Accenture Risk Analytics Network

Credit Risk Analytics

Accenture, as a leader in risk analytics,

works closely with banks and other
financial institutions in developing the
building blocks of credit risk analytics.

Lending, the core business of banks, is their

daily activity for profit generation, while
risk management serves as the controlling
function for lending. Through innovation
in the financial services industry over the
last decade, regulators have gradually
encouraged banks to develop their own risk
management tools and enhance their risk
management framework.
For managing credit risk, many banks
still use expert judgment models without
the benefit of an accurate or integrated
framework to support their often
complicated risk management needs in
a changing and evolving environment.
A banks exposure to risks which have
not been fully measured might lead
to unexpected default rates and high
write-offs, which would influence
their profit and capital requirement.

Probability of Default
Loss Given Default
Exposure at Default Modeling
Beginning in 2004, Basel II imposed a
standard methodology for credit risk
management and introduced more
flexible regulatory supervision. This in
turn led banks to move towards the
development and implementation of
accurate modeling methodologies on an
Internal Ratings-Based (IRB) approach,
and the quantitative-based measurement
of credit risk factors - Probability of
Default (PD), Loss Given Default (LGD)
and Exposure at Default (EAD). (See
Figure 1.)

Figure 1: Banks can help reduce their capital charge by using an advanced IRB (Internal Ratings-Based) approach
Pillar 1:
Minimum Capital
Requirements

Banks who move up the ladder are rewarded by a reduced capital charge

Increased Sophistication

Advanced Internal
Ratings-Based
Approach

Foundation Internal
Ratings-Based
Approach

Standardized
Approach

Banks use internal estimations

of PD, loss given default (LGD)
and exposure at default (EAD)
to calculate risk weights for
exposure classes.

Banks use internal estimations

of probability of default (PD) to
calculate risk weights for exposure
classes. Other risk components are
standardized.

Risk weights are based on

assessments by external credit
assessment institutions.

Potential Reduced Capital Requirements

Source: Accenture

Benefits of Adopting Basel

Accord Compliance
With all the relevant models in place
within a Basel II framework, banks can
enjoy a broad range of potential benefits,
including:
Improved Credit Risk Return profile due to:
Improved credit rating and monitoring
Enhanced risk-based pricing
Reduced non-performing loans and
bad debts
Optimized credit portfolio structure
Reduced Economic capital requirement
due to:
Lower risk-weighted assets through
the adoption of an internal ratings-based
approach
Improved capital allocation

Improved credit processing efficiency due to:

Streamlined and/or automated
credit processing
Improved collections management
Reduced operational losses due to:
Improved allocation of capital
This can also help banks potentially
improve their credit rating and
thereby provide a competitive edge over
competitors.

Our Approach
Gradual implementation of the complete
credit risk management framework
using PD/LGD/EAD models as the basic
building blocks can help banks realize
these benefits. Accenture typically uses a
six-step credit risk management process
consisting of:
1. Risk identification
2. Risk measurement
3. Approval and control
4. Reporting and monitoring
5. Provision and capital

The Accenture Risk Analytics Network

consists of experienced members who
are all dedicated to developing and
implementing accurate and robust PD,
LGD and EAD models for corporate,
small and medium enterprises (SME) and
the consumer sectors. This experience
includes providing credit scoring such
as application scorecard, behavior
scorecard and collection scorecard for
all consumer lending products, such
as credit cards, installment loans and
mortgages. With many successful
assignments providing IRB approach
and risk scoring to international and
regional banks, we have deep industry
insight and a broad array of industry
benchmarks to support such initiatives.

6. Portfolio management and capital

allocation
These steps take into account the
organizations data, transaction and
portfolio levels while aligning key
components such as governance,
policies and processes, and information
technology enablement to set appropriate
transaction and portfolio limits.

Probability of Default Loss

Given Default and Model
Validation
The critical role played by internal
models, industry leading practices
and regulatory requirements dictate
that financial institutions implement
an independent model validation
process to assess the quality and
accuracy of their internal models.
Independent validation of internal IRB
models is in increasing demand under
Basel II.
Banks worldwide need to invest and
implement a strong mechanism via
systems to authenticate the precision and
reliability of rating systems, processes,
and the appraisal of all relevant risk
components.
In addition, a bank must also demonstrate
to regulators the completeness of its
internal model validation process.

