Research Proposal for Business Mgmt

8/12/2019 Research Proposal for Business Mgmt

1/21

Credit riskmanagement instock marketResearch proposal

Trainees

8/10/2012


2/21

Table of ContentsAbstract ........................................................................................................................................... 3

Introduction ..................................................................................................................................... 4

Credit risk measurement ............................................................................................................. 5

Aims ................................................................................................................................................ 6

Objectives ....................................................................................................................................... 6

Literature ......................................................................................................................................... 7

Research Methodology ................................................................................................................... 9

Summary and conclusion .............................................................................................................. 17

Appendix ....................................................................................................................................... 18

Bibliography ................................................................................................................................. 20


3/21

Abstract

In the present research paper it describes the developments in the credit risk measurement

literature over the last 20 years. The paper is basically parted into two parts. In the first partthe growth of the literature on the credit-risk measurement of individual loans and portfoliosof loans is described by mode of reference to articles coming out in applicable consequencesof the Journal of Banking and Finance and other publications. Present paper coversintroduction to credit risk management and Credit risk measurement in that expert systemsand subjective analysis and accounting based credit-scoring systems.

In the second part, it describes a Modern approach reinforced across a mortality riskframework to evaluating the risk and returns on loans and bonds are demonstrated. Thismodel is presented to over few promises in examining the risk-return structures of portfoliosof credit-risk disclosed debt instruments.


4/21

Introduction

Initially it is very important to understand what are Credit risk management then Credit risk

measurement. In which it enhances regarding expert systems and subjective analysis. Creditrisk measurement has evolved dramatically over the last 20 years in response to a number ofsecular forces that have made its measurement more important than ever before. Among theseforces have been: (i) a worldwide structural increase in the number of bankruptcies, (ii) atrend towards disintermediation by the highest quality and largest borrowers, (iii) morecompetitive margins on loans, (iv) a declining value of real assets (and thus collateral) inmany markets and (v) a dramatic growth of off-balance sheet instruments with inherent

default risk exposure (see, e.g. McKinsey, 1993), including credit risk derivatives.

In response to these forces academics and practitioners alike have responded by: (i)developing new and more sophisticated credit-scoring/early-warning systems, (ii) movedaway from only analyzing the credit risk of individual loans and securities towardsdeveloping measures of credit concentration risk (such as the measurement of portfolio riskof fixed income securities), where the assessment of credit risk plays a central role (iii)developing new models to price credit risk (such as the risk adjusted return on capitalmodels (RAROC)) and(iv) developing models to measure better the credit risk of off-balancesheet instruments.

In this paper it has been traced that key developments in credit risk measurement over the past two decades and show how many of these developments have been rejected in papersthat have been published in the Journal of Banking and Finance over this period. In addition,it explores a new approach, and provides some empirical examples to measure the credit risk

of risky debt portfolios or credit concentration risk.


5/21


6/21

Aims

The main aim of the study is developments over the last 20 years to Credit risk measurement

in stock market.

Objectives

1. To identify key the factors which are important in analyzing credit risk management plays a critical role in the process.

2. To analyze a new approach to measuring the return risk trade-off in portfolios of riskydebt instruments, whether bonds or loans.

3. To evaluate the development of credit risk measurement techniques over the pastyears

4. To examine how new approach add much promise to the complex problem ofestimating the optimal composition of loan/bond portfolios.


7/21

Literature

In this paper, literature review initially starts with accounting based credit-scoring systems. It

also discusses about the approaches to developing multivariate credit-scoring systems and soon.

Accounting based credit-scoring systems

In univariate accounting based credit-scoring systems, the FI decision-maker comparesvarious key accounting ratios of potential borrowers with industry or group norms. Whenusing multivariate models, the key accounting variables are combined and weighted to

produce either a credit risk score or a probability of default measure. If the credit risk score,or probability, attains a value above a critical benchmark, a loan applicant is either rejected orsubjected to increased scrutiny.

In terms of sheer number of articles, developments and tests of models in this area havedominated the credit risk measurement literature in the JBF and in other scholarly journals. Inaddition to a significant number of individual articles on the subject, the JBF published twospecial issues (Journal of Banking and Finance, 1984, 1988) on the application of distress prediction models internationally. Indeed, international models have been developed in over25 countries; see Altman and Narayanan (1997).

