Dissertations / Theses on the topic 'Management of credit risk'

Credit risk management is becoming more and more important in recent years. Credit risk refers to the risk that an obligor fails to make payments on any type of debt at the time of maturity. Credit risk models are statistical tools to infer the future default probabilities and loss distribution of values of a portfolio of debts. This doctoral thesis focus on the application of credit risk management in different areas. To better understand the credit risk management, in the first chapter, we introduce the basic ideas in credit risk management and review the models developed in the last decades. To empirical test the performance of models reviewed in the first chapter, in the second chapter, we compare the reduce-form model with the structural model based on the China’s stock market. It turns out that both models contribute to explaining the default risk of listed firms, however, reduce-form model outperformances the structural model. The empirical results from the second chapter suggests that reduce-form model can better predict the firm’s default risk, but the correlated default risk between firms has not been answered yet. So therefore in the third chapter, we investigate the correlated default risk using copula theory which has been introduced in the first chapter. Based on the insurances firms and other financial firms in the US market, both short-term and long-term default dynamic correlations are found. Another interesting finding from the third chapter is that insurance firms which were considered to be stable actually have higher default risk. This motive us to further explore the determinants of default risk of insurance firms in the fourth chapter and new risk factors (macroeconomic and insurance-specific variables) are found.

Banking is topic, practice, business or profession almost as old as the very existence of man, but literarily it can be rooted deep back the days of the Renaissance (by the Florentine Bankers). It has sprouted from the very primitive Stone-age banking, through the Victorian-age to the technology-driven Google-age banking, encompassing automatic teller machines (ATMs), credit and debit cards, correspondent and internet banking. Credit risk has always been a vicinity of concern not only to bankers but to all in the business world because the risks of a trading partner not fulfilling his obligations in full on due date can seriously jeopardize the affaires of the other partner.

The axle of this study is to have a clearer picture of how banks manage their credit risk. In this light, the study in its first section gives a background to the study and the second part is a detailed literature review on banking and credit risk management tools and assessment models. The third part of this study is on hypothesis testing and use is made of a simple regression model. This leads us to conclude in the last section that banks with good credit risk management policies have a lower loan default rate and relatively higher interest income.

Práce pojednává o řízení rizik v leasingové společnosti. Popisuje proces hodnocení bonity klienta a faktory ovlivňující schvalování obchodních případů. Charakterizuje ratingový a scoringový model v konkrétní leasingové společnosti, hodnotí jejich nedostatky a navrhuje změny na jejich vylepšení. Obsahuje i praktický příklad komplexního hodnocení obchodního případu včetně posouzení bonity klienta prostřednictvím ratingového modelu a nástrojů finanční analýzy.

In this thesis, efforts are devoted to the stochastic modeling, measurement and evaluation of credit risks, the development of mathematical and statistical tools to estimate and predict these risks, and methods for solving the significant computational problems arising in this context. The reduced-form intensity based credit risk models are studied. A new type of reduced-form intensity-based model is introduced, which can incorporate the impacts of both observable trigger events and economic environment on corporate defaults. The key idea of the model is to augment a Cox process with trigger events. In addition, this thesis focuses on the relationship between structural firm value model and reduced-form intensity based model. A continuous time structural asset value model for the asset value of two correlated firms with a two-dimensional Brownian motion is studied. With the incomplete information introduced, the information set available to the market participants includes the default time of each firm and the periodic asset value reports. The original structural model is first transformed into a reduced-form model. Then the conditional distribution of the default time as well as the asset value of each name are derived. The existence of the intensity processes of default times is proven and explicit form of intensity processes is given in this thesis. Discrete-time Markovian models in credit crisis are considered. Markovian models are proposed to capture the default correlation in a multi-sector economy. The main idea is to describe the infection (defaults) in various sectors by using an epidemic model. Green’s model, an epidemic model, is applied to characterize the infectious effect in each sector and dependence structures among various sectors are also proposed. The models are then applied to the computation of Crisis Value-at-Risk (CVaR) and Crisis Expected Shortfall (CES). The relationship between correlated defaults of different industrial sectors and business cycles as well as the impacts of business cycles on modeling and predicting correlated defaults is investigated using the Probabilistic Boolean Network (PBN). The idea is to model the credit default process by a PBN and the network structure can be inferred by using Markov chain theory and real-world data. A reduced-form model for economic and recorded default times is proposed and the probability distributions of these two default times are derived. The numerical study on the difference between these two shows that our proposed model can both capture the features and fit the empirical data. A simple and efficient method, based on the ordered default rate, is derived to compute the ordered default time distributions in both the homogeneous case and the two-group heterogeneous case under the interacting intensity default contagion model. Analytical expressions for the ordered default time distributions with recursive formulas for the coefficients are given, which makes the calculation fast and efficient in finding rates of basket CDSs.
published_or_final_version
Mathematics
Doctoral
Doctor of Philosophy

The thesis investigates the exogenous maturity vintage model (EMV) as a framework for achieving unification in consumer credit risk analysis. We explore how the EMV model can be used in origination modelling, impairment analysis, capital analysis, stress-testing and in the assessment of economic value. The thesis is segmented into five themes. The first theme addresses some of the theoretical challenges of the standard EMV model – namely, the identifiability problem and the forecasting of the components of the model in predictive applications. We extend the model beyond the three time dimensions by introducing a behavioural dimension. This allows the model to produce loan-specific estimates of default risk. By replacing the vintage component with either an application risk or a behavioural risk dimension, the model resolves the identifiability problem inherent in the standard model. We show that the same model can be used interchangeably to produce a point-in-time probability forecast, by fitting a time series regression for the exogenous component, and a through-the-cycle probability forecast, by omitting the exogenous component. We investigate the use of the model for regulatory capital and stress-testing under Basel III, as well as impairment provisioning under IFRS 9. We show that when a Gaussian link function is used the portfolio loss follows a Vašíček distribution. Furthermore, the asset correlation coefficient (as defined under Basel III) is shown to be a function of the level of systemic risk (which is measured by the variance of the exogenous component) and the extent to which the systemic risk can be modelled (which is measured by the coefficient of determination of the regression model for the exogenous component). The second theme addresses the problem of deriving a portfolio loss distribution from a loan-level model for loss. In most models (including the Basel-Vašíček regimes), this is done by assuming that the portfolio is infinitely large – resulting in a loss distribution that ignores diversifiable risk. We thus show that, holding all risk parameters constant, this assumption leads to an understatement of the level of risk within a portfolio – particularly for small portfolios. To overcome this weakness, we derive formulae that can be used to partition the portfolio risk into risk that is diversifiable and risk that is systemic. Using these formulae, we derive a loss distribution that better-represents losses under portfolios of all sizes. The third theme is concerned with two separate issues: (a) the problem of model selection in credit risk and (b) the problem of how to accurately measure probability of insolvency in a credit portfolio. To address the first problem, we use the EMV model to study the theoretical properties of the Gini statistic for default risk in a portfolio of loans and derive a formula that estimates the Gini statistic directly from the model parameters. We then show that the formulae derived to estimate the Gini statistic can be used to study the probability of insolvency. To do this, we first show that when capital requirements are determined to target a specific probability of solvency on a through-the-cycle basis, the point-in-time probability of insolvency can be considerably different from the through-the-cycle probability of insolvency – thus posing a challenge from a risk management perspective. We show that the extent of this challenge will be greater for more cyclical loan portfolios. We then show that the formula derived for the Gini statistic can be used to measure the extent of the point-in-time insolvency risk posed by using a through-the-cycle capital regime. The fourth theme considers the problem of survival modelling with time varying covariates. We propose an extension to the Cox regression model, allowing the inclusion of time-varying macroeconomic variables as covariates. The model is specifically applied to estimate the probability of default in a loan portfolio, where the experience is decomposed the experience into three dimensions: (a) a survival time dimension; (b) a behavioural risk dimension; and (c) calendar time dimension. In this regard, the model can also be viewed as an extension of the EMV model – adding a survival time dimension. A model is built for each dimension: (a) the survival time dimension is modelled by a baseline hazard curve; (b) the behavioural risk dimension is modelled by a behavioural risk index; and (c) the calendar time dimension is modelled by a macroeconomic risk index. The model lends itself to application in modelling probability of default under the IFRS 9 regime, where it can produce estimates of probability of default over variable time horizons, while accounting for time-varying macroeconomic variables. However, the model also has a broader scope of application beyond the domains of credit risk and banking. In the fifth and final theme, we introduce the concept of embedded value to a banking context. In longterm insurance, embedded value relates to the expected economic value (to shareholders) of a book of insurance contracts and is used for appraising insurance companies and measuring management's performance. We derive formulae for estimating the embedded value of a portfolio of loans, which we show to be a function of: (a) the spread between the rate charged to the borrower and the cost of funding; (b) the tenure of the loan; and (c) the level of credit risk inherent in the loan. We also show how economic value can be attributed between profits from maturity transformation and profits from credit and liquidity margin. We derive formulae that can be used to analyse the change in embedded value throughout the life of a loan. By modelling the credit loss component of embedded value, we derive a distribution for the economic value of a book of business. The literary contributions made by the thesis are of practical significance. The thesis offers a way for banks and regulators to accurately estimate the value of the asset correlation coefficient in a manner that controls for portfolio size and intertemporal heterogeneity. This will lead to improved precision in determining capital adequacy – particularly for institutions operating in uncertain environments and those operating small credit portfolios – ultimately enhancing the integrity of the financial system. The thesis also offers tools to help bank management appraise the financial performance of their businesses and measure the value created for shareholders.