Figure 2: Validation is an integral part of the Model Life Cycle

Validate

As required by the Basel Accords, IRB

model validation is necessary to meet
external and internal compliance. While
the various aspects of model quality
can be assessed with complicated
quantitative procedures, qualitative
judgment is essential to guarantee
that the financial institutions are
using the correct model. As a
consequence, the efforts involve a
combination of in-depth knowledge
in analytical validation techniques as
well as banking industry practices.
Financial institutions are also expected to
have frameworks in place to enable:

Design &
Develop

Monitor

Model
Life Cycle

Use

Initial model validation-- review of the

model development, the processes and
the execution of the model
Ongoing model validation-- ongoing
validation of rank-order performance
using industry wide standard metrics.

Implement

Source: Accenture

Benefits of Model Validation

Process
Accenture can add value by helping
clients implement a model validation
process. The prospective benefits include:

In-depth model enhancements based on

real-world applications during validation
process

Access to experience and know-how

gained from implementing robust practice
methodologies and processes, and model
validation efforts during previous client
assignments

Prompt reporting capabilities including

from issue to outcome analysis

Access to implementation benchmarks

that can be used by clients as part of
their assessment effort
Expertise gained from working
with regional regulators on IRB
model reviews

Top-down evaluation process for

the design and implementation of risk
mitigation controls
Support through the Accenture
Analytical Network
Knowledge transfer program

Figure 3: Advantages of outsourcing independent validation

Basel II Regulatory Perspective
A key requirement for IRB compliance
A key element of board and senior
management reports
A well-defined, actionable process around
ongoing reporting on model quality with
clearly defined responsibilities, metrics
and thresholds for acceptable quality

Risk Management Unit

Department 1
Team 1

Team 2

Department 2
Team 3

Team 4

Independent
Validation Team
Team 5

Independent of model development

In order to comply with Basel II regulation, most banks should consider establishing internal
independent validation teams to meet the requirements.

The responsibilities of banks,

not supervisors

Consideration should also be given to having the independent validation teams focus on
providing effective feedback and recommendations for strengthening the models.

Our Approach

Our validation services include:

The goals of model validation are to:

Independent model validation:

reviewing model methodologies,
assumptions, data inputs, intermediate
adjustments, expert judgment and
outcomes

Improve Basel compliance through an

efficient approach to risk assessment,
confirming the model is operational as
expected
Improve model development and
best-in-class validation, by identifying
model inadequacies and determining
the situations where the model is
inappropriate
Find the tradeoff between analytical
foundation and risk judgment, on behalf
of a standard process incorporating these
essential needs (See Figure 3.)
In order to fulfill regulatory requirements
prior to the Basel II compliance
application, banks had to perform an
independent validation through an
external vendor (a common industry
practice). Today there are some practical
challenges to performing these external
independent validations.
Accenture can work with clients to
perform such validations, as well as the
regular monitoring and reporting of
internal models to external parties. (See
Figure 4.)

Program management: identifying

program managed capabilities and endto-end delivery
Model and process matching:
conducting a complete process map using
template or interview inputs
Outsourcing of independent validation,
regular model monitoring and reporting
function

Countercyclical Capital Buffer

in Basel Regulation
Pro-cyclicality in the banking industry
was said to exacerbate the impact of the
banking crisis. While this is inherent to
the industry and cannot be completely
eliminated, the Basel Committee on
Banking Supervision (BCBS) introduced
a framework for countercyclical capital
buffers beyond the minimum capital
requirement set in the Basel II framework.

Figure 4: Accentures approach to performing an independent validation

Using a common external

validation team creates value

Optimized Internal Resources Usage

Greater emphasis placed on the validation of newly implemented models. Stable
and robust models and proper monitoring can help reduce the validation effort.

