pay or not. Thus, in modeling, it seeks to make a comparison between the characteristics of customers who had their defaulted loans and the customers whose credit was fulfilled. In other words, it tries to trace the good and bad customer profile (Dantas; Souza, 2008). In the environment of financial institutions, it is necessary to also consider the systemic elements, such as those that transformed the crisis of mortgage-backed securities in the United States in a worldwide crisis. The form of recognition of credit risk in the case of default of US mortgages, combined with other factors, such as market failure, accounting at fair value, capital impairment on financial institutions and the systemic effects of prudential regulation created a downward spiral detrimental to the global financial system (Hellwig, 2008). The question about how a credit risk problem in the Subprime Crisis in the US market, has affected the entire global banking system turned out to show that wrong decisions in relation to credit and market, and failures in the financial system triggered such a situation. So when you think about the system regulation, one must go beyond the regulatory and supervisory considerations and pay attention to systemic interdependence and transparency issues (Hellwig, 2008). Such systemic crisis hit Brazil when the country was going through a time of growth and acceleration, which presupposed need for business credit for industrial expansion and / or new investments. In this context, the Central Bank had to act promptly, on the issue of credit policy practiced by the banks, to ensure that the economy does not come into stagnation (Central Bank, 2008). Thus emerges the question posed by the research: What behavior, credit risk management, submitted by the Bank of Brazil, Caixa Federal, Bradesco and Itaú, in the period between 2006 and 2011? To answer the question-problem, it defined the purpose of this study to compare the credit risk management of major public and private banks Brazilians, before and after the global crisis of 2008, to see if there were different behaviors in lending between public and private banking agents. This study is justified by the possibility of expanding the theoretical and empirical research on the behavior of public and private banks Brazilians, for the management of credit risks in the market uncertainties generated by a phenomenon of global magnitude. To present the study developed, this paper was divided into five sections: introductory aspects covered here; the theoretical framework, comprising a reflection on the credit risk and its APPRAISAL and regulation in the banking sector and the impact of the global financial crisis in lending in Brazil; methodological procedures; search results; and finally the closing remarks followed the references. 2. Credit Risk in the Banking Sector Studies on the credit risk gained relevance with the pioneering work of Altman (1968), called Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy, which established a methodology for the counterparty credit rating and ways to predict a situation default. The concern from various sectors and also of the researchers on the management of credit risk can be proven in numerous studies, developed since then, both nationally and internationally. Aboutratings with predictive models, there are studies Kanitz (1976), Matias (1978) and Altman et al (1979), Bruni, Famá and Murray (1988), Minardi and Sanvicente (1999), Altman and Sabato (2007). Insolvency prediction models, specifically, there are studies of Beaver (1966) and Ohlson (1980), which became reference in the literature. Other studies have tried to analyze the models of credit risk assessment, among them the Perera (1997), Crouhy, Galai and Mark (2004), Silveira (2007), AssafNeto and Brito (2008), Selau (2008) Santos and Santos (2009), Yanaka and Holand

3 132 Júlio César da Silva, Silvio Aparecido Teixeira, Rodrigo Moreira Casagrande and Francisco Carlos Fernandes (2009), AmaralJúnior and TávoraJúnior (2010), Bhamra, Kuehn and Strebulaev (2010) and Chen, Cheng and Wu (2011). The granting of chalksaid is basic function of banks, so credit risk plays an important role in the composition of the risks of an institution and can be found both in operations where there is release of money for clients such as those where there is only the possibility of using previously granted limits. The main types of a bank credit operation are: loans, financing, bond discounts, depositors advance, advance exchange, leasing operations (leasing), guarantees and sureties. Aggregate credit risk modeling has gained importance through the development of models for market risk assessment and the need for banks to adhere to the practices suggested by the New Basel Capital Accord for Banking Supervision with respect to capital allocation in credit portfolios. An established consensus in economic thinking understands that the financial system exhibits special dynamics when compared to other sectors of the economy, especially when it comes to systemic risk (Souza, 2007). In this context, the International Agreement of Capital Measurement and Capital Requirements, known as Basel I (1988), signed by the member countries Basel Commitee on Banking Supervision (BCBS) sought to strengthen the international financial system and to promote convergence of national capital requirement standards for internationally active banks. Basically, it proposed measures for the own capital requirements of financial institutions due to the risk presented in its financial operations (Souza, 2007). After the Basel I agreement, due to the financial crisis that highlighted the fragility of the international financial system from destabilizing speculation possibilities or even manipulation, new supervisory and regulatory measures, global scope, they have been established. These measures have been celebrated in a new agreement, known as the New Basel Accord or Basel II, which was based on three pillars: minimum capital requirements; banking supervision; and market discipline (Fortune, 2009). In Basel II,in the case of credit risk, specifically, an already standardized approach and built in the first agreement seeks to introduce the use of assessment credits granted (rating) for less complex banks. Banks with more advanced capabilities of risk management, developing stricter standards supervision, can make use of an internal approach based on