Graphical User Interface, which is known as GUI, is a way for a person to communicate and interact with a system through icons or other visual indicators. A well designed and intuitive user interface is critical to the success of a system since it encourages a natural interaction between a user and a system, thus conveying information more clearly and efficiently to the user.The aim of this study is to design and develop a user interface that is used in a financial technology company in their credit risk assessment process. The current user interface contains a visualization of an individual credit assessment flow together with a lot of data that is generated in the process. Some of the data is not properly visualized, which leads to confusion among end users.In order to optimize the user experience, a user-centered design approach was used combined with a heuristic evaluation. A new user interface was designed and implemented and according to the heuristic evaluation result, the usability was greatly improved. The new interface is able to help the company to visualize their credit risk assessment process in a better way and facilitate credit officers to make credit decisions. The result could also provide insights to other companies or organizations in presenting their data more clearly and effectively.
Grafiskt användargränssnitt, som även kallas GUI, är ett sätt för en person att kommunicera och interagera med ett system genom ikoner eller andra visuella indikatorer. Ett väl utformat och intuitivt användargränssnitt är avgörande för framgången för ett system, eftersom det uppmuntrar till en naturlig interaktion mellan en användare och ett system och därmed förmedlar information tydligare och effektivare till användaren.Syftet med denna studie är att designa och utveckla ett användargränssnitt som används i ett finansiellt teknikföretag i deras kreditriskbedömningsprocess. Det nuvarande användargränssnittet innehåller en visualisering av ett individuellt kreditbedömningsflöde tillsammans med mycket data som genereras i processen. En del av data är inte korrekt visualiserade, vilket leder till förvirring bland slutanvändare.För att optimera användarupplevelsen användes en användarcentrerad designmetod i kombination med en heuristisk utvärdering. Ett nytt användargränssnitt designades och implementerades och enligt det heuristiska utvärderingsresultatet förbättrades användbarheten kraftigt. Det nya gränssnittet kan hjälpa företaget att visualisera sin kreditriskbedömningsprocess på ett bättre sätt och underlätta kreditansvariga att fatta kreditbeslut. Resultatet kan också ge andra företag eller organisationer insikter om att presentera sina uppgifter tydligare och mer effektivt.

This thesis studies the Chinese banks’ credit risk assessment using the Post Keynesian approach. We argue that bank loans are the major financial sources in emerging economies and it is uncertainty, an unquantifiable risk, rather than asymmetric information about quantifiable risk, as held by the mainstream approach, which is most important for the risk attached to credit loans, and this uncertainty is particularly important in China. With the universal existence of uncertainty, borrowers and lenders have to make decisions based on convention and experience. With regard to the nature of decision-making, this implies the importance of qualitative methods rather than quantitative methods. The current striking problem in Chinese banking is the large amount of Non-Performing Loans (NPLs) and this research aims to address the NPLs through improving credit risk management. Rather than the previous literature where Western models are introduced into China directly or with minor modification, this work advocates building on China’s conventional domestic methods to deal with uncertainty. We briefly review the background of the Chinese banking history with an evolutionary view and examine Chinese conventions in the development of the credit market. Based on an overview of this history, it is argued that Soft Budget Constraints (SBC) and the underdeveloped risk-assessing mechanism contributed to the accumulation of NPLs. Informed by Western models and experience, we have made several suggestions about rebuilding the Chinese convention of credit risk assessment, based on an analysis of publications and interviews with Chinese bankers. We also suggest some further development of the Asset Management Companies (AMCs) which are used to dispose of the NPLs.

I explore whether inter-firm linkages affect firms' credit risk. After controlling for the endogeneity between a firm's credit risk and its dependence on customers and suppliers, I find that supply-chain relationships affect firms' credit risk. My results indicate firms with exposure to major customers have lower ratings, and the level of firm dependence on major customers is negatively associated with firms' credit ratings. Further, I show when a firm's customers also depend on it, this mitigates the negative effect of dependence on credit risk. Finally, I document a negative association between a customer's reliance on its dependent suppliers and the customer's credit rating. Overall, my results provide insights regarding how inter-firm relationships between corporate customers and suppliers affect credit risk.

This thesis is concerned with the assessment, perception and insurance of credit risk. The thesis aims to make contributions both within these areas, and at specific points of interface between them. No attempt is made to develop a single unifying thesis. Rather, a series of partial models are developed, both theoretical and empirical, that develop and connect particular facets of financial economics. The first model demonstrates how movements in market risk produce movements in lender risk-assessment effort. It is demonstrated that deleterious movements in market-wide risk can actually produce a fall in assessment effort. The capricious nature of risk assessment causes changes in the lender's perception of the weights placed on determinants. This has important implications for borrowers' attempts to minimize risk premiums. Time-variability of signal-weights is tested using structural break tests on ordinary least squares and fixed effects panel models. Results suggest a fluid relationship between risk and determinants. Central to empirical investigation is the measurement of perceived risk. A critique of potential measures rejects the use of interest rate spreads - the most commonly used measure - on the basis that they do not take into account the possibility of credit rationing. A model is then constructed to reproduce the standard explanation of credit rationing - Adverse Selection induced Credit Rationing Equilibrium (ASCRE). This model is then extended to include classificatory risk assessment. Assessment is found to reduce the scope for ASCRE, and to cause favourable selection. Credit insurance is then included, and it is found that insurance cover makes risk assessment less of an imperative to lenders, and reduces the utility losses from raising interest rates. The parallel implication is that credit insurance weakens ASCRE, to the extent that full insurance with flat-rate premiums removes the possibility of ASCRE altogether. If the terms of insurance are made contingent on the terms of the loan, a new form of credit rationing emerges: Contingent Insurance induced Credit Rationing Equilibrium (CICRE). CICRE is separate, but not mutually exclusive, to ASCRE. A theoretical model of the demand for loan insurance is developed, and empirically estimated, in the context of the UK mortgage market. Inter alia, the model examines the role of auto-perception of risk determining credit insurance demand. Results reveal the take-up of credit insurance to be relatively insensitive to the borrower's perception ofhis/her own risk.

