𝝈

FINANCIAL RISK MANAGEMENT

CREDIT RISK EVALUATION AND MEASUREMENT
Credit Risk
• Credit is an agreement where one party receives
something of value and agrees to pay for the
good or service at a later date. The word “credit”
is derived from the ancient Latin word credere,
which means “to believe” or “to entrust.” The
creditor must have knowledge of the borrower’s
character and reputation as well as his financial
condition
Credit Risk
• Generally, there is not a definitive yes or no
answer to whether a borrower can and will pay
back a loan. As such, the lender must address the
question of likelihood. The lender must assess the
likelihood that the borrower will pay back the
loan in accordance with the terms of the
agreement.
Credit Risk
• NOTE: Borrower, obligor, counterparty, and
issuer are all used to signify the party receiving
credit. Lender, creditor, and obligee are primarily
used to signify the party granting credit.
Aspects of Credit Risk
• The concept of credit risk encompasses a range of risk
measures. Those relating to default include default risk,
recovery risk, and exposure risk. Those relating to valuation
include migration risk, spread risk, and liquidity risk.
Additional measures include concentration risk
and the correlation with pure financial risks (e.g., interest rate,
exchange rate, and inflation risks).
• Default risk, or counterparty risk, relates to a borrower’s
inability to make promised payments. Recovery risk is the risk
that the recovered amount, in the event of default, is less than
the full amount that is due. Exposure risk measures the risk
that a credit exposure at the time of default increases relative to
its current exposure
Aspects of Credit Risk
• Migration risk looks at the risk that the credit quality
and market value of an asset or position could deteriorate
over time. To mitigate this risk, a periodic assessment of
the credit quality of assets is necessary, and institutions
may need to make credit provisions and record gains and
losses.
• Spread risk is the risk that spreads may change during
adverse market conditions as investors require different
risk premiums, leading to gains and losses. Liquidity
risk is the risk that asset liquidity and values deteriorate
during adverse market conditions, lowering their market
value.
Aspects of Credit Risk
• Credit default risk is the probability that a borrower will not
pay back a loan in accordance with the
terms of the credit agreement. The risk can result from:
-Default on a financial obligation.
-An increased probability of default on a financial obligation
-A more severe loss than expected due to a greater than expected
exposure at the time of a default.
-A more severe loss than expected due to a lower than expected
recovery at the time of a default.
Aspects of Credit Risk
• Credit events include:
-Bankruptcy.
-Failure to pay.
-Restructuring.
-Repudiation.
-Moratorium.
-Obligation default
Credit Risk Evaluation
• -The borrower’s (or obligor’s) capacity and
willingness to repay the loan
• -The external environment and its effect on the
borrower’s capacity and willingness to repay the
borrowed funds
• -The characteristics of the credit instrument
• -The quality and adequacy of risk mitigants such as
collateral, credit enhancements, and loan guarantees.
Credit Risk Evaluation
• The borrower’s (or obligor’s) capacity and willingness
to repay the loan. Questions the lender must consider
include:
-What is the financial capacity to pay?
-Is it likely the borrower can fulfill its financial obligations
through the maturity of the loan?
-Are there outside forces that affect the borrower’s capacity
and/or willingness to pay? For example, does the ownership
structure of the firm, relationships within and outside the
firm, and other obligations of the firm affect the borrower’s
ability to pay?
-How does the business itself affect the borrower’s capacity
to pay? Are there credit risk characteristics tied to this
particular industry or sector?
Does the firm have a niche within the industry or sector?
Credit Risk Evaluation
• The external environment and its effect on the borrower’s
capacity and willingness to repay the borrowed funds.
Factors such as the business climate, country risk, and
operating conditions are relevant to the lender.
• Are there cyclical changes that will affect the level of credit
risk? Will political risks affect the likelihood of repayment?
Credit Risk Evaluation
• The characteristics of the credit instrument. The credit instrument might
be a bond issue, a bank loan, a loan from a finance company, trade credit,
or other type of debt agreement/security. Concerns include:
• Risk characteristics that are inherent in the credit instrument, including
legal risks and obligations that are specific to the instrument.
• The maturity (also called “tenor”) of the instrument.
• Is the debt secured or unsecured? Is there collateral backing the loan? Are
there loan guarantors?
