Credit Risk Dissertation

point of reference luck DissertationCREDIT RISKEXECUTIVE SUMMARYThe in store(predic cause) of banking will undoubtedly rest on fortune circumspection dynamics. Only those banks that entertain efficient essay management system will survive in the food mart in the long run. The major(ip) cause of life-threatening banking problems all over the courses continues to be immediatelyly tie in to lax realization standards for borrowers and counterparties, poor portfolio pretend management, or a lack of sparing aid to deterioration in the impute standing of a banks counterparties.Credit encounter is the oldest and biggest take a chance that bank, by virtue of its very nature of business, inherits. This has however, acquired a greater signifi force outce in the parvenu-fangled past for diverse reasons. in that location have been m whatsoever(prenominal) traditional adventes to measure consultation stake like logit, bi bilinear opportunity assume al maven with departure of time new approaches have been authentic like the Credit+, KMV Model.Basel I Accord was in word formd in 1988 to have a framework for regulative hood for banks entirely the wizard size fit in all approach led to a shift, to a new and comprehensive approach -Basel II which adopts a trio column approach to insecurity management. Banks use a number of techniques to mitigate the citation lucks to which they ar exposed. RBI has prescribed adoption of comprehensive approach for the purpose of CRM which allows stuffed offset of security of collateral against pictorial matters by effectively reducing the exposure amount by the esteem ascribed to the collateral.In this study, a star nationalized bank is taken to study the mensurations taken by the bank to impose the Basel- II Accord and the entire framework developed for recognition assay management. The bank chthonic the study uses the book of facts scoring method to tell the attribute fortune of exposure involved in motley lends/advances. The bank has set up special package to evaluate distributively case under various parameters and a monitor system to continuously track each additions performance in accordance with the valuation parameters.CHAPTER 1INTRODUCTION1.1 RationaleCredit Risk Management in todays deregulated market is a big challenge. Increased market irritability has brought with it the need for suffer depth psychology and specialized applications in managing recognition risk. A well defined policy framework is required to help the in operation(p) rung find the risk- return, determine a probability to each, fix the prob fitting blemish, value the acceptability of the exposure, equipment casualty the risk and monitor them right to the point where they are give off.Generally, Banks in India evaluate a proposal through the traditional tools of project financing, computing maximum permissible limits, assessing management capabilities and prescribi ng a ceiling for an indus endeavor exposure. As banks move in to a new high powered world of pecuniary operations and trading, with new risks, the need is felt for much sophisticated and diverse instruments for risk assessment, monitor and controlling risk exposures. It is, thitherfore, time that banks managements equip them fully to grapple with the demands of creating tools and systems capable of assessing, monitoring and controlling risk exposures in a more scientific manner. match to an estimate, Credit Risk takes about 70% and 30% remaining is shared between the other both primal risks, namely securities industry risk (change in the market price and operational risk i.e., failure of internal controls, etc.). Quality borrowers (Tier-I borrowers) were able to access the capital market directly without going through the debt route. Hence, the commendation route is now more open to lesser mortals (Tier-II borrowers). With margin takes going d give, banks are unable to abs orb the level of loan losings. eve in banks which regularly fine-tune address policies and streamline credit processes, it is a real challenge for credit risk managers to correctly identify pockets of risk absorption, quantify conclusion of risk carried, identify opportunities for diversification and balance the risk- reverberation trade-off in their credit portfolio. The management of banks should strive to embrace the notion of un truety and risk in their balance opinion poll and instill the need for approaching credit administration from a risk-perspective a marking the system by placing well drafted strategies in the hands of the ope judge staff with payable material support for its successful implementation.There is a need for Strategic approach to Credit Risk Management (CRM) in IndianCommercial Banks, oddly in view of(1) Higher NPAs level in comparison with global benchmark(2) RBI s stipulation about dividend dispersion by the banks(3) revise NPAs level and CAR norm s(4) parvenu Basel Capital Accord (Basel -II) revolution1.2 OBJECTIVES To understand the conceptual framework for credit risk. To understand credit risk under the Basel II Accord. To read the credit risk management practices in a Leading Nationalised Bank1.3 RESEARCH METHODOLOGYResearch Design In straddle to have more comprehensive definition of the problem and to amaze familiar with the problems, an extensive literature survey was d sensation to collect secondary selective information for the location of the various variables, probably contemporary issues and the clarity of concepts.Data Collection Techniques The data collection technique employ is interviewing. Data has been collected from both primary and secondary sources. old Data is collected by making personal visits to the bank.Secondary Data The details have been collected from research papers, working papers, w constructe papers earn by various agencies like ICRA, FICCI, IBA etc articles from the inter exonerate and various journals.1.4 LITERATURE REVIEW* Merton (1974) has utilize options determine moulding as a engineering to evaluate the credit risk of enterp startle, it has been drawn a lot of attention from western academic and business circles.Mertons Model is the theoretical foundation of structural stickers. Mertons vex is not entirely found on a strict and comprehensive theory but also used market reading stock price as an important chance variable toevaluate the credit risk.This makes credit risk to be a real-time monitored at a much higher(prenominal) frequency.This advantage has do it widely applied by the academic and business circle for a long time.Other Structural Models try to refine the original Merton Framework by removing one or more of unrealistic assumptions.* downhearted and Cox (1976) postulate that heedlessnesss occur as soon as pisseds asset note rank falls at a lower place a certain threshold. In contrast to the Merton approach, heedlessness flowerpo t occur at each time. The paper by Black and Cox (1976) is the startle of the supposed First pass Models (FPM). First passage gravels specify thoughtlessness as the first time the firms asset value hits a lower barrier, allowing omission to take place at any time. When the remissness barrier is exogenously fixed, as in Black and Cox (1976) and Longstaff and Schwartz (1995), it acts as a safety covenant to protect bondholders. Black and Cox introduce the possibility of more complex capital structures, with subordinated debt.* Geske (1977) introduces liaison-paying debt to the Merton pretense.* Vasicek (1984) introduces the distinction between short and long term liabilities which now represents a distinctive skylark of the KMV model.Under these models, all the relevant credit risk elements, including inadvertence and convalescence at disrespect, are a function of the structural characteristics of the firm asset levels, asset excitability (business risk) and supplemen t (fiscal risk).