fees. Within this approach, some of the major credit risk elements such as probability of default the lender will be estimated internally. The dynamics of regulations governing financial institutions can affect substantially, and adversely, their operations and revenues. This regulation is represented by the following key points (on which the institutions have no control): a) changes in reserve requirements and reserve requirements; b) changes in the minimum limits for loans to property, federal and rural sectors; c) changes in tax regulations; d) changes in the benchmark interest rate by the Central Bank (SarloNeto, 2004). The credit risk relates to the recording of the allowance for doubtful accounts, represented by an account that reduces the customer account, trade accounts receivable or accounts receivable, resulting from sales or operating services of the company. Your goal is to cover possible losses by not receiving the credits. As Marion (1998, p. 243), "the receipt of the duplicate is not for granted since the company is subject to credit risk." Consequently, if the company has losses with its clients, the balance of accounts receivable will be reduced, that is, the company does not receive the registered importance, but the amount least possible losses. The model for calculating the credit risk is more sophisticated than theused to market risk, mainly in the Brazilian case. The absence of secondary markets for loan contracts and difficulties to observe the default probabilities hinder the analysis of correlations between the borrowers, a necessary requirement for an aggregate view of credit risk (Silveira, 2007). For one complete risk assessment of commercial banks, it is essential to measure the combined impact of interest rates and credit risk. Credit and interest rate, together, are the two main risks faced, and how they are intrinsically related, cannot be measured separately (Drehmann; Sorensens, stringa, 2008).

4 The Subprime Crisis and Reflections in Credit Risk Management of Brazilian Banks: A Comparative Study in the Largest Public and Private Banks in Brazil Impacts of the Global Financial Crisis of 2008 in the Bank Lending in Brazil The world economy grew considerably between 1980 and 2000, despite the economic crisis. During this period, there was a moratorium of Mexico in 1981; the so-called black October in 1987, when the Dow Jones index plummeted 22%; Tequila crisis in Mexico in 1994; the Asian crisis in 1997; and the crisis in Russia a year later. In Brazil, there was the real mega devaluation in 1999, but it is also worthy to record the Argentine crisis of 2002 (Toledo Filho et al, 2009). However, the fact is that we had not experienced in the recent history of the economy after the 1970s, a systemic crisis along the lines of that erupted in Freitas and Cintra (2008) explain that after the crisis of technology companies, which took place from 2000 to 2002, banks sought new alternatives in assets to trade and increase their profits. As the real estate market in the United States, was very heated, the generation of new mortgages has fulfilled this role, and what we saw was aconsumer credit expansion by the old mortgage refinancing. It began thus to form a bubble, with a significant rise in prices of American real estate. Fever consumption, in real estate, attracted borrowers seen as higher risk, to the extent that mortgages are now offered to people who did not have reasonable assurance of payment and who apparently did not care about the worsening rates of operations (Freitas; Cintra, 2008). Due to the consecutive increase in basic rates of interest of the US economy since 2004, mortgage holders no longer are able to pay their benefits, a phenomenon known as "bursting of the housing bubble." As a result, there was a sharp drop in property prices. On September 15, 2008, exactly, the world took note of the bank Lehman Brothers. This event threw the international financial crisis, generated by the bursting of the subprime mortgage crisis and its financial pyramid built on a borrowing base (Guillen, 2011). The bankruptcy of Lehman Brothers caused the risk of perception were specifically related to the development of the real economy and the problemsspecific financial system. The prospect of global recession became imminent,generating the deterioration of banks' loan portfolios and infecting the global financial system (Eichengreen, et al, 2009). In times of crisis, when the financial fragility is increasing, changes in macroeconomic conditions could lead banks to develop excessive prudence and postures credit rationing measures and relocation portfolios, which confirms the theory that competitive pressures and the pursuit of profit lead banks to lend excessively on expansion phase of the economy and dam the granting of credit in periods of contraction (Freitas, 1997). Banks act as drivers of productive activities to the extent that fuel producers and consumers towards production and consumption. On the other hand, it must consider that banks are companies that work toward the expansion of its profits. In times of economic crisis, the asymmetry of information and the search for the reduction of transaction costs lead banks to develop strategies forrecompose their portfolios, which ultimately reduce substantially the resources for loans. (Frascaroli; Paes, Ramos, 2010). This systemic crisis hit Brazil in a time of prosperity, steady and rapid growth, observed over six quarters, which ensejava precisely the need for credit for companies seeking to expand their production lines, with new investment expansion (Central Bank, 2008). As an immediate reaction to the risk posed by the global crisis was spreading, private banks have adopted credit retraction strategies, in addition to the enhancement interest charged on credit lines. As the banking market works, including, as the creator of currency, as obtains funds to surplus agents and lends to the deficit, the credit crunch has as a direct consequence the slowdown in economic activity caused by the neck to appear on the release of funds to companies that wish to expand their activity levels (Lodi, 2010). According to Almeida (2010), the first months of 2009 had severe credit bottleneck in the Brazilian market. In the first quarter, taking as reference the month of December 2008, the new loans with free resourcesbanks, ie, accumulated in the month awards for companies, declined 30%.