• Is the debt subordinated or senior to other obligations? What is the priority
assigned to the creditor?
• How do loan/bond covenants increase or decrease the credit risk for each
party?
• Can the borrower repay the loan early without penalty? Can the lender call
the loan? Can the security be converted to another form (e.g., a convertible
bond)?
What is the denominated currency of the obligation?
• Are there any contingent risks?
Credit Risk Evaluation
• The quality and adequacy of risk mitigants such as
collateral, credit enhancements, and loan guarantees. Secured
lending (i.e., using risk mitigants in the lending process) is
generally the preferred method of lending. If there is collateral,
a bank or other lender may not have to force a delinquent
borrower into bankruptcy but may instead sell the collateral to
satisfy the financial obligation. Secured lenders are also
generally in a better position than unsecured lenders in the
event of bankruptcy.
Credit Risk Evaluation
• The use of collateral not only mitigates losses in the event of
default, but also lowers the probability of default because the
obligor typically does not want to lose the collateral.
Historically, banks have substituted collateral for analysis of
the borrower’s ability to pay. In some sense, the use of
collateral eliminates the need for credit analysis, or at the very
least makes the credit decision simpler. A lender can normally
put a market value on collateral and determine if it is sufficient
to cover potential losses. Three issues regarding risk mitigants
include:
Credit Risk Evaluation
• Is the collateral pledged to, or likely to be pledged to, another
loan?
• Has there been an estimation of the value of the collateral?
• If there is a loan guarantor, has there been sufficient credit
analysis of the third party’s willingness and ability to pay in
the event the borrower does not pay? A guarantor accepts
liability for debt if the primary borrower defaults. The bank is
able to substitute analysis of the guarantor’s creditworthiness
for that of the primary borrower. Typically, the guarantor has a
greater ability to pay than the
primary borrower (e.g., a parent guaranteeing a child’s car loan
or a parent company guaranteeing a loan to a subsidiary)
Credit Risk Evaluation
• The willingness to repay a loan is a subjective attribute.
Lenders must make unverifiable judgments about the
borrower. In some cases, intuition, or “gut feelings,” are
necessary to conclude whether a borrower is willing to repay a
loan. As such, qualitative credit analysis
techniques are largely used to evaluate the borrower’s
willingness to repay.
Credit Risk Evaluation
• Qualitative techniques include:
• Face-to-face meetings with the potential borrower to assess
the borrower’s character are routine in evaluating willingness
to pay.
• “Name lending” involves lending to an individual based on
the perceived status of the individual in the business
community.
Credit Risk Evaluation
• Gather information from a variety of sources about the
character and reputation of the potential borrower. Old-
fashioned lending relied on first-hand knowledge of the people
and businesses in a town. In this case, lenders knew (or
thought they knew) potential borrowers. It is more difficult in
the modern world, where lending decisions are centralized, to
know customers personally.
Credit Risk Evaluation
• Extrapolating past performance into the future. Lenders
often assume that a pattern of borrowing and repaying in the
past (e.g., a credit record compiled from past history
with the borrower and data garnered from credit bureaus) will
continue in the future.
• Historical lending norms relied on the moral obligation of
borrowers who could pay to repay their debts. Thus, gauging
the borrower’s willingness to pay was a critical component of
credit analysis.
Credit Risk Evaluation
• However, in modern society, the moral obligation to pay if one
is capable of paying has been replaced by the legal obligation
to pay.
• In other words, in terms of credit analysis, determining the
capacity to pay is more important than determining the
willingness to pay because the legal system will force those
who can pay to honor their commitment.
• Courts can seize the assets of those who will not fulfill their
financial obligations.
Credit Risk Evaluation
• In corrupt or ineffective states, a borrower will not suffer, even if
able to pay but not doing so.
• The willingness to pay is more important in countries with less-
developed financial markets and legal systems. Creditors must
evaluate the legal system and the strength of creditors’ rights in
emerging markets, along with the prospective borrower’s ability and
willingness to repay the obligation.
• The creditor must also consider the costs associated with taking
legal action against a delinquent borrower. If costs are high, the
creditor may be unwilling to take action regardless of the strength of
the enforcement of creditor rights. As such, the willingness to pay