* Kim, Ramaswamy and Sundaresan (1993) have suggested an alternative approach which still adopts the original Merton framework as far as the heedlessness process is concerned but, at the same time, removes one of the unrealistic assumptions of the Merton model namely, that default foundation occur only at maturity of the debt when the firms assets are no longer sufficient to come home debt obligations. Instead, it is assumed that default may occur anytime between the issuance and maturity of the debt and that default is triggered when the value of the firms assets reaches a lower threshold level. In this model, the RR in the event of default is exogenous and independent from the firms asset value. It is generally defined as a fixed ratio of the outstanding debt value and is indeed independent from the PD.The blast to overcome the shortcomings of structural-form models gave rise to reduced-form models. Unlike structural-form models, reduced-form models do not co ndition default on the value of the firm, and parameters related to the firms value need not be estimated to implement them.* Jarrow and Turnbull (1995) assumed that, at default, a bond would have a market value pay off to an exogenously specified fraction of an otherwise equivalent default-free bond.* Duffie and Singleton (1999) followed with a model that, when market value at default (i.e. RR) is exogenously specified, allows for closed-form solutions for the term-structure of credit spreads.* Zhou (2001) attempt to combine the advantages of structural-form models a clear economic mechanism stub the default process, and the ones of reduced-form models unpredictability of default. This model links RRs to the firm value at default so that the variation in RRs is endogenously generated and the correlational statistics between RRs and credit ratings reported first in Altman (1989) and Gupton, render and Carty (2000) is justified.Lately portfolio view on credit spillagees has emerged by recognising that changes in credit whole step tend to comove over the business pass and that one can diversify part of the credit risk by a clever composition of the loan portfolio across regions, industries and countries. Thus in order to assess the credit risk of a loan portfolio, a bank must not only investigate the creditworthiness of its customers, but also identify the concentration risks and feasible comovements of risk factors in the portfolio.* CreditMetrics by Gupton et al (1997) was publicized in 1997 by JP Morgan. Its methodology is found on probability of moving from one credit role to some other within a given time persuasion (credit migration analysis). The estimation of the portfolio Value-at-Risk payable to Credit (Credit-VaR) through CreditMetrics A rating system with probabilities of migrating from one credit quality to another over a given time horizon (transition matrix) is the come across component of the credit-VaR proposed by JP Morgan. Th e specified credit risk horizon is usually one year. A rating system with probabilities of migrating from one credit quality to another over a given time horizon (transition matrix) is the key component of the credit-VaR proposed by JP Morgan. The specified credit risk horizon is usually one year.* (Sy, 2007), states that the primary cause of credit default is loan offense due to insufficient liquid or cash flow to service debt obligations. In the case of unsecured loans, we assume delinquency is a necessary and sufficient condition. In the case of collateralized loans, delinquency is a necessary, but not sufficient condition, because the borrower may be able to refinance the loan from positive equity or net assets to prevent default. In general, for secured loans, both delinquency and insolvency are assumed necessary and sufficient for credit default.CHAPTER 2THEORECTICAL FRAMEWORK2.1 CREDIT RISKCredit risk is risk due to uncertainty in a counterpartys (also called an obligors or credits) ability to meet its obligations. Because there are many types of counterpartiesfrom respective(prenominal)s to sovereign governmentsand many different types of obligationsfrom political machine loans to derivatives transactionscredit risk takes many forms. Institutions manage it in different ways.Although credit losses naturally fluctuate over time and with economic conditions, there is (ceteris paribus) a statistically measured, long-run average loss level. The losses can be divided into two categories i.e. expected losses (EL) and unexpected losses (UL).EL is based on three parameters The likelihood that default will take place over a specified time horizon (probability of default or PD) The amount owned by the counterparty at the moment of default (exposure at default or EAD) The fraction of the exposure, net of any recoveries, which will be broken following a default event (loss given default or LGD).EL = PD x EAD x LGDEL can be add up at various different levels (e.g. single loan or entire credit portfolio), although it is typically calculated at the transaction level it is normally mentioned all as an sheer(a) amount or as a percentage of transaction size. It is also both customer- and facility-specific, since two different loans to the same customer can have a very different EL due to differences in EAD and/or LGD.It is important to note that EL (or, for that matter, credit quality) does not by itself constitute risk if losses always equaled their expected levels, then there would be no uncertainty. Instead, EL should be viewed as an anticipated cost of doing business and should thus be corporal in loan pricing and ex ante provisioning. Credit risk, in fact, arises from variations in the actual loss levels, which give rise to the so-called unexpected loss (UL). Statistically speaking, UL is just now the standard deviation of EL.UL= (EL) = (PD*EAD*LGD)Once the bank- level credit loss dissemination is constructed, credit economic c apital is simply determined by the banks tolerance for credit risk, i.e. the bank needs to decide how much capital it wants to hold in order to avoid insolvency because of unexpected credit losses over the next year. A safer bank must have sufficient capital to withstand losses that are larger and rarer, i.e. they extend save out in the loss distribution tail. In practice, therefore, the choice of confidence interval in the loss distribution corresponds to the banks target credit rating (and related default probability) for its own debt. As Figure below shows, economic capital is the difference between EL and the selected confidence interval at the tail of the loss distribution it is equal to a multiple K (often referred to as the capital multiplier) of the standard deviation of EL (i.e. UL).The shape of the loss distribution can transfigure considerably depending on product type and borrower credit quality. For example, high quality (low PD) borrowers tend to have proportionally less EL per unit of capital charged, importee that K is higher and the shape of their loss distribution is more skewed (and vice versa).Credit risk may be in the following forms * In case of the direct bring * In case of the guarantees and the letter of the credit * In case of the treasury operations * In case of the securities trading businesses * In case of the cross border exposure2.2 The need for Credit Risk RatingThe need for Credit Risk Rating has arisen due to the following1. With raze of State control, deregulation, globalisation and allowing things to shape on the basis of market conditions, Indian Industry and Indian Banking face new risks and challenges. Competition results in the survival of the fittest. It is therefore necessary to identify these risks, measure them, monitor and control them.2. It provides a basis for Credit Risk price