Charles University in Prague Faculty of Social Sciences Institute of Economic Studies Credit Risk in the Macroprudential Framework: Three Essays DISSERTATION Author: PhDr. Jakub Seidler Supervisor: prof. Ing. Oldřich Dědek, CSc Academic Year: 2011/2012 Abstract This thesis focuses on proper credit risk identification with respect to macroprudential policies, which should mitigate systemic risk accumulation and contribute to higher financial stability of the financial sector. The first essay deals with a key credit risk parameter - Loss Given Default (LGD). We illustrate how the LGD can be estimated with the help of an adjusted Mertonian structural approach. We present a derivation of the formula for expected LGD and show its sensitivity analysis with respect to other company structural parameters. Finally, we estimate the five-year expected LGDs for companies listed on Prague Stock Exchange and find that the average LGD for the analyzed sample is around 20-50%. The second essay examines the issue of how to determine whether the observed level of private sector credit is excessive in the context of the "countercyclical capital buffer", a macroprudential tool proposed in the new regulatory framework of Basel III by the Basel Committee on Banking Supervision. An empirical analysis of selected Central and...

The aim of this thesis is to introduce and summarize the process of scoring model development in general and then basic statistical approaches used to resolve this problem, which are in particular logistic regression, neural networks and decision trees (random forests). Application of described methods on a real dataset provided by PROFI CREDIT Czech, a.s. follows, including discussion of some implementation issues and their resolution. Obtained results are discussed and compared.

This thesis analyses what drives sovereign credit risk when contagion is con- trolled for. CDS spreads are used as a measure of credit risk and bond yields are used to estimate interconnectedness of the examined countries. The main contribution lies in the use of high-frequency data and a robust wavelet based estimator in addition to spatial econometric model. The aim of this thesis is to test for presence of contagion and to evaluate which fundamentals are decisive for market perception of sovereign credit risk. Another goal is to evaluate the possibility of a structural break caused by the Greek debt restructuring. The results show that the restructuring did bring change. Contagion is present during the post-crisis period and it diminishes as the economies recover. Sim- ilarly, fundamentals are of higher importance in the post-crisis period when compared with the following period. JEL Classification C22, C31, C33, G01, G32, G33 Keywords spatial econometrics, CDS spreads, sovereign credit risk, financial contagion, realised covari- ance Author's e-mail josef.zahlava@gmail.com Supervisor's e-mail petr.gapko@seznam.cz

Evidence from various countries over the past two decades proves that swings in house prices have been concomitant with financial instability. The history of financial crises shows that the six biggest banking crises in advanced economies were accompanied by housing busts. Despite the abundance of literature on the forces behind the financial crisis, and in particular studies investigating the connections between financial stability and disturbances in the real estate market, fundamental questions still wait for convincing answers, such as: (i) To what extent is regional heterogeneity in property price increases reflected in dissimilarity in the evolution of credit default? (ii) What role do borrower-related factors such as housing affordability and household indebtedness, and financial market-related factors such as financial developments, play on the growth of bad loans as a main concern for banking sector? (iii) To which extent do banks’ lending behaviour and property prices undermine the stability of the banking sector, and what are the directions of causality between credit defaults, property prices and banks’ lending behaviour? The goal of this thesis is to investigate these issues and explain the practical implications of the findings. This thesis contains three empirical essays. The first essay explores the nexus between house prices and non-performing loans (NPLs), concentrating on the extent to which geographical variations in house prices are translated into regional variations in credit defaults. The stochastic dominance approach has been used for this purpose, with 372 individual US banks. The stochastic dominance analyses disclose symmetric behaviour between NPLs and the scale of house price increments. The essay is further extended by employing Arellano and Bond’s (1991) GMM model to explore the effect of GDP, unemployment rates, lending interest rates and house prices on the growth of NPLs. The outcomes of the GMM estimations reveal a high explanatory power of economic growth, unemployment and lending interest rates on NPLs. In an additional analysis, a generalised panel threshold model is estimated to check for the presence of a threshold point, above which different impacts of house prices might be found. The threshold model specifications provide a threshold point, in relation to which two different impacts of house prices on the evolution of NPLs are estimated. A general consensus in the literature attributes credit defaults to a wide-ranging spectrum of drivers that take into consideration borrower-related factor, lender-related factors and factors related to financial and real estate markets. The second essay attempts to answer the second question mentioned above, by investigating the impact of borrower-related factors, lender-related factors and financial market-related factors in driving NPLs. The impact of these factors on the evolution of impaired loans is explored by estimating fixed effect models then the analysis is extended to dynamic models using the GMM procedure on an annual balanced panel dataset. Household vulnerability, financial developments and housing affordability are found to be significant contributors to the growth of NPLs. The interaction mechanism between the real estate market and the financial system has often been blamed for being the root of financial crises, through the accumulation of housing market bubbles that leads to the ultimate collapse of the financial markets. The third essay, using the Autoregressive Distributed Lag technique, looks for the presence of cointegrating relationships between mortgage defaults, property prices and bank lending in Hong Kong. Our findings reveal evidence of cointegrating relationships between bank lending, property prices and mortgage defaults in the long term, which governs the correction mechanism between these variables. These outcomes call for more effort to be devoted to maintaining a balanced relationship between these factors. The essay also finds evidence of short-term dynamics between these variables. Importantly, loan-to-value is found to play the most effective role in curbing mortgage default risk in the portfolios of the Hong Kong banking sector.