This report is for MA575: Market and Credit Risk Models and Management, given by Professor Marcel Blais. In this project, three different methods for estimating Value at Risk (VaR) and Expected Shortfall (ES) are used, examined, and compared to gain insightful information about the strength and weakness of each method. In the first part of this project, a portfolio of underlying assets and vanilla options were formed in an Interactive Broker paper trading account. Value at Risk was calculated and updated weekly to measure the risk of the entire portfolio. In the second part of this project, Value at Risk was calculated using semi-parametric model. Then the weekly losses of the stock portfolio and the daily losses of the entire portfolio were both fitted into ARMA(1,1)-GARCH(1,1), and the estimated parameters were used to find their conditional value at risks (CVaR) and the conditional expected shortfalls (CES).

Frail credit risk management practices have dragged financial intermediaries into financial crisis or bankruptcy if not well managed. The study seeks to appraise the intent to which Nigerian banks have meritoriously managed credit risk after the 2005 bank recapitalization exercise. It also seeks to establish other factors on why some banks to fail the 2009 stress test conducted by Central Bank of Nigeria. The study found that the failure to effectively manage credit risk as a result of increase capital inflow into the banking system and excessive lending contributed immensely to the 2009 banking crisis. The research also identified lax credit risk management practices as a major factor that caused the crisis. Furthermore, banks to develop and implement their credit I scoring models for assessing, monitoring and reviewing of credit portfolios and other credit granted.

Credit risk is one of the most general risks that exist in the financial market and a major risk faced by financial institutions. Credit risk management (CRM) is to identify, measure, monitor, and control risk arising from the possibility of default in loan repayments. The primary objective of CRM of rural commercial banks (RCBs) is to maintain risk within acceptable parameters and satisfy the regulatory requirements. CRM has long been the focus of governments, regulatory authorities and financial institutions. This thesis examines the importance of CRM for RCBs, which has been overlooked in the literature, and attempts to develop a CRM framework for RCBs. It has four specific research objectives: 1) to discuss the differences between RCBs and city based-commercial banks; 2) to examine the importance of CRM for RCBs and identify the approaches available for banks to manage credit risks; 3) to identify the key factors that have influenced the credit evaluation and assessment, as well as credit risk control in the context of China's RCBs; and 4) to propose a practicable CRM framework that suits the characteristics of Chinese RCBs. This study adopts qualitative analysis and case study approaches to identify key factors contributing to the failure of RCBs' customers, resulting in loan defaults and banks' credit risk. The quantitative-based CRM tools available for large financial institutions do not meet the requirements of RCBs because the main customers of RCBs are small and medium-sized enterprises (SMEs) and farming households and there is a lack of financial data and credit rating relating to these customers. In addition to normal risks faced by financial institutions, RCBs in China are also exposed to risks specifically to rural commercial banking business and in particular, farming-related loans and services. This study proposes a CRM framework for RCBs in China. The framework is based on the identification of business failures of RCBs' customers and factors contributing to the failures of SMEs and farming households. The framework is divided into five steps. The first step is to distinguish business failure and closure. The second step is to identify factors contributing to the failure of customers, which should be considered from environmental, operational, financial and guanxi aspects. The third step is to use PCA to identify principal factors. The fourth step is to design a credit risk analysis model with an analysis of these principal factors. The final step is to use the credit risk analysis model to manage credit risks of their portfolios and individual loans provided to SMEs and farming households. The CRM framework has been confirmed by practitioners through interviews conducted in the case bank. Interviews raise a number of issues relating to the development of a CRM model and assessment of credit risk of SMEs in China. The case study through an analysis of documents of the case bank reveals the importance of CRM and organisational structure in risk management and CRM. The case study presents evidence of lacking of practical methods in managing credit risk by RCBs in China. The proposed framework expects to address the problem. This study has made several contributions to the literature that studies CRM in financial institutions in general and RCBs in particular. This study critically identifies the current lack of studies specifically addressing the RCBs' CRM, and proposes a CRM framework for RCBs. The framework considers financial and non-financial variables to analyse SMEs and farming household for which financial information is very limited. Using nonfinancial variables along with financial variables as predictors of business failure significantly improves credit analysis quality and accuracy. Also, this study recognises guanxi as risk potentials affecting the business of SMEs and farming households and includes guanxi risks in the framework. The consideration of guanxi in credit risk analysis fits well with China's business environment.

This dissertation comprises three essays. In the first essay, we provide results for the valuation of European style contingent claims for a large class of specifications of the underlying asset returns. Our valuation results obtain in a discrete time, an infinite state-space setup using the no-arbitrage principle. Our approach allows for general forms of heteroskedasticity in returns. It also allows for conditional non-normal return inno- vations, which is critically important because heteroskedasticity alone does not su¢ ce to capture the option smirk. The resulting risk-neutral return dynamics are from the same family of distributions as the physical return dynamics. Our framework nests the valuation results obtained by Duan (1995), and Heston and Nandi (2000) by allowing for a time-varying price of risk and non-normal innovations. In the second essay, we develop a methodology to study the linkages between equity and corporate bond risk premia and apply it to a large panel of corporate bond transaction data. We find that a significant part of the time variation in bond default risk premia can be explained by equity-implied bond risk premium estimates. We compute these estimates using a recent structural credit risk model. In addition, we show by means of linear regressions that augmenting the set of variables predicted by typical structural models with equity-implied bond default risk premia significantly increases explanatory power. This, in turn, suggests that time-varying risk premia are a desirable feature for future structural models. In the third essay, we first document empirically that embedded put option values are related to proxies for term structure risk, default risk and illiquidity. In a second step, we develop a valuation model that simultaneously captures default and interest rate risk. We use this model to disentangle the reduction in yield spread enjoyed by putable bonds that can be attributed to each risk. Perhaps surprisingly, the most imp
Cette thèse comprend trois essais. Dans le premier essai nous avons développé des résultats pour l'évaluation des actifs contingents de type Européen pour une vaste classe de spécification du rendement de l'actif sous-adjacent. Notre méthode est obtenue dans une économie à temps discret et espace infini en utilisant seulement la condition de non arbitrage dans le marché. Notre approche permet une forme générale d'heteroskedasticité pour les rendements. Les résultats pour les cas d'homoskedasticité sont retrouvés comme des cas spéciaux. Notre approche permet d'accommoder les cas où l'innovation dans la dynamique du rendement est conditionnellement non normale. Cette flexibilité est extrêmement importante car l'heteroskedasticité seulement n'est pas su¢ sant pour cap- turer le phénomène du "smirk" dans les prix des options. Nos résultats emboîtent ceux obtenue dans Duan (1995) et Heston et Nandi (2000). Dans le deuxième essai nous avons développé une méthodologie pour étudier le lien entre la prime de risque dans les obligations corporatives et celle de l'actif risqué de la firme. Nous avons appliqué notre méthode sur une large base de données des transactions des obligations corporatives. Nous avons trouvé qu'une importante partie de la variation temporelle du risque de défaut dans ces obligations peut être expliquer par des estimées de la prime de risque du défaut reconstruite à partir de l'actif risqué de la firme seulement. En plus, nous avons démontré à l'aide des régressions linéaires qu'augmentant la série des variables prédites par le modèle structurel par notre estimé de la prime du risque de défaut ajoute une explication significative. Dans le troisième essai nous avons montré empiriquement que la valeur des obligations corporatives du type" puttable" est reliée aux risques de défaut, de liquidité et celui dû aux taux d'intérêts. Dans la deuxième étape de ce projet nous avons développé un mo