There are at least four methodological approaches to developing multivariate credit-scoringsystems: (i) the linear probability model, (ii) the logit model, (iii) the probit model, and (iv)the discriminant analysis model. By far the dominant methodologies, in terms of JBF publications, have been discriminant analysis followed by logit analysis. In the inauguralissue (JBF, June 1977), Altman et al. (1977) developed the now commonly used andreferenced ZETA discriminant model. Stripped to its bare essentials, the most commonform of discriminant analysis seeks to and a linear function of accounting and marketvariables that best distinguishes between two loan borrower classification groups repayment and non-repayment. This requires an analysis of a set of variables to maximize the between group variance while minimizing the within group variance among these variables.Similarly, logit analysis uses a set of accounting variables to predict the probability of borrower default, assuming that the probability of default is logistically distributed i.e., the


8/21

cumulative probability of default takes a logistic functional form and is, by definition,constrained to fall between 0 and 1.

Martin (1977) used both logit and discriminant analysis to predict bank failures in the

19751976 period, when 23 banks failed. Both models gave similar classifications in termsof identifying failures/non-failures. West (1985) used the logit model (along with factoranalysis) to measure the financial condition of FIs and to assign to them a probability of being a problem bank. Interestingly, the factors identified by the logit model are similar to theCAMEL rating components used by bank examiners. Platt and Platt (1991a) use the logitmodel to test whether industry relative accounting ratios, rather than simple firm specificaccounting ratios, are better predictors of corporate bankruptcy. In general, the industry

relative accounting ratio model outperformed the unadjusted model. (Similar findings to thishave been found in the context of relative accounting ratio based discriminant analysismodels (see Izan, 1984) Lawrence et al. (1992) use the logit model to predict the probabilityof default on mobile home loans. They and that payment history is by far the most important predictor of default. Smith and Lawrence (1995) use a logit model to and the variables thatother the best prediction of a loan moving into a default state (calculated from a Markovmodel of default probabilities).

Finally, as noted earlier, by far the largest number of multivariate accounting based credit-scoring models have been based on discriminant analysis models. Altman et al. (1977)investigate the predictive performance of a seven variable discriminant analysis model (thatincludes the market value of equity as one variable). A private firm version of this model alsoexists. In general, the seven variable model the so-called ``Zeta model'' is shown toimprove upon Altman's (Altman, 1968) earlier five variable model. Also, Scott (1981)compares a number of these empirical models with a theoretically sound approach. He

concludes that the ZETA model most closely approximates his theoretical bankruptcyconstruct. A large number of other mainly international applications of discriminant analysiscredit related models are to be found in the two special JBF issues on credit risk, mentionedabove.


9/21

Research Methodology

The intention of present research analysis is to describe the developments over the last 20years to Credit risk measurement in stock market. Discuss the growth of the literature on thecredit-risk measurement of individual loans and portfolios of loans by mode of reference toarticles coming out in applicable consequences of the Journal of Banking and Finance andother publications. Primary data were collected for the research.

For this it is providing a clear depth view of analysis methodology such as the quantitativeand qualitative analysis is considered.

Quantitative Analysis objective is to develop and employ mathematical models, theoriesand/or hypotheses pertaining to phenomena. The process of measurement is central toquantitative research because it provides the fundamental connection between empiricalobservation and mathematical expression of quantitative relationships. Since it a huge process, it is better to adopt qualitative analysis.

Qualitative Analysis aims to analyze the developments over the last 20 years to Credit risk

measurement in stock market. The qualitative method investigates the why and how ofdecision making, not just what, where, when. Hence, smaller but focused samples are moreoften needed, rather than large samples.

In the present research paper it is mainly concentrates on primary research methodology,Thus it is good to carry on with secondary research methodologies.

Fixed income portfolio analysis

Since the pioneering work of Markowitz (1959), portfolio theory has been applied tocommon stocks. The traditional objectives of maximizing returns for given levels of risk orminimizing risk for given levels of return have guided efforts to achieve effectivediversification of portfolios. Such concepts as individual stock and portfolio betas to indicaterisk levels and to calculate efficient frontiers, with optimal weightings of the portfolio'smember stocks, are now common parlance among investment professionals and in textbooks(e.g., Elton and Gruber, 1995). This is not to say that these concepts are widely used to the

exclusion of more traditional industrial sector, geographical location, size, or some otherdiversification strategy. The necessary data in terms of historical returns and correlations of


10/21

returns between individual stocks are usually available to perform the portfolio optimizationanalysis.

