Credit

Risk
Managem
ent
 Module 1

Introduction
Contents
1.1
Introduction .............................................................................................1/2
1.2 Credit Assessment
Methods ............................................................... 1/11
1.3 Expected Losses and Unexpected
Losses ......................................... 1/20
1.4 Controlling Credit
Risk ....................................................................... 1/25
1.5 The Credit Policy
Manual ................................................................... 1/33
Learning
Summary ....................................................................................
..... 1/37
Review
Questions ...................................................................................
........ 1/39
Case Study 1.1: Determining the Credit Risk of a
Portfolio ...................... 1/43

Learning Objectives
This module introduces the key ideas for managing credit risk.
Managing credit risk is a complex multidimensional problem and
as a result there are a number of different approaches in use,
some of which are quantitative while others involve qualitative
judgements. Whatever the method used, the key element is to
understand the behaviour and predict the likelihood of particular
credits defaulting on their obligations. When the amount that can
be lost from a default by a particular set of firms is the same, a

Credit Risk Management Edinburgh Business School 1/1

Module 1 / Introduction
higher likelihood of loss is indicative of greater credit risk. In
cases where the amount that can be lost is different, we need to
factor in not just the probability of default but also the expected
loss given default.
Determining which counterparty may default is the art and
science of credit risk management. Different approaches use
judgement, deterministic or relationship models, or make use of
statistical modelling in order to classify credit quality and predict
likely default. Once the credit evaluation process is complete, the
amount of risk to be taken can then be determined.
After completing this module, you should:
• understand the nature of credit risk, and in particular:
 what constitutes credit risk
 the causes of credit risk
 the consequences of credit risk
• understand the nature of the credit assessment problem, and
that:
 credit risk can be viewed as a decision problem
 the major problem in assessment is in misclassifying
credit risk  understand the different techniques used to
evaluate credit risk, namely:
 judgemental techniques
 deterministic models
 statistical models
• be able to set up and undertake the credit review process 
know the basic contents of a credit policy manual.

1.1 Introduction
The company is fundamentally sound. The balance sheet is strong.
Our financial liquidity has never been stronger… My personal belief is
that Enron stock is an incredible bargain at current prices and we will
look back a couple of years from now and see the great opportunity
that we currently have.
Enron Chairman Kenneth Lay on 26 September 2001 in an
online chat with employees, as reported by Reuters. Less than three
months later the company filed for bankruptcy.

As the quotation above indicates, a transaction – in this case buying

a firm’s stock (or shares) – exposes the buyer to unforeseen
outcomes. While buying stock can be considered inherently risky, in
that you are taking a chance on the firm’s performance, most firms
have similar problems: to gain orders, they may need to lend money

1/2 Edinburgh Business School Credit Risk Management

 Module 1 / Introduction
to customers through granting credit terms on sales; they may ask
other firms to undertake work on their behalf; or they may place
surplus cash on deposit with a financial institution. Hence, as a result
of transactions of various kinds, credit risk and credit risk
management are key issues for most firms.
The possibility that a contractual arrangement is not adhered to
means that there is a risk of non-performance. This has the capacity
to hurt the objectives of a firm when what it considered will happen,
in fact, does not. Money can be lost if the customer fails to pay or if
the financial institution in which money is deposited goes bankrupt.
Companies with whom the firm has placed orders may themselves
become insolvent and fail to deliver on their promise.
Credit risk can be defined as ‘the potential that a contractual
party will fail to meet its obligations in accordance with the agreed
terms’. Credit risk is also variously referred to as default risk,
performance risk or counterparty risk. These all fundamentally
refer to the same thing: the impact of credit effects on a firm’s
transactions. There are three characteristics that define credit risk:
1. Exposure (to a party that may possibly default or suffer an adverse
change in its ability to perform).
2. The likelihood that this party will default on its obligations (the
default probability).
3. The recovery rate (that is, how much can be retrieved if a default
takes place).
Note that, the larger the first two elements, the greater the
exposure. On the other hand, the higher the amount that can be
recovered, the lower the risk. Formally, we can express the risk as:

Credit risk ExposureProbability of default 1

Recovery rate ﴾1.1﴿
Given the above, credit risk management is the process of
controlling the potential consequences of credit risk. The process
follows a standard risk management framework: namely
identification, evaluation and management. That is, the cause of the
risk has to be identified, the extent of the risk has to be evaluated and
decisions have to be made as to how this risk is to be managed. 1

1.1.1 Credit Risk

What is credit risk? Well, the easiest way to consider credit risk is to
think of your own situation. Take the case where an acquaintance,
someone you may have known at school or in a social situation, turns
1 The elective courses on Financial Risk Management and Strategic Risk Management
examine the process of risk management in detail.

Credit Risk Management Edinburgh Business School 1/3

Module 1 / Introduction
to you and asks you to lend them some money. Not a trivial amount to
pay for their bus fare home but a sufficient amount so that, if they do
not repay you as promised, you are left significantly out of pocket.
What do you do? Do you lend the individual the money? They may not
repay you. Therefore, it is better to refuse. Then again, you may lose
out on a possible profitable opportunity.
The crux of the decision is whether the individual honours the
promise to repay or defaults. The desirable result is that the loan is
repaid (with interest). The undesirable outcome that you wish to
avoid is that the individual fails to repay the loan or, in the parlance
of credit, defaults.
Note how the example raises all sorts of issues. If you knew the
individual better, you might be more inclined to go with the lending
decision (that is, if you knew the person’s circumstances and their
ability to repay). The past experience of others who have lent money
to the individual might be useful to know. You may also wish to
compare the individual to others who have borrowed money in a
similar situation. As a result, you may be able to obtain a statistical
estimate of the likelihood that the individual will repay you (or,
equivalently, will default on the loan).
Your views as to whether you would be wise to lend the money to
this acquaintance might change if the individual produced a
guarantee to support the loan, or some collateral (that is, something
you could call upon if the individual were unable or unwilling to meet
the obligation).
Whatever your thoughts, the decision requires you to make a
judgement on the uncertain future outcome. This might take the form
of a gut feeling (or what professionals would term expert
judgement), or you might be able to rely on a formal assessment
model.
In commerce, every time an individual or a firm borrows and hence
makes a promise to pay, a financial asset is created. This promise can
be informal and take the form of a verbal agreement or can be based
on a formal written contract. The promise can involve the purchase of
an asset, product or a service from the provider. The promise can also
be, as in the above example, to repay a loan. Regardless of the
purpose of the transaction, the value of the promise will depend on
the ability and willingness of the person or firm to make good on the
promise.
Some financial assets are backed only by the general credit and good faith of
the borrower to repay. Others are backed by legal obligations that would force
payment or the forfeiture of a specific asset. Such collateralised promises include
liens, mortgages, leases and auto loans. Other contractual arrangements provide

1/4 Edinburgh Business School Credit Risk Management

 Module 1 / Introduction
for a third-party guarantee and hence, because the guarantor is also pledging
their credit, require analysis of both the initial party and the quality of the
guarantee.

Box 1.1: What are Credit Events?

The market that has developed in credit risk transfer, known as credit
derivatives, has led to the formalisation of what constitutes credit events. The
following list gives the principal types as usually defined in the documentation for
these instruments.
Bankruptcy or insolvency under corporate law: in this case the legal entity
is dissolved, having been declared bankrupt or insolvent by a commercial court.
Credit upon a merger: assets are transferred (with a negative impact on the
ability to pay) or the legal entity is consolidated into another company.
Cross-acceleration: other obligations of the legal entity become due prior to
maturity owing to a breach of contract. (In debt contracts, covenants – which are
legally binding requirements on the borrower – are breached, which leads to
repayment being mandated.)
Cross-default: obligations of the legal entity have been declared in default.
Under debt contracts, failure of one set of obligations to perform leads to the
automatic declaration of other obligations to be in default, even if these
obligations have not experienced any breach of contract (covenant).
Currency convertibility: foreign exchange controls in a particular country or
countries prevent repayments in the affected currency or currencies. Note that
currency convertibility is usually linked with currency risk. Country risk is the risk
associated with lending to a particular country, whereas default risk is usually
company specific.
Downgrade: the situation where an obligor has a downward, hence adverse,
change made in the independently and publicly available credit opinion (known
as a credit rating) on the obligor. This can include the cancellation of the
available rating. A downgrade may lead to holders having to sell the obligor’s
debt securities.
Restructuring: the legal entity defers or reschedules outstanding debt(s):
reduces the interest payable, postpones payments, changes the obligation’s
seniority or extends its maturity.
Failure to pay: the legal entity is not able to or does not make the contractual
payment.
Government action: any action(s) by a government or an agency of
government that results in outstanding claims becoming unenforceable against
the legal entity.
Market disruption: the situation where the tradable securities of
the obligor cease trading.

Credit Risk Management Edinburgh Business School 1/5

Module 1 / Introduction
Moratorium on debts: the legal entity declares a standstill on its
existing debts and interest payments.
Obligation acceleration: a contractual obligation becomes
payable before its due maturity owing to default by the legal entity.
This is similar to, but not identical to, cross-default, since it refers to
a direct obligation with an affected party (whereas cross-default
refers to a third party).
Obligation default: an event of default has occurred in the legal
entity’s obligations.
Repudiation: the legal entity disaffirms or disclaims its debts.

1.1.2 Assessing Credit Risk

Assessing credit risk requires us to model the probability of a
counterparty defaulting in full, or in part, on its obligation. We can
picture the credit decision in terms of the basic risk management
model. This involves a decision either (A) to extend credit, which
provides a reward but entails a risk, or (B) to refuse credit. The
situation facing the credit manager is shown as a decision problem in
Figure 1.1. The requirement is to balance the gain from taking the
credit risk by extending credit against the potential loss. In the
decision problem the alternative is to refuse credit and not obtain any
reward.

Credit pays Revenue – cost

A: Extend credit (1 – )

Credit defaults – Cost or

replacement value

Decision

B: Refuse credit 0

Figure 1.1 The credit decision as a decision problem

Note: In this model, boxes designate decisions and circles designate
outcomes
The credit risk decision facing a firm relates to (1) the gain if no
default happens against (2) the potential loss from extending credit
based on the likelihood that default takes place and the amount that
is lost if default occurs. The probability that the credit defaults is
given as (ρ). There are only two possible outcomes: the credit
performs according to expectations or the credit defaults. If the

1/6 Edinburgh Business School Credit Risk Management

 Module 1 / Introduction
credit defaults, the cost to the credit manager will be the cost or the
replacement value for what has not been provided.2
For instance, in deciding whether to provide trade credit, a firm
faces a decision as to which applications to proceed with; what limit
to set on the amount of credit extended and whether this needs to be
modified over time; what action should be taken if there is a delay in
repayment; and which counterparties should be actively solicited for
business.
Although the nature of the credit analysis decision can be readily
described, the steps required to effectively manage the process are
more complicated. In essence, the problem relates to the risk that
counterparties will not honour their obligations when the moment
comes for them to perform under their contract.
Determining which counterparty may default is the art of credit
risk management. Different approaches use judgement, deterministic
or relationship models, or make use of statistical modelling, in order
to classify credit quality and predict likely default frequency. Once
the credit evaluation process is complete, the amount of risk to be
taken can then be determined.
In applying the decision model, the effect of credit exposure is
measured by the cost of replacing the cash flows if the other party
defaults. But losses do not arise only from failure to pay. Losses also
result from credit risk when credit rating agencies downgrade firms
or businesses. Where these obligations are traded securities, there is
often a decrease in their market value. In addition, where the
transaction is cross-border, country risk needs to be included in the
risk assessment.