Credit risk management is a significant fragment in financial institutions' security precautions against the downside of their investments. A major quandary within the subject of credit risk is the modeling of simultaneous defaults. Globalization causes economises to be affected by innumerous external factors and companies to become interdependent, which in turn enlarges the complexity of establishing reliable mathematical models. The precarious situation is exacerbated by the fact that managers often suffer from the lack of data. The default correlations are most often calibrated by either using financial and/or market information. However, there exists circumstances where these types of data are inaccessible or unreliable. The problem of scarce data also induces diculties in the estimation of default probabilities. The frequency of insolvencies and changes in credit ratings are usually updated on an annual basis and historical information covers 20-25 years at best. From a mathematical perspective, this is considered as a small sample and standard statistical models are inferior in such situations. The first part of this thesis specifies the so-called entropy model which estimates the impact of macroeconomic fluctuations on the probability of defaults, and aims to outperform standard statistical models for small samples. The second part specifies the CIMDO, a framework for modeling correlated defaults without financial and market data. The last part submits a risk analysis framework for calculating the uncertainty in the simulated losses. It is shown that the entropy model will reduce the variance of the regression coefficients but increase its bias compared to the OLS and Maximum Likelihood. Furthermore there is a significant difference between the Student's t CIMDO and the t-Copula. The former appear to reduce the model uncertainty, however not to such extent that evident conclusions were carried out.
Kreditriskhantering är den enskilt viktigaste delen i banker och finansiella instituts säkerhetsåtgärder mot nedsidor i deras investeringar. En påtaglig svårighet inom ämnet är modelleringen av simultana konkurser. Globalisering ökar antalet parametrar som påverkar samhällsekonomin, vilket i sin tur försvårar etablering av tillförlitliga matematiska modeller. Den prekära situationen förvärras av det faktum att analytiker genomgående saknar tillräcklig data. Konkurskorrelation är allt som oftast kalibrerad med hjälp av information från årsrapporter eller marknaden. Dessvärre existerar det omständigheter där sådana typer av data är otillgängliga eller otillförlitliga. Samma problematik skapar även svårigheter i skattningen av sannolikheten till konkurs. Uppgifter såsom frekvensen av insolventa företag eller förändringar i kreditbetyg uppdateras i regel årligen, och historisk data täcker i bästa fall 20-25 år. Syftet med detta examensarbete är att ge ett övergripande ramverk för kreditriskhantering i avsaknad av finansiell information och marknadsdata. Detta innefattar att estimera vilken påverkan fluktueringar i makroekonomin har på sannolikheten för konkurs, modellera korrelerade konkurser samt sammanfatta ett ramverk för beräkning av osäkerheten i den estimerade förlustdistributionen. Den första delen av examensarbetet specificerar den så kallade entropy modellen. Denna skattar påverkan av makroekonomin på sannolikheterna för konkurs och ämnar att överträffa statistiska standardmodeller vid små datamängder. Den andra delen specificerar CIMDO, ett ramverk för beräkning av konkurskorrelation när marknads- och företagsdata saknas. Den sista delen framlägger ett ramverk för riskanalys av förlustdistributionen. Det visas att entropy modellen reducerar variansen i regressionskoefficienter men till kostnad av att försämra dess bias. Vidare är det en signifikant skillnad mellan student’s t CIMDO och t-Copula. Det förefaller som om den förstnämnda reducerar osäkerheten i beräkningarna, men inte till den grad att uppenbara slutsatser kan dras.

Small, Medium and Micro Enterprises (SMMEs) make a significant contribution to the South African Economy. Regardless of size, these businesses have the ability to create employment, make a generous contribution to tax collections, uplift communities and serve as a beacon of hope for those trapped in the cycle of poverty and unemployment. However, SMMEs lack access to much-needed financial resources that are critical for their growth. Development Finance Institutions (DFIs) aim to bridge the gap between the SMME’s financial needs and the development of the respective SMME businesses, by providing funding to entrepreneurs with potentially viable businesses and ideas. Debt funding to these SMMEs are based on sound commercial lending principles that take various non-quantitative variables into account. The sustainability of SMMEs is a primary concern to all participants in the economy, as it is known that SMME failure rates are high Therefore, the primary objective of this study was to investigate the impact that the credit risk management practices of DFIs have on the sustainability of SMMEs, by examining a case study of a typical DFI. An electronic questionnaire survey was considered as an appropriate measurement method for this study. The targeted population of the study included SMMEs in the Eastern Cape that are Trust for Urban Housing (TUHF) clients and 23 SMMEs were identified as part of the study sampling frame. A total number of 14 questionnaires were returned out of the 23 targeted SMMEs - giving a response rate of 61%. The quantitative data was processed using the STATISTICA program, leading to appropriate descriptive statistical analyses. In order to better understand the impact of credit risk management practices on the sustainability of SMMEs, a hypothesis was formulated and linear regression analysis was used to establish the statistical significance of certain credit risk principles and sustainability characteristics. The results of the empirical study revealed that credit risk management practises do impact on the sustainability of SMMEs. Further, by testing the hypothesis, it was also revealed that certain sustainability variables are regarded as more important than others.

Arguably, the credit risk models reported in the literature for the retail lending sector have so far been less developed than those for the corporate sector, mainly due to the lack of publicly available data. Having been given access to a dataset on defaulted mortgages kindly provided by a major UK bank, this work first investigates the Loss Given Default (LGD) of mortgage loans with the development of two separate component models, the Probability of Repossession (given default) Model and the Haircut (given repossession) Model. They are then combined into an expected loss percentage. Performance-wise, this two-stage LGD model is shown to do better than a single-stage LGD model (which directly models LGD from loan and collateral characteristics), as it achieves a better Rsquare value, and it more accurately matches the distribution of observed LGD. We next investigate the possibility of including macroeconomic variables into either or both component models to improve LGD prediction. Indicators relating to net lending, gross domestic product, national default rates and interest rates are considered and the interest rate is found to be most beneficial to both component models. Finally, we develop a competing risk survival analysis model to predict the time taken for a defaulted mortgage loan to reach some outcome (i.e. repossession or non-repossession). This allows for a more accurate prediction of (discounted) loss as these periods could vary from months to years depending on the health of the economy. Besides loan- or collateral-related characteristics, we incorporate a time-dependent macroeconomic variable based on the house price index (HPI) to investigate its impact on repossession risk. We find that observations of different loan-to-value ratios at default and different security type are affected differently by the economy. This model is then used for stress test purposes by applying a Monte Carlo simulation, and by varying the HPI forecast, to get different loss distributions for different economic outlooks.

In this dissertation, I consider a range of topics in bankruptcy, credit risk and asset pricing. The first chapter proposes a structural-equilibrium model to examine some economic implications arising from voluntary filing of Chapter 11. The results suggest that conflict of interests (between debtors and creditors) arising from the voluntary filing option causes countercyclical losses in firm value. The base calibration shows that these losses amount to approximately 5% of the ex-ante firm value and are twice those produced by a model without incorporating the business cycles. Furthermore, besides countercyclical liquidation costs as in Chen (2010) and Bhamra, Kuehn and Strebulaev (2010), countercyclical pre-liquidation distress costs and the conflict of interests help to generate reasonable credit spreads, levered equity premium and leverage ratios. The framework nests several important models and prices the firm's contingent claims in closed-form. The second chapter proposes a structural credit risk model with stochastic asset volatility for explaining the credit spread puzzle. The base calibration indicates that the model helps explain the credit spread puzzle with a reasonable volatility risk premium. The model fits well to the dynamics of CDS spreads and equity volatility in the data. The third chapter develops a consumption-based learning model to study the interactions among aggregate liquidity, asset prices and macroeconomic variables in the economy. The model generates reasonable risk-free rates, equity premium, real yield curve, and asset prices in equity and bond markets. The base calibration implies a long-term yield spread of around 185 basis points and a liquidity premium of around 55 basis points for an average firm in the economy. The calibrated yield spread and liquidity premium are consistent with the empirical estimates.