One might expect that these very same techniques would (and could) be applied to the xed

income area involving corporate and government bonds and even to bank loans. There has been, however, very little published work in the bond area 1 and a recent survey of practices by banks found fragmented and untested efforts. 2 The objective of effective risk reducingmethods is, however, a major pre-occupation of FIs, with bank loan research departments andregulators spending considerable resources to reduce the likelihood of major loan losses that jeopardize the very existence of the lending institution.

Recent bank failures attributed to huge loan losses in the US, Japan, Europe and LatinAmerica has raised the level of concern. Still, conceptually sound diversification techniqueshave eluded most bank and bond portfolio managers, probably for valid reasons. And, despiterecent analytical attempts (e.g. Credit Metrics, 1997), effective portfolio managementtechniques of loans/bonds is still an unresolved area.

It is the objective of this section of present paper to outline a method that will avoid the majordata and analytical pitfalls that have plagued fixed income portfolio efforts and to provide a

sound and empirically feasible portfolio approach.The empirical examples will involve corporate bonds but there is a confidence that themethodology is applicable as well to commercial and industrial loans.

Return-risk framework

The classic mean variance of return framework is not valid for long-term, fixed income portfolio strategies. The problem does not lie in the expected return measure on individual

assets, but in the distribution of possible returns. While the fixed income investor can lose allor most of the investment in the event of default, positive returns are limited. This problem ismitigated when the measurement period of returns is relatively short, e.g., monthly, and thelikely variance of returns is small and more normal. Simultaneously it will return to measuresof portfolio risk both for short-term returns and the more challenging buy-and hold, long-termstrategy.


11/21

Return measurement

The measurement of expected portfolio return is actually quite straightforward for fixedincome bond and loan assets. The investor (or FI) is promised a fixed return (yield-to-

maturity or yield-to-worst) over time and should subtract, from this promised yield, theexpected losses from default of the issuer.

For certain measurement periods, the return will also be influenced by changes in interestrates but it has been assumed that for purposes of exposition, that these changes are randomwith an expected capital gain of zero. Likewise, we acknowledge that investors can infercapital gains or losses from the yield curve and also from whether the bonds are trading at a premium or discount from par.

The expected annual return is therefore

Where EAR is the Expected annual return, YTM the Yield-to-Maturity (or Yield-to-Worst)and EAL the Expected Annual Loss. It has derived the EAL from prior work on bondmortality rates and losses (Altman, 1988, 1989). Each bond is analyzed based on its initial (or

existing)For Table 1 refer Appendix.

bond rating which implies an expected rate of default for up to ten (or longer) years afterissuance. Tables 1 and 2 list cumulative mortality rates and cumulative mortality losses,respectively, covering the period 19711994. 4 Table 3 annualizes these mortality rates andlosses. So, for example, a 10-year BB (S&P rated) bond has an expected annual loss of 91 basis points per year. If the newly issued BB rated bond has a promised yield of 9.0% with aspread of 2.0% over 7.0% risk-free US Treasury bonds, then the expected return is 8.09% peryear, or a risk premium of 109 basis points over the risk-free rate.

If measurement periods were quarterly returns instead of annual, then the expected returnwould be about 2.025% per quarter. Again, the expected return measure is focused primarilyon credit risk changes and not on yield curve implications.

The latter is obviously more relevant to government bond portfolios. The problem of

measuring expected returns for commercial loans is a bit more complex. Since most loans do


12/21

not have a risk rating attached by the rating agencies, 5 the loan portfolio analyst must utilizea proxy measure. It advocate using the bank's own risk rating system, or a rating replicationsystem, as long as each of the internal ratings is linked with the public bond ratings, e.g.,those used by Altman, Moody's or S&P in their cumulative default studies.

For Table 2 refer Appendix.

Here it will also show that these proxy risk measures, either from internal systems or fromcommercially available systems, 6 are critical ingredients in the compilation of historicalcorrelations of risk and return measures between assets in the portfolio. The expected portfolio return (Rp) is therefore based on each asset's expected annual return, weighted by

the proportion (Xi) of each loan/bond relative to the total portfolio,where

Portfolio risk and efficient frontiers using returns the classic mean return-variance portfolio

framework is given in Eq. (3) when there is a utilize of a short holding period, e.g., monthlyor quarterly, and historical data exist for the requisite period to calculate correlation of returnsamong the loans/bonds

For Table 3 refer Appendix.

where Vp is the variance (risk) of the portfolio, Xi the proportion of the portfolio invested in bond issue i, the standard deviation of the return for the sample period for bond issue i, andqij the correlation coefficient of the quarterly returns for bonds i and j

Case Study

Bear Stearns was a large investment bank, securities trader, and brokerage firm operating

globally with headquarters in New York. The firm had been in operation for 85 years whenits utsized position in subprime mortgages raised questions from investors, clients, and


13/21

counterparties about the banks balance sheet and the quality of its assets. A failed hedge

fund sponsored by a subsidiary of the bank in 2007 had brought unwanted questions aboutsubprime loans in general in an increasingly wary market.