1.1.3 Additional Factors in Analysing Credit Quality

Credit risk has a number of sub-issues including concentration risk
and settlement risk. Concentration risk arises when there are a
large number of exposures to parties that share similar
characteristics. Settlement risk arises when a clearing agent or third
party processes transactions for other parties. Indeed, financial
institutions have recognised the importance of measuring credit
concentration risk in addition to the credit risk of individual loans.
That is, they examine not only individual risks but also the total credit
risk characteristics at the portfolio level.
Settlement risk is particularly important for financial institutions
that process a large number of high-value transactions. Prior to 1974,
2 Obviously, as per Equation 1.1, there is the potential for some recovery, so the cost is
simply the amount times (1 × Recovery rate). Higher recovery rates mitigate the
credit risk.

Credit Risk Management Edinburgh Business School 1/7

Module 1 / Introduction
settlement risk was not considered a major issue by banks until a
small German bank, Bankhaus Herstatt, defaulted on interbank
payment transactions after receiving payment from counterparties.
Herstatt’s failure prompted the establishment in 1974 of the Basel
Committee on Banking Supervision (BCBS).3
As a business grows, a greater number of exposures are acquired, increasing
credit risk. As the scale increases, it is necessary for the firm to consider how
concentrated its portfolio of credit customers is. In particular, two sub-types of
credit risk – country risk and industry risk – affect multinational enterprises.
Country risk arises from having exposure to individuals and institutions in
countries that have legal systems, business codes and standards that differ
from those of the lender. There are four factors relevant to this. The first is
political risk, which arises when a country’s government is challenged
externally or from within national borders. Political risk is more problematic in
long-term lending agreements than for short-term transactions. An example of
this problem is lending to countries of the former Soviet Union. Some of these
countries are highly unstable politically, and a change of regime might mean
that the new government repudiates or seeks to renegotiate contracts already
entered into.
The second factor in country risk is economic risk, namely the depressed or
declining economic stability in a country. In this situation it is sensible to
question the quality of loans or credit to such a country and also to implement a
limit to any such agreement. In 1997, for example, the South East Asian Crisis
meant that both corporations and banks had to revise their policy towards
counterparties in the affected countries.
The third factor is currency risk, which always arises with cross-border
lending. If a British company extended credit to, say, an Australian company in
Australian dollars and the value of the Australian dollar declined, the value of
the debt, in British pounds, would decline if the loan had not previously been
hedged to remove the currency effect.4
Finally, the fourth factor is the enforcement risk from the legal system in
the debtor country. Because a creditor has to go through a foreign legal system,
it has been known for debtors to use their domestic legal process to stall or
attempt to avoid paying, claiming that rules from their home country apply.

3 This is the Group of Ten (G10) countries’ coordination effort to set joint bank capital
adequacy standards. The BCBS is situated at the Bank for International Settlements
(BIS) in Basel, Switzerland. The BIS is colloquially known as the central banker’s
bank, since it acts as a clearing house for the major central banks in the world. See
www.bis.org for further details.
4 Hedging is the process of removing a risk through the use of financial markets transactions. A
loan in Australian dollars could be hedged by using forward foreign exchange contracts to
convert the future cash flows from the loan back into British pounds. Therefore, regardless of the
future behaviour of the Australian dollar, the British company would be guaranteed a fixed
amount of British pounds. These kinds of transaction are a common feature of the international
capital markets. For a fuller discussion, see the elective course Derivatives.

1/8 Edinburgh Business School Credit Risk Management

 Module 1 / Introduction
Another important credit risk is industry risk, which is a form of
concentration risk. This applies particularly when the domestic or international
economy is in recession and the poor economic conditions particularly affect
certain industries. Industry structure may have credit consequences because of
the supply chain within which most firms operate. For instance, a steel
producer is involved with car manufacturers. This has two important
consequences. If car sales decline, this affects manufacturers of motor vehicle
components, together with the car manufacturers. Consequently, a producer
with all its output destined for one industry finds it impossible to avoid industry
risk exposure to that industry.
1.1.4 Evaluating the Credit Decision
The decision model given in Section 1.1.2 provides a theoretical
framework for assessing the risk. One approach to evaluating the risk
is to work out the payoffs from the choices facing the firm (or
individual). The ex ante payoff of the two situations is:

Extend credit: PV Revenue Costs 1 PV Cost

﴾1.2﴿
Refuse credit: 0
The reward is the revenue earned less the costs; the risk is the full
or partial loss if a default takes place. In evaluating the desirability of
taking on credit risk, different situations will have different levels of
risk based on (a) the probability of default ( ρ) and (b) the amount of
loss that is expected or actually incurred. Thus lending, for instance,
has a higher exposure (the amount at risk) for the same sized
contract than does entering into an off-balance-sheet derivative
agreement such as a swap. If a firm goes into liquidation, the loan
might be an almost total write-off, but the risk on the swap relates to
the difference between the original value and its replacement cost.
Part of the evaluation process therefore needs to calculate the
exposure that will arise if default takes place, and this exposure
needs to be controlled by setting overall credit limits per individual
counterparty and by industry type and country, if applicable.
Firms gain a reward by accepting the risk that the credit may
default. Alternatively, nothing is hazarded if credit is refused. Since
most organisations have to take risks to earn a return, the decision is
slightly more complex than this simple model would suggest. In their
day-to-day activities, firms seek to trade off credit risk against the
potential gains and losses. These arise not just from accepting bad
credits but also in rejecting good ones.
At this point it should be noted that modern finance theory
suggests that rejecting credit is not necessarily the appropriate
response to poor credit quality. The tenets of modern theory about

Credit Risk Management Edinburgh Business School 1/9

Module 1 / Introduction
risk postulate that the required return should be adjusted for the risk
taken. If the risk has been correctly estimated, then, for large
organisations over the medium term and where portfolio
diversification effects apply, losses will be compensated by gains
elsewhere. Finance theory would, in addition, suggest that only the
systematic risk component need be priced. One aim of the credit
modelling process should be to provide estimates of the likely risk.
The decision can then be made as to whether to provide a credit line
at an appropriate risk-adjusted price to compensate for the risk or to
find ways to reduce the degree of exposure but still enter into the
transaction.5 This is a more sophisticated approach than is used in
most organisations, which tend to adopt a ‘yes’ or ‘no’ view to
extending credit and also seek to control their exposure via limits on
the amounts at risk.
For instance, if the company earns a margin of 20 per cent on
sales, then it will break even; that is, it will be indifferent towards
extending credit or refusing credit if:
PV Revenue Costs 1 PV Cost 0 ﴾1.3﴿
PV 20 1 PV 100 0
20 20 100
20 120
0.167

So, with a probability of loss of 0.167, the firm is indifferent

between accepting a credit risk and refusing it. Note that if the firm
is risk averse then it will need an expected positive payoff to
compensate it for risk taking.
The firm can do two things to increase its sales and yet not incur
unacceptable credit losses. First, it can be indifferent to a higher
level of losses if its margin is higher. If the margin were 30 per cent,
then it would be indifferent at a loss probability of 0.23. On the other
hand, if it can reduce its loss given default, say to 80 rather than 100,
then again the firm can be indifferent with a loss probability of 0.2.

1.1.5 Errors in the Credit Evaluation Process

Two types of error can arise when evaluating a credit decision. The
type I error is advancing credit to a lesser-quality credit (that is, a
‘bad credit’ that has mistakenly been classified as a ‘good credit’) and
thereby incurring an unanticipated loss. The type II error arises from
misclassifying a good credit as a bad credit and thereby forgoing an
opportunity to earn profit. The different risks can be portrayed in

5 Note that such risk pricing is common in the banking industry, which tends to be at
the forefront of developments in risk management.

1/1 Edinburgh Business School Credit Risk Management

0
 Module 1 / Introduction
terms of the actual credit quality (here simply called ‘good’ or ‘bad’
credit) versus the analysed credit quality, as shown in Figure 1.2.
Actual credit quality

Good Bad

Accept firm Assessing bad

as good firm as good
means

Analysed credit
Good
unexpected risk
of loss
(Type I error)

quality
Opportunity
loss from
rejecting
Bad good firm
as bad
(Type II Reject firm
error) as bad

Figure 1.2 The credit assessment problem

Note: The analysis of credit quality is in relation to the expected level of loss
from credit effects
In practice, the credit analyst will devote more time to avoiding type I errors;
that is, to assessing bad credits as good ones. The financial consequences of
accepting bad risks that have mistakenly been classified as good ones are greater
than if some good risks are mistakenly rejected. This is because the costs of
extending credit in a situation where there is a credit event are far greater than
the opportunity for profit forgone by refusing credit to the good risk. This is due to
the uncertainties in loss recovery rates and the opportunity costs involved. That
said, a credit evaluation model that habitually rejects high-quality good credits as
bad means excessive opportunity losses from forgone business. Hence the
probability of default of a particular kind of credit needs to be carefully factored
into any analytic framework.

Box 1.2: Credit Class Default Rates and Class Transition

Information about credit losses is collected from a number of sources and firms,
such as banks and consumer credit companies, whose business directly involves
credit lending. Such firms keep records of their historical performance. Most of
the information collated by individual companies is not in the public domain.
However, the US credit rating agency Standard & Poor’s does provide the default
history for the companies it has rated.
Standard & Poor’s classifies companies of ‘investment grade’ (that is, firms where
there is a reasonable level of security) with a credit rating of triple-B (BBB) or
above. The best-rated credits/firms are classified as triple-A (AAA). Firms or

Credit Risk Management Edinburgh Business School 1/1

1
Module 1 / Introduction
credits below triple-B (that is, double-B (BB) or below) are defined as ‘speculative
grade’. These firms have a higher likelihood of default over any given period.
Table 1.1 shows the cumulative default history for rated firms over time. These
ratings relate to large, publicly traded corporations and their securities and
hence may not be typical of the corporate sector as a whole, given the size bias.
The table shows that there is a 0.0024 probability that a triple-B rated company
(that is, a company classified at the lowest level of investment grade) will default
over the coming 12 months. However, the cumulative probability of such a credit
defaulting over 10 years is 0.0685. That is, over a 10-year period nearly 7 per
cent of such rated companies are likely to default. Table 1.1 shows that credit risk
increases with time and inversely to the creditworthiness of the obligor. Note that
the risk rises asymptotically: a drop in credit class more than doubles the risk of
default. So a firm rated single-A has a one-year probability of default of 0.0004,
whereas for a firm rated triple-B the probability of default, 0.0024, is 6 times
higher.
Table 1.2 shows the propensity of firms in a given class to change credit rating.
This change of creditworthiness, as measured by ratings, is known as an
‘upgrade’ if the credit rating is raised or as a ‘downgrade’ if the credit rating is
lowered. What is evident in the table is that a firm of a given credit quality is
most likely to remain in its existing class, but there is some propensity, reflecting
changes in the underlying corporate dynamics, for the company to change credit
class. Note that the default probabilities for Table 1.2 differ
somewhat from those in Table 1.1 owing to the way the two
tables have been compiled.

