0% found this document useful (0 votes)
35 views52 pages

Report 19 PDF

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 52

OPERATIONAL RISK MEASUREMENT

FOR THE INDIAN BANKING SECTOR:


ALTERNATIVE MEASURES
by
Romar Correa & Swati Raju
University of Mumbai (Department of Economics)
200809

1
INTRODUCTION
Banks, as major financial intermediaries, possess an inherent
advantage in assessing the riskiness of potential borrowers.
At the same time, the substantial credit, market and
operational risks they face in their operations can have a
significant impact on their earnings. Thus, risk management
which is synonymous with bank management is the core
activity for any bank. The ability of a bank to be successful/
survive under adverse economic conditions is related to the
quality of its risk, its management processes, and to its capital
adequacy. Capital adequacy not only helps prevent individual
bank failures but also helps create a sound and safe banking
system. While regulation exists in several industries, active
prudential regulation through the imposition of minimum
regulatory capital requirements (MRCR) is special to the
financial sector.
The discussion on the optimal level of capital requirements
or the minimum regulatory capital requirements has been
continuing since the 1970s. Till then, the optimal level of
capital that a bank should maintain was considered to be an
internal decision of the bank as there was no explicit
relationship expressed between the level of capital adequacy
and the level of risk faced by a bank. Further, the safety and
soundness of the banking system was promoted through a
control on the level of competition through restrictions on
entry and expansion of bank branches. A shift, however, in
banking to a deregulated, competitive and internationally
active and inter-dependent system saw regulators expressing
a felt need for active prudential regulation in terms of
development of uniform capital adequacy ratios for
internationally active banks. Common minimum regulatory
capital, apart from providing a safe and sound banking system,
was also expected to ensure a level playing field among banks
from different countries. The Basel I Accord of July 1988
was the first formal documentation of active prudential
regulation and established an explicit link between the level

of bank risk and its capital adequacy. The Accord prescribed


a minimum level of regulatory capital for internationally active
banks which required banks to hold a minimum regulatory
capital of 8% or greater of the sum of risk-weighted bank
assets (the Cooke Ratio). Further, it defined the concept of
regulatory capital and its division into Tier I (core) and Tier
II (supplementary) capital and also specified the risk weights
for the different assets. The 1996 Amendment of the Basel I
Accord addressed the issue of market risk and the regulatory
capital requirement for market risk was incorporated in the
capital adequacy framework through the introduction of a
Tier III in supplementary capital.
It is pertinent to note here that the definition of regulatory
capital is open to refinements as can be observed from the
changes introduced in the minimum required capital
requirement under Basel II which introduced an additional
capital requirement for operational risk and made
modifications to credit risk.
The Basel II Accord is based on three mutually reinforcing
pillars Pillar 1 Minimum Capital Requirements; Pillar 2
Supervisory Review, and Pillar 3 Market Discipline. The idea
is that the enforcement of minimum capital requirements
along with an effective and strong supervisory process and
disclosure of risk and market discipline can ensure a sound
financial system.

Pillar I Minimum Capital Requirements


The Basel II Accord introduced substantial changes under
Pillar I to the regulatory capital requirements as regards credit
risk and introduced an additional capital charge for operational
risk. The total regulatory capital requirement is an addition
of minimum capital requirement under credit risk, market
risk, and operational risk. Banks under Basel II can choose
from among any three of the following approaches for credit

374 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
risk the Standardized Approach and between two
sophisticated Internal Ratings-Based Approaches
Foundation and Advanced.
The Standardized Approach is the simplest and defines fixed
risk weights for all credit exposures. The substantial difference
from Basel I is that exposures to the same class of risk can be
assigned different risk weights depending on the assets
external credit rating. Banks could alternatively adopt either
of the more sophisticated Internal Ratings-Based Approaches
where the risk weights assigned are a function of the
probability of default of the borrower, loss given default,
exposure at default and effective maturity. In the Foundation
Internal Rating-Based Approach, the bank combines the
internal estimates for probability of default with fixed
parameters set by the regulator for the other three parameters,
while in the Advanced Internal Ratings Based Approach all
the four parameters of concern may be provided by the bank.
Vital to the Internal Ratings Based Approach is a validation
of the internal process by the supervisor/regulator.
The other substantial change introduced under Basel II was
the introduction of a capital charge for operational risk. Banks
can choose from any of three approaches Basic Indicators
Approach, the Standardized Approach and the Advanced
Management Approach whilst determining the capital
charge for operational risk. Each of the three approaches is
increasingly complex and sophisticated. (A detailed
discussion on operational risk is in the next section)
The overall approach under Basel II, thus, has been to give
banks the choice to adopt approaches with different levels of
sophistication with the more advanced approaches resulting
in a lower level of minimum regulatory capital requirement
as compared to the simpler approaches. Since regulators were
not willing to accept a reduction in average capital
requirements, a trade-off between the approach adopted by a
bank and the level of capital requirement was acceptable to
regulators as banks that adopted the more sophisticated
approaches would have the advantage of better risk
measurement and management systems. Thus, under Basel
II banks could have differing capital requirements dependent
on the approach selected.

Pillar 2 Supervisory Review Process


The focus under this segment is on the role of processes in
bank risk management. It draws attention to the role of the
national regulator in bringing about improvements in banks
risk management techniques and procedures, provisioning
policy, and their capital management processes. The Basel
Committee on Banking Supervision has specified the
following four key principles for supervisory review:

(i)

Banks should have a process for assessing and


maintaining overall capital adequacy with respect to
their risk profile.
(ii) Supervisors should review banks internal capital
adequacy assessments and strategies, taking proper
actions if they are unsatisfied with those processes.
(iii) Supervisors should expect banks to operate above Pillar
1 minimum regulatory capital requirements and should
have the ability to require banks to hold additional
capital; and
(iv) Supervisors should look for early intervention to prevent
a bank from falling below the minimum regulatory
capital requirement and take remedial action if the bank
is under-capitalized.

Pillar 3 Market Discipline


This facet of the Accord focuses on strengthening market
discipline. It concerns the pressure put on bank managements
by financial markets to provide a sound banking system. The
focus under this segment is on qualitative and quantitative
disclosure requirements with an aim to increase the
transparency of a banks risk profile.
Section II of the paper contains the conceptual framework of
the pricing and measurement of operational risk, wherein
Section II.A focuses on the approaches proposed by the Basel
II Accord, while Section II.B discusses the alternative
approaches to operational risk capital measurement. Section
III discusses the Basel II Accord and its relevance to India
while the computational results of operational risk capital
charge under different approaches and its impact on Tier I
capital of banks and the sensitivity analysis of gross income
of a bank to the gross income from its business lines is
contained in Section IV. Section IV.A presents the results for
the Basel II Approaches while Sections IV.B and IV.C contain
the results of the Alternative Approaches and sensitivity
analysis respectively. Section V concludes the paper.

II Operational Risk: Concept and Measurement


Management of credit and market risks has traditionally been
at the centre of bank risk management. Operational risk must
be distinguished from credit risk and market risk. For one
thing, there is no equivalent to the concept of risk exposure.
That is to say, Operational Risk does not correspond in a
simple fashion to any financial indicator. Secondly, the
distribution of Operational Risk is more fat-tailed than that
of credit risk. In addition, Operational Risk is endogenous
relative to credit and market risk. In other words, the scope
for reduction of risk are greater in the case of Operational
Risk. Operational Risk is founded on the premise that a bank,

Chapter 1: Introduction

independent of outside factors, will fail to meet one or more


operational targets in a given year. Operational Risk and its
management has garnered substantial attention since the mid1990s as a consequence of banking crises resulting from
human error, fraud and/or missing controls (e.g. Barings
Bank, Daiwa Bank and Allied Irish) and due to the intent of
the Basel Committee on Banking Supervision since 1999 to
introduce a new regulatory capital charge for Operational Risk
in addition to the minimum regulatory capital requirement
for credit and market risk. Further, technology and increased
product complexity has led to a greater focus on the
management of Operational Risk rather than its mere
measurement.
It is appropriate to begin with a compendium of definitions
and approaches along with their limitations (Bonsn, Escobar
and Flores, 2007). The Commonwealth Bank of Australia
(1999) defines Operational Risk as all risks which would
generate volatility in a banks reserves, expenses and the value
of its business. Others would confine the measure to
unpredictability in its cost structures and exclude its revenue
structure. The proposed checklist for the European Union is
as follows (Oesterreichische Nationalbank, 2006): Interest
income and non-interest income include interest receivable
and similar income, income payable and similar changes,
income from shares and other variable/fixed-yield securities,
commissions and fees receivable, commissions and fees
payable, net profit and net loss on financial operators and
other operating income. The following data is not to be used
in computing the indicator: Realized profits/losses from the
sale of non-trading items, income from extraordinary or
irregular items and income derived from insurance. Care is
taken to ensure that the indicator is calculated before the
deduction of provisions for operating expenses. The latter
include fees paid for outsourcing services provided by third
parties which are not a parent or a subsidiary of the bank or a
subsidiary of a parent which is also the parent of the bank. If
revaluation of trading book items is part of the profit-andloss statement, revaluation must be included in the calculation
of the indicator. A limitation of these concepts is that indicators
of interest and non interest income only reflect the volume of
business in each line but not the level of Operational Risk.
The risk adequacy of the capital requirement calculated on
the basis of these indicators is inadequate as bank-specific
loss data is not used. Consequently, it is not possible to effect
a control of Operational Risk tailored to their causes and
targeted risk management. In addition, the potential
diversification effects between business lines is not factored
by aggregating the amounts of capital. According to the Basel
Committee on Banking Supervision (BCBS), the financial