should never be completely ignored in credit analysis.
Credit Risk Evaluation
• The ability of a borrower to repay a loan is an objective
attribute. Quantitative credit analysis techniques are largely
used to evaluate the borrower’s ability to repay. The primary
quantitative technique used in financial analysis is examining
the past, current, and forecasted financial statements of the
prospective borrower. This forms the core of the quantitative
credit analysis used to determine a borrower’s capacity to meet
its financial obligations.
Credit Risk Evaluation
• There are limitations associated with quantitative data, which
include:
Limitations on the Historical nature of the data. Financial
data is typically historical and thus may not be up-to-date or
representative of the future. Also, forecasted financial data is
notoriously unreliable and susceptible to miscalculations
and/or misrepresentations.
Credit Risk Evaluation
• Difficult to make accurate projections using historical data.
Financial statements attempt to represent the economic reality
of a firm in a highly abbreviated report. As such, some
information is lost in translation that is critical to the loan
decision. The rules guiding financial reporting are created by a
diverse group with varying interests and are often decided by
compromise.
Credit Risk Evaluation
• Firms may use the latitude in financial reporting to deceive
interested parties. Even if the reports are accurate, financial
data is subject to interpretation. There can be a range of
conclusions drawn from the same data due to the variety of
needs, perspectives, and experiences of the various analysts. This
means there is a subjective, qualitative component to an
objective, quantitative exercise.
• Given the shortcomings of financial reporting, lenders should
not ignore qualitative analysis. The quality of management, the
motivation of the firm’s management, and the incentives of
management are relevant for both nonfinancial and financial
firms.
• Qualitative skills are necessary to assist in determining
the willingness of the entity to repay debt (e.g.,
reputation, repayment track record).
• It is critical for analysts to think beyond numbers and
apply considerable judgment, reasoning, and experience
in determining which factors are relevant for making
decisions (e.g., management competence, bank’s credit
culture, and the robustness of credit review process).
• The ability to analyze the quality, reliability, and
consistency of reported earnings is also necessary. In
addition, an understanding of the regulatory
environment of banks and the impact(s) of any
regulatory changes is important (e.g., central bank given
more authority to regulate banks)
• Quantitative skills are necessary to assist in determining the
ability of the entity to repay debt. A banking credit analyst
must be able to read and interpret financial statements in
order to perform a wide range of ratio analysis.
• The ratios to be analyzed depend on which measures of
financial performance are relevant (i.e., liquidity, solvency,
profitability). For example, return on equity (ROE) is a
commonly used measure because it considers efficiency and
leverage in addition to profitability.
• Analysts must also understand statistical concepts (e.g.,
sampling, confidence intervals, correlation) in order to
properly interpret data to arrive at reasonable conclusions
under uncertainty. Analysts should have an understanding of
monetary policy and an ability to compute and interpret
macroeconomic data (e.g., GDP growth rates), both of which
impact the general banking industry.
Credit Risk Evaluation
• Credit quality analysis from an experts-based approach
will apply frameworks such as the
four Cs of credit (Character, Capital, Coverage,
Collateral) proposed by Altman/NYU,
• LAPS (Liquidity, Activity, Profitability, Structure)
from Goldman Sachs, and
• CAMELS (Capital Adequacy, Asset Quality,
Management, Earnings, Liquidity, Sensitivity) from JP
Morgan. As Porter (1980, 1985) emphasized, qualitative
features need to be factored into any analysis along with
quantitative components.
Credit Risk Evaluation
• The types of qualitative items that may be found in a credit
analysis questionnaire include things like corporate structure
(incorporation date, group members), business information
(competitive forces within the industry, growth forecasts),
• Management quality (degree of involvement, experience),
strategy (business plans, nonrecurring transactions such as
mergers and transfers),
• Financial position sustainability (liquidity risk, debt maturity
concentration), quality of information given to the bank by the
company (availability of financial projections, relationship
history), and other risks (geographic focus, client base quality).
Due to the enormous breadth of qualitative factors, a best practice
would be to only collect qualitative information that cannot be
quantified.
Credit Analysis Comparison for Different Customers

