Scribd
Upload
31 views

Lecture 2 Scoring Model

1. The document discusses Altman's Z-score model, which uses multiple discriminant analysis and five key financial ratios to assess the likelihood of corporate default. 2. These ratios are combined into a Z-score, with scores above 2.99 indicating a safe financial status, scores between 1.8-2.99 indicating a grey area, and scores below 1.8 indicating distress. 3. While the Z-score model is 72-80% accurate in predicting bankruptcy, it only assesses past financial performance and does not reflect changing creditworthiness over time. More advanced models have since been developed, such as the ZETA model.

Uploaded by

soniasafdar97
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
31 views

Lecture 2 Scoring Model

1. The document discusses Altman's Z-score model, which uses multiple discriminant analysis and five key financial ratios to assess the likelihood of corporate default. 2. These ratios are combined into a Z-score, with scores above 2.99 indicating a safe financial status, scores between 1.8-2.99 indicating a grey area, and scores below 1.8 indicating distress. 3. While the Z-score model is 72-80% accurate in predicting bankruptcy, it only assesses past financial performance and does not reflect changing creditworthiness over time. More advanced models have since been developed, such as the ZETA model.

Uploaded by

soniasafdar97
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 18

FINANCIAL MANAGEMENT

Week 2: Scoring model


Learning objectives

• Altman Z-score model to assess financial health


General Motors
• One of the biggest auto companies; more than 100 years
of history; global presence everywhere.

• In 2005, it was one of the most prominent issuers in the


corporate bond market with great rating.

• In 2009? Bankrupt!
• GM announced its bankruptcy under Chapter 11, declaring a debt
of $172 billion and assets only of $82 billion.
• It is the fourth largest in US history, following Lehman Brothers,
Washington Mutual, and WorldCom.

• How did it happen?


More about GM
• https://www.youtube.com/watch?v=xHnPD3ttBlA
Scoring model
• General idea:
• (1) identify those characteristics of business failures
• (2) and quantify the marginal effect of those factors on corporate
distress (like assigning weight to your model)

• For example, equal weight


• Y(score)=0.25*X1+ 0.25*X2+0.25*X3+0.25*X4

• What if factors have different weights?


Model implementation

• Two key questions:

• (1) Which financial and economic factors are most


important in detecting default?

• (2) What weights should be attached to those selected


ratios?
Methodology: multiple regression
• Y = b0 + b1 X1 + b2 X2 + …+ bn Xn

• Outcome variable: Y=1 if the firm defaulted, and 0


otherwise.

• Explanatory variables: a linear combination of explanatory


variables, which best separates default vs. non-default
• Does so by choosing a new line that maximizes the similarity
between members of the same group and minimizing the similarity
between members belonging to different groups

• Examples include discriminant analysis and logistic


regressions
Altman’s Z-score model (1968)
• Edward I. Altman applied multiple discriminant analysis (MDA)
to derive a combination of financial ratios, which can predict
the likelihood of financial distress.

• The discriminant function is Z=V1X1+V2X2+…+VnXn, where Z is


a discriminant score, V1….Vn are discriminant coefficients, and
X1,….Xn are independent variables (financial ratios).

• Z-Score model is a linear analysis in that five measures are


objectively weighted and summed up to arrive at an overall
score that then becomes the basis for classification of firms
into one of the a priori groupings (distressed and
nondistressed)
Cut-off value
• Uses the Discriminant Function line to score new
observations (prospects) and classify them into one
of two groups based on a cut-off value

Cut-off
Value

Z
R2 R1
Cut-off value
• The main challenge for credit risk managers is to define the
most appropriate and efficient thresholds (cut-off) for each
scoring model.
• For example, any scoring model has a “gray” area where it is not able to
separate with an acceptable level of confidence between expected
“good” clients and expected “bad” ones.

• In order to maximize the benefits of a scoring model, the


optimal cut-off should be set taking into account the
misclassification costs related to the type I and type II error
rates as Altman et al. point out.

• The optimum cut-off value cannot be found without a careful


consideration of each particular bank risk preference
Altman’s sample collection
• Initial sample with 66 firms and two groups are made: 33
distressed firms (those filed bankruptcy) vs. 33
nondistressed firms (those still survive as of the anlaysis).

• The distressed firms are manufacturers that file a


bankruptcy from 1946-1965.

• For the nondistressed firms, he matches size and industry


to control their potential effects on financial distress
Variable selections
• 22 financial variables are initially selected and then grouped
into five categories: liquidity, profitability, leverage, solvency,
and activity.
• Current ratio= current asset/current liability
• Leverage= debt/asset debt/equity

• After iterative process of reducing the number of variables, 5


variables are selected from the 22 financial ratios that jointly
provide the “best” prediction of corporate bankruptcy.

• That is, we could not significantly improve upon our results by


adding more variables, and no model with fewer variables
performed as well. Judgment from analysts is also needed.
Risk Management and Financial Institutions 4e,
Chapter 19, Copyright © John C. Hull 2015 13

Altman’s final discriminant function


• X1=Working Capital/Total Assets
• X2=Retained Earnings/Total Assets
• X3=EBIT/Total Assets earning before interest and taxes
• X4=Market Value of Equity/Book Value of Liabilities
• X5=Sales/Total Assets

Z = 1.2X1+1.4X2+3.3X3+0.6X4+0.99X5
Note: above original Altman Z is commonly used for
manufacturing public companies because of “sales”
14

Zones of Discrimination:

Z > 2.99 - “Safe” Zone


1.8 < Z < 2.99 - “Grey” Zone
Z < 1.80 - “Distress” Zone
Implication of Z-score
• A health firm should maintain
• Enough liquid assets to pay bills
• Sufficient internal funds (retained earnings) to meet growth
opportunity
• Good amount of returns for its investors
• Good confidence from stock market
• Efficient operating or sales
Advantage of Altman Z-score
 72% accurate in predicting bankruptcy two years prior to
the event.

 80-90% accurate in predicting bankruptcy one year prior


to the event.

 Reflects the health or weakness of balance sheets.

 It is a starting point of analyzing corporate default


Problems of Z-score
 Backward looking

 The actual relation between Z-score and probability of default


is unclear both on average and across the credit cycle

 The changes in the credit worthiness of the firm can only be


monitored at quarterly intervals at the least.

 Some other sophisticated regression techniques may be used


like logistic regression to better model probability of default
scores.
The ZETA Credit Risk Model
• A second generation model with several enhancements
to the original Z-Score approach.

• The new model, which we call ZETA, was effective in


classifying bankrupt companies up to five years prior to
failure on a sample of corporations consisting of
manufacturers and retailers.

• It is a linear model of 7 variables as predictors, including


return on assets, stability of earnings, interest coverage
ratio, cumulative profitability, liquidity, capitalization, and
firm size

You might also like