375

impact of a loss event includes all out-of-pocket costs and


excludes opportunity costs and foregone revenues.
The Basel Committee on Banking Supervision has defined
operational risk as the risk of loss resulting from inadequate
or failed internal processes, people and systems or from
external events. This definition includes legal risk, but
excludes strategic and reputational risk (Basel Committee,
2004). Operational risk, thus, is the risk arising from
operational loss and operational losses in turn emerge from
operational errors. In other words, operational risk is
concerned with the risk to a banks performance due to the
management of the bank as opposed to the financing of the
bank. Consequently, when operational risk is not addressed
systematically it can result in inconsistent performance and
earnings surprises for the stakeholders. Thus, operational risk
exposures can have an impact on banks revenues and net
worth.
Operational risk, thus, generates operational losses and the
losses generated are a cost to the bank. Hence, the pricing
and the consequent measurement of the operational risk
capital charge has to be adequate to cover for these losses.
Therefore, to price operational risk appropriately, a bank
would need a measure of expected loss which is based on its
history of operational losses. This would require the
development of an internal as well as external database on
the sources and types of operational loss exposure in which
every single operational loss event is recorded from which
the mean and standard deviation of losses for the relevant
time period can be computed. However, the creation of such
a comprehensive database is fraught with several problems,
namely, identifying the range of business activities across
which operational loss events might be classified and
allocation of a loss event to a particular area of activity.
Besides, the database can only provide ex-post guidelines on
the potential sources of operational risk events but not exante signals. Further, at times, bank managements could write
off small operational losses against revenue which raises
issues regarding the credibility of the operational loss
database. The problem of credible databases can be redressed
through the creation of an independent external database
which can be shared among banks. Another matter concerns
the appropriate time frame over which data needs to be
collected for a study of operational loss exposures. Apart from
the issues related to data collection, there are statistical issues
such as the nature of the underlying distribution of loss
exposures. The distribution of loss exposures most likely
would be non-normal and skewed as there would be a large
number of small losses and infrequent and decreasing
numbers of large operational loss events.

376 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Correlations between different types of operational loss events
could pose another statistical problem and one needs to
investigate whether these correlations are stable. Among the
problems here are the assumptions to be made about the
correlation of operational loss events. The Basel II Accord
assumes that all operational risk losses occur simultaneously.
The simple summation of high percentile value at risks implies
the simultaneous occurrence of a set of worst-case scenarios.
The issue here is that it is hard to compute the degree of
correlation between various risk types and/or banks because
of the absence of historical data. Both top down and bottom
up approaches, elaborated upon below, rely on historical data.
The latter are based on loss events in individual processes
whereas the former operate at the level of the bank moving
down business lines. Top-down models are likely to superior
at estimating capital requirements. According to the factor
approach, an attempt is made to identify important
determinants of Operational Risk (OR) either at the level of
the banks or at the level of individual business lines. A formula
like
m

OR = t i Fi +
i =1

is used. The Fis are the risk factors. The approach covers risk
indicators, CAPM-like models, and predictive models. In the
risk indicators methodology, regression analysis is used to
identify risk factors like the volume of operations, credit
ratings, and employee turnover. CAPM-based models are used
to connect the volatility of returns to OR variables. With
predictive models, discriminant analysis is used to single out
the elements that lead to OR losses.
Correlation, if any, between operational, market, and credit
risk must be taken into consideration as it will have an impact
on a banks capital requirements. For instance, it can lower
bank capital requirements. A failure to take into consideration
such correlations might result in a bank being over-capitalised.
The Accord specifies three distinct approaches to compute
capital requirements for Operational Risk based on increasing
risk sensitivity and allows banks to adopt different approaches
to different operations. Banks, however, will not be allowed
to revert to a simpler approach from a sophisticated approach
(except under particular circumstances). The idea is to ensure
that banks do not cherry pick among approaches to reduce
their capital charges. Further, each approach has certain
qualifying qualitative and quantitative standards (King, 2001;
Saita, 2007; Tripe, 2000).

II.A Operational Risk Capital Charge: Basel II


Approaches
The three approaches for computing operational risk capital
charge are (i) the Basic Indicator Approach (BIA), (ii) the
Standardized Approach (SA) and (iii) the Advanced
Measurement Approach (AMA), each of which is detailed
below. The identification and measurement of operational risk
can be viewed as following either the top down or the bottom
up mechanism depending on the method used to calculate
the risk charge. In the top down approach, financial data is
extracted from the balance sheet and Profit & Loss statement.
This method may not result in the proper capturing of risks
nor does it help in risk mitigation. This approach corresponds
with the Basic Indicator and the Standardized Approaches of
the Basel II Accord. The third approach of the Accord, the
Advanced Measurement Approach, is consistent with the
bottom up approach in which the regulatory capital
requirement will be defined by the estimate generated by the
internal operational risk measurement system. No eligibility
priors are needed for using the Basic Indicators Approach
because that approach is the default position designed for
small local banks. According to the standardized approach,
the business activities of a financial institution are divided
into standardized business lines and assigned relevant
indicators, net interest income and net non-interest income.
In business line mapping, banks must enunciate principles
and provide documentary evidence for mapping net income
from their own current activities into the standardized
framework. The principles include :
1. The mutually exclusive and exhaustive nature of the
mapping from activities into business lines and 2. Costs
generated in one business line and imputed to a different
business line, may be reallocated to the business line to which
they pertain. For instance, a formulation based on internal
transfer costs between the two activities may be used.
Internationally active banks or those exposed to significant
Operational Risk are expected to use more elaborate
approaches than the Basic Indicators Approach. Lars
Svensson notes the Basel Paradox here, that internationally
active banks might not meet the eligibility criteria to use
sophisticated approaches (Mussa, 2007)
(i) The Basic Indicator Approach (BIA)
The BIA is the simplest of the three approaches to calculating
operational risk capital charges. This approach uses a single
indicator, gross income, as a proxy for a banks overall
operational risk exposure. Minimum capital requirements
under BIA is a percentage (equal to fifteen percent) of the
average of positive gross income (GI) over the preceding three

Chapter 1: Introduction

377

years. Years with negative gross income are excluded. Gross


income is computed as net interest income plus net noninterest income, gross of any provisions and operation
expenses,

broad indicator which serves as a proxy for the scale of


business operations and the operational risk exposure within
each of these business lines. A detailed mapping of the
activities under each business line is provided in Annexure I.

K BIA = GI1,...n * n

Table 1. Standardized Approach Business Units


and Business Lines

where

Business Units

KBIA capital charge under the Basic Indicators Approach

Investment Banking

GI

annual gross income of a bank in a given year

the number of previous three years in which gross


income (GI) is positive.

The advantage of this approach lies in its simplicity and ease


of implementation. The BIA can be applied universally and
allows for easy comparison across banks. However, while
this approach is suitable for small banks, internationally active
banks with substantial operational risk exposure would need
to adopt a more sophisticated risk management and
measurement approach within the overall framework.
(ii) The Standardized Approach (SA)
This approach is a refinement over the Basic Indicators
Approach and can reflect better the differing risk profiles
across banks as reflected by their broad business activities.
Under the Standardized Approach, a banks operational risk
capital charge is sensitive to the risk arising from the various
business lines.
Under this approach, a banks activities are divided into a
number of standardized business units and business lines.
The SA can, hence, better reflect the differing risk profiles
across banks as reflected by their broad business activities.
However, similar to the BIA, the capital charge for the
different business lines is standardized by the supervisor. The
business lines proposed under the SA reflect an industry
initiative to collect internal loss data in a consistent manner.
A broad financial indicator has been specified for each
business line and the indicator would reflect the size/volume
of a banks activity. The indicator can serve as a rough proxy
for the amount of operational risk within each of these
business lines. The operational risk capital charge within each
of these business lines is calculated by multiplying a banks
broad financial indicator by a beta factor. The beta provides
a rough proxy for the relationship between the Operational
Risk loss experience for the industry for a given business
line and the financial indicator representing a given banks
activity in that particular business line. Table 1 below presents
the eight businesses into which a banks activities can be
decomposed. Within each business line, gross income is the

Banking

Others

Business Lines

Beta Factors

Corporate Finance

18%

Trading & Sales

18%

Commercial Banking

15%

Retail Banking

12%

Payment and Settlement

18%

Retail Brokerage

12%

Asset Management

12%

Agency Services

15%

For instance, the operational risk capital charge for the


business line of corporate finance under investment banking
would be calculated as follows:
Kcorporate finance = corporate finance * Gross income
where
Kcorporate finance capital requirement under corporate finance
business line
corporate finance Beta factor for the corporate finance
business line.
It is relevant to observe that each business line has its assigned
beta factor and a respective financial indicator. Gross income
here refers to the gross income of a bank for that particular
business line and does not refer to the gross income of the
whole bank. Ideally, the beta factor for each business line
should be calibrated as per its loss experience and the goal of
the Basel Committee is to re-calibrate the SA when credible
risksensitive data (loss experience information) is available.
Such a mapping will enable each bank to map its structure
into the regulatory framework. Since most banks are in the
process of developing an internal loss database or may not
choose to incur the investment required to develop an internal
loss database for all business lines (especially for those
business lines that present a less material operational risk),
in the interim the SA could provide the required framework
other than the BIA to calculate their regulatory capital charge.
Alternatively, the SA can provide a basis on which a bank
can move to more sophisticated approaches like the Advanced
Management Approach that would help develop better risk
management techniques within banks.
The Basel Committee on Banking Supervision has set
qualifying criteria that a bank must satisfy before it can adopt

378 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
the Standardized Approach (items 660-663 of BCBS Report,
June 2006). The total capital charge is computed as a three
year average of the simple summation of the regulatory capital
charges across each of the business lines in each year. For
any given year, the negative capital charges (resulting from
negative gross income) for a business line may offset positive
capital charges in other business lines without limit. However,
if the aggregate capital charge across all business lines is
negative for a year, then the input for the numerator for that
year would be zero. The total capital charge may be expressed
as:

K = max jY j , 0 max
i =1

j =1
3

The treatment of negative values here is not different: In each


year, a negative capital requirement in one business line
resulting from negative gross yield may be imputed to the
whole. When, however, in a given year, the aggregate capital
charge accruing to all business lines is negative, the element
in the numerator for the year is zero.
A criticism that has been made in this context is that risk
management in terms of risk control is not sufficient because
the capital requirement is not determined by the actual
Operational Risk but by the level of net interest income and
net non interest income. The assumption is that, in general,
higher income can only result from accepting higher
Operational Risk. However, improved performance can also
result from superior risk management techniques.
(iii) Advanced Management Approach (AMA)

Banks may be allowed by national supervisors to adopt the


Alternative Standardized Approach (ASA) in which retail
banking and commercial banking operational risk exposures
can be calculated by applying betas to a percentage m of
outstanding loans rather than gross income. Capital charges
for retail and commercial banking are calculated differently.
Instead of using gross income as the indicator of exposure,
the value of loans and advances is used. Gross income is
substituted by a number that amounts to 0.035 times the value
of loans and advances. Thus,

computed separately for retail and commercial banking. K is


the capital charge against retail banking, the beta assigned to
it is 0.12 and L is the total outstanding loans and advances
(retail banking, non risk-weighted and gross of provisions)
averaged over the previous year. A bank would be allowed at
the discretion of the supervisor to adopt the ASA only when
it is able to convince the supervisor that the ASA would
provide a better basis for the calculation of regulatory capital.
A bank would not be allowed to revert to the Standardised
Approach from the ASA without obtaining the prior sanction
of the supervisor (BCBS Report 2001, 2006; Saita, 2007).

This approach identifies potential risk areas for each line of


business based on historical data and the frequency of their
occurrence and size of loss. The AMA provides discretion to
individual banks on the use of internal loss data while the
method to calculate the required capital charge is uniform to
all banks and established by the supervisor. Further, banks
while adopting this approach will need to satisfy several
quantitative and qualitative criteria (item 664-674, BCBS,
June 2006) which would ensure the integrity of the
measurement approach, data quality, and internal
measurement processes. As the AMA is the most sophisticated
in the spectrum of approaches available to measure
operational risk capital, the Basel Committee believes that
the adoption of this approach will incentivize banks to develop
a credible internal loss database. The Committee recognizes
that the industry is currently at a nascent stage in developing
the database necessary for the implementation of the AMA
and, consequently, some re-calibration would be required at
a later date. Thus, under the AMA, the regulatory capital
requirement will equal the risk measure generated by the
banks internal Operational Risk measurement system and
its adoption is subject to regulator approval

An illustrative specification from the European Union


Directive referred to is as follows:

Under the AMA the operational risk capital charge is to be


determined through the following procedures:

K = 0.035L

Corporate finance
Trading and sales
Payments and settlement
Commercial Banking
Agency services
Retail brokerage
Retail banking

1
2
6
4
7
3
5

Asset management

(i)
18%

15%
12%

A banks activities are categorized into a number of


business lines and a broad set of operational loss types
is defined and applied across business lines.
(ii) Within each business line/loss type combination, the
supervisor specifies an exposure indicator (EI) which
is a proxy for the size of each business lines operational
risk exposure.
(iii) In addition to the EI, banks measure based on their
internal loss data a parameter that would represent the

Chapter 1: Introduction

probability of loss event (PE) and a parameter that would


represent the loss given that event (LGE), is calculated.
The product of the EI*PE*LGE is used to calculate the
expected loss (EL) for each business line.
(iv) The supervisor would supply a factor the gamma term
for each business line which would translate the
expected loss into a capital charge. The overall capital
charge for a particular bank is the simple sum of all the
resulting products for the different business lines.
(v) Finally, to facilitate supervisory validation, banks would
provide the supervisor with the individual components
of the expected loss calculation, namely the EI, PE and
LGE, based on which the supervisor would calculate
the expected loss (EL) and then adjust for the
unexpected loss through the gamma term so as to
achieve the desired level of safety and soundness.
The business lines under the AMA would be similar to those
under the SA and the operational risk measure under the AMA
must guarantee standards which are comparable with credit
risk under the Internal Ratings Based Approach. Further, the
regulatory capital requirement should be the sum of expected
and unexpected losses unless the bank can show that the
expected losses have been already captured and accounted
for by its internal business processes. The risk from the
different business lines are additive with no diversification
benefit. However, the bank may be allowed to use internallydetermined correlations after regulatory approval by the
supervisor that the system of determining the correlations is
satisfactory. Thus, crucial to the success of the Advanced
Measurement Approach is accurate risk mapping which would
help identify key risk indicators which, in turn, can provide
anticipatory signals and enable better monitoring and control
action through building of internal and external loss databases.
The Basel II Accord, further, includes guidelines for
constructing internal loss databases or for avoiding double
counting when operational losses are already included under
regulatory capital for credit risk (e.g. collateral management
failures). A bank would further monitor for changes in risk
control and accordingly adjust operational risk estimates.
Further, under the AMA a bank is permitted to consider riskmitigation benefits which cannot be greater than 20% of the
total AMA operational risk capital requirement. The
possibility of loss data being hidden by operational
management can be supplemented by having an appropriate
organizational structure such as independent operational risk
management, and internal loss databases can be compared
with available external databases and scenario analysis to
measure high severity losses. Scenario Analysis contributes
by suggesting what might happen, even if it (the loss event)
never happened before (Saita, 2007, p.125).

379

The strategy under AMA can, thus, be summarized as: (i)


Map potential risks (ii) Measure risk (iii) Implement risk
mitigation measures and (iv) Predict and forecast risks.
Alternatively, operational risk estimates should be derived
by a proper combination of (i) internal data (ii) relevant
external data (iii) scenario analysis and (iv) business
environment and internal control systems. (BCBS, 2004,
2006; Bhatia, 2002; Saita, 2007).
As banks move from the BIA to the AMA, their capital charges
are lowered. The regulatory capital requirement will be
calculated on the basis of the banks own Operational Risk
model. One of the objectives of the Basel II Accord is to
align regulatory capital with the economic capital determined
by the internal models of banks. Economic capital is the
amount of capital that a bank must hold to protect itself, at
chosen confidence intervals, from insolvency due to
unexpected losses over a period of time. Under AMA, banks
must quantify Operational Risk capital requirements for seven
types of risk and eight business lines, giving a matrix with
fifty-six elements. Ignoring correlation, these estimates are
reduced to a summary statistic of the Operational Risk of the
bank.

II.B Operational Risk Capital Charge: Alternative


Approaches
Operational risk emanates not just from the activity of bank
lending (which forms the core of credit risk) but from several
other activities undertaken by a bank. Consequently,
operational risk costs must be built into fees and commissions
for activities such as deposit processing, cheque issuance and
all other activities that can generate operational errors/losses.
The alternative approaches to the pricing of operational risk
in banks focus on the volatility of non-interest expense
account by looking at two ratios: (i) Cost to Asset Ratio and
(ii) Cost to Income Ratio. Further, the use of ratios insulates
the operational risk capital charge from the growth observed
in bank balance sheets. The operational risk capital charge
under these approaches is linked to the volatility in the cost
to asset ratio and the cost to income ratio. Tripe (2000) has
suggested that the operational risk capital charge for a bank
can be computed using a multiple of the standard deviation
(say three standard deviations) of the cost to asset ratio relative
to average total assets and to total income for the cost to
income ratio. He writes this multiple having been selected
for convenience rather than to reflect any particular theoretical
rationale, it is not considered inconsistent with likely practical
approaches (p.11). The rationale for employing multiple
standard deviations can be attributed to the non-normal,
skewed, and fat tailed distribution of the loss exposures.

380 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Hence, multiple standard deviations would cover a smaller
proportion of the possible outcomes than that under a normal
distribution. A fall-out of this would be that banks would
possibly be under capitalized as compared to the actual risk
exposure.
(i) Cost to Asset Ratio
The cost to asset ratio can be defined as the ratio of operating
(non-interest) costs to average total assets. The operational
risk capital charge is linked to volatility in the cost to income
ratio. The ratio of cost to average assets of a bank reflects a
banks business mix. For instance, a bank with a focus on
corporate lending or placements in the inter-bank market would
have a lower cost to asset ratio as compared to a bank with a
greater focus on retail lending. However, a problem with this
ratio is the complications for cross-border comparisons (which
is very relevant for internationally active banks). Banks in
different countries can have different forms of constructing
their balance sheets and this can have an impact on the cost
to asset ratio more than the cost to income ratio. Further, it is
important to note that if banks are undertaking substantial
efforts to reduce costs, then the variations in the cost to asset
ratio will be around a decreasing mean rather than around a
stationary mean making comparisons difficult. Also, the cost
to asset ratio does not capture the non-interest income which
is a significant pointer to the operational risks in a bank. Further,
banks at times may adjust operating losses (small losses)
against revenues. Hence pricing operational risk by studying
the volatility in the cost to income ratio would be an alternative
to the cost to asset ratio.
(ii) Cost to Income Ratio
The cost to income ratio is also known as the Efficiency
Ratio or Expense to Income Ratio. The ratio is used by bank
managements and market analysts to assess bank
performance. The components of the ratio are cost and
income and, hence, the measure is indirectly related to bank
profitability. A reduction in costs for a given level of income
will reflect increased profits and vice versa. Increased profits,
in turn, will result in improved return on equity and share
prices of the bank which is of great interest to investors.
Further, most bank costs have been reducing in response to
margin squeezes, thus lowering both costs and income. Hence,
volatility in a banks cost to income ratio might be a better
measure of volatility in a banks cost performance.
The cost to income ratio is the ratio of non-interest (operating)
costs excluding bad and doubtful debt to the net interest
income plus non-interest income of the bank. Non-interest
costs are perceived as those costs which are most amenable
to management decisions and considered to be that part of a