i.e. fixation of rate of interest on lending to different borrowers based on their credit risk rating thereby balancing Risk Reward for the Bank.3. The Basel Accord and consequent Reserve Bank of India guidelines requires that the level of capital require to be maintained by the Bank will be in proportion to the risk of the loan in Banks Books for measurement of which proper(a) Credit Risk Rating system is necessary.4. The credit risk rating can be a Risk Management tool for prospecting fresh borrowers in addition to monitoring the weaker parameters and taking remedial action.The types of Risks Captured in the Banks Credit Risk Rating ModelThe Credit Risk Rating Model provides a framework to evaluate the risk emanating from following main risk categorizes/risk areas* Industry risk * Business risk * Financial risk * Management risk * installing risk * hurl risk2.3 WHY CREDIT RISK MEASUREMENT?In upstart years, a revolution is brewing in risk as it is both managed and measured. There are seven reasons as to why certain surge in interest1. Structural increase in bankruptciesAlthough the almost recent recession hit at different time in different countries, most statistics show a significant increase in bankruptcies, compared to prior recession. To the extent that there has been a permanent or structural increase in bankruptcies worldwide- due to increase in the global competition- accurate credit analysis become even more important today than in past.2. DisintermediationAs capital markets have expanded and become accessible to small and mid sized firms, the firms or borrowers left behind to raise funds from banks and other traditional financial institutions (FIs) are likely to be smaller and to have weaker credit ratings. Capital market growth has produced a winners curse effect on the portfolios of traditional FIs.3. More Competitive MarginsAlmost paradoxically, despite the decline in the average quality of loans, interest margins or spreads, especially in wholesale loan markets have become very thin. In short, the risk-return trade off from lending has gotten worse. A number of reasons c an be cited, but an important factor has been the enhanced competition for low quality borrowers especially from finance companies, much of whose lending activity has been concentrated at the higher risk/lower quality end of the market.4. Declining and Volatile Values of CollateralConcurrent with the recent Asian and Russian debt crisis in well developed countries such as Switzerland and Japan have shown that property and real assets value are very hard to predict, and to realize through liquidation. The weaker (and more uncertain) collateral set are, the riskier the lending is likely to be. Indeed the current concerns about deflation worldwide have been accentuated the concerns about the value of real assets such as property and other animal(prenominal) assets.5. The Growth Of Off- counterbalance Sheet DerivativesIn many of the very large U.S. banks, the notional value of the off-balance-sheet exposure to instruments such as over-the-counter (OTC) swaps and forwards is more than 10 multiplication the size of their loan books. Indeed the growth in credit risk off the balance sheet was one of the main reasons for the fundament, by the Bank for world-wide Settlements (BIS), of risk based capital requirements in 1993. Under the BIS system, the banks have to hold a capital requirement based on the mark- to- market current set of each OTC Derivative contract plus an add on for potential future exposure.6. TechnologyAdvances in computer systems and related advances in reading technology have given banks and FIs the opportunity to test high powered modeling techniques. A survey conducted by internationalist Swaps and Derivatives companionship and the Institute of International Finance in 2000 found that survey participants (consisting of 25 commercial banks from 10 countries, with varying size and specialties) used commercial and internal databases to assess the credit risk on rated and unrated commercial, retail and mortgage loans.7. The BIS Risk-Based Capi tal RequirementsDespite the importance of above six reasons, probably the greatest incentive for banks to develop new credit risk models has been dissatisfaction with the BIS and central banks post-1992 imposition of capital requirements on loans. The current BIS approach has been described as a one size fits all policy, irrespective of the size of loan, its maturity, and most importantly, the credit quality of the borrowing party. Much of the current interest in fine tuning credit risk measurement models has been provide by the proposed BIS New Capital Accord (or so Called BIS II) which would more fast link capital charges to the credit risk exposure to retail, commercial, sovereign and interbank credits.Chapter- 3Credit Risk Approaches and Pricing3.1 CREDIT RISK MEASUREMENT set aboutES1. CREDIT SCORING MODELSCredit gain ground Models use data on observed borrower characteristics to calculate the probability of default or to sort borrowers into different default risk classes. B y selecting and combining different economic and financial borrower characteristics, a bank manager may be able to numerically establish which factors are important in justifying default risk, evaluate the relative degree or importance of these factors, improve the pricing of default risk, be better able to screen out bad loan applicants and be in a better position to calculate any reserve needed to meet expected future loan losses.To employ credit scoring model in this manner, the manager must identify target area economic and financial measures of risk for any particular class of borrower. For consumer debt, the objective characteristics in a credit -scoring model might include income, assets, age strain and location. For incarnate debt, financial ratios such as debt-equity ratio are usually key factors. After data are identified, a statistical technique quantifies or scores the default risk probability or default risk classification.Credit scoring models include three broad t ypes (1) linear probability models, (2) logit model and (3) linear discriminant model. running(a) PROBABILITY MODELThe linear probability model uses past data, such as accounting ratios, as inputs into a model to explain repayment engender on old loans. The relative importance of the factors used in explaining the past repayment performance then forecasts repayment probabilities on new loans that is can be used for assessing the probability of repayment.Briefly we divide old loans (i) into two observational stems those that defaulted (Zi = 1) and those that did not default (Zi = 0). Then we relate these observations by linear regression to s set of j casual variables (Xij) that reflects quantative information about the ith borrower, such as leverage or earnings. We estimate the model by linear regression ofZi = jXij + misconductWhere j is the estimated importance of the jth variable in explaining past repayment experience. If we then take these estimated js and multiply them by t he observed Xij for a future borrower, we can attain an expected value of Zi for the probability of repayment on the loan.LOGIT MODELThe objective of the typical credit or loan review model is to replicate judgments make by loan officers, credit managers or bank examiners. If an accurate model could be developed, then it could be used as a tool for reviewing and classifying future credit risks. Chesser (1974) developed a model to predict noncompliance with the customers original loan arrangement, where non-compliance is defined to include not only default but any workout that may have been arranged resulting in a settlement of the loan less favorable to the tender than the original agreement.Chessers model, which was based on a technique