Table 1.1 Cumulative default rates (in per cent)

Years

Rating 1 2 3 4 5 7 10
AAA 0.00 0.00 0.04 0.07 0.12 0.32 0.67
AA 0.01 0.04 0.10 0.18 0.29 0.62 1.39
A 0.04 0.12 0.21 0.36 0.57 1.01 2.59
BBB 0.24 0.55 0.89 1.55 2.23 3.60 6.85
BB 1.08 3.48 6.65 9.71 12.57 18.09 27.09
B 5.94 13.40 20.12 25.36 29.58 36.34 48.81
CCC 25.26 34.79 42.16 48.18 54.65 58.64 62.58
Source: Derived from historical information froms.
Standard & Poor’

Table 1.2 Rating transition matrix (in per cent)

Rating at Ratin at endf (%)
AA CCC Defaul

1/1 Edinburgh Business School Credit Risk Management

2
 Module 1 / Introduction

start of g o period B
period (%) AAA
A BBB BB t
AAA 93.1 6.3 0.5 0.1 0.1 - - -
AA 0.6 91.0 7.6 0.6 0.1 0.1 0.0 0.0
A 0.1 2.1 91.4 5.6 0.5 0.2 0.0 0.1
BBB 0.0 0.2 4.4 89.0 4.7 1.0 0.3 0.4
BB 0.0 0.1 0.4 6.1 82.7 7.9 1.2 1.5
B - 0.1 0.3 0.4 5.3 82.1 4.9 7.0
CCC 0.1 - 0.3 0.6 1.6 10.0 55.8 31.6
Default - - - - - - - 100.0
Source: Derived from historical information from Standard &
Poor’s.

If the analyst can correctly identify the credit quality of the

counterparty, then steps may be taken to protect the lender. For
instance, in the case of a financial institution that holds a loan, asset
or instrument, or credit position with the counterparty, this may be
closed out, insurance purchased, or the loan sold off to another (less
perceptive) institution. For a supplier extending trade credit, a (high-
risk) customer can be required to pay cash or provide suitable
collateral to offset the credit risk.

1.2 Credit Assessment Methods

Given the decision to be made, it is necessary to analyse the credit in
order to determine its quality. In the parlance of credit risk
management, we would want to determine the counterparty’s
creditworthiness. This is an obligor’s ability and willingness to
honour its agreement with the party extending the credit (that is, the
party ‘at risk’ from non-performance of the legal entity).6
In order to establish the status of the counterparty, credit analysts will
typically use a combination of financial or accounting data and non-financial
variables, as well as a number of different models, or analytical tools. Some of
the methods involve a subjective approach, such as judgemental methods;
others are more systematic in that they use quantitative techniques to evaluate
a credit against objective benchmarks. We can distinguish a number of different

6 Note that, in some kinds of transactions, the credit risk goes in both directions. So, for a swap
transaction (that is, an off-balance-sheet instrument involving a future exchange of contractual
cash flows), the performance risk lies on both sides. That said, in most cases credit risk is carried
by one of the parties to the transaction.

Credit Risk Management Edinburgh Business School 1/1

3
Module 1 / Introduction
approaches and their underlying methodology, which are summarised in Table
1.3.7

Table 1.3 Different approaches to the credit evaluation process

Approach Methodology
Judgemental methods Apply the assessor’s experience and
understanding of the case to the decision to
extend or refuse credit
Expert systems Use a panel approach to judge the case or
(e.g. lending formalise judgemental decisions via lending
committees) system and procedures
Analytic models Use a set of analytic methods, usually on
quantitative data, to derive a decision
Statistical models Use statistical inference to derive appropriate
(e.g. credit scoring) relationships for decision making
Behavioural models Observe behaviour over time to derive
appropriate relationships for reaching a decision
Market models Rely on the informational content of financial
market prices as indicators of financial solvency

Process space
Credit Risk Management

Problem Analytic space Decision

space space

Problem Knowledge Effect Statistical Decision

definition models models models making

Judgmental approach Systemic approach

Problem evaluation

Data/information sources

Business environment

Environment space

Figure 1.3 Credit evaluation process

One way to characterise the credit risk management process is shown in Figure
1.3. The schematic should be seen as a continuum from left to right, where the

7A detailed discussion of these models is undertaken in later modules of this course.

1/1 Edinburgh Business School Credit Risk Management

4
 Module 1 / Introduction
first step is to define the problem. In most cases this is simply a no-default/default
variable. In some applications, it might be more complex, since we may want to
monitor and evaluate changes in credit quality and not simply non-performance.
Irrespective of how the problem is defined, the problem is then analysed. The
different methods given in Table 1.3 can be mapped into the analytic space. These
require data and/or information from the business environment (for instance,
company reports, news reports, financial statements, market prices of the firm’s
securities, payment history and so on). The different analytical approaches can be
loosely grouped into (1) knowledge models, which have a degree of subjectivity
(for instance, the use of expert judgement by an analyst), (2) effect models, which
combine some elements of subjectivity and systemic analysis (ratio analysis would
fall into this category), and (3) statistical models, which can be considered more
systemic in approach (credit scoring models are of this kind). The results of the
analysis are used in the decision space, namely to reach a decision as to whether
to grant, or not grant, credit.
A detailed analysis of the structure and characteristics of the various models is
reserved for later modules of this course.
Box 1.3: Wise Advice on the Credit Decision Process
from Lenders
Rouse (2002), in his book on bank lending, suggests that the
professional credit risk manager apply the following ‘lending
principles’ to the credit decision:
• Take time to reach a decision.
• Do not be too proud to ask for a second opinion.
• Get full information from the customer and do not make
unnecessary assumptions (i.e. do not lend to a business you do
not fully understand).
• Do not take a customer’s statements and representations at face
value and do ask for evidence to support the statements.
• Distinguish between facts, estimates and opinions when forming
a judgement.
• Think again when your gut reaction suggests caution, even
though the factual assessment looks satisfactory.
Rouse argues that adherence to the above principles facilitates well-
informed, thoroughly analysed and documentary-supported credit
decisions. At the same time, in implementing the credit decision
process, Rouse also considers it important to avoid unnecessary
bureaucracy and delays.
The basic methodology behind classifying credit risk is to
undertake financial detective work in order to determine the
likelihood of default. Because analysing firms is a complex,
multidimensional process, this likelihood is often not normally

Credit Risk Management Edinburgh Business School 1/1

5
Module 1 / Introduction
expressed as a probability but rather as an opinion. The process can
be divided into two principal areas: qualitative assessments and
quantitative assessments. These are then often combined to
produce the credit assessment. Note that, although they are
described as separate activities, the two are usually carried out in
parallel, insights gained from one aspect of the examination leading
to investigation in the other. So while the different methods used to
assess credit quality are discussed separately, they are often
combined in the decision-making process.

1.2.1 Classifying Risk and Predicting Default

For operational purposes, it is often the case that the credit
assessment is used to classify a particular firm into a given credit
class. These are called credit ratings or credit opinions. Different
commercial credit assessors and firms using their own credit
assessments use different rating systems. The rating system used by
rating agencies such as Standard & Poor’s has four categories of
investment credit quality and three categories of speculative credit
quality. The intention is to group cases in a consistent way such that,
for decision-making purposes, all firms in a particular group will be
treated as equivalent.
Since all firms within a particular group can be considered as
having the same degree of creditworthiness, the group experience
can be applied to any new credit being analysed. The process is
shown schematically in Figure 1.4.

1/1 Edinburgh Business School Credit Risk Management

6
 Module 1 / Introduction

Firm K to be graded for creditworthiness

EVALUATION

Quantitative Qualitative
Comparator firms
assessment assessment

ASSESSMENT

Decision

CREDIT CLASS

AAA AA A BBB BB B C D

Firm K Firm K is determined to be quivalent

to single-Acredit quality with characteristics
that match other single-Arated firms

Figure 1.4 Classifying credit quality

The procedure in Figure 1.4 is for a particular firm (K) to be
analysed – in this case using both a formal quantitative approach and
a more judgemental qualitative approach in order to determine its
credit category. In doing so, it is compared to similar firms whose
credit quality has already been determined, and hence the firm is
deemed to be like a particular type of credit for default prediction
purposes. In Figure 1.4, firm K is categorised as having a single-A
creditworthiness and hence (referring back to Table 1.1) would have
a one-year default probability of 0.0004.
Given different default rates and types of firms, the number of
classes of credit quality can be greater or fewer depending on the
granularity of the model. For instance, many banks use a scale of 10
credit classes, with the highest credit quality being that of the state
of the country of incorporation and the lowest being default.

1.2.2 Qualitative and Quantitative Credit Assessment Methods

As the previous section indicates, there are a number of approaches
to the evaluation of credit. We can categorise the list in Table 1.3 into
four categories, which are, to some extent, overlapping. We can
consider these to be (1) expert systems, (2) rating systems, (3) credit
scoring models, and (4) market-based models. In practice, credit

Credit Risk Management Edinburgh Business School 1/1

7
Module 1 / Introduction
analysts use a combination of methods to evaluate firms and to
predict their future creditworthiness
The first method is expert systems. These range from the simple
judgement of the credit analyst to more formal models. Some of these
involve templates or processes. Expert systems or qualitative models
are based on judgements as to what constitutes good and bad credit
quality, a typical model of this type being the commonly used 6 Cs of
credit.8 Such models have been built up over time based on
individuals’ and organisations’ collective experience of the credit
process and are reflected in a set of operating procedures. Although
not rigorous, such models can be useful in complex situations and as
checklists when carrying out an assessment.
One important area for the use of expert systems is in financial analysis. This
is the process of examining the financial statements of a firm with a view to
understanding the nature, activity and risks that are inherent in the business.
When formalised, the use of expert systems tends to be combined with the
second approach, rating systems, where the credit quality of a firm or an
individual is categorised into a pool of cases that are considered to have the
same degree of creditworthiness.
For instance, Dun & Bradstreet, the credit reporting agency specialising in
analysing small and medium enterprises, publishes a composite credit appraisal
based on firm size (defined as net worth) as a proxy for financial capacity and
dividing firms within a given net worth band into ‘high’, ‘good’, ‘fair’ and
‘limited’ quality. These composite rankings are determined from a range of
factors that have a bearing on credit quality, including elements such as
number of employees, the firm’s payment history and so on.
The analytic model approach is based on using financial information and
makes use of accounting relationships that, when taken together, provide a
picture of the credit quality of the entity. The best-known model of this type is
the DuPont system, named after the company that developed the approach.
At a more systemic and formal level, rating systems develop into the third
category of credit assessment method, known as credit scoring models.
These scoring models provide a rating system that is formalised into a
mathematical or statistical model, and all credits are assessed using the same
data and methodology. As such, they are more rigorous and transparent in their
approach than rating systems that still depend on judgement, although they are
designed to provide the same level of decision support.
The statistical model assessment model is based on multivariate statistical
inference techniques. In such models a large sample of good and bad credits
are analysed in order to establish a differential equation. The best-known of
these models is the Z-score method used for predicting the probability of
8 These are discussed in Module 4 and refer to character, capacity, capital, collateral, conditions
and compliance.