banks costs which can be controlled. A focus on non-interest


costs would ensure that fluctuations in the level of interest
rates do not affect the volatility seen in this ratio. The use of
the net interest income term in the denominator will reduce
the volatility that could arise from fluctuations in the general
level of interest rates.
The rationale for the exclusion of bad and doubtful debt can
be attributed to the following: (i) Bad and doubtful debt
largely reflects bad credit decisions made in the past rather
than current performance. (ii) The cost to income ratio would
be adversely affected by major write-offs, if any, undertaken
at points of time in the future. (iii) Such assets can distort the
ratio as well as reflect high levels of operating costs and low
levels of income.
Since the cost to income ratio is affected by changes in both
costs as well as incomes, the ratio needs to be interpreted
with caution. An increase in the ratio on account of falling
income needs to be studied so as to determine whether the
fall in income is because of the banks inability to generate
income, thereby indicating inefficiency or is attributable to a
change in competitive conditions which reduce margins across
the board or a change in overall economic conditions which
restrict opportunities to undertake profitable business from
which a bank can earns fees.
The ratio is also sensitive to individual bank structure in terms
of the strategy adopted for deposit mobilisation, spread of
bank branch networks and business mix. Another factor that
can have an impact on the cost to income ratio is banks holding
excess capital or banks that are over capitalized. Banks
holding excess capital are in a position to undertake greater
wholesale lending or investment at low cost thus increasing
their gross income with the same level of operating costs
resulting in a lower cost to income ratio. Tripe (1998) has
detailed, through several illustrations, the impact of differing
bank structures on the cost to income ratio.
(iii) Range
Range between the maximum and minimum cost ratios can
be looked at as an alternative to standard deviation for
measuring the operational risk capital requirement. Such an
estimate can be adopted when observations are sufficiently
large and is based on a methodology like back simulation.
However, an estimate based on range would be larger than
that under standard deviation as the range extends on both
sides of the mean (Tripe, 1998, 2000).

III The Basel II Accord and India


The Reserve Bank of India (RBI) has adopted a consultative
approach as regards the implementation of Basel II. A Steering

Chapter 1: Introduction

Committee comprising of senior officials from fourteen banks


(private, public, and foreign) has been constituted alongside
a representation of the Indian Banks Association. Keeping
in view the RBI's goal to be consistent and harmonious with
international standards, it has been decided that at a minimum,
all banks in India will adopt the Standardized Approach for
credit risk and the Basic Indicator Approach for operational
risk with effect from March 31, 2007. After adequate skills
are developed, both in banks and at supervisory levels, some
banks may be allowed to migrate to the Internal Ratings Based
Approach after obtaining the specific approval of the Reserve
Bank of India. Further, the Reserve Bank of India will review
the capital requirements produced by the Basic Indicator
Approach for general credibility and, in the event that
credibility is lacking, appropriate supervisory action under
Pillar 2 will be considered. Banks will be encouraged to move
along the spectrum of available approaches as they develop
more sophisticated operational risk measurement systems and
practices (Reserve Bank of India, 2005). While the RBI has
mandated the BIA approach, banks are aware that they must
progress toward the AMA approach. One problem that has
been recognised is that low frequency, high impact data, by
definition, must be collated. Therefore, a data exchange by
the Indian Banks Association has been proposed along the
lines of the Global Operational Loss Database (GOLD) set
up by the British Bankers Association. In India, business
continuity planning (BCP) is a part of Operational Risk
(Reserve Bank of India, 2008A). The context is the increased
leverage of technology. In that case, disaster recovery is an
important component of the BCP programme directed towards
the recovery of technology. An effective BCP must factor in
the possibility of disasters covering an entire region and the
resulting attrition of staff. Thus, the BCP methodology
includes the IT continuity template and formulating recovery
time objectives (RTO) based on Business Impact Analysis.
The model must be robust enough to contend with the most
stressful situations. The recovery point objectives (RPO) for
data loss in the case of each critical business will have to
identified along with strategies to deal with them.

381

IV Pricing Operational Risk: Measurement for


Indian Banks
This section presents the estimates of operational risk capital
measurement for Indian banks and its consequent impact on
Tier I capital under a broad spectrum of approaches, namely,
the Basic Indicators Approach, the Standardized Approach,
the Cost to Asset Ratio and the Cost to Income Ratio. The
study estimates the operational risk capital charge at an
individual bank level as well as bank group-wise depending
on the availability of data. When the estimates have been
presented at the individual bank level, it covers 30 banks
including the 27 public sector banks (19 nationalized Banks
and 8 banks of the State Bank group) and the 3 leading new
private sector banks viz. ICICI Bank, HDFC Bank and UTI/
Axis Bank. The period of the study varies as per the Approach
and the availability of data. The time period of the study spans
2005-2008 for three of the four approaches - the Basic
Indicators Approach, the cost to asset ratio and the cost to
income ratio, while it spans the period 2006-2007 for the
Standardized Approach. The data has been sourced from
various issues of the Reserve Bank of Indias Basic Statistical
Returns of Scheduled Commercial Banks in India and
Statistical Tables Relating to Banks in India and Report on
the Trend and Progress of Banking in India.

IV.A Basel II Approaches


(i) The Basic Indicators Approach
This approach is to be adopted by all banks in India from
March 2007. This approach, as detailed above, would require
banks to keep aside a percentage ? (equal to 15%) of positive
gross annual income over the past three years excluding any
year where gross income is negative as operational risk capital
charge. Table 2 presents the additional capital charge
requirement for operational risk under the Basel II Accord
for the major bank groups.

Table 2. Operational Risk Capital Charge and Impact on Tier I Capital


Bank Group-wise (Basic Indicators Approach)

Year

Gross Income
(Rs. Crore)

Annual Capital
Required

Capital Charge
(Rs. Crore)

Networth
(Rs. Crore)

Current Tier I
(%)

Estimated Tier I
(%)

Public Sector Banks


2002

48095

7214

57454

7.8

2003

58611

8792

65582

8.3

2004

71822

10773

79225

8.2

2005

75802

11370

8926

85946

7.4

6.6

2006

79276

11891

10312

115044

8.8

8.1
Table contd...

382 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
2007

85909

12886

11345

135626

8.3

7.6

2008

97039

14556

12049

174853

7.4

6.9

2009

13111

Nationalised Banks
2002

29888

4483

37395

7.1

2003

36952

5543

42450

7.9

2004

45343

6801

51475

8.1

2005

46298

6945

5609

61181

7.6

6.9

2006

46829

7024

6430

71012

9.0

8.2

2007

53945

8092

6924

84385

8.5

7.8

2008

60310

9047

7354

104325

7.5

7.0

2009

8054

State Bank Group


2002

18207

2731

20059

9.5

2003

21660

3249

23133

9.3

2004

26480

3972

27750

8.5

2005

28690

4304

3317

32764

7.9

7.1

2006

30787

4618

3842

37660

8.5

7.6

2007

30278

4542

4298

42942

7.7

6.9

2008

34437

5166

4488

61706

7.2

6.7

2009

4775

Old Private Sector Banks


2002

4451

668

5411

10.4

2003

4952

743

6295

10.6

2004

5569

835

7291

10.9

2005

4852

728

749

7926

9.0

8.1

2006

5346

802

769

9671

9.3

8.5

2007

5987

898

788

10737

11.2

10.3

2008

6838

1026

809

15315

14.1

13.4

10514

7.5

2009

909

New Private Sector Banks


2002

4058

609

2003

8205

1231

13404

8.2

2004

10163

1524

14846

7.0

2005

11507

1726

1121

18981

9.9

9.4

2006

16838

2526

1494

24314

8.8

8.3

2007

23035

3455

1925

33075

8.1

7.7

2008

32665

4899

2569

49332

10.3

9.8

2009

It is important to mention here that even while banks in India


are to introduce the operational risk capital charge from March
2007, we have estimated this capital charge from 2005. Apart
from giving a historical perspective to the impact of the
operational risk capital charge, the exercise will also provide
a useful continuation to an early study of ICRA in this area.
ICRA in its estimates had suggested that in 2005 scheduled

3627

commercial banks would need Rs.120 billion as additional


capital requirements for operational risk. Of this, a substantial
amount would be needed by public sector banks comprising
the nationalized banks and the State Bank group followed by
the new private sector banks and the old private sector banks.
While the pattern of capital requirements remains the same
at the end of March 2007, the focus here is on the period