5 134 Júlio César da Silva, Silvio Aparecido Teixeira, Rodrigo Moreira Casagrande and Francisco Carlos Fernandes With the obvious diagnosis that the pace of credit growth was negatively affected by the significant reduction of external sources of funding and the cooling of the level of domestic economic activity, associated, among other factors, the deterioration in expectations of economic agents, the Central Bank He had to act and used the following mechanisms: foreign currency injection in the national economy and increasing the liquidity of the currency, through the release of funds from compulsory deposits (Central Bank, 2008). With these interventions, Brazil's Central Bank was able to circumvent restrictions identified in the credit market, so that at the end of the year, credit operations increased 31.1% over the previous year and amounted to around 41.3% of Gross Domestic product. It is noteworthy that among the most important countercyclical measures adopted by the economic team of the Brazilian government, a major decision, which came mitigate the effects of the fall in liquidity and compensate for the shortage of international credit lines, was the guidance transmitted to the Bank of Brazil, Caixa Federal and BNDES, to broaden the extension of credit (Central Bank, 2008). Almeida (2010) points out that at the beginning of the crisis, the credit from public banks accounted for around 35% of total loans, a percentage that would rise to 41% (percentage of February 2010). To the extent that there was a greater demand for government banks, by investors for the investment of funds, especially with a view to an alleged greater security, they thus managed more resources to expand its financing and put into practice strategy envisioned by the economic team of the Brazilian government. The Central Bank report (2008) indicates that this scenario, the total volume of financial system credit operations, considered the free resources and targeted, totaled R $ billion in December 2008, with annual growth of 31.1% over the same month last year when the expanded by 27.8%. As a result, total loans increased to represent 41.3% of Gross Domestic Product (GDP) in 2008 from 34.2% in 2007 and 24.9% in December Research Methods and Procedures This study is characterized as a descriptive research and was held in the largest financial institutions credit in Brazil. According to Raupp and Beuren (2009), this type of research describes the characteristics of a given population or phenomenon. The studies were conducted on the basis of established procedures for documentary research and were based on secondary data of the financial statements and rankings provided by the Central Bank of Brazil, Sisbacen system in the Central Bank website, and in their own financial statements posted on the websites of banks searched. According to Raupp and Beuren: The documentary research is based on materials that have not yet received an analytic treatment or can be reworked according to the research objectives [...] its notability is justified when it can organize information are dispersed, giving him a new importance as a source of information (Raupp; Beuren, 2009, p 89.). The choice of the sample population was intentional, because the objective of the study was to develop a comparative analysis of the two largest private national banks and the two largest national public bank business profile, according to rankings published by Sisbacen. As it is a comparison between the periods before and after the crisis of 2008, the rankings published between 2008 and 2011 were used for the definition of banks to be analyzed, and according to Tables 01 and 02. According ColautoBereun (2009, p. 126), "The sampling typicality or intentional is to select sample based on information available".

