Individual loan credit risk


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Individual loan credit risk

  1. 1. BUS FINANCE 826
  2. 2. Overview <ul><li>The analysis and measurement of credit risk on individual loans. This is important for purposes of: </li></ul><ul><ul><li>Pricing loans and bonds </li></ul></ul><ul><ul><li>Setting limits on credit risk exposure </li></ul></ul>
  3. 3. Credit Quality Problems <ul><li>Problems with junk bonds, LDC loans, residential and farm mortgage loans. </li></ul><ul><li>More recently, credit card loans and auto loans. </li></ul><ul><li>Crises in Asian countries such as Korea, Indonesia, Thailand, and Malaysia. </li></ul>
  4. 4. Web Resources <ul><li>For further information on credit ratings visit: </li></ul><ul><li>Moody’s </li></ul><ul><li>Standard & Poors </li></ul>Web Surf
  5. 5. Credit Quality Problems <ul><li>Over the 90s, improvements in NPLs for large banks and overall credit quality. </li></ul><ul><li>Recent exposure to borrowers such as Enron. </li></ul><ul><li>New types of credit risk related to loan guarantees and off-balance-sheet activities. </li></ul><ul><li>Increased emphasis on credit risk evaluation. </li></ul>
  6. 6. Types of Loans: <ul><li>C&I loans: secured and unsecured </li></ul><ul><ul><li>Spot loans, Loan commitments </li></ul></ul><ul><ul><li>Decline in C&I loans originated by commercial banks and growth in commercial paper market. </li></ul></ul><ul><li>RE loans: primarily mortgages </li></ul><ul><ul><li>Fixed-rate, ARM </li></ul></ul><ul><ul><li>Mortgages can be subject to default risk when loan-to-value declines. </li></ul></ul>
  7. 7. Consumer loans <ul><li>Individual (consumer) loans: personal, auto, credit card. </li></ul><ul><ul><li>Nonrevolving loans </li></ul></ul><ul><ul><ul><li>Automobile, mobile home, personal loans </li></ul></ul></ul><ul><ul><li>Growth in credit card debt </li></ul></ul><ul><ul><ul><li>Visa, MasterCard </li></ul></ul></ul><ul><ul><ul><li>Proprietary cards such as Sears, AT&T </li></ul></ul></ul><ul><ul><li>Risks affected by competitive conditions and usury ceilings </li></ul></ul>
  8. 8. Other loans <ul><li>Other loans include: </li></ul><ul><ul><li>Farm loans </li></ul></ul><ul><ul><li>Other banks </li></ul></ul><ul><ul><li>Nonbank FIs </li></ul></ul><ul><ul><li>Broker margin loans </li></ul></ul><ul><ul><li>Foreign banks and sovereign governments </li></ul></ul><ul><ul><li>State and local governments </li></ul></ul>
  9. 9. Return on a Loan: <ul><li>Factors: interest payments, fees, credit risk premium, collateral, other requirements such as compensating balances and reserve requirements. </li></ul><ul><li>Return = inflow/outflow </li></ul><ul><li>k = ( f + ( L + M ))/(1-[ b (1- R )]) </li></ul><ul><li>Expected return: E(r) = p (1+k) </li></ul>
  10. 10. Lending Rates and Rationing <ul><li>At retail: Usually a simple accept/reject decision rather than adjustments to the rate. </li></ul><ul><ul><li>Credit rationing. </li></ul></ul><ul><ul><li>If accepted, customers sorted by loan quantity. </li></ul></ul><ul><li>At wholesale: </li></ul><ul><ul><li>Use both quantity and pricing adjustments. </li></ul></ul>
  11. 11. Measuring Credit Risk <ul><li>Qualitative models: borrower specific factors are considered as well as market or systematic factors. </li></ul><ul><li>Specific factors include: reputation, leverage, volatility of earnings, covenants and collateral. </li></ul><ul><li>Market specific factors include: business cycle and interest rate levels. </li></ul>
  12. 12. Credit Scoring Models <ul><li>Linear probability models: </li></ul><ul><li>Z i = </li></ul><ul><ul><li>Statistically unsound since the Z’s obtained are not probabilities at all. </li></ul></ul><ul><ul><li>*Since superior statistical techniques are readily available, little justification for employing linear probability models. </li></ul></ul>
  13. 13. Other Credit Scoring Models <ul><li>Logit models: overcome weakness of the linear probability models using a transformation (logistic function) that restricts the probabilities to the zero-one interval. </li></ul><ul><li>Other alternatives include Probit and other variants with nonlinear indicator functions. </li></ul>