The primary objective of the

countercyclical capital buffer is to
achieve the broader macro prudential
goal of protecting the banking sector
from periods of excess aggregate credit
growth that have often been associated
with the build-up of system-wide risk.
As such, the common reference point
put forward by the BCBS for taking
countercyclical buffer decisions is the
credit to GDP guide.1 This buffer is up
to an additional 2.5 percent of riskweighted assets (RWAs). The direct
implication is that the minimum capital
requirement will increase by 30 percent,
from eight percent to 10.5 percent at a
maximum and at the national supervisors
discretion, depending on the different
level of Credit to GDP ratio.
The Credit to GDP2 ratio for Hong Kong
from 1995 to 2010 and its long-run
trend are shown in Figure 3. Using the
countercyclical capital buffer rule (buffer
derived from the Credit to GDP ratio for
Hong Kong), we can observe the capital
buffer that would have been required in
the years 1990, 1996, 1997, 1998, 1999,
2009 and 2010.
During the Asian Financial Crisis (1998
to 1999), it is likely that the high Credit
to GDP ratios were due to very low GDP
figures rather than significant systematic
credit risk, and therefore the additional
capital buffer might be misleading.
In order to understand the benefits, an
example is provided to illustrate the
impact of countercyclical capital buffer.

Re-allocation of internal risk analytics resources can help create more value:
Resources can be released for new risk management research, such as LR or
counterparty risk.

Compliance Purpose
Standardized independent validation program/Approval: Market consistency,
Acceptable thresholds.
Cost Efficiency: Economy of scale.
Consultants with adequate experience can provide advice during tests.
Provide industry benchmarks: Peer to peer comparison for all participants.

It is important to note that the BCBS has caveats with respect to its use, not the least of which is that the common reference point could give misleading signals if used as
a standalone measure. The BCBS proposed supervisory judgment is also exercised when countercyclical buffer decisions are made. The key role given to judgment by relevant
national authorities, and the designation of which will be left to each jurisdiction, could result in an unleveled playing field.

The precise methodology can be found in the Countercyclical capital buffer proposal, Issued for comment on September 10, 2010. More recent information on this topic can be
found at: http://www.bis.org/publ/bcbs187.htm and http://www.bis.org/publ/bcbs189.htm.

Benefits of Countercyclical
Capital Buffer
We constructed a hypothetical3 corporate
segment portfolio of HK$ 150 billion
from 1000 obligors in Hong Kong (HK)
for the years 1995 to 2006 and used HK
Credit to GDP ratio to derive the timing
and magnitude of the countercyclical
capital buffer. (See Figure 5.)

The graph in Figure 6 plots the internal

rating distribution for different years
and Figure 7 presents the assumed
average LGD per rating grade, which
is a downward trending curve where
greater collateral coverage requirements
were imposed to higher default
risk customers. With the exposure
amount fixed at HK$ 150 billion, the
distribution of the exposure across
rating grades is merely determined
by customer rating distribution.

For the impact analysis, we constructed

the rating distribution for the internal
model based upon Point-in-Time (PIT)
and Through-the-Cycle (TTC) models.
By comparing the impact of the
internal model capital requirements,
the PIT and TTC models, we can see
the differences attributable to the
countercyclical capital buffer.

Figure 5: Countercyclical capital buffer distribution

200

2.5

2.5 2.5

190

2.0

180

Credit to GDP Ratio

2.5

2.0

170
1.5

160
150
140

1.0

0.8

130
120

0.5

0.3

110
100

Credit to GDP (LHS)

HP Trend (LHS)

Source: Accenture

This hypothetical portfolio is built from Accentures corporate experience with the rating migration of an internal
rating model, average LGD and EAD across rating grades.