1/1 Edinburgh Business School Credit Risk Management

8
 Module 1 / Introduction
corporate failure based on discriminant analysis techniques. Statistical models
are also used in a more general process known as credit scoring.
The final category of credit evaluation models is market-based models.
These are formal models, as with credit scoring models, but the information
used to determine credit quality is derived from financial market prices. These
models make use of the information processing that takes place in financial
markets to model the probability of default. For example, if investors have
reservations about the future creditworthiness of a particular listed company,
they tend to sell the shares. This will have the effect of reducing the share price
as these concerns get translated into the market price. A credit assessment
model that can capture this effect is using the combined understanding and
information processing of all investors in the market. Thus this last type of
model uses a wider information set than the first three.

Credit Risk Management Edinburgh Business School 1/1

9
Module 1 / Introduction
Box 1.4: Analytic Templates for Credit Assessment
The process of credit assessment is discussed in depth in later
modules. This section seeks to illustrate the judgemental
method, which is inherently qualitative in its approach. It also
demonstrates the many facets of credit risk. In order to
facilitate analysis, to ensure that all key areas are addressed
and to provide a consistent template for reporting, the
practice is to group a credit’s attributes into categories. These
are often in the form of mnemonics. A popular judgemental
template used in the banking industry is CAMPARI and ICE.
This is, in essence, a summary of banks’ good lending
practices.
CAMPARI, which relates to the performance risk in lending,
stands for:
C character (of firm and its
managers) A ability (of
managers/directors)
M means (of repayment based on financial resources of
credit) P purpose (of credit)
A amount (in absolute and relative terms)
R repayment (how, when, likelihood)
I insurance (what will ensure repayment – if anything)

Character refers to the integrity of the business and its

management. Honest borrowers of good character are more
likely to meet their obligations.
Ability refers to the legality of the contract between the bank
and customer. A company’s directors must act within the legal
authority granted to them in their Articles of Incorporation.
Means refers to the borrower’s financial, technical and
managerial means.
Purpose refers to the reason for granting credit, which must
be unambiguous and acceptable to the lender. For example,
an acceptable purpose would be borrowing to fund faster
growth of a company.
Amount refers to the quantity of the loan, which should be
sufficient to cover the purpose of the borrowing.
Repayment relates to the ability of the borrower to repay the
loan, by considering the source of repayment. This repayment
ability is obviously of critical importance in lending and should
be demonstrated not through projected future accounting
profits but from projected cash generation. In deciding the

1/2 Edinburgh Business School Credit Risk Management

0
 Module 1 / Introduction
form of lending, a credit provider would also need to consider
the repayment structure being considered, e.g. bullet (a one-
off lump sum repayment of the principal) or amortising (that
is, principal repayment
through instalments).
Insurance refers to a safety net that the bank can rely on if the
loan is not repaid. This might be collateral or the security provided
in the loan, the conditions under which the loan is granted or third-
party credit enhancement.
ICE is the lender’s rewards for assuming the performance risk and
stands for:
I interest (paid on borrowing)
C commissions (paid to the lender)
E extras (cost of granting credit)

Interest refers to a key factor, namely the overall interest cost to

the customer. This will comprise two elements: firstly, the
underlying cost of funds (which could be fixed at the outset or
variable) and, secondly, the margin. It is usually the case that, the
higher the risk of a transaction, the greater the interest cost. Note
that in bank terms this is simply an application of risk pricing.
Commissions refers to all other fees, such as commitment fees,
payable to the bank for agreeing to provide a facility for a particular
time period.
Extras relate to additional hidden costs, such as legal fees,
associated with the provision of a loan.
Note that the total return to the bank will be the interest margin
earned between what it can borrow at and the rate it lends less the
extras associated with granting the loan. The decision the bank then
has to make is the same as that in Figure 1.1.
To summarise, for lending purposes the main reason for carrying out
a credit assessment such as that in the template above is always to
ascertain the solvency and creditworthiness of the borrower. This
necessitates a multidimensional study of the industry the business
is involved in, its management, financial situation and market
position.

1.2.3 Credit Evaluation Process

When a new credit application arrives, it is analysed using one of the
methods described in the previous section (such as the CAMPARI

Credit Risk Management Edinburgh Business School 1/2

1
Module 1 / Introduction
template). Any credit decision will be focused on the business,
financial and structural risks, if applicable, and will keep the
following key credit considerations in mind:
• The firm’s competitive environment (positioning amongst peer group,
innovation, brand image, relative cost position and other factors that impact
on the firm’s competitive advantage).
• Industry risks, namely technology, regulatory requirements, barriers to
entry for competitors and possible substitutes.
• Capital structure, which would include, depending on the purposes of the
credit analysis, the level of capital expenditure requirement, any off-balance-
sheet liabilities, accounting and tax issues, debt leverage, repayment
structure of debt, debt service capacity and the interest basis (that is,
whether the interest payment is fixed or is priced on a periodically resettable
basis).
• Financial flexibility, which includes elements such as financial needs, plans
and alternatives, ability to tap capital markets, and debt covenants. Financial
flexibility can include issues such as bank relationships, committed lines of
credit, current and future debt capacity and other issues.
The analysis of the competitive environment and industry risks, coupled with
the analysis of the bargaining power of suppliers and buyers, will give a clear
picture of the forces shaping industry competition. This is often based on
Porter’s (1985) five forces analysis.9 A firm’s external opportunities and threats
can be accounted for, recognising the firm’s and the industry’s product life
cycle and future prospects.10 An analysis of political, economic, social and
technological aspects (known as a PEST analysis) can be undertaken to give a
full picture of industry developments.
It is also very important to analyse the strategy a potential credit is planning
to follow (for example, whether the firm intends to be a low-cost producer or
use a differentiated or focused business strategy) and to examine the quality of
its management team (perhaps one of the most important aspects of
creditworthiness), its brand, and other intangible factors that can determine
the difference between success and failure of a business. Non-financial
information used by credit analysts includes:
• a lender’s own reports on reputation and experience of customers received
from local market resources;
• reports prepared by credit appraisal agencies (if available), financial
analysts, information providers or credit rating agencies and bureaus;
• environmental risks reports; and

9 The five forces are industry rivalry, power of suppliers, power of customers, threat of substitutes
and threat of new entrants.
10 High-technology industries are prone to major market shifts that can destabilise firms and lead
to large write-downs in the economic value of assets that may have been used as collateral.

1/2 Edinburgh Business School Credit Risk Management

2
 Module 1 / Introduction
• call memos and reports prepared from on-site reports and relationship
managers.
For more sophisticated analysis, it may be possible to investigate the
likelihood of credit migration (that is, the risk of a credit downgrade over some
given time frame) and the risk of default (which can be considered the
uncertainty surrounding a credit’s ability to service its debt and obligations).
Measures of expected default frequency (EDF) provide benefits in the
evaluation process. They:
• offer a greater degree of accuracy in the evaluation of the credit risk;
• quantify risk for appropriate pricing and hence improve profitability (that is,
having an estimate of the expected default frequency facilitates the use of
risk pricing);
• focus credit analysis resources into those areas where they can add the most
value; and
• provide early-warning indicators of serious credit deterioration.
Three factors in the business and capital structure of a firm
contribute to a credit’s loss given default (LGD):
• Value of assets in the firm. Namely, the market value of the firm’s assets;
that is, the discounted future cash flows of the firm generated by its assets.
• Asset risk: the uncertainty or risk value of the asset value; this is a measure
of the firm’s business and industry risks.
• Leverage: the extent of the firm’s contractual obligations and claims on
future cash flow. Higher levels of leverage increase the likelihood a firm will
not be able to service its fixed obligations, such as debt and creditors. In the
jargon of finance, such firms are deemed to be in financial distress and
obligations of these firms are deemed to be distressed debt. Once a firm
enters legal bankruptcy or insolvency proceedings, it is unequivocally in
default.
The above factors affect credit risk because they affect the amount
that can be recovered given default.
For consumer credit (when lending to individuals) there is not the
same ready explanation of default risk. In consumer credit
applications, factors such as home ownership (which may be
considered a proxy for asset values) and lifestyle (married, single,
children, length of time in job) may be seen as indicative of asset risk
and leverage. However, while these partly explain default probability,
there is no good underlying theoretical rationale. In the case of
consumer credit analysis, analysts use variables that have been good
indicators of credit risk in the past to predict future credit behaviour.

Credit Risk Management Edinburgh Business School 1/2

3
Module 1 / Introduction
1.3 Expected Losses and Unexpected Losses
A key issue in understanding credit risk over time is the concept of
expected losses and unexpected losses. The easiest way to
understand these concepts is to illustrate them. In Figure 1.5, the
loss history from credit defaults is shown over time. Using the
historical time series, we have the expected losses, which is simply
the average loss rate over time. That is, if we wanted to price for
future losses, we would know that there was an expected loss rate
(EL) to factor into our pricing. Typically, this is the average loss rate
determined from past data on losses. However, there is also volatility
in the loss rate. In other words, the loss experience varies over time.
This variation might be due to chance and/or the underlying
economic situation for the industry or for the economy as a whole, or
as a result of other factors unique to the group of credits being
analysed. Whatever the cause, this means that there is an element of
unexpected loss if the actual loss experience differs from that
predicted. This is known as unexpected losses. What Figure 1.5
shows is that, the wider the dispersion of unexpected losses, the
greater the degree of risk from unexpected credit events. Hence it is
not sufficient to know what the expected losses from credit events
are; it is also necessary to know the distribution of such losses over
time.

Loss rate

Unexpected loss (UL)

Anticipated volatility in loss rate

UL

EL

Expected loss (EL)

Anticipated average loss rate Distribution of
losses over time

Time

Figure 1.5 Expected losses and unexpected losses

1/2 Edinburgh Business School Credit Risk Management

4
 Module 1 / Introduction

Box 1.5: Recovery Rates

If a firm suffers a loss, what recovery can be
expected? At worst, there is a 100 per cent write-off
of the amount due. At best, everything is recovered.
Obtaining probabilities for default is difficult. Getting
good estimates of recovery rates is harder still. There
is no good publicly available source of such
information. Some data on recovery rates can be
obtained from traded securities. Historical estimates
of recovery rates suggest that there is an expected
loss regardless of seniority but that the median or
average amount recovered increases with priority.
Table 1.4 shows statistics for recovery rates on bonds
and loans, as quoted by Altman et al. (2003).

Table 1.4 Recovery at default of traded corporate

bonds and bank loans
Seniority Number Median Average Standar
of recovery recovery d
issues (per (per cent) deviatio
cent) n
Loans
Senior secured 155 73.0 68.5
Senior unsecured 28 80.5 55.0

Bonds
Senior secured 220 54.5 52.8
Senior unsecured 910 42.3 34.9
Senior subordinated 395 32.4 30.2
Subordinated 248 32.0 29.0
Discount 136 18.3 20.9
Total 1909 40.1 34.3
Notes to table: Data for bonds is from 1974 to 2003 and
bank loans from 1989 to Q3 2003. Recovery rates are, for
bonds, based on the difference in bond prices just after
default and, for loans, 30 days after default.