6 The Subprime Crisis and Reflections in Credit Risk Management of Brazilian Banks: A Comparative Study in the Largest Public and Private Banks in Brazil 135 Table 01: Ranking of the four largest domestic banks by total assets - September / 2008 Ranking Bank TD TC Total assets No. Agencies Basel Immobilizati Employees No. index on Index 1 Brazil W 1 444,702, ,744 4, Itaú W 3 400,957,932 69,393 2, Bradesco W 3 365,295,511 77,919 3, Cashier I 1 276,059, ,817 2, Source: Adapted from Sisbacen (2012). Table 02: Ranking of the four largest domestic banks by total assets - September / 2011 Ranking Bank TD TC Total assets No. Agencies Basel Immobilizati Employees No. index on Index 1 Brazil W 1 907,743, ,640 5, Itaú W 3 810,464, ,653 3, Bradesco W 3 636,399,735 94,516 3, Cashier I 1 507,306, , Source: AdaptedSisbacenof (2012). Tables: TD (Document Type) C - conglomerate, I - Establishment Independent. TC (Control Type): 1 - Federal Public, 2 - State Public 3 - National Private. Thes financial institutions were selected based on the analysis of the rankings showed that the Bank of Brazil and CaixaEconomica Federal as the two largest public banks, while Itau and Bradesco as the two largest national private banks in total assets. This ranking was repeated before and after the subprime crisis. After this classification, the financial statements and notes have been raised, the period from 2006 to 2011, banks selected for the collection of data from credit transactions and risk in recorded level in the Balance Sheet and the exercises Income Statement specifically, in a reduction of active, known as Allowance for Doubtful Accounts, and its counterpart in the result. Other information, such as details of the amounts recorded by level of risk, ie the risk according to the credit rating was collected in the notes. Subsequently, to achieve the research objectives, the data were analyzed quantitatively and processed in SPSS software (Statistical Package for Social Sciences); the two groups, public and private banks, were analyzed separately, before and after the subprime crisis, evaluating and including the handling of credit risk rating for the level of transactions in the period. For the definition of risk levels, respecting the minimum percentage established by Resolution No. 2682/99 of the Central Bank, were used in a deterministic way, the following criteria: Low Level: Ratings from A to B, comprising risks ranging from 0% to 1%; Middle Level: Ratings of C to D, whose risks are provisioned between 3% to 10%; and Level High to operations Rating E to H, with risks above 30%. 4. The Results In this section, the main data collected and the results for the recognition of credit risk recorded in the Balance Sheet and the Income Statement exercises, evidenced in the Notes, as described in the search procedures are presented. Initially, Table 03 shows the percentage of allowance for doubtful accounts, segregated by level of risk, ie, Low Level, Middle Level and High Level, per bank and per period, in relation to the crisis. Thus, the data were divided into eleven preceding quarters and eleven quarters subsequent to the subprime crisis, in order to highlight the percentage level behavior of the accrued value as a risk in relation to the amount of credit transactions in the period.

7 136 Júlio César da Silva, Silvio Aparecido Teixeira, Rodrigo Moreira Casagrande and Francisco Carlos Fernandes Table 3: Exposure by credit risk level (allowance for loan losses in relation to loans) per bank and per period banks Brazil Cashier Bradesco Itaú Before Crisis After the Crisis TRIM Low Medium High Trim Low Medium High 1/ % 55.59% 5.19% 4/ % 50.77% 4.11% 2/ % 57.17% 5.25% 1/ % 42.12% 4.46% 3/ % 56.30% 5.32% 2/ % 43.18% 4.64% 4/ % 51.40% 4.15% 3/ % 41.95% 4.60% 1/ % 51.86% 4.06% 4/ % 42.54% 4.17% 2/ % 52.34% 4.03% 1/ % 44.16% 4.04% 3/ % 52.56% 4.08% 2/ % 54.32% 3.90% 4/ % 52.59% 4.10% 3/ % 53.62% 3.70% 1/ % 51.04% 4.18% 4/ % 53.37% 3.38% 2/ % 49.78% 4.02% 1/ % 53.32% 3.18% 3/ % 49.38% 3.92% 2/ % 52.75% 3.13% 3/ % 51.67% 3.17% 1/06 2/06 3/06 4/06 1/07 2/07 3/07 4/07 1/08 2/08 3/08 1/06 2/06 3/06 4/06 1/07 2/07 3/07 4/07 1/08 2/08 3/08 1/06 2/06 3/06 4/06 1/07 2/07 3/07 4/07 1/08 2/08 3/08 Source: research data % 45.55% 47.51% 48.53% 47.71% 47.56% 47.51% 46.51% 46.08% 44.74% 41.68% 62.06% 61.11% 60.91% 59.89% 60.35% 59.52% 58.04% 58.62% 58.15% 57.89% 60.44% 45.50% 48.10% 47.48% 49.68% 50.35% 51.71% 51.72% 52.20% 50.74% 50.92% 49.39% 29.08% 45.85% 44.49% 43.17% 43.82% 44.19% 44.21% 46.24% 46.92% 48.57% 51.74% 33.74% 34.53% 34.39% 35.26% 34.99% 35.97% 37.37% 37.17% 37.91% 38.22% 35.59% 49.21% 47.90% 46.64% 44.63% 43.90% 42.53% 43.32% 43.07% 44.89% 45.17% 46.69% 11.10% 8.60% 8.00% 8.30% 8.46% 8.26% 8.28% 7.25% 7.00% 6.70% 6.58% 4.20% 4.36% 4.70% 4.85% 4.66% 4.51% 4.59% 4.20% 3.94% 3.89% 3.97% 5.29% 4.00% 5.89% 5.69% 5.75% 5.76% 4.97% 4.74% 4.38% 3.92% 3.92% 4/11 4/08 1/09 2/09 3/09 4/09 1/10 2/10 3/10 4/10 1/11 2/11 3/11 4/11 4/08 1/09 2/09 3/09 4/09 1/10 2/10 3/10 4/10 1/11 2/11 3/11 4/11 4/08 1/09 2/09 3/09 4/09 1/10 2/10 3/10 4/10 1/11 2/11 3/11 4/ % 42.77% 44.50% 45.05% 47.36% 51.82% 51.72% 51.35% 52.86% 55.20% 51.45% 57.22% 59.40% 60.68% 61.81% 60.22% 58.35% 57.96% 58.17% 57.62% 57.18% 56.38% 54.79% 53.96% 54.08% 53.35% 52.03% 47.46% 54.94% 52.56% 52.80% 53.34% 52.57% 54.25% 54.13% 54.95% 62.97% 62.23% 63.26% 62.12% 49.61% 51.11% 49.70% 49.19% 47.02% 42.95% 42.84% 43.55% 42.20% 39.95% 43.77% 37.90% 35.72% 34.56% 34.42% 35.66% 37.13% 36.95% 36.72% 37.64% 38.42% 39.26% 40.94% 41.91% 41.91% 42.56% 43.78% 46.75% 39.69% 40.62% 39.21% 39.09% 40.64% 39.74% 40.00% 39.28% 31.31% 31.72% 30.30% 31.28% 3.00% 6.12% 5.80% 5.76% 5.62% 5.23% 5.44% 5.10% 4.94% 4.84% 4.78% 4.87% 4.88% 4.76% 3.77% 4.12% 4.52% 5.08% 5.11% 4.74% 4.40% 4.36% 4.27% 4.13% 4.01% 4.09% 4.19% 5.78% 5.37% 6.82% 7.99% 7.57% 6.79% 6.01% 5.87% 5.78% 5.71% 6.06% 6.44% 6.61%