called logit analysis, consisted of the following six variables.X1 = (Cash + groceryable Securities)/ amount AssetsX2 = scratch Sales/(Cash + Marketable Securities)X3 = EBIT/Total AssetsX4 = Total Debt/Total AssetsX5 = Total Assets/ Net WorthX6 = working Capital/Net SalesThe estimated coefficients, including an intercept term, areY = -2.0434 -5.24X1 + 0.0053X2 6.6507X3 + 4.4009X4 0.0791X5 0.1020X6Chessers classification rule for above equation is If P 50, limit to the non compliance group and If P50, assign to the compliance group.LINEAR DISCRIMINANT MODELWhile linear probability and logit models project a value foe the expected probability of default if a loan is made, discriminant models divide borrowers into high or default risk classes contingent on their observed characteristic (X).Altmans Z-score model is an application of multivariate Discriminant analysis in credit risk modeling. Financial ratios measuring probability, liquidity and solvency appeared to have significant discriminating power to separate the firm that fails to service its debt from the firms that do not. These ratios are weighted to produce a measure (credit risk score) that can be used as a metric to differentiate the bad firms from the set of technical ones.Discriminant analysis is a multivariate statistical technique that analyzes a set of variables in order to differentiate two or more groups by minimizing the within-group variance and maximizing the between group variance simultaneously. Variables taken wereX1Working Capital/ Total AssetX2 Retained Earning/ Total AssetX3 Earning before interest and taxes/ Total AssetX4 Market value of equity/ Book value of total LiabilitiesX5 Sales/Total AssetThe original Z-score model was revised and modified several(prenominal) times in order to find the scoring model more specific to a particular class of firm. These resulted in the private firms Z-score model, non manufacturers Z-score model and appear Market Scoring (EMS) model.3.2 New ApproachesTERM STRUCTURE DERIVATION OF CREDIT RISKOne market based method of assessing credit risk exposure and default probabilities is to analyze the risk premium inherent in the current structure of yields on incorporated debt or loans to si milar risk-rated borrowers. Rating agencies categorize corporate bond issuers into at least seven major classes according to perceived credit quality. The first four ratings AAA, AA, A and BBB indicate investment quality borrowers.deathrate RATE APPROACHRather than extracting expected default rates from the current term structure of interest rates, the FI manager may analyze the historic or past default experience the mortality rates, of bonds and loans of a similar quality. Here p1is the probability of a grade B bond surviving the first year of its issue hence 1 p1 is the marginal mortality rate, or the probability of the bond or loan dying or defaulting in the first year while p2 is the probability of the loan surviving in the second year and that it has not defaulted in the first year, 1-p2 is the marginal mortality rate for the second year. Thus, for each grade of corporate buyer quality, a marginal mortality rate (MMR) curve can show the historical default rate in any speci fic quality class in each year after issue.RAROC MODELSBased on a banks risk- heading capacity and its risk strategy, it is thus necessary bearing in encephalon the banks strategic orientation to find a method for the efficient allocation of capital to the banks individual siness areas, i.e. to define indicators that are suitable for balancing risk and return in a sensible manner. Indicators fulfilling this requirement are often referred to as risk alter performance measures (RAPM).RARORAC (risk adjusted return on risk adjusted capital, usually abbreviated as the most commonly found forms are RORAC (return on risk adjusted capital),Net income is taken to suppose income minus refinancing cost, operational cost, and expected losses. It should now be the banks goal to maximize a RAPM indicator for the bank as a whole, e.g. RORAC, taking into account the correlation between individual transactions. Certain constraints such as volume restrictions due to a potential lack of liquidit y and the maintenance of solvency based on economic and regulatory capital have to be observed in reaching this goal. From an organizational point of view, value and risk management should therefore be joined as closely as possible at all organizational levels.OPTION MODELS OF DEFAULT RISK (kmv model)KMV Corporation has developed a credit risk model that uses information on the stock prices and the capital structure of the firm to estimate its default probability. The starting point of the model is the proposition that a firm will default only if its asset value falls below a certain level, which is function of its liability. It estimates the asset value of the firm and its asset volatility from the market value of equity and the debt structure in the option theoretic framework. The resultant probability is called Expected default Frequency (EDF). In summary, EDF is calculated in the following three stepsi) Estimation of asset value and volatility from the equity value and volatili ty of equity return.ii) enumeration of distance from defaultiii) Calculation of expected default frequencyCredit METRICSIt provides a method for estimating the distribution of the value of the assets n a portfolio subject to change in the credit quality of individual borrower. A portfolio consists of different stand-alone assets, defined by a stream of future cash flows. Each asset has a distribution over the possible range of future rating class. Starting from its initial rating, an asset may end up in ay one of the possible rating categories. Each rating category has a different credit spread, which will be used to discount the future cash flows. Moreover, the assets are correlated among themselves depending on the industry they expire to. It is assumed that the asset returns are normally distributed and change in the asset returns causes the change in the rating category in future. Finally, the make-believe technique is used to estimate the value distribution of the assets. A number of scenario are generated from a multivariate normal distribution, which is defined by the appropriate credit spread, the future value of asset is estimated.CREDIT Risk+CreditRisk+, introduced by Credit Suisse Financial Products (CSFP), is a model of default risk. Each asset has only two possible end-of-period states default and non-default. In the event of default, the lender recovers a fixed proportion of the total expense. The default rate is considered as a continuous hit-or-miss variable. It does not try to estimate default correlation directly. Here, the default correlation is assumed to be determined by a set of risk factors. qualified on these risk factors, default of each obligator follows a Bernoulli distribution. To get unconditional probability generating function for the number of defaults, it assumes that the risk factors are independently gamma distributed stochastic variables. The final step in Creditrisk+ is to obtain the probability generating function fo r losses. Conditional on the number of default events, the losses are entirely determined by the exposure and recovery rate. Thus, the distribution of asset can be estimated from the following input datai) Exposure of individual assetii) Expected default rateiii) Default ate volatilitiesiv) Recovery rate given default3.3 CREDIT PRICINGPricing of the credit is essential for the survival of enterprises relying on credit assets, because the benefits derived from extending credit should surpass the cost.With the introduction of capital adequacy norms, the credit risk is linked to the capital-minimum 8% capital adequacy. Consequently, higher