An important issue is that, once a default occurs,

although recovery can take place, there can be a
considerable time delay between when the obligation
was due and when a liquidator or bankruptcy court

Credit Risk Management Edinburgh Business School 1/2

5
Module 1 / Introduction

makes payment. Table 1.5 shows the ultimate

recovery rates on bank loan defaults discounted for
the time delay in repayments.
Table 1.5 suggests that even the best secured
creditor, namely senior bank debt, is likely to suffer
an average loss of 21.2 per cent. Given the high
standard deviation of the data, a creditor could be
confident of reclaiming only 28 per cent with 95 per
cent confidence.11
Table 1.5 Nominal and discounted recovery rates
on bank loans and bond obligations
Seniority Number Ultimat Ultimate Standar
of e discounte d
cases nominal d deviatio
recover recovery n
y (per cent)
(per
cent)
Senior bank debt 750 87.3 78.8
Senior secured notes 222 76.0 65.1
Senior unsecured notes 419 59.3 46.4
Senior subordinated 350 38.4 31.6
notes
Subordinated notes 343 34.8 29.4
Notes to table: Data is for 1988 to Q3 2003. The recoveries
are discounted at each instrument’s pre-default interest
rate.
1.3.1 Catastrophic Losses
For solvency purposes a firm extending credit needs to know not just
about expected losses and unexpected losses but also the potential
for catastrophic losses. In Figure 1.6 we have two business
activities that service different markets that, in turn, have different
loss characteristics. Business 1 has a low level of expected losses but
a long tail of large unexpected losses. Business 2, in contrast, has a
higher level of expected losses but a lower element of unexpected
losses. Modelling the loss distribution indicates that, at the 99
percentile level, Business 2 suffers less loss than Business 1. 12
Therefore, although expected losses are higher, the variability in

11 That said, the distribution is skewed since the series is bounded with the range between the
minimum and maximum recovery being 100 per cent.
12 The 99 percentile level means that all but 1 in 100 losses will be less than this
amount. The remaining 1 per cent will be higher than this.

1/2 Edinburgh Business School Credit Risk Management

6
 Module 1 / Introduction
losses for Business 2 is lower. As such, since expected losses are
anticipated, whereas unexpected losses are not, Business 2 has lower
credit risk than Business 1.
Note that there is also a corollary to the above, namely the
existence of catastrophic losses. These are long-tail losses. For
Business 1, while the expected losses are low, there is a relatively
high incidence of significant long-tail losses above the 99 percentile
level, due to the skewed nature of the loss distribution. These are
known as catastrophic losses because of their potentially serious
impact on the business. Business 1 has a much greater exposure to
such exceptional catastrophic losses than Business 2, despite the
latter’s higher expected losses. When losses are expected, they can
be priced into any transaction bearing credit risk. Unexpected losses,
however, require a different approach. Therefore, it is necessary for
credit management purposes for the analyst to understand the full
distribution of credit loss in order to effectively manage the inherent
credit risk.
The issue of catastrophic losses arises where there is a
concentration of credit in a particular industry, to a particular
country and/or to a particular type of debtor. This is because such
exposure is likely to have a common underlying risk factor.
That is, loans across the portfolio are highly correlated and will tend
to behave in the same manner. For instance, firms lending to
Argentina are exposed to common factors that affect less-developed
countries (LDCs) and Latin America, and to internal economic and
political developments in the country itself.

Credit Risk Management Edinburgh Business School 1/2

7
Module 1 / Introduction
Probability

Distribution of losses for Business 1

Distribution of losses for Business 2

Expected loss Unexpected loss Catastrophic

Business 1 Business 1 loss

Expected loss Unexpected loss Catastrophic

Business 2 Business 2 loss

99 percentile 99 percentile
Business 1 Business 2
Business 2 Business 1

Aggregate losses in a given time period

Figure 1.6 Loss distribution showing expected losses, unexpected

losses and catastrophic losses

1.3.2 Credit Evaluation and the Business Cycle

Credit losses may be cyclical in the sense that credit events are more
likely in periods of economic downturn. Hence it is important to
understand what is being analysed and how the resultant decision
can be affected by the business cycle. Good credit anticipates
changes in economic conditions.
It may be argued that the underlying aim of credit management
should be to control the level of credit risk. In this model, the credit
manager’s key activity is to identify high-risk elements. For many
firms making decisions about credit, an important issue is how the
firm decides, during periods of above-normal profitability within an
economic expansion (that is, when actual loss experience is below
expected losses), whether a particular counterparty would be a high
credit risk if its business was in decline or under adverse economic
conditions. On the other hand, if a business is still profitable in
recession, the chances of its survival and continued prosperity appear
good.
The above is not a trivial point; a key contention raised by the Basel
Committee on Banking Supervision (2000) was that a major cause of

1/2 Edinburgh Business School Credit Risk Management

8
 Module 1 / Introduction
serious banking problems is ‘directly related to lax credit standards
for borrowers and counterparties, poor portfolio management, or a
lack of attention to changes in economic or other circumstances that
can lead to a deterioration in the credit standing of a bank’s
counterparties’.

1.3.3 Relevance and Timeliness

In order that market participants receive up-to-date information
concerning the prospects of a business or company, information must
be released at regular intervals. Transforming this information into
the credit assessment process is a major element of maintaining
credit quality. This is particularly the case with credit exposures that
arise during the normal course of business. Given that the financial
status of credits can change over time, it is important to use up-to-
date information and to review new information on a timely basis.
In terms of the management of credit exposures, this necessitates a
reporting process that updates exposures when new information is
available and that monitors developments over time. Typically, for
matters such as trade credit, this takes place every month, or even
more frequently. For longer-dated exposures such as the loans banks
make to customers, the review process is likely to be once a year.

1.3.4 Comparability
Comparability of credit risks between individual cases is desirable
from a processing perspective and to facilitate cross-credit analysis.
However, different assessors often adopt different terminology, use
different criteria for ranking firms and individuals, and define non-
performance in different ways. In addition, companies that are
incorporated in different jurisdictions are subject to differing national
accounting practices. For example, the generally accepted
accounting practice in the UK (UK GAAP) requires goodwill to be
written off, whereas under US GAAP goodwill can be amortised over
the useful life of the asset. This means that the financial statements
of firms in the two countries are not fully comparable.13

13 The movement to a common reporting standard, known as International Financial

Reporting Standards, or IFRS, should help reduce the problem, though there may
continue to be some variations in the way different jurisdictions convert these to
generally accepted accounting standards, in areas where the new IFRS rules provide
a degree of latitude.

Credit Risk Management Edinburgh Business School 1/2

9
Module 1 / Introduction
1.4 Controlling Credit Risk
A key issue for credit risk management is controlling risk. The
traditional approach to managing credit risk, as discussed earlier, is
to evaluate the risk by assessing the borrower’s ability to repay. This
involves examining historical financial statements of the counterparty
and projecting future ability to pay. The core issue is whether the
credit officer in the firm believes the counterparty can reasonably
expect to meet their repayments, in full and on time. The credit risk
can be controlled in various ways, as discussed below.
As such, credit risk differs from other types of risk being managed
in that, essentially, it is an event risk. An obligor will either be good
or will be in default on its obligations. As mentioned earlier, there are
variations on this: in some cases the credit risk management decision
also includes the possibility of credit downgrades. For example, a
bond fund may only be eligible to invest in investment grade
securities. Any bonds that fall below this category would have to be
sold and proceeds reinvested in eligible bonds. So such a fund is not
just subject to default risk but is also exposed to changes in the credit
standing of its investments. Any factors that can influence the value
of the counterparty and affect the value of the transaction go by the
generic name of credit triggers or credit events. Box 1.1 gave a
list of such triggers identified by the credit derivatives market (credit
derivatives are contracts used in financial markets for transferring
credit risk from one party to another). The five most common are
given in Box 1.6.

Box 1.6: The Five Most Common Credit Events

1. Failure to meet payment obligations when due (after any grace
period and above any payment requirement specified at the
contract’s inception)
2. Bankruptcy (or moratorium for sovereign entities)
3. Repudiation
4. Material adverse debt restructuring
5. Obligation acceleration or obligation default

1.4.1 Managing the Loss Given Default

Managing the credit risk means managing the amount of loss if a
default should take place, known as the loss given default (LGD).
There are a number of different ways in which firms can do this. One
method mentioned earlier is through the use of collateral and
collateralisation. With this approach, the party taking credit
provides a security against default. In the event of default, the lender

1/3 Edinburgh Business School Credit Risk Management

0
 Module 1 / Introduction
has a claim on a specific asset of the borrower. The benefit (for the
lender) from using collateral is that, in the event of default, the
lender is spared the expense of the default process and dealing with
the borrower’s other creditors, assuming that the collateral pledged
sufficiently covers the loan amount.
For example, it is a requirement in futures markets for all market
participants to provide margin (also, interestingly enough, sometimes
called a performance bond) when entering into transactions. This is a
deposit that covers potential future losses and hence means that
there is no performance risk to other parties in the futures markets. 14
As well as collateral, a lender should be interested in the total level of a
borrower’s capital. The reason being that an adequate supply of capital
provides a cushion against bankruptcy or insolvency, should the credit
experience temporary losses. Where a particular credit is unacceptable due to
its high risk of non-performance, it is possible in some cases to obtain a credit
guarantee. This is third-party insurance that, in effect, underwrites the risk of
default. Such guarantees are often requested when borrowing is being
undertaken by a subsidiary of a company that is thinly capitalised (that is, the
subsidiary is mostly financed via debt) and the parent company is then asked to
guarantee the transaction.
Another approach, common in derivatives transactions between major banks
and other financial institutions, is to reduce the exposure via netting
agreements. In such an agreement only the net proceeds rather than gross
proceeds are due to the other party. So, for instance, in a swap contract, one
party is due to pay a sum to the other, conditional on receiving in return the
contracted payment by the other party. With a netting agreement only the
difference is paid across. Hence where there are a large number of positions
between the two parties (this may be because both are active market makers)
only the net sum of the long and short positions in the swap contracts with the
counterparty is paid. Note that some of these contracts will have a positive
value (an asset) while others will have a negative value (a liability). By using
netting agreements, asset positions (which carry credit risk on the other party)
can be offset against liability positions (where the position is reversed).
Since 1993, when the first credit derivative contract was transacted, there
has been a growing market in off-balance-sheet transactions that transfer the
credit risk from one party (the protection buyer) to another (the protection

14 This is an overstatement since there is the possibility that the margin is insufficient
to cover future losses. However, to take account of this potential additional risk, a
futures exchange and its clearing house (which undertakes to act as the
counterparty to the futures transactions) will both be well capitalised and counter-
indemnified against losses by the brokerage firms that trade the market through an
insurance fund. To date, these credit-reducing institutional structures have stood the
test of a number of major market crises (such as the October 1987 market crash)
without failing, and hence, as

Credit Risk Management Edinburgh Business School 1/3

1
Module 1 / Introduction
seller) so that all the credit risk of the particular reference asset is transferred
to the protection seller.*

Box 1.7: The Seven Stages of Credit Risk

Stage 1: We only make good loans
Credit approval, monitoring and pricing decisions are decentralised and
judgemental. Good loans are accepted; bad ones rejected.
Stage 2: Loans should be graded
The relative riskiness of loans is formally recognised, with three to four grades
for good loans and the same for bad. But, due to grade definitions, most good
loans fall into one category.
Stage 3: Return on equity is the name of the game
Business unit managers receive bonuses linked to their unit’s return on equity
(ROE) performance. But measurement techniques lack the appropriate
adjustments for credit risk and this can lead to managers originating many
high-yielding (and high-risk) assets too cheaply.

a result, market participants consider there to be little or no credit risk from trading futures.
Note that, if required, the exchange also has the right to request additional margin from market
participants (for instance, during a period of unexpected market instability) in order to
strengthen the degree of protection.
*
For a full discussion of the economic function of such off-balance-sheet instruments see the elective
course Derivatives.
Stage 4: We need to price for risk
Key risk measurement advances allow successful implementation of
the ROE culture. These include expanding the loan grading scale to
ten levels, each explicitly calibrated to an expected loss level, and
introducing different risk adjustments into the customer, product and
business line profitability measurement systems, based on
unexpected losses.
Stage 5: Manage the loan book like an investment portfolio
Modern portfolio theory is applied to the management of the loan
book. A portfolio manager and statisticians are appointed. But this
can lead to conflict between customer-focused functions and
portfolio managers, while initial results may be disappointing if
model inputs are inaccurate (as they are likely to be).
Stage 6: Our shareholders demand risk/return efficiency
Advances – including better risk discrimination (say, 15–20 grades),
appropriate default correlation measurement, and implementation of
techniques to quantify unexpected loss contributions – allow the
setting of limits on exposures and volatility, target weights for
sectors and expected asset returns.