8 The Subprime Crisis and Reflections in Credit Risk Management of Brazilian Banks: A Comparative Study in the Largest Public and Private Banks in Brazil 137 Then the data were analyzed with SPSS software (StatisticalPackage for Social Sciences) for best interpretation based on descriptive statistics. Descriptive statistics, pointed in Table 4 shows the exposure to credit risk results from the major secreted in the immediately preceding period (eleven quarters) and posterior (eleven quarters) divided by the risk level of the transactions. Table 4: Descriptive analysis of the exposure to credit risk for the period (before and after the subprime crisis) Descriptives Previous Period Crisis Later period Crisis Low risk Medium risk High risk Low risk Medium risk High risk Mean % Lower Confidence Bound Interval for Upper Mean Bound % Trimmed Mean Median variance Std. Deviation Minimum maximum Source: research data. average tests show that, in general, the two groups, ie, public and private banks decreased in average exposure to risk in the upper and middle levels in the post-crisis period ( and 0, to and previous), with the risk of transfer to the average Low risk identified as the response ( to earlier). The results were confirmed in the confidence interval for the average of 95%, also evidence other assumptions, such as no departure from the median, which further reinforces the confidence in the data. With results for the period, we tried to repeat the descriptive statistics and average tests now by type of bank, ie public banks versus private banks. Table 5 shows the results. Table 5: Descriptive analysis of the exposure to credit risk by bank type (public and private) Descriptives Publicbanks Privatebanks Lowrisk Mediumrisk Highrisk Lowrisk Mediumrisk High risk Mean Lower 95% Confidence Bound Interval for Upper Mean Bound 5% TrimmedMean Median variance Std. Deviation Minimum maximum Source: research data. The analysis shows that public banks in the period investigated before and after the subprime crisis had greater exposure to risk in the portfolio with medium and high ratings ( on average, respectively) than the private banks ( and on average, respectively). Private Banks have sought to align its strategy to the low level of risk, which have an average of

9 138 Júlio César da Silva, Silvio Aparecido Teixeira, Rodrigo Moreira Casagrande and Francisco Carlos Fernandes against of public banks. The data points throughout the period examined, from 2006 to 2011, the greatest risk exposure occurred in public banks, while private transferred part of their risks to the risks of lower levels. To continue the study, we tested, then the behavior of these numbers segregated by period, in order to point out the response behavior of the risk before and after the 2008 subprime crisis Therefore, significance tests were conducted to sign posts for the two samples: before and after the crisis and bank type (public and private), as shown in Table 6. Table 6: Wilcoxon test for level of exposure, before and after the subprime crisis compared within each group of banks (public and private) Public banks Private banks Test Statistics Low Risk Medium High Risk Low Risk Medium High Risk (Before Risk (Before (Before (Before Risk (Before (Before After x) After x) After x) After x) After x) After x) Z - 276th a a a a a Asymp. Sig. (2-tailed) , The. Based on positive ranks. B. Based on negative ranks. w. Wilcoxon Signed Ranks Test Source: research data. The significance tests showed that the middle and upper levels, public banks have changed between the periods before and after the crisis, with p and (for the significance below 0.05). Once inside the private banks, the behavior was a significant change to the risks in Low ratings (0.000) and high (0.033), with no significant change in the Middle (0.249) level. Thereafter, for evidence of difference between groups, the changes significance tests were conducted on a comparative basis among the bank groups in both periods analyzed, and the results are shown in Table 7. Table 7: Mann-Whitney U test for level of exposure, before and after the subprime crisis compared between groups of banks (public and private) Test Statistics Before Crisis After the Crisis Low risk Medium risk High risk Low risk Medium risk High risk Z Asymp. Sig. (2-tailed) The. Group Variable: Type Source:research data. As noted, Mann-Whtiney tests matched for independent samples, show that the risk exposure levels before the crisis were significantly different between groups at the three levels: Low, Medium and High. For the period after the crisis, in response to the risks arising from the international scene, the low risk showed no significant difference between the groups, but this was perceived in the upper and middle levels. Infers that the different reaction between the analyzed banks due to the credit risk management behavior developed in the pre- and post-crisis Public banks took greater risks (high and medium levels), while private banks They opted for conservatism. One factor that can be considered as a search limiter is earnings management practiced by banks, as these institutions have special interest in transmitting a constant image in the results, avoiding the production of high volatility in reported profits, which could adversely affect the perception of investors about future patterns of results and the risks involved in its operations (Goulart, 2007).