Chapter 1: Introduction

since March 2007 when Indian banks would start


implementing the Operational Risk capital charge under BIA.
Table 2 details the additional capital requirement on account
of the Operational Risk capital charges for the Indian banking
sector excluding foreign banks in India and estimates indicate
that the requirements on account of operational risk capital
charges would be of the order of Rs.14,058 crores at the end
of March 2007 and Rs.15,427 crores in March 2008. The
requirement is estimated to be Rs.17,647 crores for March
2009. Furthermore, the requirement at the end of March 2009
is about 1.63 times greater than the requirement at the end of
March 2005 which was approximately Rs.10,796 crores. A
substantial proportion nearly 80.70% (at Rs.11,345 crores)
of the overall additional capital requirements on account of
Operational Risk capital charge would form the requirement
of the public sector banks at the end of March 2007. While
the public sector banks have seen an increase in the absolute
amount of capital requirements, the percentage as a proportion
of the overall capital requirements for public sector banks
has since declined and is estimated to be around 74.30% at
Rs.13,111 crores in 2009 of the overall additional capital
requirement on account of the Operational Risk capital charge.
Compare this with the new private sector bank group whose
additional capital requirements for Operational Risk, though
substantially lower than the public sector banks, witnessed
an increase both in percentage terms as well as in the absolute
amount of capital required since March 2007. At end March
2007, the capital requirements of new private sector banks
was estimated to be Rs.1925 crores, which was 13.69% of
the overall requirement for the banking sector (excluding the

foreign banks) and the percentage share is estimated to be


around 20.55% (Rs.3627 crores) at the end of March 2009.
The bank group of the old private sector banks has seen an
increase in absolute amount of capital requirements from
Rs.788 crores at the end of March 2007 to Rs.909 crores at
the end of March 2009. The percentage share of this group
however remains stable in the range of 5.15 5.61% over
the period March 2007-09. The impact of this additional
capital requirement has had the expected impact on lowering
Tier I capital across all bank groups. On an average for 2007
and 2008, the impact of lowering the Tier I capital has been
the highest for the old private sector banks at around 0.45%
followed by the public sector banks at 0.6% and the lowest
for the new private sector banks at 0.45%. The Operational
Risk capital requirement for each of the individual 30 banks
on an average for 2007-09 is encapsulated in Table 3 while a
detailed analysis is in Annexure III. Table 3 highlights the
fact that the Operational Risk capital charge for 23 of the 30
banks under consideration would, on an average for 200708, be below Rs.500 crores. Five banks would require an
operational risk capital charge in the range of Rs.500Rs.1000
crores, while the State Bank of India at Rs.2203 crores and
ICICI Bank at Rs.1423 crores would need an additional capital
requirement of above Rs.1000 crores. A graphical presentation
of the comparative requirements of Operational Risk capital
charges of each individual public sector bank vis--vis the
overall requirement of the public sector banks and of each of
the three leading private sector banks vis--vis the overall
requirement for the new private sector banks for the period
2005-2009 is presented in Annexures VI and VII respectively.

Table 3. Operational Risk Capital Requirement of Banks: Average 2007-09 Basic Indicators Approach
(Rs. Crore)

Below 250

250-500

BoM (184)

ALLA (315)

BoB (694)

ICICI (1423)

DENA (175)

ANDHRA (261)

BoI (643)

SBI (2203)

PSB (129)

CENTRAL (456)

CANARA (759)

UNITED (213)

CORP (267)

PNB (925)

VIJAYA (198)

INDIAN (330)

HDFC (594)

SBBJ (198)

IOB (414)

SBH (240)

ORIENTAL (327)

SB-IND (117)

SYND (377)

SBM (146)

UCO (300)

SBP (209)

UNION (466)

SBS (91)

AXIS (306)

SBT (197)

Figures in parentheses represent the Operational Risk capital charge.

383

500-1000

Above 1000

384 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Considering that a substantial burden of the additional capital
requirement on account of Operational Risk capital charges
lies on the public sector banks, it would be interesting to study
and compare the impact of the Operational Risk capital
charges on the regulatory Tier I capital. Tables 4, 4A and 4B
summarize the impact on Tier I capital of the additional capital
charges for Operational Risk for nationalized banks, the State
Bank group and the three major new private sector banks
respectively for 2005-08. As can be gleaned from Table 2,
the introduction of the additional capital charges for
Operational Risk has lowered the Tier I capital of banks. As
of end March 2005, if the operational risk capital charge was
assumed to be active, then all others remaining the same, the
estimates from Table 4A indicate that nearly 9 of the 19

nationalized banks would have had a Tier I capital below 6%


while two banks from the State Bank group the State Bank
of Indore and State Bank of Travancore (Table 4B) - would
have a Tier I capital of less than 6%. The Reserve Bank of
India on February 15, 2005, issued guidelines to all Scheduled
Commercial Banks indicating the approach to be adopted
for the implementation of the Basel II Accord in Indian banks.
Subsequently, several nationalized and private banks boosted
their Tier I capital through raising equity in 2005 and 2006
and the Tier I capital of the following banks showed
substantial improvement, namely Allahabad Bank, Andhra
Bank, Bank of Baroda, Dena Bank, Oriental Bank of
Commerce, Punjab and Sind Bank, Punjab National Bank,
Syndicate Bank and Union Bank of India.

Table 4A. Impact of Operational Risk Capital Charge on Tier I Capital


Basic Indicators Approach Nationalized Banks

Tier I Capital
Year

Below 5%

Between 5-6%

Between 6-7%

Between 7-9%

Above 9%

2005

ORIENTAL

ALLA

BoI

ANDHRA

CORP

PSB

CENTRAL DENA

BoM

BoB

UNITED

IOB

CANARA

INDIAN

SYND

VIJAYA

PNB

UCO
UNION
2006

DENA

BoI

ALLA

SYND

BoM

CANARA

ANDHRA
BoB

UCO

CENTRAL

IOB

CORP

UNION

UNITED

INDIAN

VIJAYA

ORIENTAL
PSB
PNB

2007

2008

BoI

CANARA

ALLA

ANHDRA

BoM

UNITED

BoB

CORP

CENTRAL

VIJAYA

IOB

INDIAN

DENA

PSB

ORIENTAL

SYND

PNB

UCO

UNION

BoM

CENTRAL

CANARA

ALLA

CORP

UCO

VIJAYA

DENA

ANDHRA

INDIAN

SYND

BoB

UNION

BoI

UNITED

IOB
PSB
PNB
ORIENTAL

Chapter 1: Introduction

In addition, as indicated by the spread of banks at the end of


March 2007, when banks were to adopt the additional
regulatory capital requirement for Operational Risk under the
Basic Indicators Approach, regulatory Tier I capital adequacy
would be below 6% for 6 of the 19 nationalized banks while
nearly 10 of the 19 nationalized banks would have
comfortable levels of Tier I capital (greater than 7%). A
worrying scenario in the nationalized banks in 2008 is

385

observed for the Bank of Maharashtra and UCO Bank whose


Tier I capital would be below 5% and two other banks
Central Bank of India and Vijaya Bank would have a Tier I
capital in the range of 5-6%.
Annexure VIII contains a graphical presentation of the
Operational Risk capital charge and the impact on Tier I
capital at an individual bank level for nationalized banks

Table 4B. Impact of Operational Risk Capital Charge on Tier I Capital


Basic Indicators Approach - State Bank Group

Tier I Capital
Year

Below 5%

2005

Between 5-6%

Between 6-7%

Between 7-9%

Above 9%
SBP

SB-IND

SBH

SBI

SBT

SBM

SBBJ
SBS

2006

SB-IND

SBI

SBM

SBBJ

SBT

SBH

SBP

SBS
2007

SBM

SB-IND
SBT

SBI
SBBJ
SBH
SBP
SBS

2008

SBM

SBBJ

SBI

SBH

SBS

SB-IND
SBP
SBT
Table 4C. Impact of Operational Risk Capital Charge on Tier I Capital
Basic Indicators Approach Axis Bank, HDFC Bank and ICICI Bank

Tier I Capital
Year

Below 5%

Between 5-6%

Between 6-7%

2005

Between 7-9%

Above 9%

AXIS

HDFC

ICICI
2006

AXIS

HDFC
ICICI

2007

AXIS

HDFC
ICICI

2008

AXIS
HDFC
ICICI

386 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table 4B presents a similar spread for the State Bank of India
and its associate banks. The State Bank of India and some of
its associate banks like the State Bank of Saurashtra, the State
Bank of Bikaner and Jaipur, State Bank of Hyderabad and
the State Bank of Patiala, have shown comfortable levels of
capital adequacy under Tier I capital in 2007. At the end of
March 2008, though, only two banks in the State Bank group
viz. the State Bank of India and the State Bank of Saurashtra
show comfortable levels of Tier I capital greater than 7%
while 5 of the associate banks have Tier I levels in the range
of 6%-7%. The State Bank of Mysore is the lone associate
bank of the State Bank group to have a Tier I capital of less
than 6% after the implementation of the Operational Risk
capital charge in 2007.
Table 4C gives the similar spread for the 3 major new private
sector banks considered in this paper. With the exception of
Axis Bank in 2007, all banks have a Tier I capital greater
than 9% after accounting for the capital requirement under
Operational Risk.
(ii) The Standardized Approach
We attempt to provide an estimate for Operational Risk capital
charge under the Standardized Approach (SA) (though it is
not to be adopted and is the second in the spectrum of
approaches suggested by the Basel II Accord). This approach
is a refinement of the Basic Indicators Approach and under
this approach a banks business is divided into eight business
lines and different percentages ? are applied to each business
lines gross income so as to arrive the Operational Risk capital
charge. Operational Risk under the SA is sensitive to the
decomposition of income from different business lines.
Data constraints limit the computation of the Operational Risk
capital charge under this approach to 2007 and 2008 for banks
group-wise, namely, nationalized banks, the State Bank group
and private sector (old and new) banks. Under the SA, a
banks business lines can be broadly classified into 3 major
lines Investment Banking, Banking and Others. Each of
these 3 lines at Level 1 are, in turn, decomposed into the
following (for details on Activity groups corresponding to
each business line see Annexure II).
Investment Banking (i) Corporate Finance (ii) Trading
&Sales
Banking (i) Retail Banking (ii) Commercial Banking (iii)
Payment & Settlement and (iv) Agency Services
Others (i) Asset Management (ii) Retail Brokerage and
(iii) Insurance.