  14. 14. Altman’s Linear Discriminant Model: <ul><li>Z=1.2X 1 + 1.4X 2 +3.3X 3 + 0.6X 4 + 1.0X 5 </li></ul><ul><ul><li>Critical value of Z = 1.81. </li></ul></ul><ul><ul><li>X 1 = Working capital/total assets. </li></ul></ul><ul><ul><li>X 2 = Retained earnings/total assets. </li></ul></ul><ul><ul><li>X 3 = EBIT/total assets. </li></ul></ul><ul><ul><li>X 4 = Market value equity/ book value LT debt. </li></ul></ul><ul><ul><li>X 5 = Sales/total assets. </li></ul></ul>
  15. 15. Linear Discriminant Model <ul><li>Problems: </li></ul><ul><ul><li>Only considers two extreme cases (default/no default). </li></ul></ul><ul><ul><li>Weights need not be stationary over time. </li></ul></ul><ul><ul><li>Ignores hard to quantify factors including business cycle effects. </li></ul></ul><ul><ul><li>Database of defaulted loans is not available to benchmark the model. </li></ul></ul>
  16. 16. Term Structure Based Methods <ul><ul><li>If we know the risk premium we can infer the probability of default. Expected return equals risk free rate after accounting for probability of default. </li></ul></ul><ul><ul><li>p (1+ k ) = 1+ i </li></ul></ul><ul><ul><li>May be generalized to loans with any maturity or to adjust for varying default recovery rates. </li></ul></ul><ul><ul><li>The loan can be assessed using the inferred probabilities from comparable quality bonds. </li></ul></ul>
  17. 17. Mortality Rate Models <ul><ul><li>Similar to the process employed by insurance companies to price policies. The probability of default is estimated from past data on defaults. </li></ul></ul><ul><ul><li>Marginal Mortality Rates: </li></ul></ul><ul><ul><li>MMR 1 = (Value Grade B default in year 1) (Value Grade B outstanding yr.1) </li></ul></ul><ul><ul><li>MMR 2 = (Value Grade B default in year 2) (Value Grade B outstanding yr.2) </li></ul></ul>
  18. 18. RAROC Models <ul><ul><li>Risk adjusted return on capital. This is one of the more widely used models. </li></ul></ul><ul><ul><li>Incorporates duration approach to estimate worst case loss in value of the loan: </li></ul></ul><ul><ul><li> L = -D L x L x (  R/(1+R)) where  R is an estimate of the worst change in credit risk premiums for the loan class over the past year. </li></ul></ul><ul><ul><li>RAROC = one-year income on loan/  L </li></ul></ul>
  19. 19. Option Models: <ul><ul><li>Employ option pricing methods to evaluate the option to default. </li></ul></ul><ul><ul><li>Used by many of the largest banks to monitor credit risk. </li></ul></ul><ul><ul><li>KMV Corporation markets this model quite widely. </li></ul></ul>
  20. 20. Applying Option Valuation Model <ul><li>Merton showed value of a risky loan </li></ul><ul><li>F(  ) = Be -i  [(1/d)N(h 1 ) +N(h 2 )] </li></ul><ul><li>Written as a yield spread </li></ul><ul><li>k(  ) - i = (-1/  ) ln [N(h 2 ) +(1/d)N(h 1 )] </li></ul><ul><li>where k(  ) = Required yield on risky debt </li></ul><ul><li>ln = Natural logarithm </li></ul><ul><li>i = Risk-free rate on debt of equivalent maturity. </li></ul>
  21. 21. *CreditMetrics <ul><li>“If next year is a bad year, how much will I lose on my loans and loan portfolio?” </li></ul><ul><li>VAR = P × 1.65 ×  </li></ul><ul><li>Neither P, nor  observed. </li></ul><ul><li>Calculated using: </li></ul><ul><ul><li>(i)Data on borrower’s credit rating; (ii) Rating transition matrix; (iii) Recovery rates on defaulted loans; (iv) Yield spreads. </li></ul></ul>
  22. 22. * Credit Risk + <ul><li>Developed by Credit Suisse Financial Products. </li></ul><ul><ul><li>Based on insurance literature: </li></ul></ul><ul><ul><ul><li>Losses reflect frequency of event and severity of loss. </li></ul></ul></ul><ul><ul><li>Loan default is random. </li></ul></ul><ul><ul><li>Loan default probabilities are independent. </li></ul></ul><ul><li>Appropriate for large portfolios of small loans. </li></ul><ul><li>Modeled by a Poisson distribution. </li></ul>
  23. 23. Pertinent Websites <ul><li>For more information visit: </li></ul><ul><li>Federal Reserve Bank </li></ul><ul><li>OCC KMV </li></ul><ul><li>Card Source One </li></ul><ul><li>FDIC </li></ul><ul><li>Credit Metrics </li></ul><ul><li>Robert Morris Assoc. </li></ul>Web Surf
  24. 24. Pertinent Websites <ul><li>The Economist </li></ul><ul><li>Fed. Reserve Bank St. Louis </li></ul><ul><li>Federal Housing Finance Board </li></ul><ul><li>Moody’s </li></ul><ul><li>Standard & Poors </li></ul>Web Surf
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.