The growth of micro-credit along with the excellent conditions to carry out microfinance activity in the economy and financial system of the Republic of Peru are pushing for Microfinance Institutions (IMF) increased competition with banks in this segment business. Like in commercial banks, in microfinance questions such as: is this customer profitable?, What is the credit limit that I must accept to his/her application?, What interest rate should I charge to him/ her?, How I can reduce the risk default?, etc., are matters to be assessed properly. We propose a method that could facilitate improvement in customer qualification between failed and not failed. To this end, we propose a methodology that analyzes credit risk in the provision of microcredit through the design of a credit scoring model that we apply to a Development Agency for Small and Micro Enterprise (EDPYME), which is an IMF under the supervision by the Banking and Insurance Superintendency (SBS). / El crecimiento del número de microcréditos junto con las excelentes condiciones para llevar a cabo la actividad microfinanciera en la economía y sistema financiero de la República de Perú están impulsando a las instituciones de microfinanzas (IMF) a una mayor competencia con las entidades bancarias por este segmento de negocio. Al igual que en la banca comercial, en microfinanzas preguntas tales como: ¿conviene este cliente?, ¿cuál es el límite de crédito que debo aceptar a su solicitud?, ¿qué tasa de interés debo cobrar?, ¿cómo puedo reducir el riesgo de impago?, etc., son cuestiones que deben valorarse de forma adecuada. Este trabajo plantea un método que podría facilitar una mejora en la calificación de los clientes fallidos y no fallidos. Para ello, se propone una metodología que analiza el riesgo de crédito en la concesión de microcréditos mediante el diseño de un modelo de credit scoring aplicado a una entidad de desarrollo de la pequeña y micro empresa (EDPYME), IMF sometida a supervisión por la Superintendencia en Bancay Seguros (SBS).

Alors que les modèles de credit scoring sont largement utilisés depuis plus de cinquante ans et sont considérés comme un outil indispensable dans la prise de décision dans d'innombrables institutions financières du monde entier, la littérature et les empiriques disponibles sur ce sujet restent encore très limitées. Notre objectif est de combler cette lacune en présentant une analyse approfondie sur les modèles de credit scoring et le processus de prise de décision d’octroi de crédit, avec diverses applications sur des données réelles et extensives provenant de différents pays. Notre thèse comporte trois chapitres. Chapitre 1 commence par présenter le processus de développement d’un modèle de credit scoring, et fournit une application sur des données réelles d'une banque de détail basée en France. Visant à donner de nouvelles perspectives sur les pays émergents, Chapitre 2 analyse le marché du crédit à la consommation en Chine et enquête sur l'utilisation des modèles de credit scoring dans un tel marché prometteur. Chapitre 3 va plus loin que la littérature méthodologique précédente et se concentre sur les différentes techniques d'inférence des refusés qui peuvent corriger le biais de sélection lors de la construction d'un modèle de crédit scoring basé uniquement sur les dossiers acceptés. Ces chapitres présentent les différents aspects du crédit scoring, pour lesquels les principales problématiques de credit scoring seront traitées. / While credit scoring has been broadly used for more than fifty years and continued to be a great support on decision-making in countless businesses around the world, the amount of literature, especially empirical studies, available on this subject is still limited. Our aim in this thesis is to fill this gap by providing a profound analysis on credit scoring and credit decision processes, with various applications using real and extensive sets of data coming from different countries. The thesis is organized in three chapters. Chapter 1 starts by presenting the credit scoring development process, and provides an application to real data from a France-based retail bank. Aiming at providing new insights regarding emerging countries, Chapter 2 analyzes the Chinese consumer lending market and investigates the use of credit scoring in such a promising market. Chapter 3 goes further than the previous methodological literature and focuses on reject inference techniques which can be a way to address the bias when developing a credit-scoring model based solely on accepted applicants. These chapters provide a round tour on credit scoring, after which major issues in credit scoring are treated.