For most banks in India, a one-to-one correspondence with


these business lines in terms of data availability is difficult.
For instance, income from payment and settlement and agency
services is included under Commission, Brokerage and
Exchange, an item under Other Income in a banks Profit &
Loss account. Further, most banks undertake asset
management and insurance as joint ventures and the income
from this activity could be reflected to an extent under the
item Income from Investments under Interest Earned in a
banks Profit and Loss account. Consequently, the estimation
of Operational Risk capital charge has been restricted to Level
1 and the following heads of activity have been considered,
viz. (i) corporate finance (ii) trading and sales under
investment banking; (iii) Retail Banking and (iv) Commercial
Banking under Banking. It is appropriate, here, to discuss
the correspondence between the income from the different
activity groups of a bank and available data. Data on income
from corporate finance is the income a bank obtains from
investments and other income, while income from trading
and sales is the profit a bank makes from the sale of land,
sale of investment assets and sale of foreign exchange, income
from commission and brokerage and net repo income of the
bank. Net repo income is the difference between interest
earned on balances with the RBI and other inter-bank funds
and the interest expended on borrowings from the RBI and
other inter- bank funds. Some computations had to be
undertaken to derive the income from retail and commercial
banking and is obtained as under:
Income from Retail Banking = credit outstanding against each
occupation for individuals * the weighted average of lending
rate for each of the occupations
Credit outstanding under commercial banking = Total credit
outstanding for each occupation credit outstanding against
individuals
Income from commercial banking = credit outstanding under
commercial banking activity * by the weighted average of
lending rate for each of the occupations
The Operational Risk capital charge is then obtained by
applying the relevant Beta to the gross income from a
particular business line. Table 5 presents the Operational Risk
capital charge and its impact on Tier I capital at the level of
the bank group viz. nationalized banks, the State Bank group
and private sector (old and new) banks under the Standardized
Approach).

Chapter 1: Introduction
Table 5. Operational Risk Capital Charge and Impact on Tier I Capital
Bank Group Wise (Standardized Approach)
Year

Annual

Capital

Capital

Charge

Networth Current Estimated


(Rs. Crore) Tier I

Requirement (Rs. Crore)

Tier I

(Rs.Crore)

Nationalised Banks
2004

11273

51475

8.1

2005

13144

61181

7.6

2006

15292

71012

9.0

2007

16261

13236

84385

8.5

7.2

2008

Na

14899

104325

7.5

6.4

State Bank Group


2004

6524

27750

8.5

2005

7302

32764

7.9

2006

7857

37660

8.5

2007

8241

7228

42942

7.7

6.4

2008

Na

7800

61706

7.2

6.3

387

the State Bank group on an average would need around


Rs.7500 crores and the requirement for the private sector
banks would be lower at around Rs.6000 on an average over
2007-08. Further, the estimated decline in the Tier I capital
for all the bank groups studied is much greater than under
the Basic Indicators Approach (Table2). Also, the extent of
the impact is almost similar across the bank groups the
estimated Tier I capital after taking into consideration the
Operational Risk capital charge is lower by 1% to 1.3% in
2007 and 2008.

IV.B Alternative Approaches


The alternative approaches, as discussed in Section II.B
above, refer to the computation of the Operational Risk capital
charges using the cost to asset and the cost to income ratios.
Both these measures focus on the volatility of non-interest
expenses. While the literature suggests the using of quarterly
data, for comparison with the Basel II Approaches, the
analysis has been done using annual data.
Tables 6A, 6B and 6C show the spread of banks across
different ranges of Tier I capital, when the cost to asset ratio
is used to compute the Operational Risk capital charge
(Detailed results are in Annexure IV).

Private Sector Banks


2004

4332

22137

7.0

2005

5280

26907

9.9

2006

6880

33985

8.8

2007

8473

5497

43812

8.1

7.1

2008

Na

6877

64647

10.3

9.2

The results of Table 5 indicate the additional capital


requirement that has to be set aside by the different bank
groups under the Standardized Approach would be higher
than that estimated under the Basic Indicators Approach
(Table 2). The pattern of capital requirements though remains
similar to that under the Basic Indicators Approach, wherein
the public sector banks would require a larger amount of
additional capital requirements, with the nationalized banks
required to set aside on an average nearly Rs.14,000 crores,

Table 6A underscores the severe impact of Tier I capital on


the nationalized banks using the cost to asset ratio as a measure
of the Operational Risk capital charge compared to that under
the Basic Indicators Approach (see Table 4A). In 2005, the
number of banks that had a Tier I capital below 6%, if we
assumed the Operational Risk capital charge were applicable,
remained the same at 9 banks as under the BIA the
composition had reversed under the cost to asset ratio measure
7 banks had Tier I capital below 5%, while 2 banks were in
the 5%-6% range. Likewise, the performance for the later
years also continued to reflect the severity.

Table 6A. Impact of Operational Risk Capital Charge on Tier I Capital


Cost to Asset Ratio - Nationalized Banks

Tier I Capital
Year
2005

Below 5%

Between 5-6%

Between 6-7%

Between 7-9%

Above 9%

ALLA

BoM

BoI

ANDHRA

CORP

CENTRAL

UCO

CANARA

BoB

UNITED

DENA

INDIAN

PNB

ORIENTAL

IOB

VIJAYA

PSB
SYND
UNION
Table contd...

388 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
2006

DENA
PSB

BoI
BoM
CENTRAL
SYND
UNION

CANARA
UCO

ALLA
INDIAN
IOB
PNB
UNITED

ANDHRA
BOB
CORP
ORIENTAL
VIJAYA

2007

CENTRAL
DENA
PSB
SYND

BoI
BoM
UCO
UNION
VIJAYA

CANARA
UNITED

ALLA
ANDHRA
BoB
IOB
ORIENTAL

CORP
INDIAN

2008

BoM
CENTRAL
VIJAYA

PSB
SYND
UCO
UNION
UNITED

ALLA
CANARA
DENA

ANDHRA
BoB
BoI
CORP
IOB
ORIENTAL
PNB

PNB

Some banks continued to be poor performers as regards


Tier I capital (below 6%) using the cost to asset ratio viz
Bank of Maharashtra, Central Bank of India, Dena Bank,
Syndicate Bank, UCO Bank and Vijaya Bank. The surprise as
indifferent performer under the cost to asset ratio has been
Punjab and Sind Bank which has shown Tier I capital below
6%, while United Bank of India and Vijaya Bank showed a
continuous decline in Tier I levels since 2005 to 2008.
However, as compared to the State Bank group and the 3
leading new private banks, a substantial number of nationalized
banks maintained a Tier I capital greater than 9% as can be
seen from Tables 6B and 6C. Besides, the performance of
some nationalized banks such as Andhra Bank, Allahabad Bank,
Bank of Baroda, Corporation Bank, Oriental Bank of
Commerce under the cost to asset approach for operational
risk capital charge is similar to their performance under the
Basic Indicators Approach.
The impact of the cost to asset ratio as a measure of
determining Operational Risk capital charges on the
regulatory Tier I capital for State Bank of India and its

INDIAN

associates is in Table 6B and the impact has been extreme


like that for the nationalized banks. There is no bank from
the State Bank group that has a Tier I capital of above 9%
while the State Bank of India, except for 2007, has had a
relatively comfortable position on Tier I in the range of 7%9%. The impact of the Operational Risk capital charge on
Tier I capital for the associate banks in this group has been
mixed. The associate banks that have scored badly on Tier I
capital in this group (as compared to under BIA) have been
the State Bank of Patiala whose Tier I capital has been
declining since 2005 and the State Bank of Bikaner and Jaipur,
while the State Bank of Mysore has worsened (as compared
to the BIA) and its Tier I has consistently being below 5%
under the cost to asset ratio measure.
The performance of the 3 major new private sector banks
vis--vis Tier I capital is in Table 6C. The performance of
HDFC Bank has been stable and its Tier I capital like under
the BIA has been in the range of 7-9% while mixed
performance is observed for Axis Bank and ICICI Bank.

Table 6B. Impact of Operational Risk Capital Charge on Tier I Capital Cost to Asset Ratio State Bank Group

Tier I Capital
Year

Below 5%

2005

SBM

Between 5-6%

SBT

SBBJ

SBH

SBI-IND

Between 6-7%

Between 7-9%

Above 9%

SBI

SBP
SBS
Table contd...

Chapter 1: Introduction

389

Table contd...
2006

SBM

SBBJ

SB-IND

SBI

SBT

SBP

SBH
SBS

2007

SBBJ

SBP

SBI

SBM

SBH

SB-IND
SBS
SBT

2008

SBBJ

SBH

SBM

SB-IND

SBP

SBS

SBI

SBT
Table 6C. Impact of Operational Risk Capital Charge on Tier I Capital
Cost to Asset Ratio Axis Bank, HDFC Bank and ICICI Bank

Tier I Capital
Year

Below 5%

Between 5-6%

Between 6-7%

2005

Between 7-9%

ICICI

Above 9%

AXIS
HDFC

2006

AXIS

HDFC

ICICI

HDFC

ICICI
2007

AXIS

2008

AXIS
HDFC

The second alternative approach adopted to estimate the


operational risk capital charge is the cost to income ratio.
Tables 7A, 7B and 7C highlight the impact on the Tier I capital

ICICI

of nationalised banks, the State Bank and its associate banks


and the three leading private sector banks. (Detailed Results
are in Annexure V).

Table 7A. Impact of Operational Risk Capital Charge on Tier I Capital


Cost to Income Ratio Nationalised Banks

Tier I Capital
Year

Below 5%

Between 5-6%

Between 6-7%

Between 7-9%

Above 9%

2005

ALLA
CENTRAL
DENA
ORIENTAL
PSB
SYND
UCO

BoI
BoM
IOB
UNION
VIJAYA

ANDHRA
CANARA
INDIAN

BoB
PNB

CORP
UNITED

2006

CENTRAL
DENA
PSB
UCO

BoI
SYND

CANARA
PSB
UNION
VIJAYA

ALLA
INDIAN
IOB
ORIENTAL
PNB
UNITED
VIJAYA

ANHDRA
BoB
CORP

Table contd...

390 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
2007

BoM

BoI

CENTRAL

SYND

CANARA

ALLA

ANDHRA

PSB

BoB

CORP

DENA

UNION

IOB

INDIAN

UCO

VIJAYA

ORIENTAL
PNB
UNITED

2008

BoM

SYND

BoI

ALLA

CENTRAL

UNITED

CANARA

ANDHRA

UCO

VIJAYA

DENA

BoB

IOB

CORP

PSB

INDIAN

UNION

PNB
ORIENTAL

The distribution of nationalised banks (Table 7A) for Tier I


regulatory capital maintained is similar to that under the cost
to asset ratio more so in the range below 5% and above 7%
of Tier I capital, while some changes are observed for the
middle range of 5-7% of Tier I. While Central Bank of India
continues to score poorly under the cost to income measure,
its Tier I capital is the lowest in the range of 1.30 to 3.57%.
The other banks that have not fared well continue to be Dena
Bank, Bank of Maharashtra, UCO Bank, Syndicate Bank.