Bear Stearns had a reputation as an aggressive trading bank willing to take risks. The firmwas proud of its reputation as a company run by employees with a blue collar background.Fortune magazine listed the firm as one the most admired securities firms in 2005 through2007 in its annual survey of most admired from companies.

Bear Stearns was one of the most highly leveraged firms on Wall Street. Throughout itshistory it was both innovative and creative which, at times, caused it to take some risky positions. Thefirms management was known to focus on immediate opportunistic returnswith little long term strategic planning.

Bear Stearns had pioneered the securitization of subprime mortgages but despite the growingevidence of weaknesses in this market the bank increased its exposure in 2006 and 2007 togain market share. The collapse of the company in March 2008 and its eventual sale toJPMorgan was a key ingredient in recognizing the weaknesses of risk management in the

industry as a whole that led to the meltdown in the financial services industry in September2008 and the subsequent global financial crisis and ensuing recession.

Key Issues

Weak supervision can lead to a bank taking undue risk and failing to maintain sufficientcapital against the constellation of risks it faces

Identifying when an institution is too big to fail when nodepositors interests are at stakeOversight of a financial holding company where there is no legal authority Meetinglegitimate and reasonable requirements of host supervisors

This case study demonstrates the need for proper supervisory oversight and strengthenedcoordination among supervisors and Self-Regulatory Organizations. It also underscores thefutility of attempting to organize a supervisory program designed solely to meet thelegitimate requirements of a foreign supervisor without taking the responsibility seriously.


14/21

Its primary customers included corporations, financial institutions, hedge funds,governments,and individuals. The companys business included corporate finance , mergersand acquisitions, institutional equities and fixed income sales, trading and research, privatewealth management services, derivatives, foreign exchange futures and trading, assetmanagement and custody services. Through a major subsidiary, Bear Stearns SecuritiesCorporation, it conducted global clearing services to broker dealers and professional tradersand was a major security lender.

By year-end 2007 its balance sheet showed $395 billion in assets supported by $11.1 billionin equity a leverage ratio of around 36 to 1. Notional contracts amounted to around $13.4trillion in derivative financial instruments of which around 14% were in listed futures and

option contracts.

Bear Stearns was the fifth largest investment bank in the United States.

THE PROBLEM

Bear Stearns sponsored two hedge funds through its subsidiary, Bear Stearns AssetManagement. The main fund, the High-Grade Structured Credit Strategies Fund, was madeup of complex derivatives backed by home mortgages. During most of its life it was highly

profitable but as the housing market began to stutter in late 2006 the returns suffered. Thisfund was leveraged at 35 times its invested funds.

As the market worsened the returns of the two funds sank. In urging investors to stay put thefund managers promised an eminent turnaround of the market (two Bear Stearns executiveswere subsequently indicted for misleading investors).

In June 2007, Bear Stearns pledged a collateralized loan of around $3.2 billion to arrest the

deteriorating value of the High Grade Structured Credit Strategies Fund. It also wasnegotiating with other lenders to lend additional money to the other fund Bear Stearns HighGrade Structured Credit Enhanced Leverage Fund. Both funds were invested almostexclusively in very thinly traded collateralized debt obligations. As the market downturnaccelerated the funds were left with billions of dollars of money losing securities that wereunmarketable. Investors were trying to cut their losses and flee. Lenders, such as MerrillLynch and J.P Morgan were threatening to seize the collateral. Bear Stearns managers tried to

convince investors to allow more time for the situation to turn around and invest more moneyto plug the widening gap under the theory that the housing market dip was only a temporary


15/21

blip. The investors attempted to get Bear Stearns to invest its own money to arrest the lossesestimated at the time to be around $1.6 billion which was dismissed out of hand.