1/3 Edinburgh Business School Credit Risk Management

2
 Module 1 / Introduction
Stage 7: Diversification is paramount
Portfolio management realises that diversification is paramount to
achieving risk/return efficiency. This may lead to conflict with
customer-centred functions, which benefit from larger transactions
and specialisation. The conflict can be resolved by a formal
separation of portfolio management and origination.

1.4.2 Diversification
Diversification is a fundamental ingredient in financial risk
management since it allows the credit grantor to spread its risk.
Although one must accept that all commercial and consumer loans
exhibit some degree of credit risk, diversification can significantly
reduce the risk. If such risks are uncorrelated, a portfolio of just 10
separate credits of 10 000 versus one credit of 1 000 000 will have a
significantly lower risk of loss. If the probability of a loss is 0.2, there
is no recovery given loss (to simplify the analysis), and the correlation
between loss events in the portfolio is zero, then the expected value
from having just one credit exposure will be as shown in Table 1.6.
Table 1.6 Expected value with one credit exposure
Probability Payoff Expected value
0.2 0 0
0.8 1 000 000 800 000
800 000
A spreadsheet version of this table is available on the EBS course website.
The standard deviation (σ) of the loss will be:

0.2 0 800 000 0.8 1 000 000 800 000 400 000
﴾1.4﴿

If we now have two credit exposures for the same amount, the expected value
remains the same, as shown in Table 1.7. Now there are four possible states of the
world: both credits default with probability of 0.2, but using probability arithmetic
this joint probability is 0.2 × 0.2 = 0.04; there are two states of the world where
one credit defaults and the other is good (0.2 × 0.8) × 2 = 0.32; and one state of
the world where both credits are good (0.8 × 0.8). In this result, the expected
value remains unchanged at 800 000, but the standard deviation of the loss will
now be lower, as shown in Equation 1.4.

Table 1.7 Expected value with two credit exposures

Probability Payoff Expected value
0.04 0 0

Credit Risk Management Edinburgh Business School 1/3

3
Module 1 / Introduction
0.32 500 000 160 000 0.64 1 000 000 640 000
800 000
A spreadsheet version of this table is available on the EBS course website.

The standard deviation (σ) of the loss with two credits that are not correlated
will be:

0.04 0 800 000 0.32 500 000 800 000 ﴾1.5﴿

0.64 1 000 000 800 000
282 843
As we add more credits, the expected value remains unchanged, but at the same
time the standard deviation of the expected value is reduced. The effect on the
standard deviation from adding more credits to the portfolio is shown in Table 1.8.

Table 1.8 Effect of diversification on the distribution of losses

Number of credits in the Standard deviation
portfolio of the portfolio
1 400 000
2 282 843
3 230 940
10 126 500
100 40 000

Simply put, the greater the number of credits in such a portfolio, the closer the
actual loss experience will be to the expected loss experience. Diversification
reduces the loss volatility of the portfolio (i.e. it reduces the extent of unexpected
losses).
In order for diversification to be effective, it should include
industrial, geographical and international factors. For example, there
is evidence that, in past decades, banks made inadequate attempts to
ensure their portfolios were well-diversified and such loan books had
excessive concentration into credits that had common underlying risk
factors; hence the loss experience was more volatile than
diversification would have led the credit manager to expect. 15 For
instance, banks had undiversified portfolios of loans as a result of
making advances to different customers dependent on the same
source of income. This would be the case in a farming community
where loans were made to farmers and also to farm equipment
suppliers. A domino effect could be triggered if farmers experienced

15 Note that concentration in a portfolio is also colloquially known as a hot spot.

Reducing the credit risk in such ill-diversified portfolios is one of the rationales for
the development of credit derivatives.

1/3 Edinburgh Business School Credit Risk Management

4
 Module 1 / Introduction
hard times, affecting their ability to pay their suppliers and, in turn,
affecting the suppliers’ ability to meet repayments to the bank.

1.4.3 Covenants and Monitoring

A covenant is a restriction on, or requirement imposed on, the
borrower, and agreed to under the terms of a contract. An example
would be where a corporation borrows using a guarantee and the
guarantor, to improve the financial stability of the borrower, requires
them to conform to set conditions. These conditions, or covenants, as
they are known, might include a minimum level of liquid assets or
working capital, limits on dividends, or conditions requiring the
borrower corporation to maintain the quality of assets pledged as
collateral in the facility.16
It is common for specific covenants of the kind indicated above to
be included in loan agreements to maintain or protect the capital
level of the firm. These covenants are of three types: affirmative
covenants, which require the contracted party to undertake certain
actions (such as maintaining the value of the collateral); negative
covenants, which restrict the actions of the contracted party (for
instance, limiting the payment of dividends during the contract
period); and information covenants, which require the contracted
party to provide agreed information to the party at risk in the
contract.
Covenants allow the lender or party at risk from credit events to
take action. Ensuring that the contracted party is performing to
expectations requires monitoring. This oversight aims to identify
problems as and when they occur and to trigger the appropriate
recovery process. A breach of covenant allows the at-risk party to
take action, for example by allowing for acceleration of repayment,
declaration of insolvency or the liquidation of collateral.
Consequently controlling credit risk requires the credit manager to
actively monitor the financial condition of counterparties and be
prepared to act, if conditions deteriorate. If a credit does get into
difficulties (known as financial distress), then such problem
accounts can be managed to maximise the amount that can be
recovered. In some cases, selling the amount due to a third-party
specialist recovery firm removes the problem, although this is likely
to be at a substantial discount to its contracted value.

16 Note that a lot of trade credit has no covenant protection being granted on ‘open
account’ credit lines. Such trade credit is based on the standard contract used in
commercial law and is only secured on the general willingness of the counterparty to
repay. It ranks pari passu with all other unsecured credit in any winding-up of the
company. Secured creditors (with access to collateral) would get priority.

Credit Risk Management Edinburgh Business School 1/3

5
Module 1 / Introduction
By properly monitoring exposures, problems can be detected at an
earlier, less critical stage, when there are still options open to solve
the problem without incurring substantial losses. Good credit control
uses a range of signals to try to identify such credits prior to the
critical stage. Although firms differ in the exact way they would
handle such a credit problem, all firms should have clearly laid-out
plans for dealing with this kind of situation.

1.4.4 Risk Pricing

As mentioned earlier, correctly pricing the credit risk to take into
account expected losses is important. The type of analysis discussed
in Section 1.1.4 needs to be applied to situations involving credit risk.
For instance, banks must ensure that the loan rate (also known as
the price of the loan) exceeds a risk-adjusted rate to compensate for
expected losses and is inclusive of administration fees, establishment
costs, the bank’s desired profit and other costs of being in business.
Therefore, under this model the loan rate would comprise the market
rate of interest for the currency being borrowed, administrative
costs, plus a risk premium to compensate for expected default. The
less creditworthy a potential borrower is considered to be, the higher
the risk premium.17

1.4.5 Loan Loss Provisioning

A common practice among financial institutions engaged in lending is
to provision against expected losses. The provision of loan losses
reserves is a mechanism used by such lenders to recognise in a
timely fashion impending losses on troubled loans. The fact that a
certain proportion of credits will default is acknowledged and
accepted by financial institutions. In the same way, an industrial and
commercial corporation would have a reserve for expected bad debts.
Further, on occasions where changes in the business cycle or local
factors have an adverse effect on the loan book or default experience,
such reserves or provisions can be used to mitigate the consequences
on the lender. For example, a downturn in a local economy may lead
to defaults in real estate loans. These tend to occur in clusters due to
the concentration effect of such loans. As a result, it may cause the
lender financial distress to write off a considerable loss within one

17 Note that, for a variety of reasons, in practice there is likely to be a loan rate above
which a bank will not lend and hence a minimum credit quality that is acceptable.
Acceptable credit quality borrowers are known as prime. Credit quality below
acceptable lending quality is called sub-prime.

1/3 Edinburgh Business School Credit Risk Management

6
 Module 1 / Introduction
reporting period; hence the reason for the loan loss provisions. These
protect the lender’s earnings stream and smooth over the losses.18

Box 1.8: BIS Survey on How Bank Rating Systems

Differ
Mixture of risk factors
Banks appear to consider similar types of risk factors when
assigning a credit rating, but their relative importance and the
mix of quantitative and qualitative considerations differ
between banks, and even between different borrower types
within the same bank.
Judgement versus modelling
Banks use different approaches to assign internal ratings. At
one extreme are systems focused on the judgement of expert
personnel, and at the other those based solely on statistical
models. Each will probably require a different approach to
supervisory review and validation.
Borrower versus transaction type
The vast majority of banks surveyed assign ratings based on
an assessment of the borrower. Approximately half also
consider the risk contributed by the specific characteristics of
the transaction.
Use of information
The information gleaned from ratings is used in broadly similar
processes at the banks surveyed, including management
reporting, pricing and limit setting.
Quantifying loss
Data sources and techniques for quantifying loss
characteristics per grade (default probability, loss given
default, etc.) differ between banks. In addition, in those banks
surveyed, they appear to have greater difficulty in attributing
loss given default estimates to their exposures than in
assessing default probability.
Differing definitions
Data providers and banks use differing definitions of ‘default’
and ‘loss’ in assigning ratings and quantifying loss
characteristics, which represent a source of inconsistency
and/or measurement error that will need to be considered in
the credit risk management process.

18 It is a distinctive feature of financial reporting that firms can create provisions and
set these against earnings and then release these provisions at a later date to be
included in the reported earnings for that period.