Praticamente todas as grandes instituições brasileiras que trabalham com concessão de crédito utilizam-se de modelos para avaliar o risco de inadimplência dos potenciais contratantes de produtos de crédito. Qualquer avanço nas técnicas, que resulte no aumento da precisão de um modelo de previsão, acarreta ganhos financeiros para a instituição. Neste trabalho são apresentados, em um primeiro momento, conceitos de crédito e risco. Posteriormente, a partir de uma amostra de dados, fornecida por uma grande instituição financeira brasileira, estão desenvolvidos três modelos, aplicando-se três técnicas para a classificação de clientes: Regressão Logística, Redes Neurais e Algoritmos Genéticos. Em uma etapa final, são avaliadas e comparadas a qualidade e performance dos modelos desenvolvidos, onde é apontado qual o modelo que melhor se ajusta aos dados. Os resultados obtidos pelos modelos de regressão logística e rede neural são satisfatórios e bastante próximos, sendo o primeiro ligeiramente superior. O modelo embasado por algoritmos genéticos apresenta também bons resultados embora num patamar inferior aos dois já citados. Este trabalho ilustra os procedimentos a serem adotados por uma empresa para identificar o melhor modelo de concessão de crédito que tenha boa aderência aos seus dados. A adoção do melhor modelo detectado permite o direcionamento da estratégia da instituição, podendo aumentar a eficiência do seu negócio. / Most of the large Brazilian institutions which work with credit concession use credit models to evaluate the risk of consumer loans. Any improvement in techniques that results in the precision increase of a prediction model, will provide financial gains to the institution. The first phase of this study introduces concepts of credit and risk. Subsequently, with a sample set of applicants from a large Brazilian financial institution, three credit scoring models are built applying three different techniques: Logistic Regression, Neural Networks and Genetic Algorithms. Finally, the quality and the performance of these models are evaluated and compared, and the best one is identified. The results obtained by the logistic regression model and neural network model are good and very similar, but the first one is slightly better. The results obtained with the genetic algorithm model are also good, but a little bit inferior. This study shows proceedings to be adopted by a financial institution in order to identify the best credit model to evaluate the risk of consumer loans. The use of the proper model will help the definition of an adequate business strategy and increase profits.

Submitted by Felipe Montenegro (fsmontenegro@bol.com.br) on 2018-09-30T16:50:09Z No. of bitstreams: 1 Dissertação Felipe Santiago Montenegro VF.pdf: 1151283 bytes, checksum: 7f4de8e37ff17ff90e6925d9fc5ffc0a (MD5) / Approved for entry into archive by Joana Martorini (joana.martorini@fgv.br) on 2018-10-01T21:33:33Z (GMT) No. of bitstreams: 1 Dissertação Felipe Santiago Montenegro VF.pdf: 1151283 bytes, checksum: 7f4de8e37ff17ff90e6925d9fc5ffc0a (MD5) / Approved for entry into archive by Suzane Guimarães (suzane.guimaraes@fgv.br) on 2018-10-02T13:24:09Z (GMT) No. of bitstreams: 1 Dissertação Felipe Santiago Montenegro VF.pdf: 1151283 bytes, checksum: 7f4de8e37ff17ff90e6925d9fc5ffc0a (MD5) / Made available in DSpace on 2018-10-02T13:24:09Z (GMT). No. of bitstreams: 1 Dissertação Felipe Santiago Montenegro VF.pdf: 1151283 bytes, checksum: 7f4de8e37ff17ff90e6925d9fc5ffc0a (MD5) Previous issue date: 2018-09-24 / O mercado de títulos de dívida offshore (bonds) é amplamente utilizado por empresas brasileiras a fim de obter recursos a taxas mais atraentes, melhorando seu perfil de endividamento assim como o retorno para o acionista. Outra vantagem neste tipo de captação é a possibilidade de as empresas mitigarem exposições cambiais inerentes ao seu negócio. O principal objetivo deste trabalho é entender como o mercado offshore avalia o risco de crédito das empresas brasileiras exportadoras de celulose. Foram utilizados dados de mercado da Bloomberg para avaliação do risco de crédito dos bonds, câmbio e risco-país. Como proxy para o risco de crédito foram utilizados dados de ZSpread de bonds com duration próxima. No caso do risco-país, foi utilizado o CDS do Brasil de 5 anos, inferior ao vencimento dos bonds analisados. Para o preço da celulose foi utilizada a mensuração da FOEX, amplamente reconhecida como referência de preço nesse mercado. O fator risco-país mostrou-se tão importante quanto o preço da celulose na avaliação do risco de crédito das empresas pelo mercado, conclusão que é condizente com os modelos reduced form. O câmbio mostrou ter uma relação positiva com o Z-Spread. / The bond market is widely used by Brazilian companies in order to obtain resources at attractive rates, improving their debt profile as well as the return to the shareholder. Another advantage in this type of funding is the ability of companies to mitigate currency exposures inherent in their business. The main objective of this work is to understand how the offshore market evaluates the credit risk of Brazilian pulp exporters. Bloomberg market data were used to assess the credit risk of bonds, foreign exchange and country risk. As proxy for credit risk the Z-Spread bond data was used, considering assets with similar durations. In the case of sovereign risk, Brazil’s 5-year CDS was used, lower than the maturity of the bonds analyzed. For the price of pulp, the measurement of FOEX, widely recognized as the reference price in this market, was used. The sovereign risk factor was as important as the price of pulp in the assessment of the companies' credit risk in the market, a conclusion that is consistent with the reduced form models. The exchange rate showed a positive relation with the ZSpread.