Organisations or individuals that lend money (banks and micro lenders) or that sell goods on credit (retailers) are classified as credit providers. The debtor enters into a contractual agreement with a credit provider, or creditor, with the obligation to repay the loan amount, fees and interest according to a predetermined schedule. The contractual agreement, also known as a credit agreement, is as a general rule very complex. Legislation protecting debtors in various ways is an international phenomenon. In South Africa, the National Credit Act, Act 34 of 2005 (NCA) was enacted in 2005. The NCA changed the playing field for credit providers participating in the South African consumer credit market to a great extent. Consumer lending is the sleeping giant of the financial sector. The key to successfully unlock this enormous market is the credit provider's ability to accurately assess the creditworthiness of a potential customer during the customer acquisition phase. The creditworthiness of the customer is related to the risk of default, i.e. a debtor's non-payment of debt in terms of the credit agreement. The risk of default is also known as credit risk. Real People Investment Holdings (Pty) Ltd (RPIH) classifies credit risk as the single largest risk the Group is exposed to. They recognise that the intelligent and responsible management of credit risk makes it the Group's largest profit driver. Credit risk scorecards are excellent decision aids during the customer acquisition phase. The characteristics and behaviour of merchants submitting credit applications to RPIH for assessment have a definite impact on the credit risk of the Group. The merchant plays a pivotal role in the debtor-creditor-supplier business model. The merchant influences the customer's sales experience and subsequent level of satisfaction with the transaction. A satisfied customer constitutes a lower level of credit risk for the creditor, in this case RPIH. The research is conducted with a positivistic paradigm. The cross-sectional study approach is used. The merchant is the unit of analysis. A sample of 77 merchants is selected from the population of 244 merchants who submitted credit applications to RPIH during the observation period. Questionnaires are used as the data collection method in this research project. The predictive ability of fourteen merchant related characteristics are demonstrated through this empirical study.

In this thesis, we study random bridges of a certain class of Lévy processes and their applications to credit risky asset pricing. In the first part, we construct the compound random bridges(CLRBs) and analyze some tools and properties that make them suitable models for information processes. We focus on the Markov property, dynamic consistency, measure changes and increment distributions. Thereafter, we consider applications in credit risky asset pricing. We generalize the information based credit risky asset pricing framework to incorporate prematurity default possibilities. Lastly we derive closed-form expressions for default trends and intensities for credit risky bonds with CLRB as the background partial information process. We obtain analytical expressions for specific CLRBs. The second part looks at application of stochastic filtering in the current information based asset pricing framework. First, we formulate credit risky asset pricing in the information-based framework as a filtering problem under incomplete information. We derive the Kalman-Bucy filter in one dimension for bridges of Lévy processes with a given finite variance.

Tito Mboweni (2011) said one of South Africa’s biggest tests is the overwhelming the skills shortage. He was echoing the views of Higher Education Minister Blade Nzimande who himself said “South Africa could not afford to have an economy "constrained by a severe lack of skills". There are numerous initiatives that having been undertaken by government in an attempt to solve the skills shortage problem. However, these initiatives are not aimed at the tertiary education system. The tertiary education system is the focus of this study as the author investigates how the NMMU Business School can play a significant role in addressing the skills shortage in the credit risk management sector. Following a literature review, surveys were completed by the NMMU Business School MBA students (ninety of them completed it) and personal interviews were conducted with three Provincial HR managers from South Africa’s “four big banks” in Nelson Mandela Bay (Nedbank, Standard Bank and ABSA). The study found that the skills shortage is indeed a problem. The study found that reasons including the legacy left by apartheid and students pursuing the wrong degrees were highlighted as some of the reason for this skills shortage. An opportunity for the NMMU Business School was identified to support the banking industry in addressing credit risk management skills shortage. The benefits include financial reward and more importantly an opportunity to differentiate the Business School and the courses offered at the school from the rest. Some of the recommendations included sourcing of the best practices from institutions like the Millpark Business School on effective partnering with the banking industry as well as a proactive approach to be adopted by the banking industry in terms of lobbying support from other potential role players for example but not limited to, student bodies, BankSeta and the smaller banks in the industry.

Thesis: S.M. in Management Studies, Massachusetts Institute of Technology, Sloan School of Management, 2019
Cataloged from PDF version of thesis.
Includes bibliographical references (page 40).
Credit risk is often a critical risk in the financial sector. Therefore, how a financial institution manages its credit risk is an important determinant of profitability and solvency. In this regard, the identification and measurement of credit risk is the first component of efficient risk management. Correct and timely credit ratings are important for risk management systems, and for informing regulators about financial system risks. Credit risk is the main risk faced by the Paraguayan financial sector. Effectively managing it requires banking supervision and regulation in line with international best practices. As a step in that direction, this research assesses the Paraguayan banking regulation of credit risk and compares it to the principles and the best practices about credit risk management issued by the Basel Committee. I propose principles to guide the implementation of statistical models for better measurement of credit risk in Paraguayan financial institutions.
by Bruno Gomez.
S.M. in Management Studies
S.M.inManagementStudies Massachusetts Institute of Technology, Sloan School of Management

This thesis discusses three topics in the area of quantitative finance in relation to risk and credit portfolio management. Chapter 2 investigates the issue of estimating and testing the goodness-of-fit of a model for a dependence break. The dependence is modelled by copulas and an unknown break of dependence structure is allowed for by including a dummy variable in the copula. The model is selected by minimizing the Akaike Information Criterion (AIC) of each candidate breaking point. The candidate models are estimated by a well-established two-step Maximum Likelihood (ML) approach, namely "Inference Function for Margin" (IFM). Moreover, we examine 5 single-factor copulas and compare them to each other by AIC criteria. A parametric bootstrap goodness-of-fit test is also proposed. Empirically, the dependence structures of stock indices between the US-UK and US-Japan markets during the Subprime crisis are examined. We found breaks in both dependence structures. In Chapter 3, a new general approach is developed for optimizing a credit portfolio by minimizing the default risk of a whole credit portfolio subject to a certain target premium. The approach is rooted in concepts from Modem Portfolio Theory. The default risk is measured by a quadratic form of weights and a matrix containing information about default correlations between any two single-names and default intensities of each single-name. The default correlation and the default intensities are modelled by a new binomial intensity model. A Genetic Algorithm (GA) approach is also introduced to optimize a credit portfolio with the purpose of overcoming limitations of the analytical method and the traditional numerical method based on the first order condition. Empirically, the approach is applied to optimize Credit Default Swap (CDS) portfolios consisting of members of iTraxx and CDX indices. In Chapter 4, we focus on modelling counterparty risks of two important financial instruments: the Interest Rate Swap (IRS) and the CDS. Analytical solutions are derived for the theoretical fair prices of the IRS and the CDS under various assumptions of defaults of counterparties. Also a Monte Carlo approach is proposed as a numerical solution for the fair prices. Numerical experiments are designed to study the effects of various factors on the fair price. Empirically, we examine the counterparty risk of a CDS portfolio, composed of randomly selected single-names from iTraxx series 10.

In recent years, most developed countries have suffered a severe recession due to a financial crisis starting in the US with mortgages loans. The lack of credit risk management has been pointed out as one of the causes of this bank panics. To avoid a similar situation, the credit card companies need to have proper risk management tools. This thesis presents a credit scoring system which aims at setting credit lines and thus, controlling credit risk. It includes three types of models: application scorecards, early detection scorecards and behavioral scorecards. They have been built on real and recent data coming from a German credit card company. The models have been built with a training sample and validated accordingly, using logistic regression. Information value and validation charts have been used for comparing the models. In the scoring process described, the scorecards are used in a sequential order. The author shows that minimizing losses might not be optimal in order to maximize profit. Finally, the author presents possible extensions to the research. The author hopes that the microeconomic analysis of the mechanics of a particular lender’s credit allocation process described in this thesis can play some part in preventing future financial crisis.

The question of whether ratings agencies convey new information to financial markets when they assign new ratings or change previous ratings has been debated for at least 40 years. In this study I first examine equity market, bond market and CDS market reactions to long and short term rating changes from S&P, Fitch and Moody's. I find that not all the credit rating changes affect the market but only those classified as unanticipated. Subsequently, I study whether the regulatory setting, in which the Credit Ratings Agencies work, can possibly affect the financial markets reactions. Lastly I show that the probability of a future rating change is severely affected by different factors proportional hazard rate models.