Credit Risk Management Edinburgh Business School 1/3

7
Module 1 / Introduction

Data availability
Data availability remains a challenge to banks’ efforts to
quantify risk, although some banks are making progress in
collecting and analysing internal data for certain market
segments covering the past few years.
Source: Basel Committee on Banking Supervision (2000). Range of
Practice in Banks’ Internal Rating Systems. Bank for International
Settlements: Basel.
1.5 The Credit Policy Manual
Turning a credit evaluation process into a functional activity requires
a detailed set of guidelines, procedures and processes. This goes by
the generic name of a credit policy manual. Every firm that has a
credit department or a credit manager needs to have a credit policy
manual to formalise its decision processes regarding the dayto-day
management of credit decisions and the resultant collection
challenges when accounts are slow to pay, or fail to pay amounts
when due. The thinking behind creating a manual, which is really a
formalisation of credit risk management procedures, is to be able to
recognise and detail policy on important credit risk management
issues, and to ensure consistent thinking and action on these issues
by people engaged in credit risk management.
As a document that formalises the management of credit risk, a
credit policy manual should provide decision rules and guidelines on
important aspects of the credit-granting process being performed
within the credit department and as discussed in earlier sections of
this module.
It is important to remember that, by formalising the credit risk
management process in a set of procedures, this will affect other
elements of a firm’s operations, such as marketing and sales, the
buying department and corporate treasury. Consequently, it will be a
combined document based on agreed policies from the firm’s senior
management, sales and the other affected departments.
For the most part, credit policies will not change very often.
However, as a matter of good practice, firms should review the
manual annually, including in the review the views from senior
management and affected departments, as mentioned above, to
ensure that the procedures it details are up to date and reflect
current thinking and practice.
No two companies will have the same set of credit risk
management policies; however, the following list of components
represents a typical set of policies that most firms would adopt to
manage their credit exposures.

1/3 Edinburgh Business School Credit Risk Management

8
 Module 1 / Introduction
1.5.1 Credit Management Mission Statement
The mission statement will provide a summary of the overall
objectives of the credit risk management process detailed within the
manual. For example: ‘The credit function’s mission is to help sell the
firm’s products and services to customers using best practice credit
risk assessment and collection procedures and services.’
The mission statement might include a statement about credit
philosophy. For instance:

The firm develops, markets and sells products and services within the
home entertainment field. These products are generally high-margin
and short shelflife items. Because of the product and the nature of the
industry, company management has maintained a credit philosophy
of being liberal on sales and conservative on collections. This means
that the company is willing to accept a larger degree of risk in order
to make our product available to a wider audience.
1.5.2 Goals of Credit Management
In establishing the mission statement, the credit policy manual will
set out the goals of the credit management process. This can be done
by either listing specific goals or by making the goals more general in
nature. These goals will be based upon many factors, including the
company’s credit philosophy (that is, its attitude to assuming credit
risk by offering credit sales to customers). It will also be a factor in
relation to sales targets and financial performance. Other factors,
over which the firm may not have control, include competition in its
markets and business conditions.
As part of the goals, there may be specific, observable objectives,
such as setting a maximum number of days outstanding for credit
items, collection systems, and the frequency of bad and doubtful
debts. Note that such objectives will dictate the type and extent of
credit risk being taken by the firm. For instance, if the company is
seeking a low level of bad debts, it will most likely have to confine its
credit to highquality firms with low-default probability. This will
impact on its ability to offer credit and hence grow sales with lesser
creditworthy customers. Attitudes to credit risk will also dictate the
frequency and scope of credit reviews of existing and prospective
customers.

1.5.3 Responsibilities
As part of developing a sound process of credit granting, the various
responsibilities and limitations on discretion of staff involved in credit
risk management and credit granting have to be clearly laid out. As
with many other types of operation, there is a potential conflict of

Credit Risk Management Edinburgh Business School 1/3

9
Module 1 / Introduction
interest between those individuals who are rewarded by creating new
business and those who act to control risks. Hence proper lines of
responsibility and awareness of these issues should lead to the
separation of the credit evaluation and credit-granting functions.
Also, who finally signs off on a credit decision and who has the
authority to override the normal criteria for acceptance and rejection
needs to be thought through. Since refusing a credit application has
important repercussions on customers and staff alike, it is important
that all parties within the firm understand clearly who has
responsibility and power to authorise. This is even more important in
the case where an existing customer has to have their credit line
withdrawn.

1.5.4 Credit Management Policies

The credit policy manual will provide a variety of policies and
processes that govern the credit function, including credit terms, the
processes required for opening new accounts, processing
applications, methods and techniques for credit investigation, the
creation and dissemination of credit reports, setting lines of credit,
and other factors that are involved in the credit management process.
It should also include monitoring of accounts to ensure that they
are compliant with credit terms and include issues such as collection
policy and procedures and, for financial institutions, dealing with
events of default. As with granting credit, the processes and
approach that will be adopted in cases of delinquency need to be
mapped out. For instance, in many banks, once an account becomes
delinquent it is transferred from the relationship manager to a
specialised unit whose function is to maximise the amount that can
be recovered from the credit. It should also be clear who within the
organisation has the authority to initiate enforcement actions.
There will be criteria for transferring accounts between categories,
namely good accounts, accounts that may be doubtful and defaulted
accounts. The process for provisioning and writing off debts should
also be detailed. For instance, the policy might be that non-paying
accounts can be written off to bad debt only after the customer has
filed for bankruptcy, gone out of business or been placed with a
collection agency or collection attorney and no payments have been
received for a period of six months.19

19 Note that for many financial institutions, such as banks, that are subject to
regulation there may be externally imposed criteria as to how to categorise bad and
doubtful debts. For instance, the banking industry has a 90-day rule on overdue
interest for determining whether a loan should be considered impaired and hence
falls into the doubtful or bad debt category.

1/4 Edinburgh Business School Credit Risk Management

0
 Module 1 / Introduction
1.5.5 Additional Policies
Additional policies can be identified for issues such as compliance,
regulation and the law regarding lending and credit. There should be
criteria laid down for the exchange of credit information in order to
obtain bank and trade experience, record keeping, credit
organisations, customer visitation, travel, interaction with other
departments, international credit, security and other costs of
administering the credit department.

Box 1.9: Credit Implications of the Enron Collapse

The opening quotation of this module highlighted the fact that, even
though Enron appeared sound and a good credit, it went bankrupt.
Enron was a highprofile credit default partly due to its rapid change
from being an acceptable counterparty to being in default, and
partly due to the reasons that lay behind the firm’s collapse. What
impact did its demise have on its trading partners? Because of the
huge amount of publicity surrounding the event, a lot of firms went
public with their exposures and potential losses (before any
recoveries) from the default. The default’s impact on energy trading
(that is, trading in oil, gas, coal and electricity, in which Enron was a
major participant) could have seriously affected the market.
The International Swaps and Derivatives Association (ISDA), the off-
balancesheet derivatives trade association that establishes the legal
relationships between counterparties in financial derivatives, and of
which Enron was a board member, indicated that it considered itself
to be confident that the effectiveness of its documentation,
combined with the fact that energy markets were deep and liquid,
would be sufficient for the market to continue post Enron.
US energy-trading firms at the time of Enron’s default participated in
a global market that was estimated to have a value of US$60 trillion.
To facilitate trading and to minimise post-contractual disputes, the
majority of transac-

Credit Risk Management Edinburgh Business School 1/4

1
Module 1 / Introduction

tions in the market derived their contractual documentation

from Master Agreements, drawn up by the ISDA.
As one industry participant put it, commenting on Enron’s
condition and the likely impact on other parties, ‘The benefits
of netting are once again being appreciated.’
Of the energy companies that traded with Enron prior to its
financial unravelling, the following firms reported on their
counterparty exposures.
El Paso Corporation indicated its maximum natural gas and
power net trading exposure to Enron was approximately
US$50 million. The company further added that it did not
expect any adverse earnings impact from Enron’s difficulties.
Mirant announced that its current pre-tax exposure to Enron
was approximately US$50 million to US$60 million. It also
indicated that it had begun limiting its exposure risk early on
in the Enron crisis.
Exelon Power Team’s direct net exposure to Enron, based on
the current book of business and existing market prices, was
less than US$10 million and the direct gross exposure (i.e. for
current energy sales from Exelon to Enron) was less than
US$20 million.
St. Mary Land & Exploration had oil and gas production hedges
coming due the current financial year to the magnitude of
US$420 000. On a mark-tomarket basis (that is, on a
replacement cost basis), its 2002 undiscounted hedges on
which Enron was liable to pay amounted to US$3.1 million and
to US$650 000 in 2003.
Tractebel SA, the energy arm of Suez SA of France and
Electrabel, Tractebel’s European subsidiary, reported that they
had anticipated Enron’s situation and substantially reduced
their exposure early on, resulting in a negligible Electrabel
exposure and maximally covered positions for Tractebel, which
operates mainly in the US.
RWE AG, Europe’s fourth-biggest electricity company, said its
open trading positions with Enron Corp. amounted to €8.9
million but had been much higher in the past.
What is evident is that trading firms in the energy market,
when concerned about the creditworthiness of Enron as
counterparty, reduced their exposures. In addition, because of
netting agreements in place, such firms were only exposed to
the net cash flows across all their transactions with the
company. Such limited exposures by these firms indicated that
they had considered the key issues in credit risk management

1/4 Edinburgh Business School Credit Risk Management

2
 Module 1 / Introduction

and had adopted these in their trading procedures. As a result,

these firms were able to escape the worst consequences that
resulted from Enron’s failure.
Learning Summary
This module has introduced the key concepts for managing credit
risk. Credit risk arises from changes in the financial solvency of firms
and individuals. An event of default occurs when the obligor fails to
perform under the terms of the contract. In this case, the lender or
party with the credit is exposed to a potential or actual loss. The
degree of loss will depend on how much can be recovered given the
credit event or default.
Many factors affect the potential exposure to credit events and
hence creditrelated losses. The key element in determining the
acceptability of risk taking in regard to credit exposures is in
assessing the probability of default. This involves analysing and
assessing counterparties based on a variety of techniques. Even so,
there is the potential for exposure to unexpected and – at times –
catastrophic losses from credit events. For this reason, firms need to
control these credit risks through setting out policies on evaluation,
management and having the correct procedures in place.

Introduction
• Credit risk is the risk of loss from exposure to firms that undergo
credit events. This might be that the obligor defaults, but in some
cases it is that adverse changes in credit quality can lead to losses.
There are a great many events that can have a credit impact,
which complicates the definition, analysis and management of the
process.
• Credit risk can be seen as an informational problem. The credit
giver does not know enough about the quality of the credit taker
and how the obligor will perform in the future.
• As a task, credit risk management involves identifying the source
of risk, selecting the appropriate evaluation method or methods
and managing the process. This will mean setting an appropriate
cut-off point that balances the conflicting demands of the
organisation with regard to credit exposure.
• Credit risk management can be seen as a decision problem. The
assessment involves determining the benefit of risk taking versus
the potential loss.
• Decisions about extending credit are complex and subject to
change, but at the same time are critical elements of risk control
within most organisations.

Credit Risk Management Edinburgh Business School 1/4

3
Module 1 / Introduction
• While it is easy to outline the credit analysis decision,
implementing an effective approach is more complicated. At its
simplest, it requires an assessment of the likelihood that a
particular counterparty will default on a contract and of the loss
given default (LGD).
• As a process, credit decisions usually involve some classification of
creditworthiness into categories or classes as a precaution against
credit exposure to high-risk counterparties. This allows new
credits to be analysed by comparison to preclassified credits whose
default history is known.
Credit Assessment Methods
• Credit appraisal can involve a number of techniques that can be used
individually but are more often combined as part of the assessment process.
These techniques can be categorised as either qualitative or quantitative in
approach.
• There are basically three separate methodologies: judgement, deterministic
models based on past experience or knowledge of the risks, and statistical
models that may be either static or dynamic, or involve monitoring behaviour
over time.
• Two basic methods exist for analysing credit quality: traditional quantitative–
qualitative credit analysis and decision models based on deterministic or
statistical processes. Each offers a different insight into the credit risk problem.