The report notes that the CSE program requirements were inadequate and, in any event, red

flags surrounding Bear Stearns were simply missed. There were a number of significantfindings:

Bear Stearns was compliant with the programs requirements but collapsed anyway whichraises disturbing questions about the adequacy of the requirements

SEC was aware that Bear Stearns had a heavy concentration of mortgage -backed securitieswhen it voluntarily joined the program in 2006 but the concentration continued to increase

and the agency took no action

The CSE prog ram did not require a leverage ratio limit for member firms

The responsible division for oversight at the SEC was aware that the risk management ofmortgages at Bear Stearns had numerous shortcomings, including lack of expertise by riskmanagers in the mortgage-backed securities area, but took no action

Following the collapse of two hedge funds in June 2007 significant questions were raised

about senior managements lack of attention in handling the crisis but the SEC failed toassess the seriousness of the problem

The Division responsible for reviewing regulatory returns failed to conduct a review of BearStearns most recent 10 -K filing in a timely manner which deprived investors of materialinformation to make well-informed decisions

Bear Stearns practice of using high leverage to take big investment bets was not criticized

The oversight of Bear Stearns did not include actions to assess Bear Stearns Board ofDirectors and Senior Managements tolerance for risk although this seems to be a prudentand necessary oversight procedure

SEC regulations required that Bear Stearns use external auditors to do certain defined auditwork;

It gave permission for the firm to let employees do the work instead the report noted there

was no connection between Bear Stearns collapse and the poor oversight by the SEC or the


16/21

gross shortcomings of the Consolidated Supervised Entities program. In a sense this is true but ignores the fact that Bear Stearns might have benefitted from more aggressive supervisionespecially with respect to strengthening capital and reducing thecompanys high leverage.Since the CSE program was voluntary presumably Bear Stearns could have opted out of the program rather than comply with regulatory requirements the management may not agreewith.


17/21

Summary and conclusion

In this paper it has sought to accomplish two objectives. In Sections 1 and 2 it has been traced

that the development of credit risk measurement techniques over the past 20 years andshowed how any of these developments have been mirrored in published articles in the JBF.In Section 3, it developed a new approach to measuring the return risk trade-off in portfoliosof risky debt instruments, whether bonds or loans. In particular, it has been showed that thisnew approach added much promise to the complex problem of estimating the optimalcomposition of loan/bond portfolios.

Clearly, over the next 20 years one can foresee significant improvements in data bases onhistorical default rates and loan returns. With the development of such data bases will comenew and exciting approaches to measuring the ever present credit risk problems facing FImanagers.


18/21

Appendix

Table 1 Cumulative mortality rates and cumulative mortality losses

Table 2Cumulative mortality rates and cumulative mortality losses


19/21


20/21

BibliographyAltman, E. K. (1992). The implications of corporate bond rating drift. Financial Analysts

Journal June, , 6475.

Brewer, E. K. (1992). The impact of standby letters of credit on bank risk: A note. Journal of Banking and Finance , 10371046.

Chirinko, R. G. (1991). A framework for assessing credit risk in depository institutions:Toward regulatory reform. Journal of Banking and Finance , 785804.

Hull, J. W. (1995). The impact of default risk on the prices of options and other derivativesecurities. Journal of Banking and Finance , 299322.

Jagtiani, J. S. (1995). The eect of bank capital requirements on bank off -balance sheetnancing. Journal of Banking and Finance , 647658.

Lawrence, E. S. (1992). An analysis of default risk in mobile home credit. Journal of Bankingand Finance , 299312.

McAllister, P. M. (1994). Commercial loan risk management, credit-scoring and pricing:Theneed for a new shared data base. Journal of Commercial Bank Lending , 620.

Platt, H. P. (1991a). A note on the use of industry-relative ratios in bankruptcy prediction. Journal of Banking and Finance , 11831194.

Scott, J. (1981). The probability of bankruptcy: A comparison of empirical predictions andtheoretical models. Journal of Banking and Finance. , September, 317344.

Sharpe, W. (1994). The Sharpe Ratio. Journal of Portfolio Management Fall, , 4958.

Smith, L. L. (1994). Forecasting losses on a liquidating long-term loan portfoli. Forecastinglosses on a liquidating long-term loan portfoli , 959985.

Sommerville, R. T. (1995). Banker judgement versus formal forecasting models: The case ofcountry risk assessment. Journal of Banking and Finance , 281297.

West, R. (1985). A factor-analytic approach to bank condition. Journal of Banking and Finance , 253266.


21/21

Wilcox, J. (1973). A Prediction of business failure using accounting data. Journal of Accounting Research, , Vol. 2.

Research Proposal for Business Mgmt

Documents

Transcript of Research Proposal for Business Mgmt