In the following paper I have tried to find the correlation between type of ownership and effective risk management in the operations of microfinance institutions in India. The results found are consistent with the current findings of how the type of ownership does not impact both the financial or social performance of MFIs. Dataset of 72 MFIs was acquired from the Microfinance Information Exchange on MFIs and evaluated using an OLS regression. The results show that the type of ownership insignificantly impacts both the credit and liquidity risk ratios of MFIs. It is possible that the impact of ownership type is more evident in other aspects of operations. In the future, a study on type of ownership and exposure to strategic and market risks could be a way forward.

LBO transactions were structured increasingly aggressive in the years prior to the outbreak of the financial crisis (2007), as reflected in rising debt proportions. This was followed by many LBO credits experiencing difficulties to adhere to their loan documentations. Bank credit functions played a role in this, giving rise to an investigation into whether their work could be more effective. To identify areas for improvement in their work of evaluating LBO credits and – if such areas can be identified – deduce some potential measures how to address them was the aim of this research. The research is timely as evaluations of LBO credits continue to be required heavily. Many of the credits structured around 2006/2007 will soon require refinancing and in parallel new transactions come to the market. A literature review on the risk factors and cycles relating to LBOs, the simple techniques of portfolio management and LBO credit management practices found that all the ingredients required for effective LBO credit management are available. This is in conflict with the observation of increasing credit risk inherent in these transactions in the years just prior to the outbreak of the financial crisis of 2007. Based on this, 18 experts were interviewed. The results were analysed quantitatively and qualitatively. To enhance robustness, results were discussed with four senior credit executives as well as a focus group discussion of the credit function of one bank. The exploratory results of this research suggest that there is strong awareness of the risk factors in LBOs. The systematic risk in LBOs in form of an LBO cycle however is not considered to a significant degree in credit analysis/credit monitoring. Some important risk factors also receive relatively little attention in credit analysis/credit monitoring and aspects of portfolio management are not used strongly at the level of credit functions. Finally, an area of improvement that had been identified was the utilization of results; i.e. the need to draw consequences from observations made with view to risk factors. Due to the limited scope of the study, updating the results with more recent data as well validation and triangulation of results remain recommended.

碩士
國立中興大學
高階經理人碩士在職專班
100
This paper examines the effect of the idiosyncratic risk of a firm on its credit risk and the relationship between the credit risk and accrual or real earnings management under considering idiosyncratic risk. We further investigate the effects of composition of real earnings management on the credit risk of a firm. In sensitivity analysis, we examine the effects of the credit risk and the idiosyncratic risk of a firm on managerial behavior of income smoothing. The findings indicate that net cash flows of external financing, debt financing and equity financing activities are negatively related to the credit risk of a firm whether the idiosyncratic risk are measured by market model or Fama-French three factors model, implying that the higher financed funds, the higher credit rating of a firm is. It is because the firm has more investment opportunities and is in growth stage. The idiosyncratic risk of a firm is positively related to the credit risk. Next, both accrual and real earnings managements are positively related the credit risk under considering the idiosyncratic risk, and the idiosyncratic risk is also positively related to credit risk. As for the compositions of real earnings management, abnormal cash flows are negatively related to the credit risk but both abnormal production cost and abnormal discretionary expense are unrelated to credit risk. In this case, we also find the idiosyncratic risk of a firm is positively correlated with the credit risk. Moreover, in sensitivity analysis, the evidence indicates that the idiosyncratic risk is negatively related to earnings smoothing and the firm with higher credit risk prefers to income smoothing strategy.

碩士
國立臺灣大學
商學研究所
86
The main configuration of this research is to use a portfolio credit risk approach of CreditMetrics quantifies credit risks that arise due to increa sed exposure to an obligor or a group of correlated obligors. In this research, the credit risk includes not just default or insol vency risk but also changes in credit spreads and thereby market values, cha nges in credit ratings, and generic changes in credit quality. And foll ow this concept, this research has some issues as follows： 1、Use certain model to measure credit risk of any bonds、lo ans、receivables and derivatives whose value exposures are affected by the credit risk. 2、Use Black & Scholes and Vasicek model to extend the a sset pricing under the credit risk, which includes bonds pricing、loans pric ing、receivables pricing and derivatives pricing. 3、Introduce how to use this credit risk model to help obligees or investors to manage their assets'' credit exposures. 4、Introduce credit derivatives and develop new credit derivatives to help obligees or investors to hedge、 diversify a and gain access to credit exposures. 5、Use the result of the credit risk model and asset pricing to develop the pricing model of credit derivatives.

Exploring the main factors that determine bond spreads with respect to Treasury rates is one of the most critical issues in the corporate debt market. Credit risk has long been perceived as the most important determinant of bond spreads (Fisher, 1959). One of the most critical parameters in credit risk models is asset volatility, which includes idiosyncratic and systematic components. However, these models do not distinguish between them. Chapter 2 investigates the impact of idiosyncratic volatility on bond portfolio spreads between 2000 and 2010. While the prediction of traditional asset pricing models is that firm-specific risk should be diversified away at aggregate level, I find idiosyncratic volatility plays an incremental role in explaining bond portfolio spreads beyond the market factors. Recovery is an important measurement of credit risk additional to default probability. Chapter 3 focuses on the estimation of firm recovery after bankruptcy using the Leland and Toft (1996) model. Using a large sample of Chapter 11 filings from 1996 to 2007, I find that the recovery derived from the Leland and Toft model has strong explanatory power on the debt recovery observed in the market. Recent literature finds that all extant credit risk models significantly underestimate bond spreads, especially for investment grade bonds of short maturity. Chapter 4 identifies a heretofore ignored component, perceived accounting misstatement, by regressing bond spreads on the proxy of accounting misstatement propensity, while controlling for issuers’ default risk and bond illiquidity risk between January 1994 and June 2002. My thesis deepens the understanding of bond price discovery mechanisms and presents an important challenge for future research to incorporate the strong empirical relationship between idiosyncratic volatility and bond yields in asset pricing models. My thesis also sheds light on the accurate prediction of debt recovery, which is important to the valuation and hedging of risky debt and credit derivatives. Furthermore, my thesis assists in solving the credit spread puzzle by identifying a new risk factor. Overall, my thesis provides new insights into research on the corporate debt market and has important implications for academic scholars and market practitioners.
Thesis (Ph.D, Management) -- Queen's University, 2013-04-30 20:22:12.594

Counterparty credit risk (CCR) is the risk that a counterparty in a deal will not be able to meet their contractual obligations in the future. While CCR is an important task for any risk desk, it has often been underestimated due to the miss-conception that some counterparties were deemed to be either too big to fail or too big to be allowed to default. This was highlighted by the 2008 nancial crisis that saw respected banks, such as Lehman Brothers, and nancial service providers, such as AIG, default on their obligations. Since then there has been renewed interest in CCR, with the focus being on actively pricing and hedging it. In this work CCR is invistigated including its intersection with other forms of risk. CCR mitigation techniques are explored, followed by the formal quanti cation of CCR in the form of credit value adjustments (CVA). The analysis of CCR is then applied to interest rate derivatives, more speci cally forward rate agreements (FRAs) and interest rate swaps (IRSs). The e ect of correlation on unilateral and bilateral CVA between counterparties, including risk factors such as the interest rate, is investigated. This is invistigated under two credit risk modelling frameworks, the structural and intensity based frameworks. It is shown that correlation has a none-negligible e ect on both unilateral and bilateral CVA for FRAs and IRSs. Correlation structures, namely the Gaussian and the Student-t copula, are used to induce dependency in order to understand their e ect on both unilateral and bilateral CVA. It is shown that the choice of copula does not have signi cant e ect on either unilateral or bilateral CVA.