Submitted by Jean Fontes (jeanrfontes@yahoo.com.br) on 2018-06-28T00:22:20Z No. of bitstreams: 1 MFEE_Dissertação_Jean_Fontes.pdf: 17252759 bytes, checksum: 5fa63ebb1abfe8902987b38b857d341c (MD5) / Approved for entry into archive by GILSON ROCHA MIRANDA (gilson.miranda@fgv.br) on 2018-07-06T13:43:51Z (GMT) No. of bitstreams: 1 MFEE_Dissertação_Jean_Fontes.pdf: 17252759 bytes, checksum: 5fa63ebb1abfe8902987b38b857d341c (MD5) / Made available in DSpace on 2018-07-16T18:54:16Z (GMT). No. of bitstreams: 1 MFEE_Dissertação_Jean_Fontes.pdf: 17252759 bytes, checksum: 5fa63ebb1abfe8902987b38b857d341c (MD5) Previous issue date: 2018 / The post-2008 financial crisis intensified and improved risk management around the world. From 2014 to 2017, Brazil experienced a severe period of economic crisis culminating in the largest recession in history in 2016. The objective of this work is to measure the impact of this crisis on the credit spread in the secondary market of debentures and the consequent probability of default implicit of these assets. The work analyzes the data of the private credit curve in Brazil for the AAA, AA and A Ratings published daily by ANBIMA based on Nelson and Siegel (1987) parametric model with revision proposed by Diebold and Li (2006). Based on these data, we extracted the daily probability of default implicit using the reduced form of the Duffie and Singleton model (1999) proposed by Xu and Nencioni (2000). This study seeks to identify the perception of agents of the credit market in relation to the increase of risk in the current Brazilian economic scenario. The study concluded that there was a significant increase in the credit spread to the apex in 2016, decreasing during 2017 with the more favorable economic scenario and the fall in interest rates. However, the model data showed high daily volatility. Regarding Probability of Default, there was a great evolution in the perception of credit risk by agents, but there was a certain delay in the pricing of this risk when compared to other economic indicators. In the comparison of the model data with the calculated default probability data for each individual asset, a large difference was observed between assets with the same rating level and the average of the model data. The data of this model can be used in future work to set up portfolios with a better return risk ratio, besides attesting the usefulness of this tool to the economic agents to price their operations and to fulfill their expectations. / Os eventos pós-crise financeira de 2008 intensificaram e aperfeiçoaram o gerenciamento de risco em todo mundo. De 2014 a 2017, o Brasil vivenciou um grave período de crise econômica culminando na maior recessão da história em 2016. O objetivo deste trabalho é dimensionar o impacto dessa crise no spread de créditos no mercado secundário de debêntures e na consequente probabilidade de default implícita destes ativos. O trabalho analisa os dados da curva de crédito privado no Brasil para os Ratings AAA, AA e A divulgados diariamente pela ANBIMA com base na modelagem paramétrica de Nelson e Siegel (1987) com revisão proposta por Diebold e Li (2006). Com base nestes dados, extraiu-se a probabilidade de default implícita diária utilizando a forma reduzida do modelo de Duffie e Singleton (1999) proposta conforme Xu e Nencioni (2000). Este estudo busca identificar a percepção dos agentes do mercado de crédito privado em relação ao aumento do risco no atual cenário econômico brasileiro. O trabalho concluiu que houve relevante elevação do spread de crédito até o ápice em 2016, decrescendo ao longo de 2017 com o cenário econômico mais favorável e as quedas das taxas de juros. Porém, os dados do modelo passaram a apresentam alta volatilidade diaria. Em relação a Probabilidade de Default houve grande evolução da percepção de risco de crédito pelos agentes, porém houve um certo atraso na precificação deste risco quando comparado a outros indicadores econômicos. Na comparação dos dados do modelo com os dados de probabilidade de default calculado para cada ativo individualmente, observou-se grande diferença entre ativos com o mesmo nível de rating assim como em relação à média dos dados do modelo. Os dados deste modelo podem ser utilizados num trabalho futuro para montagem de carteiras com uma melhor relação de risco retorno, além de atestar a utilidade desta ferramenta para os agentes econômicos precificarem suas operações e balizarem suas expectativas.