Table 7B highlights the performance of the State Bank group


where the performance is substantially varied from that under
the cost to asset approach as well as the BIA. The State Bank
of Indias Tier I capital (between 6-7%) under the cost to
income approach is lower as compared to the other
approaches, while that of the State Bank of Bikaner and Jaipur
matches its Tier I capital with that under the BIA markedly
different from that under the cost to asset approach. The State
Bank of Mysore also fares better as compared to the other
two approaches.

Table 7B. Impact of Operational Risk Capital Charge on Tier I Capital


Cost to Income Ratio State Bank Group

Tier I Capital
Year

Below 5%

Between 5-6%

2005

SB-IND

SBH

SBT

Between 6-7%

Between 7-9%

SBI

SBBJ

SBM

SBP

SBS
2006

SBT

SB-IND

SBI

SBM

SBBJ
SBH
SBS
SBP

2007

SBI

SBBJ

SBH

SB-IND

SBS

SBM
SBP
SBT
2008

SBI

SBH

SBBJ

SB-IND

SBM

SBP
SBT

Above 9%

Chapter 1: Introduction

Table 7C contains the results for the three major new private
sector banks and it can be observed that under the cost to
income approach as compared to the other two approaches

391

BIA and cost to assets, HDFC Bank shows a Tier I capital


below 7% in 2006 and 2007. ICICI Bank continues to have a
Tier I capital greater than 9%.

Table 7C. Impact of Operational Risk Capital Charge on Tier I Capital


Cost to Income Ratio (Axis Bank, HDFC Bank and ICICI Bank)

Tier I Capital
Year

Below 5%

Between 5-6%

2005

Between 6-7%
ICICI

Between 7-9%

Above 9%

AXIS
HDFC
ICICI

2006

AXIS

HDFC

ICICI

2007

AXIS

HDFC

ICICI

2008

AXIS
HDFC

Thus, a comparison of the estimates of the impact of the


additional operational risk capital charges on the Tier I capital
of individual banks across the different approaches and bank
groups indicates that some banks would continue to be
comfortable and maintain high levels of Tier I capital (above
7%) after the imposition of the additional capital requirements
especially for end March 2007 and 2008, namely, Andhra
Bank, Bank of Baroda, Corporation Bank, Oriental Bank of
Commerce, Punjab National Bank, State Bank of India
(except under the cost to income approach). Canara Bank
continued to maintain a Tier I capital between 6%-7% under
all approaches, while the estimates for Bank of Maharashtra,
Central Bank of India , UCO Bank and State Bank of Mysore
show a Tier I capital below 6% under all the three approaches.
Among the three new private sector banks considered, Axis
Bank showed improved levels of Tier I capital under all the
three approaches in 2008 as compared to 2007 and the Tier I
capital of HDFC Bank is higher than 7% except for the cost
to income approach where it was between 6-7%. Likewise
ICICI Bank has been well capitalized and has had a Tier I
capital greater than 7% except for 2007 under the cost to
asset approach when its Tier I was between 6-7%. Annexure
IX presents a graphical comparison at the individual bank
level for the broad spectrum of 30 banks covering the
nationalized banks, State Bank of India and its associate banks
and three new private sector banks of the impact of the
Operational Risk capital charge on the Tier I capital under
the Basic Indicators Approach, the cost to asset ratio, and the
cost to income ratio.

IV.C Sensitivity Analysis


This section estimates the sensitivity of a banks gross income
to gross income from a particular business line at Level 1 of

ICICI

the classification of business activity for a bank as specified


under the Standardized Approach. In other words it is the
amount of change in gross earnings given a unit change in
the earnings from a single business line with all else held
constant. Sensitivity analysis, thus, helps identify the business
line/activity to which a banks income is most sensitive.
Sensitivity analysis, thus, is the partial derivative of the
earnings function with respect to a factor. It can be defined
as:
E/X = f/x
where E gross income of a bank
particular business line.

X income from a

Sensitivity analysis has been performed using the standard


elasticity approach by calculating the average elasticity of
the change in a banks gross income to a change in income
from a particular business line. While panel data analysis is
better suited to derive the sensitivity coefficients, the lack of
a sufficient number of data points restricts the use of the panel
data technique to ensure the reliability of our estimates.
Besides, the sensitivity analysis could be estimated at the level
of the bank group rather than at an individual bank level, as
data on income from retail and commercial banking which
was computed under the Standardized Approach, was
available at the level of bank group. Table 8 presents the
results of the sensitivity analysis for activities under Level 1
of the Standardized Approach (restricted in this study to just
income from corporate finance, trading & sales, retail banking
and commercial banking, since availability of data with a
one to one correspondence for the 8 business lines specified
under the SA is difficult) and within trading & sales income
sensitivity to income from foreign exchange transactions and
commission & brokerage.

392 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table 8. Sensitivity Analysis for 2005-2007: Bank Group-wise

Bank Group

Corporate Finance

Trading &

Retail Banking

Sales
Nationalised

Commercial

Foreign

Comm. &

Banking

Exchange

Brokerage

0.6390

0.2446

0.3137

2.3912

0.2884

-0.6337

State Bank

0.2562

-0.1642

0.1834

0.0229

0.7039

0.4092

Private

0.1795

0.6910

0.6789

-4.4634

0.5526

0.4993

The results of the sensitivity analysis present a mixed scenario


for the three bank groups considered nationalized banks,
State Bank group and private sector banks (old and new). All
the three bank groups show the existence of a positive
relationship between the gross income of a bank and the
income from corporate finance. Nationalized banks show a
very high degree of elasticity with reference to corporate
income. A percentage change in corporate income increases
gross income by almost 2.3%, while for the State Bank group
and the private banks (old and new) this coefficient is
comparatively smaller and a percentage increase in income
from corporate finance brings about a 0.25% and 0.18%
increase in gross income respectively. The sensitivity of gross
income to income from trading and sales is the highest for
the private bank group indicating that a 1% increase in income
from trading and sales would have an impact of increasing
gross income by about 0.69%, while for the nationalized banks
it was comparatively lower at 0.28%. The State Bank and its
associate banks surprisingly have a negative coefficient but
this could be explained by the fall in the income from trading
& sales since 2005 for the State Bank group from a high of
Rs.12858 crores in 2004 to Rs.10392 crores in 2005 and a
further decline to Rs.9911 crores in 2007. This fall in income
from trading and sales can be attributed to a fall in the profit
from the sale of investments and a decline in net repo income.
The estimate for net repo income would also include income
from the inter-bank funds market. The sensitivity of income
from retail banking is the highest for the private bank group
at 0.68% followed by the State Bank group where a 1%
increase in income from retail banking brings about only about
less than a quarter percent increase in gross income. The high
negative coefficient for the nationalized banks for retail
banking income seems unexpected and may perhaps be
attributed to the high proportion of credit outstanding to
agriculture, large credit outstanding of less than Rs.2 lakhs
and other social objectives such as priority sector lending
where recovery is slow. These factors may also explain the
substantial difference in the sensitivity coefficients observed
between the State Bank group and the private sector banks.
The sensitivity of the gross income to income from
commercial banking is the highest for the nationalized banks
at 0.64, whereas the sensitivity is much lower for the State

Bank group at 0.02, while the private banks are described by


a negative coefficient on sensitivity to income from
commercial banking. The private sector banks have seen a
decline in income from commercial banking since 2005 and
income from commercial banking in 2007 was almost half at
Rs.5935 crores as compared to 2006. Within trading and sales,
we have further looked at sensitivity of gross income to two
major sources of income for a bank, namely foreign exchange
transactions and commissions and brokerage. All the three
bank groups studied showed a positive sensitivity to income
from both these business activities. The sensitivity for both
these business activities for the private sector banks was in
the range of 0.49 to 0.55 implying thereby that a one percent
increase in income from these activities would reflect in about
half a percent increase in gross income of the banks. The
State Bank of India and its associate banks showed a high
sensitivity of nearly 0.70 for foreign exchange transactions
and 0.40 on commission and brokerage, while the nationalized
banks had the lowest sensitivity coefficients. The gross
income for all the bank groups is sensitive to income from
foreign exchange transactions, but this sensitivity at 0.77%
is the highest for the private banks followed by the State Bank
group and the nationalized banks. A similar pattern is observed
on the sensitivity of gross income to income from commission
and brokerage.