碩士
國立臺灣科技大學
企業管理系
99
This paper is to recommend banks to be always highly aware of the updated leading indicators to monitor if mortgage market is going to the bubble crisis and to ensure the quality of credit risk management based on the analysis of the following 4 dimensions: (1) Loan to Value Ratio (LVR) control, bank should not avoid any regulation requirement to focus on the business with less regulated. Blind competition could result in another wave of financial crisis. (2) Re-consider the Debt Service Ratio (DSR). With the double effects on interest raise and bubble crisis on property price, banks should reduce DSR to ensure customer’s long term repayment ability. (3) To prevent speculator from hyping the property price, banks should cooperate with FISC to identify borrower’s cash flow. (4) The quality of the land has been deteriorated due to the change of global environment and climate, banks should control the acceptable property located at hill land specifically. To cope with the Shoshika and the aging problem, two generations mortgage will be developed. Banks should be able to have qualified sons and daughters to be the guarantors to continue repaying the mortgage loan.

碩士
元智大學
管理碩士在職專班
105
Due to the vast territory of mainland China, the financial services are presently lack of convenience and popularity. Also, Chinese banks still mainly serve for local governments, large state-owned enterprises and listed companies, so that a large number of small and medium enterprises and individuals borrowing needs had not be satisfied. Since 2005, some non-financial industries have started to operate the Internet financial business. With the development of Internet technology and the rapid growth of P2P network borrowing, the convenience of the network has shortened the loan audit cycle and improve the turnover rate of funds, therefore this kind of new financial service model, P2P network borrowing, is widely accepted by the public. However, the risk management mechanism of Internet lending platform, whether policy risk or network risk has always been factors that affect the development of Internet lending.

碩士
國立交通大學
管理學院財務金融學程
100
Credit risk management or credit control, focuses on four key areas: • Deciding what credit to give a customer. • Everyday credit control. • Delayed payment and chasing debts. • Dealing with large companies Credit Control is aimed at serving the follow two purposes: to increase the sales revenue by extending credits to customers who are deemed good and to minimize the risk of losses from bad debts by restricting or denying credits to customers who are not good. This will improve the company's cash flows. Credit control is an important component in the overall profitability of many firms. The effectiveness of credit control procedures lies chiefly in the firm’s ability to judge the creditworthiness of potential customers. This is much more effective than trying to reclaim money from delinquent accounts.

博士
國立臺灣大學
經濟學研究所
96
The implementation of Basel II in 2007 has driven banks to enhance their risk management capability, and the IRB approach has become the goal that many banks are aiming for. The IRB capital calculation formula is based on many assumptions including independence between PD and LGD, no concentration risk and identical parameters applying to all countries. Popular credit risk models like CreditMetrics or CreditRisk+ also make assumptions that LGDs are constant or LGDs are independent of PDs and do not explicitly deal with the dependence issue between PD and LGD. For this reason, we try to build a joint model for PD and LGD. The PDs depend on expected stock returns, which are in turn affected by a common factor, as well as return thresholds that are determined by risk ratings. In contrast, LGDs depend on asset values and expected equity/debt ratios, in which asset values are decomposed to three categories with different guarantee powers, while expected equity/debt ratios are influenced by the common factor. The central idea of this joint model is that PD and LGD are both affected by the common factor and hence are correlated. Viewing listed companies in Taiwan as a portfolio, we could estimate all parameters of our joint model and run Monte Carlo simulation to generate portfolio’s credit loss distribution and calculate the corresponding economic capital. The simulated economic capital is larger than the economic capital under the independence assumption and is also larger than the IRB capital. In addition, to employ economic capital as the basis for risk pricing and risk adjusted performance measurement, we must allocate portfolio economic capital to individual counterparty, which requires further Monte Carlo simulations. From the simulations, we find the ranking of allocated economic capital substantially differs from the ranking of IRB capital. Our main conclusion is that banks need to explore the basic idea of IRB capital formula and use their internal data to construct economic capital models if they want to correctly measure and manage credit risks. Banks could then meet the requirement of the Pillar II in Basel II and allocate economic capital much more efficiently. Banks need to access loans according to comparative risk. Only when this is accomplished, will banks earn more risk adjusted profit. Their counterparty would also bear less financing cost. Social welfare could therefore improve.