10 The Subprime Crisis and Reflections in Credit Risk Management of Brazilian Banks: A Comparative Study in the Largest Public and Private Banks in Brazil 139 The earnings management, commonly called earnings management occurs when managers use judgment in the process of preparing financial statements and structuring transactions to alter financial information disclosed, either to evade some stakeholders on the economic performance of the company or influence the results that depend on contractual disclosed accounting numbers (Healy; Wahlen, 1999). 4. Final Considerations The World Crisis 2008, which became known as the Subprime Crisis, had a contagion effect that virtually affected all national economies, generating decreased production and world consumption, triggered by the credit crunch. In addition to the contraction of lending operations, there was an increase in profit remittances by companies and banks in their home countries, and a significant decline in asset prices. The crisis hit Brazil in a time when its economy was heated with six consecutive quarters of growth, which ensejava access to credit for the expansion of production facilities and investments in technology and human resources. In this sense, banks are the drivers of productive activities, as they encourage producers and consumers towards production and consumption. On the other hand, however, it must be considered that banks are companies that act towards the expansion of their profits. In times of economic crisis, the asymmetry of information and the search for lower transaction costs lead banks to develop strategies to recompose their portfolios, significantly reducing lending resources. (Frascaroli; Paes, Ramos, 2010). Official reports of the Bank Central (2008) demonstrate that this institution intervened in the Brazilian banking market in the period was triggered the world crisis of 2008, enabling the public banks to become credit developers for the maintenance of productive activities in the country. This study aimed to compare the management of credit risk in the two main public and private banks in Brazil, before and after the global crisis of 2008, to see if there was a different behavior in the credit risk management between public and private banking agents in said period. The results obtained from the survey showed that public banks before and after the subprime crisis had greater exposure to risk in the portfolio with medium and high ratings than private banks. Private Banks have sought to align its strategy to the low level of risk. The data indicate that in the analyzed period (2006 to 2011), exposure to higher risks occurred in public banks, while private transferred part of their risks to the risks of lower levels. There was different reaction behavior of credit risk management, pre and 2008 post-crisis among banks analyzed, as was evident that public banks took greater risks (high and medium levels), while private banks opted for conservatism. The survey results corroborate the Freitas approach (2010), according to which, during the said crisis, Brazilian private banks proved to be discouraged from getting involved with the financing of some sectors and activities considered strategic for economic and social development from the country. Thus followed the pragmatism of the market, which shows aversion to high risks and uncertain profitability. The survey indicates that Brazilian public banks have adopted counter-cyclical, basic actions to support the financing flows. By this view, public banks have shown that this is institutions that can ensure greater stability in the financial system, in that attenuated, in a way, the market tension in a global scenario of credit contraction. According to Freitas (2010, p. 20), "public banks may act anticyclically when the economy goes into a downturn phase, becoming thus in a stable source of funds for productive investments and activities social character. This kind of public banks posture showed that these institutions can and should be used to print a logical alternative to the market participants. At times, especially in times of economic crisis, it is crucial to rely on organizations that manage long-term loans to specific sectors, which are seen as