• Consumers
• Capacity: Wealth (i.e., net worth), salary, or incoming cash per
period, expenses per period, assets such as houses and cars, amount of
debt (e.g., credit card debt), net cash available to service debt (i.e., cash
flow minus household and mortgage expenses)
• Willingness: Reputation of individual, payment history
• Methods of evaluation: Credit scoring models that consider income,
duration of employment, and amount of debt for unsecured debt like
credit cards. Credit scoring and some manual input and review for large
exposures such as mortgage loans or automobile loans
• Loan size/type: Large exposures are typically secured (e.g., mortgage
loans). Smaller exposures are unsecured (e.g., credit card loans)
Credit Analysis Comparison for Different Customers

• Corporations:
• Capacity: Liquidity, cash flow combined with earnings
capacity and profitability, capital position (solvency), state of
the economy, strength of the industry.
• Willingness: Quality of management, historical debt service
• Methods of Evaluation: Detailed manual analysis including
financial statement analysis, interviews with management.
More complex than consumer analysis because companies are
so diverse in terms of assets, cash flow, financial
structure, etc.
• Loan Size/Type: Typically larger exposures (sometimes
considerably larger) than loans to consumers. Debt
may be secured or unsecured
Credit Analysis Comparison for Different Customers

• Financial Institutions
• Capacity: Similar to nonfinancial firms but bank specific.
Liquidity (the bank’s access to cash to meet obligations),
capital position historical performance including earnings
capacity over time (and ability to withstand financial stress),
asset quality (affects the bank’s likelihood of being paid back
and by extension the bank’s lender’s likelihood of being paid
back), state of the economy, strength of the industry.
• Willingness: Quality of management; qualitative analysis is
even more important for financial firms than for nonfinancial
firms
• Methods of evaluation: Similar to nonfinancial firms
• Loan Size/Type: Similar to nonfinancial firms (i.e., large).
Credit Analysis Comparison for Different Customers

• Sovereigns
• Capacity: Financial factors including the country’s external debt load
and debt relative to the overall economy; tax receipts
are important
• Willingness: Credit analysis for sovereigns is often more subjective
than for financial and nonfinancial firms because the legal system and
the enforcement of creditor rights is critical to the analysis. Sovereign
legal risk ratings are often considered
in the analysis.
• Methods of evaluation: Similar to financial and nonfinancial firms but
with increased subjective analysis of the political environment
• Loan Size/Type: Similar to nonfinancial and financial firms (i.e.,
large).
Credit Risk Measurement
Credit risk measurement requires Modelling of its drivers:
• – Distribution of default probabilities
• – Loss given default
• – Credit exposures
Drivers of Credit Risk

• Probability of Default (PD)

• Credit Exposure (CE) or Exposure at
default (EAD)
• Loss given default (LGD)
Modelling Default Risk
• Heuristics/Expert Systems - These methods are designed
to mirror human decision-making processes and procedures.
These methods are also known as “expert systems,” with a
goal of reproducing high frequency standardized decisions at
the highest level of quality at a low cost. The fundamental
idea is to learn from both successes and errors.
Modelling Default Risk
• An expert system may also incorporate “fuzzy logic”
applications. This logic applies “rules of thumb” based on feelings
and uses approximate as opposed to precise reasoning. A fuzzy
logic variable will not be confined to the extremes of zero and
one; rather, they can assume any value that exists between the two
extreme values.
Modelling Default Risk
• Numerical methods. The objective of these methods is to
derive optimal solutions using “trained” algorithms and
incorporate decisions based on relatively weak information
in very complex environments. An example of this is a
“neural network”, which is able to continuously update itself
in order to incorporate modifications to the
environment
Modelling Default Risk
• Logistic regression models (also known as LOGIT
models), which are from the Generalized Linear Model
(GLM) family, are statistical tools that are also used to
predict default. These types of models are based on
analyzing the dependencies of one or multiple
dependent variables from one or more independent
variables.
• GLMs typically have three common elements:
A systematic component, which specifies the variables
used in a linear predictor function.
A random component, which identifies both the target
variable and its associated probability function.
A link function, which is a function of the target variable
mean that the model ties to the systematic component.
Modelling Default Risk
• Assume that π represents the probability that a default event takes
place. The link function represents the logarithm of the ratio between the
default probability and the probability that the firm continues to be a
performing borrower (the ratio is known as odds). The LOGIT
(i.e., logarithm of odds) equation is therefore:
𝜋𝑖
𝐿𝑂𝐺𝐼𝑇 𝜋𝑖 = 𝑙𝑜𝑔
1 − 𝜋𝑖
•
The LOGIT function associates the expected value for the dependent
variable to the linear combination of independent variables, whereas the
relationship between the probability of default (π) and the independent
variables is nonlinear.
In the event that there is only one explanatory variable, the LOGIT function becomes:
𝜋𝑖
= 𝑒 𝑏0 +𝑏1 𝑥1
1 − 𝜋𝑖
Modelling Default Risk
• Cluster analysis looks to identify groups of similar cases in a data set.
Groups represent observation subsets that exhibit homogeneity (i.e.,
similarities) due to variables’ profiles that allow them to be
distinguished from those found in other groups. In the context of a
database with variables in columns and observations in rows, cluster
analysis serves to aggregate borrowers based on the profile of their
variables. The end result is a top-down, statistically based segmentation
of borrowers. An empirical default rate can be calculated for each
segment, which serves as the default probability for the borrower at
each segment
• Market-price methods: infer risk-neutral measurements from traded
prices (including a risk premium) of debt, equity or credit derivatives.
Modelling Default Risk
• Determining default probability can be based on (1)
analysis of historical default frequencies of a
borrower’s homogenous asset classes, (2) mathematical
and statistical tools, (3) a hybrid approach that
combines mathematical and judgmental analyses, and
(4) implicit default probabilities from market prices of
publicly listed counterparties.
Default risk is typically measured over one year.
However, cumulative default rates extending beyond
one year are important. Shorter exposures, such as
overnight lending, are also exposed to default risk.
Modelling Default Risk
1) Analyzing historical default frequencies of a borrower’s
homogenous asset classes.
Historically, credit analysis was based on subjective
analysis, and rating agencies assigned ratings and historical
default rates on past observations on an ex post basis
(i.e., after an event).
2)Using mathematical and statistical tools. Statistical
models are typically used for large portfolios with hundreds
or even thousands of positions, which allows for
segmentation into different risk classes, measuring risk on
an ex ante basis (i.e., before an event).
3) Using a hybrid approach that combines mathematical
and judgmental analyses. The mathematical results are
generated automatically, which are then corrected using
qualitative analysis.
Modelling Default Risk
• Recovery risk is a conditional metric assuming that default
has already occurred. The amount of recovery depends on
(1) the type of credit contracts used and the relevant legal
system, (2) general economic conditions, and (3) covenants.
Estimating the recovery rate on ex ante basis is challenging
due to the difficulty in collecting recovery rate data,
uniformity of information, and challenges in creating a
comprehensive model. Exposure risk is easily determined
for term loans. For revolving credit facilities, exposure
depends on borrower behavior and external events.
Measuring Credit Loss
• The distribution of expected losses due to credit risk from a portfolio of N
instruments can be
described as:
𝑁