This dissertation consists of three papers about credit risk management and covers different types of firms that can be seen in a bank portfolio. It is used a comprehensive sample maintained by the Portuguese central bank with approximately 43,000 observations on more than 16,000 firms for the period 1997-2003. Almost 2,500 of these observations relate to firms that entered into default according to Basel II definition, representing 6% of the total sample. For the empirical estimations, different methodologies and sub-samples were used to guarantee the fit of the data and the robustness of the results. The first paper (Chapter 1) studies the influence of the independent auditor's going concern evaluation by examining default following the release of the auditor's report. Professional auditing standards require the independent auditor to disclose the uncertainty in the auditor's opinion when there is substantial doubt about an entity's ability to continue as a going concern. The primary purpose of our study was to determine whether the independent auditor's going concern evaluation had information content by examining whether firms default following issuance of a GCO. We use a sub-sample on 12,199 audit reports relating to approximately 2,000 firms that are liable by law to have their accounts audited on an annual basis. Empirical estimation of a logit model controlling for accounting cash flow related and non-accounting variables shows that the likelihood of default for firms that received going concern opinion is 2.792 times that of firms that received a clean opinion. Likelihood ratio tests for omitted variable also confirm the incremental predictive ability of going concern opinion over and above accounting and non-accounting variables for the estimation and hold-out samples. In the non-defaulting group the average default rate is 6.05% and in the defaulting group it is 17.78%. The default rate for firms in the nondefaulting group that received a going concern opinion is 9.92% and for firms that received a clean opinion is 5.96%. In the defaulting group, the rate for firms that received a going concern opinion is 35.49% and for firms that received a clean opinion is 16.96%. Checks for robustness across different asset classes, age, industries and regions indicate that firms that receive a going concern opinion on average default more than those that receive a clean opinion. The second paper (Chapter 2) uses the total sample and analyzes default events for limited liability (non-listed) and full liability firms in order to understand the loan default characteristics of these two groups of firms. We test six hypotheses that relate the influence of accounting ratios and their distributional properties to default. Additionally, two hypotheses are tested relating the influence of liability status of firms and default and the incremental predictive ability of liability status over and above accounting ratios when predicting default. The following ratios - cash flow to total debt, operational cash flows to debt costs, equity to total assets and liquidity - are found to be negatively related to default. Size of the firm and age are also negatively related to default. Limited liability is positively related to default. The predicted default probability of full liability firms is 0.83 times that of limited liability firms. The likelihood ratio test for omitted variable also confirms the incremental predictive ability of liability status over and above accounting information for the estimation and hold-out samples. In the non-defaulting group, the average default rate for limited liability firms is 0.53% higher than that of full liability firms. This difference is 2.07% higher in the defaulting group. Cross tabulation by liability status and asset size shows that the advantage of full liability decreases with size. The third paper (Chapter 3) discloses a simple model based on accounting and nonaccounting ratios to determine default of privately-held firms and a method of computing default probabilities that can be used at the central bank in Portugal to calculate minimum capital requirements for the banking sector and utilize them as benchmarks in the validation of the internal models of the commercial banks. The Basel Committee on Banking Supervision advocates that central and commercial banks model and estimate default probabilities with specific country or bank portfolios. Little research exists on privately held firms and default. We use novel loan data for a sample of 31,025 accounts of privately-held firms maintained by the Portuguese central bank. Our data includes 30 accounting ratios and non-accounting information on size, age, industry and geographic regions. In relation to accounting ratios, we find interest costs to gross income, solidity and working capital to total assets to have the most significant influence on the probability of default. The accounting ratios that show lower marginal influence on the probability of default are return on investment, financial coverage, account payables and receivables, and return on equity. Additionally, we find that unlike the observation in the context of listed firms, size appears positively related to default and although age of firm appears negatively related to default, its marginal influence on default probability is very low. The analysis of the joint influence of non-accounting and accounting variables shows that size alters the relation between several variables and default. Our findings also indicate industry and geographic variations in the default data. The findings suggest that results relating to listed firms cannot be generalized by policy makers and regulators to apply equally to privatelyheld firms. The findings of the three papers all together provide valuable information not only to the credit risk management of financial institutions but also to policy makers and supervisory authorities.
Esta dissertação é constituída por três artigos sobre gestão de risco de crédito e engloba diferentes tipos de empresas que podem ser observadas na carteira de crédito de um banco. Foi usada uma extensa base de dados do Banco de Portugal, com cerca de 43.000 demonstrações financeiras de mais de 16.000 empresas, no periodo compreendido entre 1997 e 2003. Cerca de 2.500 observações pertencem a empresas que entraram em incumprimento de crédito, conforme a definição de Basileia II, representando 6% do total de observações da base de dados em estudo. No desenvolvimento das estimações empíricas, foram aplicadas diferentes metodologias e sub-grupos de observações da base de dados principal, no sentido de garantir a qualidade dos dados usados, bem como a robustez dos resultados obtidos. O primeiro artigo, no capítulo 1, estuda a influência da avaliação do auditor externo sobre a capacidade de continuidade de uma empresa através da análise dos eventos de incumprimento de crédito após a produção do relatório do auditor. Os requisitos profissionais da função de auditoria implicam que o auditor externo, na elaboração da sua opinião, tenha em consideração e informe explicitamente sempre que existam dúvidas significativas sobre a capacidade de continuidade de uma empresa. A primeira intenção do estudo foi determinar se a avaliação sobre a capacidade de continuidade de uma empresa, por parte do auditor externo, conteria informação substancial, examinando para isso as empresas que entraram em incumprimento de crédito após a emissão de dúvidas por parte do auditor. Nesse sentido, foram usados 12.199 relatórios de auditoria relativos a cerca de 2.000 empresas obrigadas por lei a ter as respectivas contas auditadas anualmente. A estimação empirica de um modelo logit, controlando informação contabilística de cash flows bem como variáveis não contabilísticas, mostra que nas empresas que receberam, na opinião do auditor externo, dúvidas sobre a capacidade de continuidade, a probabilidade de incumprimento de crédito é 2,792 vezes superior em comparação com as empresas sem esse tipo de opinião. Os testes de Likelihood ratio para a omissão de variáveis confirmam o valor incremental preditivo da opinião do auditor externo sobre a capacidade de continuidade das empresas, quando comparado com variáveis contabilísticas e não contabilísticas, na estimação de base e nas estimações através de bases de dados de comparação (out-ofsample). No grupo de empresas que não entraram em incumprimento de crédito, a taxa média de incumprimento estimada é de 6.05%, enquanto no grupo de empresas que entraram em incumprimento a taxa média de incumprimento estimada é de 17.78%. A taxa de incumprimento estimada de crédito para o grupo de empresas que não entraram em incumprimento e que receberam uma opinião de incerteza sobre a capacidade de continuidade, por parte do auditor externo, é de 9.92%, enquanto para as restantes empresas a taxa média é de 5.96%. No grupo de empresas que entraram em incumprimento de crédito, e nesse grupo, as empresas que receberam uma opinião de incerteza por parte do auditor externo, a taxa média estimada é de 35.49%, enquanto para as restantes empresas do mesmo grupo, a taxa média estimada é de 16.96%. A confirmação da robustez das conclusões através da análise de diferentes dimensões de empresas, antiguidade das empresas, sectores e regiões geográficas indicam que as empresas que receberam uma opinião de incerteza de capacidade de continuidade por parte do auditor externo, em média, entram mais em incumprimento de crédito do que as empresas que não têm qualquer opinião de incerteza por parte do auditor externo. O segundo artigo, no capítulo 2, usa a totalidade dos dados e analisa os eventos de incumprimento de crédito para sociedades anónimas não cotadas e para sociedades por quotas, no sentido de perceber as características de incumprimento de crédito para cada um dos grupos de empresas. São estudadas seis hipóteses teóricas que relacionam a influência de rácios contabilísticos e as respectivas propriedades distributivas relacionadas com o incumprimento de crédito. Adicionalmente, duas hipóteses são testadas relacionando a influência do enquadramento e responsabilidade legal e o incumprimento de crédito, bem como o incremento da capacidade preditiva do enquadramento legal relativamente aos rácios contabilísticos na estimação de incumprimento de crédito. Os seguintes rácios: cash flow sobre a divida total; cash flows operacionais sobre os custos financeiros da dívida; capitais próprios sobre o total de activos; e liquidez; têm uma relação negativa com o incumprimento de crédito. A dimensão da empresas e a respectiva idade também estão relacionados negativamente com o incumprimento de crédito. As sociedades anónimas estão relacionadas positivamente com o incumprimento de crédito comparativamente às sociedades por quotas. A probabilidade de incumprimento estimada para sociedades por quotas é 0,83 vezes a probabilidade de incumprimento das sociedades anónimas. Os testes de Likelihood ratio para a omissão de variáveis também confirmam o valor incremental preditivo do enquadramento legal, quando comparado com informação contabilística, na estimação de base e estimações através de bases de dados de comparação (out-of-sample). No grupo de empresas que não entraram em incumprimento de crédito, para sociedades anónimas, a taxa média de incumprimento estimada é 0.53% superior à taxa de incumprimento estimada para sociedades por quotas. A mesma diferença é 2,07% superior no grupo de empresas que entraram em incumprimento de crédito. A confirmação da robustez das conclusões através da análise de diferentes dimensões de empresas indicam que a vantagem das sociedades por quotas diminui com a dimensão das empresas. O terceiro artigo, no capítulo 3, apresenta um modelo baseado em rácios contabilísticos e não contabilísticos para determinar o incumprimento de crédito de empresas não cotadas e um método de estimação de probabilidades de incumprimento de crédito o qual pode ser usado pelo banco central em Portugal para calcular os requisitos minimos de capital regulamentar dos bancos, bem como a sua utilização como benchmarks na validação de modelos de rating internos da banca comercial. O Comité de Supervisão Bancária de Basileia refere a necessidade de os bancos centrais e a banca comercial modelizarem e estimarem probabilidades de incumprimento de crédito tendo em consideração as respectivas especificidades dos portfólios e países em questão. Relativamente a empresas não cotadas existem poucos estudos relacionados com incumprimentos de crédito. Nesse sentido, é usada uma base de dados do Banco de Portugal com 31.025 demonstrações financeiras de empresas não cotadas. Os dados usados incluem 30 rácios contabilísticos e informação não contabilística sobre a dimensão e idade das empresas, sector, e região geográfica. Em relação aos rácios contabilísticos: os custos financeiros sobre os resultados brutos; solvabilidade; e a formação bruta de capital fixo sobre o total de activos têm a influência mais significativa na probabilidade de incumprimento de crédito. Os rácios contabilísticos que apresentam menor influência marginal na probabilidade de incumprimento de crédito são a: rendibilidade dos activos; cobertura financeira; prazo médio de pagamentos; prazo médio de recebimentos; e rendibilidade dos capitais próprios. Adicionalmente, e ao contrário das observações no contexto de empresas cotadas, a dimensão da empresa aparece positivamente relacionada com o incumprimento de crédito, e apesar da idade da empresa surgir negativamente relacionada, a sua contribuição marginal na influência da estimação de probabilidades de incumprimento é reduzida. A análise da influência conjunta de variáveis contabilísticas e não contabilísticas mostra que a dimensão das empresas altera a relação de algumas variáveis com o evento de incumprimento de crédito. Os resultados também indicam variações sectoriais e geográficas nos dados de incumprimento de crédito. Os resultados sugerem que a informação relacionada com empresas cotadas não pode ser generalizada nem aplicada igualmente pelos reguladores às empresas não cotadas. Em suma, os resultados dos três artigos apresentam informação importante não apenas para a gestão de risco de crédito de instituições financeiras, mas também para os reguladores e autoridades de supervisão financeira.