11 140 Júlio César da Silva, Silvio Aparecido Teixeira, Rodrigo Moreira Casagrande and Francisco Carlos Fernandes higher risk, and to consider the social returns of the stock as of greater importance than proper financial return. References [1] Almeida, J.G. Como o Brasil superou a crise. São Paulo: Terra Magazine. Março Disponível em Acesso em 20 de março de [2] Altman, E.I. Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. JournalofFinance, v. 23, n. 4, p , [3] Altman, E.I.; Baidya, T.K.N.; Dias, L.M.R. Previsão de problemas financeiros em empresas. Revista de Administração de Empresas, v. 19, n. 1, p.17-28, [4] Altman, Edward I.; Sabato, Gabriele. Modeling Credit Risk for SMEs: evidence from the US Market. Abacus, v. 43, n. 3, p , [5] Amaral Júnior, J.B.; Távora Junior, J.L. Uma análise do uso de redes neurais para a avaliação do risco de crédito de empresas. Revista do Banco Nacional de Desenvolvimento Social, v. 12, n. 1, [6] Assaf Neto, A.; Brito,G.A.S. Modelo de Classificação de Risco de Grandes Empresas. Revista Contabilidade e Finanças, v. 19, n. 46. p , jan/abr, [7] Central Bank (BACEN). Relatório de Economia Bancária d Crédito Disponível em Acesso em 20 de março de [8] Banco Do Brasil S/A. Relação com Investidores. Disponível em: < nu=198> Acesso em: 06 de abril de [9] Banco Itaú S/A. Relação com Investidores. Disponível em: < om.br/portalri/content/dccindex/dccindex.aspx> Acesso em: 06 de abril de [10] Beatty, A.; Chamberlain, S. L.; Magliolo, J. Managing financial reports of commercial banks: the influence of taxes, regulatory capital, and earnings. Journal of Accounting Research, v. 33, n. 2, p , [11] Beaver, W.H. Financial Ratios as Predictors of Failure. Journal of Accounting Research, Empirical Research in Accounting: Selected Studies v. 4, n. 1, p , [12] Bhamra, H.S.; Kuehn, L.A.; Strebulaev, I.A. The levered equity risk premiun and credit spreads: a unified framework. Reviewof Financial Studies. v. 23, n. 2, p , [13] Bradesco S/A. Relação com investidor. Disponível em: Acesso em: 06 de abril de [14] Bruni, A.L.; Famá, R.; Murray, A.D. Modelos brasileiros preditivos de risco de crédito: um estudo exploratório atual sobre as suas eficácias. Periódico Tema, São Paulo, n. 32, p , jan./jun [15] Caixa Econômica Federal- CEF. Demonstrativos Financeiros. Disponível em: nstrativos/demonstrativos-financeiros. Acessoem: 06 de abril de [16] Chen, R-R; Cheng, X.; Wu, L. Dynamic interactions between interest-rate and credit risk: theory and evidence on the credit default swap term structure. ReviewofFinance. V. 30, n. 1, p DOI [17] Colauto, D.R.; Beuren, I.M. Coleta, análise e interpretação dos dados. In: BEUREN, I.M. (Org). Como elaborar trabalhos monográficos em contabilidade: teoria e prática. 3. ed. São Paulo: Atlas, 2009.

12 The Subprime Crisis and Reflections in Credit Risk Management of Brazilian Banks: A Comparative Study in the Largest Public and Private Banks in Brazil 141 [18] Crouhy, M.; Galai, D.; Mark, R.Gerenciamento de Risco: uma abordagem conceitual e prática: uma visão integrada dos riscos de crédito, operacional e de mercado. Qualitymark: São Paulo: Serasa: [19] Dantas, R.F.; Souza, S.A. Modelo de Risco e decisão de crédito baseado em estrutura de capital com informação assimétrica. RevistaPesquisaOperacional, v.28, n.2, p , mai./ago [20] Drehmann, M.; Sorensen, S.; Stringa, M. The integrated impact of credit and interest rate risk on banks: an economic value and capital adequacy perspective. Publications Group, Bank of England, London, Working paper, n. 339, Jan/2008. [21] Eichengreen, B.; Mody, A.; Nedeljkovic, M.; Sarno, L. How the subprime crises went global: evidence from bank credit default swap spreads. National Bureau ofeconomicresearch. Cambridge, apr/, [22] Figueiredo, R. P. Gestão de Riscos Operacionais em Instituições Financeiras: Uma Abordagem Qualitativa f. Dissertação (Mestrado em Administração). Universidade da Amazônia UNAMA, [23] Fortuna, E. Mercado Financeiro: produtos e serviços. 