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

0.0

Capital Buffer (RHS)

2.5

Figure 6: Internal rating distribution of a hypothetical portfolio

20%

Percentage of Obligors

18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
1

10

11

12

13

14

15

16

14

15

16

Rating Grades
1995
2001

1996
2002

1997
2003

1998
2004

1999
2005

2000
2006

Source: Hypothetical example created by Accenture

Figure 7: LGD distribution by rating grades

40%

Percentage of Obligors

38%
35%
33%
30%
28%
25%
1

10

11

12

13

Rating Grades
Source: Hypothetical example created by Accenture

The chart that follows (Figure 8)

shows the capital requirement of our
hypothetical portfolio. We can observe
that for a perfect TTC model, the
capital requirement is stable across the
economic cycle, while the PIT model is
more volatile and highly affected by
the economic cycle. The fluctuation of
the capital requirement for the internal
model is somewhere in between. The
countercyclical capital buffers in the
years 1996 to 1999 were imposed
as an additional capital requirement
for the bank using the internal
model as shown in the bar chart.

The differences in impact of the

countercyclical buffer can be clearly
observed in Figure 9. In the years 1996
and 1997, the capital requirements of
using a PIT model are 3.6 percent and 2.8
percent lower than those associated with
a typical internal model. These numbers
are more than enough to cover the
additional countercyclical capital buffer
of 2 and 2.5 percent in 1996 and 1997.
Banks using a TTC model are different
from those using a typical internal model.

For the years 1998 and 1999, banks

using a TTC model will benefit from
a lower capital requirement of about
one percent compared to those using
a typical internal model, while, for
those using a PIT model, their capital
requirement will be about one percent
higher due to more significant rating
downgrades incurred during the
crisis.4 Nevertheless, as mentioned
above, imposing a countercyclical
capital buffer may not have been
appropriate during 1998 and 1999.
From the impact analysis results, it is
clear that banks using a PIT internal
model would have been less affected
by countercyclical capital buffer due
to capital savings from the rating
upgrades of their internal models
during the credit expansion period.

Figure 8: Comparison of capital requirements for TTC (Through-the-Cycle) & PIT (Point-in-Time) models
7000

Capital Buffer

6000
5000
4000
3000
2000
1000
1995

1996

1997

1998

1999

2000

2001

2002

Capital Requirement
Model + Buffer

Model

Source: Hypothetical example created by Accenture

Increase in capital requirement will occur one year after the crisis.

PIT

TTC

2003

2004

2005

2006

Figure 9: Capital requirement differences as a percentage of internal models RWAs

% Difference of Capital Requirement

3.0%
2.0%
1.0%
0.0%
-1.0%
-2.0%
-3.0%
-4.0%
-5.0%
1995

1996

Capital Buffer

1997

1998
PIT

1999

2000

2001

2002

2003

2004

2005

2006

TTC

Source: Accenture

10

Our Approach

The impact differences of PIT-focused

internal models will not only affect the
return on equity of the banks, but it
will also impact the dividend payout
ratio, share buybacks and discretionary
bonus payments due to minimum capital
conservatism ratio.

The BCBS as well as banking supervisors

in many countries do not explicitly
prohibit the use of different types of
rating approaches. In fact, based on a
survey conducted by the BCBS5, most of
the banks find it difficult to use a TTC
rating method. However, the level of
PIT focus of the internal models used by
banks may differ depending on the rating
model design and the risk factors chosen.

We have designed a comprehensive

approach to help clients undertake a
model enhancement project, including
internal model enhancements to reduce
the impact of countercyclical buffer. (See
Figure 11.)

As shown in Figure 10, during the

countercyclical capital buffer periods
banks are required to conserve a higher
percentage of earnings for
the same level of capital ratio
(Tier 1 capital).

120%
100%
80%
60%
40%
20%

Common Equity Tier 1 Ratio (including other fully loss absorbing capital)
Capital Conservatism Ratios

Capital Conservatism Ratios with Countercyclical Buffer

Source: Accenture

Figure 11: The Accenture approach to enhancing internal ratings

Current State Analysis

Review credit portfolio
Obtain model
inventories
Review segmentation
Define scope of model
enhancement

Model Design
Review/Diagnostic
Review high level
model design
Review each of the
components of the
model building block
Analyze the PIT-ness
of the models
Analyze the source of
the PIT-ness or
TTC-ness of the model
Define areas of
enhancement