capital is required to be deployed if more credit risks are underwritten. The stopping point (a) whether to maximize the returns on possible credit assets with the existing capital or (b) raise more capital to do more business invariably depends upon pCredit Risk DissertationCredit Risk DissertationCREDIT RISKEXECUTIVE SUMMARYThe future of banking w ill undoubtedly rest on risk management dynamics. Only those banks that have efficient risk management system will survive in the market in the long run. The major cause of serious banking problems over the years continues to be directly related to lax credit standards for borrowers and counterparties, poor portfolio risk management, or a lack of attention to deterioration in the credit standing of a banks counterparties.Credit risk is the oldest and biggest risk that bank, by virtue of its very nature of business, inherits. This has however, acquired a greater consequence in the recent past for various reasons. There have been many traditional approaches to measure credit risk like logit, linear probability model but with passage of time new approaches have been developed like the Credit+, KMV Model.Basel I Accord was introduced in 1988 to have a framework for regulatory capital for banks but the one size fit all approach led to a shift, to a new and comprehensive approach -Basel II which adopts a three pillar approach to risk management. Banks use a number of techniques to mitigate the credit risks to which they are exposed. RBI has prescribed adoption of comprehensive approach for the purpose of CRM which allows fuller offset of security of collateral against exposures by effectively reducing the exposure amount by the value ascribed to the collateral.In this study, a leading nationalized bank is taken to study the steps taken by the bank to implement the Basel- II Accord and the entire framework developed for credit risk management. The bank under the study uses the credit scoring method to evaluate the credit risk involved in various loans/advances. The bank has set up special software to evaluate each case under various parameters and a monitoring system to continuously track each assets performance in accordance with the evaluation parameters.CHAPTER 1INTRODUCTION1.1 RationaleCredit Risk Management in todays deregulated market is a big challenge. Incr eased market volatility has brought with it the need for smart analysis and specialized applications in managing credit risk. A well defined policy framework is needed to help the operating staff identify the risk-event, assign a probability to each, quantify the likely loss, assess the acceptability of the exposure, price the risk and monitor them right to the point where they are paid off.Generally, Banks in India evaluate a proposal through the traditional tools of project financing, computing maximum permissible limits, assessing management capabilities and prescribing a ceiling for an industry exposure. As banks move in to a new high powered world of financial operations and trading, with new risks, the need is felt for more sophisticated and versatile instruments for risk assessment, monitoring and controlling risk exposures. It is, therefore, time that banks managements equip them fully to grapple with the demands of creating tools and systems capable of assessing, monitoring and controlling risk exposures in a more scientific manner.According to an estimate, Credit Risk takes about 70% and 30% remaining is shared between the other two primary risks, namely Market risk (change in the market price and operational risk i.e., failure of internal controls, etc.). Quality borrowers (Tier-I borrowers) were able to access the capital market directly without going through the debt route. Hence, the credit route is now more open to lesser mortals (Tier-II borrowers). With margin levels going down, banks are unable to absorb the level of loan losses. Even in banks which regularly fine-tune credit policies and streamline credit processes, it is a real challenge for credit risk managers to correctly identify pockets of risk concentration, quantify extent of risk carried, identify opportunities for diversification and balance the risk-return trade-off in their credit portfolio. The management of banks should strive to embrace the notion of uncertainty and risk in th eir balance sheet and instill the need for approaching credit administration from a risk-perspective across the system by placing well drafted strategies in the hands of the operating staff with due material support for its successful implementation.There is a need for Strategic approach to Credit Risk Management (CRM) in IndianCommercial Banks, particularly in view of(1) Higher NPAs level in comparison with global benchmark(2) RBI s stipulation about dividend distribution by the banks(3) Revised NPAs level and CAR norms(4) New Basel Capital Accord (Basel -II) revolution1.2 OBJECTIVES To understand the conceptual framework for credit risk. To understand credit risk under the Basel II Accord. To analyze the credit risk management practices in a Leading Nationalised Bank1.3 RESEARCH METHODOLOGYResearch Design In order to have more comprehensive definition of the problem and to become familiar with the problems, an extensive literature survey was done to collect secondary data for the location of the various variables, probably contemporary issues and the clarity of concepts.Data Collection Techniques The data collection technique used is interviewing. Data has been collected from both primary and secondary sources.Primary Data is collected by making personal visits to the bank.Secondary Data The details have been collected from research papers, working papers, whiteness papers published by various agencies like ICRA, FICCI, IBA etc articles from the internet and various journals.1.4 LITERATURE REVIEW* Merton (1974) has applied options pricing model as a technology to evaluate the credit risk of enterprise, it has been drawn a lot of attention from western academic and business circles.Mertons Model is the theoretical foundation of structural models. Mertons model is not only based on a strict and comprehensive theory but also used market information stock price as an important variance toevaluate the credit risk.This makes credit risk to be a real-time monitored at a much higher frequency.This advantage has made it widely applied by the academic and business circle for a long time.Other Structural Models try to refine the original Merton Framework by removing one or more of unrealistic assumptions.* Black and Cox (1976) postulate that defaults occur as soon as firms asset value falls below a certain threshold. In contrast to the Merton approach, default can occur at any time. The paper by Black and Cox (1976) is the first of the so-called First Passage Models (FPM). First passage models specify default as the first time the firms asset value hits a lower barrier, allowing default to take place at any time. When the default barrier is exogenously fixed, as in Black and Cox (1976) and Longstaff and Schwartz (1995), it acts as a safety covenant to protect bondholders. Black and Cox introduce the possibility of more complex capital structures, with subordinated debt.* Geske (1977) introduces interest-paying debt to the Merton model.* Vasicek ( 1984) introduces the distinction between short and long term liabilities which now represents a distinctive feature of the KMV model.Under these models, all the relevant credit risk elements, including default and recovery at default, are a function of the structural characteristics of the firm asset levels, asset volatility (business risk) and leverage (financial risk).