nÃo hà / Este trabalho se inicia analisando os aspectos teÃricos relacionados ao financiamento das empresas e os riscos atrelados a esta atividade de emprÃstimo realizada pelo sistema financeiro bancÃrio. Dada uma estrutura Ãtima de capital buscada pelas empresas, passa-se a analisar se este parÃmetro ou conjunto de parÃmetros à um bom indicativo para discriminar as empresas quanto ao seu risco de crÃdito analisado pelo mercado financeiro. Em relaÃÃo à gestÃo de risco, serà testado um modelo, tendo como variÃvel explicativa principal a variÃvel(ou conjunto de variÃveis) utilizada como parÃmetro de sinalizaÃÃo ao mercado de limite de risco, dentro dos conceitos de seleÃÃo adversa e modelos de sinalizaÃÃo num ambiente em que impera a informaÃÃo assimÃtrica. Assim, o uso de um sinalizador Ãtimo da estrutura de capital pelos bancos levaria a um equilÃbrio de Nash1 com informaÃÃo assimÃtrica no mercado de fundos emprestÃveis. No desenvolvimento do modelo estatÃstico utilizamos um modelo Logit em virtude da nÃo normalidade e as condiÃÃes de nÃo linearidade do modelo de probabilidade linear, entretanto, a anÃlise discriminante e Probit serÃo testados concomitantemente para efeitos comparativos entre os modelos. Outro ponto importante à a incorporaÃÃo de um modelo de decisÃo de crÃdito com o uso de programaÃÃo Linear Inteira. O uso deste modelo incorpora cenÃrios prospectivos com a taxa de juros, qualificando o ponto de corte(limites de aceitaÃÃo) para tomada de decisÃo. Ressaltamos aqui a importÃncia do uso da anÃlise fatorial no tratamento e configuraÃÃo das variÃveis explicativas, ferramenta nÃo observada para modelagem de risco nas diversas referencias deste trabalho. Diversos mÃtodos estatÃsticos univariados e multivariados, assim como critÃrios qualitativos sÃo usados na discriminaÃÃo e classificaÃÃo do risco, no entanto, o uso da AnÃlise Fatorial qualifica ainda mas as variÃveis independentes usadas, colocando-as em grupos de explicaÃÃo que captam melhor os efeitos dos diversos indicadores econÃmicofinanceiros. Neste trabalho foram revisados os principais modelos de insolvÃncia para avaliaÃÃo de risco de crÃdito no Brasil, concluindo-se com uma proposta de adoÃÃo de um modelo estatÃstico com o uso do modelo Logit e ProgramaÃÃo Linear Inteira, com o objetivo de medir o risco associado ao financiamento e emprÃstimo a clientes. / This work to research the theory about enterprises financial, financial struture, risk of the borrowe (enterprises) to repay the loan, credit of banks. In views of the optimal capital struture, credit analyses examines factors that may lead to default in the repayment of a loan. As for the risk management the general kinds of risks are described, particularly the credit risc and the credit concession models are evaluated. The risc models will have the financial demonstrations of interprises, here can be viewed as a signal, about the concept of asymmetric information. Thus, the signal to leave a nash equilibrium in this credit market. In the development of the statistic model, using the Logit Model because the problems of functional form of the linear probability model, the resÃduos is heteroscedastic and not have normal distribuition. The discriminant analyse, probit e logit will be test. Another important point in this work is the decision model. This model have predtion of interest to improve the decision with the cutoff. Referring to the factorial analyse in the statistic of the independentes variable, the use of factorial analyses is not observations in the reference. Having this purpose in mind a statistic model was developed, using logit regression with factorial analyse in variable and linear programming. This project aims at evaluating the used models and proposing the adoption of new models, for the allowance for dobtful accounts, with the objetive of mensuring the risk related to customers financing and loan activities.