Source: Accenture

The Internal Ratings-Based Approach, http://www.bis.org/publ/bcbsca05.pdf

11

Model Enhancement

Pilot Run and

Impact Analysis

Build/enhance model
building blocks, such as
industry rating and
model overlays

Rating migration
analysis

Analyze predictiveness
of the risk factors

Pilot run

Correlation analysis
Multifactor analysis
Calibration
Master scale
enhancement

Analysis impact of
rating changes on RWA
Feedback analysis
Documentation

9.875%

9.675%

9.475%

9.275%

9.075%

8.875%

8.675%

8.475%

8.275%

8.075%

7.875%

7.675%

7.475%

7.275%

7.075%

6.875%

6.675%

6.475%

6.275%

6.075%

5.875%

5.675%

5.475%

5.275%

5.075%

4.875%

0%
4.675%

Percentage of Earnings

Figure 10: Comparison of capital requirements during countercyclical buffer period and normal period

Why Accenture
There are many reasons why Accenture
is the right partner for risk analytics
initiatives. Accenture has a Risk
Analytics Network with experienced
professionals from local Asia Pacific
countries. Our people have extensive
in-market experience, with broad and
diversified modeling skills. They also
bring broad industry insights, knowledge,
and familiarity with industry specific
benchmarking standards to each client
assignment.
Accentures broad corporate knowledge
acquired through the years by working
with leading firms developing end-to-end
solutions allows us to support clients in
important transformation projects and
initiatives. Our risk analytics services give
clients access to a mature quantitative
methodology, qualitative assessment
capabilities in addition to a systematic
approach, and proprietary assets to assist
them in their risk analytics capability
development and implementation.

Accentures approach focuses on

collaboration, prioritizes what needs to
be undertaken early in an assignment
and aligns itself with a clients needs and
comfort level. This delivers quick wins to
stimulate organizational confidence, buyin and create focus and momentum.
Accentures solutions can accommodate
an adaptable business strategy, operating
model and solution architecture.
These solutions also provide the
necessary flexibility, built on scalable
platforms to meet future needs and
growth opportunities and respond to
evolving environmental challenges,
including new regulatory requirements
to seize a competitive advantage.

12

Asia Pacific Risk

Analytics Network

Tokyo
Shingo Yamamoto
shingo.yamamoto@accenture.com
Global Lead Risk Analytics
Direct: +81 3 3588 3820
Phillip Straley Mobile: +090 8812 1373
phillip.straley@accenture.com
India
Direct: +852 2249 2939
Sanjay Ojha
Mobile: +852 9186 2929
s.ojha@accenture.com
Singapore
Direct: +91 124 467 2191
Mobile: +91 995 369 0574
Christopher Loh
christopher.loh@accenture.com
Direct: +65 6410 6450
Mobile: +65 9069 3860
Beijing
Kent Tianshi Xu
kent.tianshi.xu@accenture.com
Direct: +86 10 5870 5881

13

14

About Accenture
Management Consulting
Accenture is a leading provider of
management consulting services
worldwide. Drawing on the extensive
experience of its 16,000 management
consultants globally, Accenture
Management Consulting works with
companies and governments to achieve
high performance by combining broad
and deep industry knowledge with
functional capabilities to provide
services in Strategy, Analytics, Customer
Relationship Management, Finance &
Enterprise Performance, Operations, Risk
Management, Sustainability, and Talent
and Organization.

About Accenture
Accenture is a global management
consulting, technology services
and outsourcing company, with
approximately 266,000 people serving
clients in more than 120 countries.
Combining unparalleled experience,
comprehensive capabilities across
all industries and business functions,
and extensive research on the worlds
most successful companies, Accenture
collaborates with clients to help them
become high-performance businesses and
governments. The company generated
net revenues of US$27.9 billion for the
fiscal year ended Aug. 31, 2012. Its
home page is www.accenture.com.

About Accenture Risk

Management
Accenture Risk Management consulting
services works with clients to create
and implement integrated risk
management capabilities designed
to gain higher economic returns,
improve shareholder value and
increase stakeholder confidence.
For more information about
Accenture Risk Management please visit
www.accenture.com/riskmanagement

Copyright 2013 Accenture

All rights reserved.
Accenture, its logo, and
High Performance Delivered
are trademarks of Accenture.

13-2558