* Kim, Ramaswamy and Sundaresan (1993) have suggested an alternative approach which still adopts the original Merton framework as far as the default process is concerned but, at the same time, removes one of the unrealistic assumptions of the Merton model namely, that default can occur only at maturity of the debt when the firms assets are no longer sufficient to cover debt obligations. Instead, it is assumed that default may occur anytime between the issuance and maturity of the debt and that default is triggered when the value of the firms assets reaches a lower threshold level. In this model, the RR in the event o f default is exogenous and independent from the firms asset value. It is generally defined as a fixed ratio of the outstanding debt value and is therefore independent from the PD.The attempt to overcome the shortcomings of structural-form models gave rise to reduced-form models. Unlike structural-form models, reduced-form models do not condition default on the value of the firm, and parameters related to the firms value need not be estimated to implement them.* Jarrow and Turnbull (1995) assumed that, at default, a bond would have a market value equal to an exogenously specified fraction of an otherwise equivalent default-free bond.* Duffie and Singleton (1999) followed with a model that, when market value at default (i.e. RR) is exogenously specified, allows for closed-form solutions for the term-structure of credit spreads.* Zhou (2001) attempt to combine the advantages of structural-form models a clear economic mechanism behind the default process, and the ones of reduced-form m odels unpredictability of default. This model links RRs to the firm value at default so that the variation in RRs is endogenously generated and the correlation between RRs and credit ratings reported first in Altman (1989) and Gupton, Gates and Carty (2000) is justified.Lately portfolio view on credit losses has emerged by recognising that changes in credit quality tend to comove over the business cycle and that one can diversify part of the credit risk by a clever composition of the loan portfolio across regions, industries and countries. Thus in order to assess the credit risk of a loan portfolio, a bank must not only investigate the creditworthiness of its customers, but also identify the concentration risks and possible comovements of risk factors in the portfolio.* CreditMetrics by Gupton et al (1997) was publicized in 1997 by JP Morgan. Its methodology is based on probability of moving from one credit quality to another within a given time horizon (credit migration analysis). The estimation of the portfolio Value-at-Risk due to Credit (Credit-VaR) through CreditMetrics A rating system with probabilities of migrating from one credit quality to another over a given time horizon (transition matrix) is the key component of the credit-VaR proposed by JP Morgan. The specified credit risk horizon is usually one year. A rating system with probabilities of migrating from one credit quality to another over a given time horizon (transition matrix) is the key component of the credit-VaR proposed by JP Morgan. The specified credit risk horizon is usually one year.* (Sy, 2007), states that the primary cause of credit default is loan delinquency due to insufficient liquidity or cash flow to service debt obligations. In the case of unsecured loans, we assume delinquency is a necessary and sufficient condition. In the case of collateralized loans, delinquency is a necessary, but not sufficient condition, because the borrower may be able to refinance the loan from positi ve equity or net assets to prevent default. In general, for secured loans, both delinquency and insolvency are assumed necessary and sufficient for credit default.CHAPTER 2THEORECTICAL FRAMEWORK2.1 CREDIT RISKCredit risk is risk due to uncertainty in a counterpartys (also called an obligors or credits) ability to meet its obligations. Because there are many types of counterpartiesfrom individuals to sovereign governmentsand many different types of obligationsfrom auto loans to derivatives transactionscredit risk takes many forms. Institutions manage it in different ways.Although credit losses naturally fluctuate over time and with economic conditions, there is (ceteris paribus) a statistically measured, long-run average loss level. The losses can be divided into two categories i.e. expected losses (EL) and unexpected losses (UL).EL is based on three parameters The likelihood that default will take place over a specified time horizon (probability of default or PD) The amount owned by the counterparty at the moment of default (exposure at default or EAD) The fraction of the exposure, net of any recoveries, which will be lost following a default event (loss given default or LGD).EL = PD x EAD x LGDEL can be aggregated at various different levels (e.g. individual loan or entire credit portfolio), although it is typically calculated at the transaction level it is normally mentioned either as an absolute amount or as a percentage of transaction size. It is also both customer- and facility-specific, since two different loans to the same customer can have a very different EL due to differences in EAD and/or LGD.It is important to note that EL (or, for that matter, credit quality) does not by itself constitute risk if losses always equaled their expected levels, then there would be no uncertainty. Instead, EL should be viewed as an anticipated cost of doing business and should therefore be incorporated in loan pricing and ex ante provisioning. Credit risk, in fact, ari ses from variations in the actual loss levels, which give rise to the so-called unexpected loss (UL). Statistically speaking, UL is simply the standard deviation of EL.UL= (EL) = (PD*EAD*LGD)Once the bank- level credit loss distribution is constructed, credit economic capital is simply determined by the banks tolerance for credit risk, i.e. the bank needs to decide how much capital it wants to hold in order to avoid insolvency because of unexpected credit losses over the next year. A safer bank must have sufficient capital to withstand losses that are larger and rarer, i.e. they extend further out in the loss distribution tail. In practice, therefore, the choice of confidence interval in the loss distribution corresponds to the banks target credit rating (and related default probability) for its own debt. As Figure below shows, economic capital is the difference between EL and the selected confidence interval at the tail of the loss distribution it is equal to a multiple K (often referred to as the capital multiplier) of the standard deviation of EL (i.e. UL).The shape of the loss distribution can vary considerably depending on product type and borrower credit quality. For example, high quality (low PD) borrowers tend to have proportionally less EL per unit of capital charged, meaning that K is higher and the shape of their loss distribution is more skewed (and vice versa).Credit risk may be in the following forms * In case of the direct lending * In case of the guarantees and the letter of the credit * In case of the treasury operations * In case of the securities trading businesses * In case of the cross border exposure2.2 The need for Credit Risk RatingThe need for Credit Risk Rating has arisen due to the following1. With dismantling of State control, deregulation, globalisation and allowing things to shape on the basis of market conditions, Indian Industry and Indian Banking face new risks and challenges. Competition results in the survival of the fittes t. It is therefore necessary to identify these risks, measure them, monitor and control them.2. It provides a basis for Credit Risk Pricing i.e. fixation of rate of interest on lending to different borrowers based on their credit risk rating thereby balancing Risk Reward for the Bank.3. The Basel Accord and consequent Reserve Bank of India