Expected Losses and Unexpected Losses

• In many cases, as with financial institutions, the amount of credit given by an
organisation is substantial and requires steps to control the exposure in order
to prevent unanticipated losses emerging.
• Unanticipated losses arise due to the variability of loss rates experienced over
time, for instance as a result of changes in business conditions. If the loss
experience in practice is above that expected, organisations will experience
unexpected losses over and above those anticipated. This will happen as a
result of variability in the actual loss rate against the expected loss rate.
• In some cases losses may be catastrophic, in that they far exceed any
reasonable degree of variation that historical loss experience would indicate.
Such losses can have a disproportionate effect on the organisations subject to
such a risk.

Controlling Credit Risk

• The credit analyst or manager is required to understand the ways in which bad
debts or credit losses arise and to devise methods for identifying these. This
then requires that due consideration is given to how these are effectively
managed.

1/4 Edinburgh Business School Credit Risk Management

4
 Module 1 / Introduction
• A key issue is credit control, which involves constantly managing the
creditgranting process. This can be seen as a policy that includes procedures,
guidelines and processes for managing the credit process.
• Diversification can play an important role in reducing exposure to unexpected
and catastrophic losses. However, spreading risks will be effective only if the
principles of efficient portfolio construction are followed. There is a danger that
the portfolio is ill-diversified, leading to unexpected losses.
• As with all risk management processes, the exposure to credit risks has to be
kept under constant review and action taken as required. Credit risk
management is a dynamic process that responds to new information.
• Finding the links between a firm’s financial condition, behaviour and default is
the key skill required in the management of credit or counterparty risk.
The Credit Policy Manual
 This process of credit risk management is formalised in most
organisations in a set of procedures generally called a credit policy
manual.

Review Questions
Multiple Choice Questions
1.1 Which of the following is correct? Credit quality is:
A. the thoroughness of the assessment made on a particular counterparty.
B. the financial standing and solidity of a particular counterparty.

C. the reputation of a credit assessment firm such as Moody’s Investors

Service or Standard & Poor’s.
D. the amount of cash on a firm’s balance sheet.

1.2 The risk from default is made up of:

I. the probability that the firm will default.
II. the amount or value that is involved.
III. the amount or value that can be recovered.
IV. the nature of the contract entered into.
Which of the following is correct?
A. I and II. B. I and III.
C. I, III and IV. D. All of them.

1.3 Which of the following is correct? When analysing credit quality, the major
danger from the assessor’s perspective is:
A. analysing poor-quality firms as being good quality. B. analysing
good-quality firms as being poor quality. C. analysing poor-quality firms
as being poor quality.

Credit Risk Management Edinburgh Business School 1/4

5
Module 1 / Introduction
D. None of the above.

1.4 Which of the following is correct? When a credit assessor refers to the credit’s
character, this means:
A. the reputation of the credit.
B. the credit’s willingness to honour the obligation. C. the credit’s
past activities that demonstrate good qualities.
D. All of A, B and C.
1.5 Which of the following is correct? Collateral is that element of a credit assessment that
deals with:
A. the security available when credit is extended.
B. the knock-on effects that credit problems have for the lender. C. the increased
rate applied to a loan to reflect credit quality.
D. None of the above.

1.6 What is meant by ‘concentration risk’ in the context of credit risk management?
A. The risk that a large number of counterparties default at the same time.
B. The risk that a large number of counterparties share common risk characteris- tics.
C. There is a strong positive correlation in the historical behaviour of credit-
sensitive assets in a portfolio.
D. All of A, B and C.

1.7 A company has a mark-up of 12 per cent, and a unit costs 80 to produce. At which of
the following probabilities of default will the company be indifferent between accepting
and rejecting a sale on credit grounds?
A. 0.10 B. 0.13
C. 0.15
D. The question cannot be answered from the information provided.

1.8 What is meant by ‘settlement risk’ in the context of credit risk management?
A. It is the uncertainty about the amount recovered in any settlement of a bad debt
by a lender.
B. It is the risk that transactions processed through third parties fail to be
completed on time.
C. It is the uncertainty surrounding what is the true expected loss for a particular type
or class of exposure.
D. It is the tendency of a firm in a particular credit class to change class over time.

1.9 What factors in Enron’s bankruptcy led energy-trading firms, which traded with the
company, to avoid major losses?
A. Their use of a credit policy manual.
B. The existence of netting agreements on two-way transactions with Enron
minimised their exposure.
C. Inside knowledge of developments at Enron prior to its filing for bankruptcy.
D. All of A, B and C.

1/4 Edinburgh Business School Credit Risk Management

6
 Module 1 / Introduction
1.10 If one accepts a ____ as a ____, the result will be ____ than expected ____.
Which of the following is correct?
A. poor credit good credit higher losses B. poor credit poor credit higher
profits C. good credit good credit lower profits
D. poor credit poor credit higher losses
1.11 What is meant by ‘investment grade’ in the context of Standard & Poor’s credit rating
classification system?
A. It refers to all obligors given an A rating or above. B. It refers to all obligors
given a BBB rating or above. C. It refers to all obligors given a B rating or above.

D. It is a meaningless term within Standard & Poor’s credit rating classification

system.

1.12 In Table 1.1, a double-B credit has a one-year default probability of 1.08 per cent and a
two-year cumulative probability of default of 3.48 per cent. What is the year 2
marginal probability of default (that is, the default probability for one year in year 2)?
A. 1.08 per cent B. 2.40 per cent C. 3.48 per cent
D. 3.52 per cent

1.13 Why does one observe that the likelihood of credit migration of an obligor in a given
credit class, such as that given by Standard & Poor’s, diminishes as the credit rating
falls?
A. This is false; there are no differences across ratings. B. It is simply a spurious
characteristic of the sampling method.
C. This is true; it is due to the fact that lower credit grades have a higher probabil- ity
of default.
D. It is simply a result of having a fixed scale where top-quality credits can only
deteriorate in quality.

1.14 What is the difference between statistical models and market-based models?
A. There is no difference between the two types of model.
B. Statistical models use information on a particular credit for assessment,
whereas market models use market-wide information.
C. Statistical models use private firm- or individual-specific information, whereas
market models use only publicly available information.
D. Both models use firm-specific information, but market models use traded
financial prices as inputs to the model.

1.15 Which of the following is correct? Firms within a particular credit class will:
A. have similar expected default probabilities. B. have common financial
conditions and size. C. be in the same or similar industries.
D. All of A, B and C will lead firms to be in a particular credit class.

1.16 If average losses over the last six years have been $0.95 million per year and
currentyear losses are $1.12 million, which of the following is true?

Credit Risk Management Edinburgh Business School 1/4

7
Module 1 / Introduction
A. The distribution of losses has a dispersion of $0.78–$1.12 million. B. Losses
from the current year are below expected losses. C. Losses from the current year
are above expected losses.
D. We cannot make any statements about losses from the information provided.
1.17 Which of the following is the most common credit loss event?
A. Bankruptcy of the obligor.
B. The failure to make payment when due. C. A material adverse debt
restructuring.
D. A debt repudiation.

1.18 What is a credit guarantee?

A. It is an understanding by a counterparty to provide additional information for credit
evaluation purposes.
B. It is a third-party undertaking to make good any losses if the counterparty
defaults on an obligation.
C. It is a surety deposit placed with a third party, such as a bank. D. It is a form of
credit insurance better known as ‘credit factoring’.

1.19 Which of the following is correct? A netting agreement is:

A. a way of encouraging debtors to pay early by offering a discount for prompt
payment.
B. an offset agreement where sums due and sums payable are set against each
other.
C. a special type of contract that allows party A with a liability to party B to net this
against a payment due from party C.
D. a concession in the corporate tax laws of most countries that allows for credit
losses to be netted against corporation tax due.

1.20 In managing credit risk, what is meant by ‘hot spots’?

A. They are a collection of credit exposures that share common features and
therefore reduce the benefits of risk diversification.
B. They are the exposure to poor-quality credit obligors in a firm’s credit portfolio.
C. They refer to those industries that are in recession and hence suffer from
above-average credit losses.
D. It is a credit assessment mnemonic template that means ‘healthy-or-troubled,
security, provisioning, tracking and settlement’.

1.21 A portfolio has three credits worth 10 000 each. The probability of default of these
accounts is 0.10, and if default takes place there is no recovery. The credits’
performances are uncorrelated. What is the expected value of the portfolio?
A. 21 000 B. 27
000 C. 28 770
D. 29 000

1.22 In the previous question, what is the standard deviation of the portfolio?

1/4 Edinburgh Business School Credit Risk Management

8
 Module 1 / Introduction
A. 1230 B.
5196 C. 24 345
D. 28 770
1.23 What is a loan loss provision?
A. It is an accounting entry for future expected losses. B. It is a charge
against profits for recently reported losses. C. It is a discount made
to future cash flows to reflect expected bad debts. D. It is a charge
against profit before calculating liability for corporate taxes.

1.24 In the context of credit risk management, what is meant by risk pricing?
A. It is an increase in the price to reflect expected credit losses.
B. It is an increase in the interest rate charged by banks to their customers
to cover expected credit losses.
C. It is an evaluation of the price to charge a customer taking trade credit to
reflect the probability of default.
D. All of A, B and C.

1.25 What is meant by ‘collateralisation’?

A. It is the provision of security to support a transaction.
B. It is when creditors to a defaulted company work together to minimise
losses.
C. It is a contract that bundles a credit-sensitive security with a credit
derivative to remove the credit risk.
D. It is a form of legal redress available to creditors when a firm goes
bankrupt and fraud is suspected.

Case Study 1.1: Determining the Credit Risk of a Portfolio

A firm has four receivables outstanding and expects to maintain this number
over time by replacing each transaction as it is paid.
The probability of default for each counterparty company is the same, at
0.15, and if default takes place then the firm expects to recover 40 per cent
of the amount outstanding. The contract size for each credit is 100000. The
firm has a selling margin mark-up of 18 per cent on these transactions.

1 What is the expected value of the contracts and what is the standard
deviation of losses for its portfolio under the assumption that the losses are
independent?

2 The firm wants to be 95 per cent (1.96 standard deviations) certain it will
achieve a minimum of 270 000 from the transactions. What should be its
quoted selling price for these transactions to the nearest thousand? (Note
that you will need to use trial and error to obtain the new selling price.)

Credit Risk Management Edinburgh Business School 1/4

9
Module 1 / Introduction
References
Altman, E., Resti, A. and Sironi, A. (2003). Default Recovery Rates in Credit
Risk Modeling: A Review of the Literature and Empirical Evidence.
December, Working Paper, Stern School of Business, New York University.
Basel Committee on Banking Supervision (2000). Risk Management Group
of the Basel Committee on Banking Supervision. September, p.1. Bank for
International Settlements: Basel.
Porter, M. (1985). Competitive Advantage: creating and sustaining superior
performance. New York: The Free Press.
Rouse, C.N. (2002). Bankers’ Lending Techniques (2nd ed.). Chartered
Institute of Bankers: Financial World Publishing, p.26.

1/5 Edinburgh Business School Credit Risk Management

0