17ª Ed. Rio de Janeiro: Qualitymark, [24] Frascaroli, B.F; Paes, N.L; Ramos, F.S. A Indústria Brasileira e o Racionamento de Crédito: Uma Análise do Comportamento dos Bancos sob Informações Assimétricas. Revista Economia, v.11, n.2, p , mai/ago [25] Freitas, M.C.P. Os Efeitos da Crise Global no Brasil: aversão ao risco e preferência pela liquidez no mercado de crédito. Estudos Avançados v. 23, n. 66, p , [26] Freitas, M.C.P.; Cintra, M.A.M. Inflação e deflação de ativos a partir do mercado imobiliário americano. Revista de Economia Política, v. 28, n. 3, [27] Fuji, A. H. Gerenciamento de resultados contábeis no âmbito das instituições financeiras atuantes no Brasil f. Dissertação (Mestrado em Ciências Contábeis). São Paulo: FEA-USP, [28] Goulart, A.M.C. Gerenciamento de resultados contábeis em instituições financeiras no Brasil p. Tese (Doutorado em Ciências Contábeis) Universidade de São Paulo, [29] Guillén, R.A. The effects of the global economic crisis in Latin America. Brazilian Journal of Political Economy, v. 31, n. 2, p , apr/jun [30] Hellwig, M. Systemic Risk in the Financial Sector: an analysisof the subprime-mortage financial crisis. Preprints of the Max Instititue for Research on Collectiv Goods. Bonn [31] Healy P. M.; Wahlen, J. M. A review of the earnings management literature and its implications for standard setting. AccountingHorizons, v.13, p , [32] Kanitz, S. Indicadores Contábeis e Financeiros de Previsão de Insolvência: a experiência da pequena e média empresa brasileira. Tese de Livre Docência. FEA/USP [33] Lodi, A.L.G. O papel dos bancos públicos do Brasil e da Índia no contexto da crise econômica mundial f. Dissertação (Mestrado em Economia) Instituto de Economia da UNICAMP, [34] Marion, J.C. Contabilidade empresarial. 8. ed. São Paulo: Atlas, [35] Matias, A.B. Contribuição às Técnicas de Análise Financeira: um modelo de concessão de crédito. CEPEFIN. USP, [36] Minardi, A.M.A.; Sanvicente, A.Z. Migração de Risco de Crédito de Empresas Brasileiras: uma aplicação de análise de clusters na área de crédito. InstitutoBrasileiro de Mercado de Capitais, [37] Ohlson, J.A. Financial Ratios and the Probabilistc Prediction of Bankruptcy. Journal of Accounting Research.v. 18, n. 1, p , Spring, 1980.

13 142 Júlio César da Silva, Silvio Aparecido Teixeira, Rodrigo Moreira Casagrande and Francisco Carlos Fernandes [38] Pereira, A. Guia prático de utilização do SPSS: Análise de dados para ciências sociais e psicologia. Edições SílaboLtda, 5ª Edição, Lisboa [39] Perera, L.C.J. Quantificação e Precificação de Risco de Crédito Através do Modelo de Opções. Revista de Administração de Empresas. São Paulo: v.37, n.3. p [40] Raupp, F. M.; Beuren, I. M. Metodologia da Pesquisa Aplicável às Ciências Sociais. In: BEUREN, I. M. (Org). Como elaborar trabalhos monográficos em contabilidade: teoria e prática. 3. ed. São Paulo: Atlas, [41] Santos, J.O.; Santos, J.A.R. O Modelo KMV e sua Utilidade no Processo de Análise do Risco de Crédito. Revista de Gestão da USP. São Paulo, v.16, n.2, p abr-jun/2009. [42] Sarlo Neto, A. A Reação dos preços das ações à divulgação dos resultados contábeis: evidências empíricas sobre a capacidade informacional da contabilidade no mercado acionário brasileiro f. Dissertação (Mestrado em Ciências Contábeis) - FUCAPE, Vitória, [43] Selau, L.P.R. Construção de modelos de previsão de risco de crédito f. Dissertação (Mestrado em Engenharia da Produção) Programa de Pós-Graduação em Engenharia de Produção, UFRGS, [44] Silveira, M.A.M.. Avaliação do Risco de Crédito Agregado: aplicação do CreditRisk+ em instituições brasileiras não financeiras. Dissertação (Mestrado em Finanças e Economia Empresarial) Fundação Getúlio Vargas. São Paulo: FGV, [45] SISTEMA DE INFORMAÇÕES DO BANCO CENTRAL SisBacen: 50 maiores bancos e o consolidado do Sistema Financeiro Nacional. Disponível em: /top50/ port/top50.asp. Acessado em: 19 de março de [46] Souza, G. J. G. A Interação entre a dinâmica macroeconômica e os bancos: uma perspectiva acerca do risco de crédito. Dissertação (Mestrado em Economia), Universidade Federal Fluminense. Departamento de Economia: [47] Stevenson, W. J. Estatística Aplicada à Administração. São Paulo: Harbra, [48] Toledo Filho, J.R.; Kronke, A.; Sothe, A. Impacto da Crise do Suprime na Provisão de Risco de Crédito dos Maiores Bancos Nacionais. Revista Brasileira de Gestão de Negócios. São Paulo, v. 11, n. 32, p , jul/set [49] Yanaka, G.M.; Holland, M. Basiléia II e Exigência de Capital para Risco de Crédito dos Bancos no Brasil. Texto para Discussão, v. 188, São Paulo: FGV: 2009.