guidelines requires that the level of capital required to be maintained by the Bank will be in proportion to the risk of the loan in Banks Books for measurement of which proper Credit Risk Rating system is necessary.4. The credit risk rating can be a Risk Management tool for prospecting fresh borrowers in addition to monitoring the weaker parameters and taking remedial action.The types of Risks Captured in the Banks Credit Risk Rating ModelThe Credit Risk Rating Model provides a framework to evaluate the risk emanating from following main risk categorizes/risk areas* Industry risk * Business risk * Financial risk * Management risk * Facility risk * Project risk2.3 WHY CREDIT RISK MEASUREMENT?In recent years, a revolution is brewing in risk as it is both managed and measured. There are seven reasons as to why certain surge in interest1. Structural increase in bankruptciesAlthough the most recent recession hit at different time in different countries, most statistics show a significant increase in bankruptcies, compared to prior recession. To the extent that there has been a permanent or structural increase in bankruptcies worldwide- due to increase in the global competition- accurate credit analysis become even more important today than in past.2. DisintermediationAs capital markets have expanded and become accessible to small and mid sized firms, the firms or borrowers left behind to raise funds from banks and other traditional financial institutions (FIs) are likely to be smaller and to have weaker credit ratings. Capital market growth has produced a winners curse effect on the portfolios of traditional FIs.3. More Competi tive MarginsAlmost paradoxically, despite the decline in the average quality of loans, interest margins or spreads, especially in wholesale loan markets have become very thin. In short, the risk-return trade off from lending has gotten worse. A number of reasons can be cited, but an important factor has been the enhanced competition for low quality borrowers especially from finance companies, much of whose lending activity has been concentrated at the higher risk/lower quality end of the market.4. Declining and Volatile Values of CollateralConcurrent with the recent Asian and Russian debt crisis in well developed countries such as Switzerland and Japan have shown that property and real assets value are very hard to predict, and to realize through liquidation. The weaker (and more uncertain) collateral values are, the riskier the lending is likely to be. Indeed the current concerns about deflation worldwide have been accentuated the concerns about the value of real assets such as pro perty and other physical assets.5. The Growth Of Off- Balance Sheet DerivativesIn many of the very large U.S. banks, the notional value of the off-balance-sheet exposure to instruments such as over-the-counter (OTC) swaps and forwards is more than 10 times the size of their loan books. Indeed the growth in credit risk off the balance sheet was one of the main reasons for the introduction, by the Bank for International Settlements (BIS), of risk based capital requirements in 1993. Under the BIS system, the banks have to hold a capital requirement based on the mark- to- market current values of each OTC Derivative contract plus an add on for potential future exposure.6. TechnologyAdvances in computer systems and related advances in information technology have given banks and FIs the opportunity to test high powered modeling techniques. A survey conducted by International Swaps and Derivatives Association and the Institute of International Finance in 2000 found that survey participants (consisting of 25 commercial banks from 10 countries, with varying size and specialties) used commercial and internal databases to assess the credit risk on rated and unrated commercial, retail and mortgage loans.7. The BIS Risk-Based Capital RequirementsDespite the importance of above six reasons, probably the greatest incentive for banks to develop new credit risk models has been dissatisfaction with the BIS and central banks post-1992 imposition of capital requirements on loans. The current BIS approach has been described as a one size fits all policy, irrespective of the size of loan, its maturity, and most importantly, the credit quality of the borrowing party. Much of the current interest in fine tuning credit risk measurement models has been fueled by the proposed BIS New Capital Accord (or so Called BIS II) which would more closely link capital charges to the credit risk exposure to retail, commercial, sovereign and interbank credits.Chapter- 3Credit Risk Approaches and Pri cing3.1 CREDIT RISK MEASUREMENT APPROACHES1. CREDIT SCORING MODELSCredit Scoring Models use data on observed borrower characteristics to calculate the probability of default or to sort borrowers into different default risk classes. By selecting and combining different economic and financial borrower characteristics, a bank manager may be able to numerically establish which factors are important in explaining default risk, evaluate the relative degree or importance of these factors, improve the pricing of default risk, be better able to screen out bad loan applicants and be in a better position to calculate any reserve needed to meet expected future loan losses.To employ credit scoring model in this manner, the manager must identify objective economic and financial measures of risk for any particular class of borrower. For consumer debt, the objective characteristics in a credit -scoring model might include income, assets, age occupation and location. For corporate debt, financial ra tios such as debt-equity ratio are usually key factors. After data are identified, a statistical technique quantifies or scores the default risk probability or default risk classification.Credit scoring models include three broad types (1) linear probability models, (2) logit model and (3) linear discriminant model.LINEAR PROBABILITY MODELThe linear probability model uses past data, such as accounting ratios, as inputs into a model to explain repayment experience on old loans. The relative importance of the factors used in explaining the past repayment performance then forecasts repayment probabilities on new loans that is can be used for assessing the probability of repayment.Briefly we divide old loans (i) into two observational groups those that defaulted (Zi = 1) and those that did not default (Zi = 0). Then we relate these observations by linear regression to s set of j casual variables (Xij) that reflects quantative information about the ith borrower, such as leverage or earni ngs. We estimate the model by linear regression ofZi = jXij + errorWhere j is the estimated importance of the jth variable in explaining past repayment experience. If we then take these estimated js and multiply them by the observed Xij for a prospective borrower, we can derive an expected value of Zi for the probability of repayment on the loan.LOGIT MODELThe objective of the typical credit or loan review model is to replicate judgments made by loan officers, credit managers or bank examiners. If an accurate model could be developed, then it could be used as a tool for reviewing and classifying future credit risks. Chesser (1974) developed a model to predict noncompliance with the customers original loan arrangement, where non-compliance is defined to include not only default but any workout that may have been arranged resulting in a settlement of the loan less favorable to the tender than the original agreement.Chessers model, which was based on a technique called logit analysis, consisted of the following six variables.X1 = (Cash + Marketable Securities)/Total AssetsX2 = Net Sales/(Cash + Marketable Securities)X3 = EBIT/Total AssetsX4 = Total Debt/Total AssetsX5 = Total