𝐸𝐶𝐿 = ෍ 𝑏𝑖 𝑋 𝐶𝐸𝑖 𝑋 𝑝 𝑋 (1 − 𝑓𝑖 )
𝑖=1

• Where bi is a Bernoulli random variable (1=default, 0=No default), CEi is

credit exposure at the time of default. P is the probability of default and fi is the
credit recovery rate.
Recovery Rates -Recovery rates tend to be lower when the economy is in a
recession. Suppose that default is accompanied by a declaration of bankruptcy.
The bankruptcy process creates a pecking order for a company’s creditors. The
recovery rate will vary by the position in the pecking order (status or seniority of
the debtor) and the value left after liquidation of the firm’s assets.
Measuring Credit Loss
• Exposure risk measures the amount of risk a firm is
exposed to in the event of a default. For
term loans, exposure is easily determined. For revolving
credit facilities, determining
exposure is more challenging since it depends on borrower
behavior and external events. In
this situation, exposure risk [i.e., exposure at default (EAD)]
• Recovery risk measures the risk that the amount recovered, in
the event of a default, is less than the full amount that is due.
The recovery rate is a conditional metric expressed as a
percentage which assumes that default has already occurred. It
is the complement to loss given default (LGD) such that the
recovery rate equals 1 - LGD. The amount of recovery
depends on the following factors:
-The type of credit contracts used and the relevant legal
system.
-General economic conditions. Firms operating in more
volatile sectors may see larger swings in asset values.
-Covenants. Negative covenants restricting the sale of assets
that are important to the borrower should be considered in
LGD estimations
• Alternatively then Expected Credit Loss can be calculated
as:
• ECL = PD X LGD X EAD
Measuring Credit Loss
• Concentration was traditionally mitigated by minimizing
exposure to a single borrower. Portfolio credit risk models
specifically factor in a borrower’s risk contribution to
concentration, and allow for segmentation of portfolio risk or
viewing the portfolio risk profile as a whole.
• Concentration risk arises in credit portfolios where borrowers
all face common risk factors, including interest rates, exchange
rates, and changes in technology. Facing common risks is
problematic since they simultaneously affect a borrower’s
willingness and ability to repay their obligations.
Measuring Credit Loss
• However, actual losses may be different from
expectations, resulting in unexpected losses (ULs). ULs are
problematic because they can jeopardize the viability of a
bank as a going concern. Banks can prepare for ULs by
holding sufficient equity capital to cover all risks, not just
credit risks. Capital can be replenished from profits in good
times, which can absorb ULs. Credit risk models and credit
ratings are important in determining the overall credit
contributions needed by banks
• End.