Assets/ Net WorthX6 = Working Capital/Net SalesThe estimated coefficients, including an intercept term, areY = -2.0434 -5.24X1 + 0.0053X2 6.6507X3 + 4.4009X4 0.0791X5 0.1020X6Chessers classification rule for above equation is If P 50, assign to the non compliance group and If P50, assign to the compliance group.LINEAR DISCRIMINANT MODELWhile linear probability and logit models project a value foe the expected probability of default if a loan is made, discriminant models divide borrowers into high or default risk classes contingent on their observed characteristic (X).Altmans Z-score model is an application of multivariate Discriminant analysis in credit risk modeling. Financial ratios measuring probability, liquidity and solvency appeared to have significant discriminating power to separ ate the firm that fails to service its debt from the firms that do not. These ratios are weighted to produce a measure (credit risk score) that can be used as a metric to differentiate the bad firms from the set of good ones.Discriminant analysis is a multivariate statistical technique that analyzes a set of variables in order to differentiate two or more groups by minimizing the within-group variance and maximizing the between group variance simultaneously. Variables taken wereX1Working Capital/ Total AssetX2 Retained Earning/ Total AssetX3 Earning before interest and taxes/ Total AssetX4 Market value of equity/ Book value of total LiabilitiesX5 Sales/Total AssetThe original Z-score model was revised and modified several times in order to find the scoring model more specific to a particular class of firm. These resulted in the private firms Z-score model, non manufacturers Z-score model and Emerging Market Scoring (EMS) model.3.2 New ApproachesTERM STRUCTURE DERIVATION OF CREDIT RI SKOne market based method of assessing credit risk exposure and default probabilities is to analyze the risk premium inherent in the current structure of yields on corporate debt or loans to similar risk-rated borrowers. Rating agencies categorize corporate bond issuers into at least seven major classes according to perceived credit quality. The first four ratings AAA, AA, A and BBB indicate investment quality borrowers.MORTALITY RATE APPROACHRather than extracting expected default rates from the current term structure of interest rates, the FI manager may analyze the historic or past default experience the mortality rates, of bonds and loans of a similar quality. Here p1is the probability of a grade B bond surviving the first year of its issue thus 1 p1 is the marginal mortality rate, or the probability of the bond or loan dying or defaulting in the first year while p2 is the probability of the loan surviving in the second year and that it has not defaulted in the first year, 1- p2 is the marginal mortality rate for the second year. Thus, for each grade of corporate buyer quality, a marginal mortality rate (MMR) curve can show the historical default rate in any specific quality class in each year after issue.RAROC MODELSBased on a banks risk-bearing capacity and its risk strategy, it is thus necessary bearing in mind the banks strategic orientation to find a method for the efficient allocation of capital to the banks individual siness areas, i.e. to define indicators that are suitable for balancing risk and return in a sensible manner. Indicators fulfilling this requirement are often referred to as risk adjusted performance measures (RAPM).RARORAC (risk adjusted return on risk adjusted capital, usually abbreviated as the most commonly found forms are RORAC (return on risk adjusted capital),Net income is taken to mean income minus refinancing cost, operating cost, and expected losses. It should now be the banks goal to maximize a RAPM indicator for the ban k as a whole, e.g. RORAC, taking into account the correlation between individual transactions. Certain constraints such as volume restrictions due to a potential lack of liquidity and the maintenance of solvency based on economic and regulatory capital have to be observed in reaching this goal. From an organizational point of view, value and risk management should therefore be linked as closely as possible at all organizational levels.OPTION MODELS OF DEFAULT RISK (kmv model)KMV Corporation has developed a credit risk model that uses information on the stock prices and the capital structure of the firm to estimate its default probability. The starting point of the model is the proposition that a firm will default only if its asset value falls below a certain level, which is function of its liability. It estimates the asset value of the firm and its asset volatility from the market value of equity and the debt structure in the option theoretic framework. The resultant probability is called Expected default Frequency (EDF). In summary, EDF is calculated in the following three stepsi) Estimation of asset value and volatility from the equity value and volatility of equity return.ii) Calculation of distance from defaultiii) Calculation of expected default frequencyCredit METRICSIt provides a method for estimating the distribution of the value of the assets n a portfolio subject to change in the credit quality of individual borrower. A portfolio consists of different stand-alone assets, defined by a stream of future cash flows. Each asset has a distribution over the possible range of future rating class. Starting from its initial rating, an asset may end up in ay one of the possible rating categories. Each rating category has a different credit spread, which will be used to discount the future cash flows. Moreover, the assets are correlated among themselves depending on the industry they belong to. It is assumed that the asset returns are normally distributed and ch ange in the asset returns causes the change in the rating category in future. Finally, the simulation technique is used to estimate the value distribution of the assets. A number of scenario are generated from a multivariate normal distribution, which is defined by the appropriate credit spread, the future value of asset is estimated.CREDIT Risk+CreditRisk+, introduced by Credit Suisse Financial Products (CSFP), is a model of default risk. Each asset has only two possible end-of-period states default and non-default. In the event of default, the lender recovers a fixed proportion of the total expense. The default rate is considered as a continuous random variable. It does not try to estimate default correlation directly. Here, the default correlation is assumed to be determined by a set of risk factors. Conditional on these risk factors, default of each obligator follows a Bernoulli distribution. To get unconditional probability generating function for the number of defaults, it ass umes that the risk factors are independently gamma distributed random variables. The final step in Creditrisk+ is to obtain the probability generating function for losses. Conditional on the number of default events, the losses are entirely determined by the exposure and recovery rate. Thus, the distribution of asset can be estimated from the following input datai) Exposure of individual assetii) Expected default rateiii) Default ate volatilitiesiv) Recovery rate given default3.3 CREDIT PRICINGPricing of the credit is essential for the survival of enterprises relying on credit assets, because the benefits derived from extending credit should surpass the cost.With the introduction of capital adequacy norms, the credit risk is linked to the capital-minimum 8% capital adequacy. Consequently, higher capital is required to be deployed if more credit risks are underwritten. The decision (a) whether to maximize the returns on possible credit assets with the existing capital or (b) raise mo re capital to do more business invariably depends upon p