V. Conclusion
Operational Risk, thus, generates potential losses and this
potential loss can be viewed as a cost to the bank. While
Operational Risk exists and needs to be priced appropriately,
there is a view among practitioners and analysts (Webb, 1999)
that the focus should be on the elimination of Operational
Risk rather than a focus on its measurement. However, in
practice, the costs involved in the elimination of Operational
Risk could be substantially higher than the probability of a
loss event actually occurring. Also, it may not be feasible to
foresee all the potential future loss events. Most of the banks
in India are operating at capital adequacy ratios higher that
the prescribed Basel II requirements. Still, overall capital
requirements are expected to go up on account of Operational
Risk (Reserve Bank of India, 2008B). The paper focuses on

Chapter 1: Introduction

the measurement of the Operational Risk capital charges that


banks in India have to incorporate from 31st March 2007
and the impact of this additional capital requirements on the
regulatory Tier I capital of banks. The paper estimates
Operational Risk capital charges and its Tier I impact at the
level of the bank group and for individual banks according
to the availability of data. At the level of individual banks,
the paper covers 30 banks 19 nationalized banks, 8 banks
of the State Bank group and 3 leading new private sector
banks Axis Bank, HDFC Bank and ICICI Bank over the
period since 2005. We compute Operational Risk capital
charges using two of the Basel II Approaches, namely, the
Basic Indicators Approach (at the bank group as well as
individual bank level) and the Standardized Approach (at the
level of bank groups) and alternate approaches using the cost
to income and cost to asset ratios. The additional capital
requirements on account of the Operational Risk capital
charges for the Indian banking sector excluding foreign banks
in India would be to the tune of about Rs.14,058 crores at the
end of March 2007 and Rs.15,427 crores in March 2008.
This requirement is estimated to be Rs.17,647 crores for
March 2009 under the Basic Indicators Approach. Of this
requirement, approximately 80% would be needed by the
public sector banks. The empirical estimates reveal the
introduction of the Operational Risk capital charges that will
have the expected impact of lowering the Tier I capital, but
this decline would be marginal under the Basic Indicators
Approach. Also, in 2007 and 2008 nearly 10 of the 19
nationalized banks will have a Tier I capital above 7%. The
State Bank group, though, would show a fewer number of
associate banks with a Tier I capital higher than 7% in 2008
as compared to 2007. While the State Bank of India would
continue to be in the range of 7%-9% of Tier I capital in both
2007 and 2008, the only associate bank in the same range
would be the State Bank of Saurashtra. All the three private
sector banks would have a Tier I capital higher than 7% after
the introduction of the additional capital requirement for
Operational Risk. Compared to the Basic Indicators
Approach, under the Standardized Approach, the nationalized
banks would be required to set aside, on an average, nearly
Rs.14,000 crores and the State Bank group, on an average,
would need around Rs.7500 crores and the requirement for
the private sector banks would be lower at around Rs. 6000,
on an average, over 2007-08. Further, Tier I capital would be
lower for all bank groups under the Standardized Approach
as compared to the Basic Indicators Approach. The alternate
approaches which use the cost to income and cost to assets

393

ratio for determining the Operational Risk capital charges


show a markedly different picture for Tier I capital. Under
the BIA, there were no banks in the Tier I range below 5% in
2007 and in 2008 only two nationalized banks i.e. Bank of
Maharashtra and UCO Bank were in this category - whereas
there are at least 3 nationalized banks and 2-3 associate banks
of the State Bank group with a Tier I capital lower than 5%
under the alternate approaches.
A comparison of the estimates of the impact of the additional
Operational Risk capital charges on the Tier I capital of
individual banks across the different approaches indicates that
some banks would continue to be good performers and
maintain a high level of Tier I capital (above 7%) viz. Andhra
Bank, Bank of Baroda, Corporation Bank, Oriental Bank of
Commerce, Punjab National Bank, State Bank of India
(except under the cost to income approach). Canara Bank
would maintain a Tier I capital between 6%-7% under all
approaches while the estimates for Bank of Maharashtra,
Central Bank of India, UCO Bank and State Bank of Mysore
show a Tier I capital below 6% under all the three approaches.
Among the three new private sector banks considered, Axis
Bank showed improved levels of Tier I capital under all the
three approaches in 2008 as compared to 2007, while HDFC
Bank and ICICI Bank continue to be well-capitalized. The
low Tier I levels of some of the nationalized banks (not taking
into account the Operational Risk capital charge) has resulted
in the government agreeing to infuse capital in UCO Bank,
Central Bank of India and Vijaya Bank to the tune of Rs.38
billion through investment in shares in February and for
United Bank of India to the tune of Rs.800 crores in two
tranches until March 2010. Apart from the measurement of
Operational Risk capital charges and their impact on Tier I,
the paper has computed some preliminary estimates of the
sensitivity of gross income of a bank to the income from
different business lines. Sensitivity analysis will, thus, help
identify the potential sources of Operational Risk for a bank.
Finally, the best approach to the appropriate pricing of
Operational Risk would be the Advanced Management
Approach which requires a bank to review its history of
operational loss events and create a database wherein every
loss event, the frequency of its occurrence, and the size of
the loss is recorded. Such an approach could result in
developing key risk indicators that could signal the occurrence
of a loss event and banks would have the discretion to use
their own internal loss data and be encouraged to develop
sophisticated risk measurement and risk management
techniques.

394 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
References
Basel Committee on Banking Supervision (2001) Operational
Risk (Supportive Document to the New Basel Capital
Accord)
International Convergence of Capital Measurements and
Capital Standards: A Revised Framework
(Comprehensive Version) (2006)
Bhatia, M., (2002) New Basel Accord: Operational Risk
Management - Emerging Frontiers for the Profession,
Information System Control Journal, vol. 1
Bonsn, Enrique, Toms Escobar and Franscisco Flores,
(2007) Issues in Operational Risk, Financial Markets,
Institutions and Instruments, 16, 4, pp.167-200
Gupta, V. and K. Srinivasan (2005) Basel II Accord: Impact
on Indian Banks, ICRA Rating Feature
King, J.L., (2001) Operational Risk: Measurement and
Modelling, John Wiley and Sons, New York.
Mussa, Imad A., (2007) Operational Risk Management,
Basingstoke, Hampshire: Palgrave Macmillan
Oesterreichische Nationalbank, (2006) Guidelines in
Operational Risk Management, Vienna

Reserve Bank of India (2005) Draft Guidelines for


Implementation of New Capital Adequacy Framework,
Reserve Bank of India, Mumbai
Reports on Currency and Finance 2003-08 Volume lV,
Mumbai: Reserve Bank of India (2008A)
Reports on Currency and Finance 2003-08 Volume V,
Mumbai: Reserve Bank of India (2008B)
Basic Statistical Returns of Scheduled Commercial Banks
(various issues)
Report on Trends and Progress in Banking in India (various
issues)
Statistical Tables Relating to Banks in India (various issues)
Saita, F., (2007) Value at Risk and Bank Capital Management:
Risk-Adjusted Performance, Capital Management and
Capital Allocation Decision Making, Academic Press,
London.
Sundmacher M., (2004) Operational Risk Measurement in
Banks: Arbitrage, Adjustments and Alternatives http://
papers.ssrn.com (Working Paper Series)
Tripe D., (1998) Cost to Income Ratios in Australasian
Banking, Centre for Banking Studies, Massey University
Pricing Operational Risk, Centre for Banking Studies, Massey
University (2000)

ANNEXURE I
LIST OF ABBREVIATIONS

PSB

Punjab and Sind Bank

Abbreviation

Bank Name

PNB

Punjab National Bank

ALLA

Allahabad Bank

SYND

Syndicate Bank

ANDHRA

Andhra Bank

UCO

UCO Bank

BoB

Bank of Baroda

UNION

Union Bank

BoI

Bank of India

UNITED

United Bank of India

BoM

Bank of Maharashtra

VIJAYA

Vijaya Bank

CANARA

Canara Bank

SBI

State Bank of India

SBBJ

State Bank of Bikaner and Jaipur

CENTRAL

Central Bank of India

SBH

State Bank of Hyderabad

CORP

Corporation Bank

SB-IND

State Bank of Indore

DENA

Dena Bank

SBM

State Bank of Mysore

INDIAN

Indian Bank

SBP

State Bank of Patiala

IOB

Indian Overseas Bank

SBS

State Bank of Saurashtra

ORIENTAL

Oriental Bank of Commerce

SBT

State Bank of Travancore

ANNEXURE II
Mapping of Business Lines under Standardized Approach

Business Unit
Investment Banking

Level 1
Corporate Finance

Level 2

Activity Groups

Corporate Finance

Mergers & Acquisitions, Underwriting,

Municipal/Government Finance

Privatisations, Securitisation, Research, Debt

Merchant Banking

(Government, High Yield), Equity,

Advisory Services

Syndications, IPO, Secondary Private


Placements

Trading & Sales

Retail Banking

Banking

Commercial Banking

Sales

Fixed Income, Equity, Foreign Exchanges,

Market Making

Commodities, Credit, Funding, Own Position

Proprietary Positions

Securities, Lending and Repos, Brokerage, Debt,

Treasury

Prime Brokerage

Retail Banking

Retail Lending and Deposits, Banking Services,


Trust and Estates

Private Banking

Private Lending and Deposits, Banking Services,


Trust and Estates, Investment Advice

Card Services

Merchant/Commercial/Corporate Cards, Private


Labels and Retail

Commercial Banking

Project Finance, Real Estate, Export Finance,


Trade Finance, Factoring, Leasing, Lends,
Guarantees, Bills of Exchange

Payment & Settlement External Clients

Payments & Collections, Funds Transfer, Clearing


& Settlement

Agency Services

Custody

Escrow, Depository Receipts, Securities Lending


(Customers) Corporate Actions

Corporate Agency

Issuer and Paying Agents

Corporate Trust
Asset Management

Others

Retail Brokerage

Discretionary Fund Management

Pooled, Segregated, Retail, Institutional, Closed,


Open, Private Equity

Non-Discretionary Fund

Pooled, Segregated, Retail, Institutional, Closed,

Management

Open

Retail Brokerage

Execution and Full Service

Life Insurance and Benefit Plans


Insurance

Property and Casualty Insurance


Health Insurance
Reinsurance
Brokerage & Advisory

Source: Consultative Document on Operational Risk, BCBS, 2001

ANNEXURE III
Basic Indicators Approach
Bank-wise Operational Risk Capital Charge and Impact on Tier I Capital
(Nationalized Banks, State Bank Group & Axis Bank, HDFC Bank & ICICI Bank)

Year

Gross Income

Annual Capital

Capital Charge

Networth

Current Tier

Estimated Tier I

(Rs. Crore)

Required (Rs. Crore)

(Rs. Crore)

(Rs. Crore)

I (%)

Capital (%)

2002

1116

167

1002

6.22

2003

1433

215

1190

6.35

2004

1836

275

1552

6.26

2005

2004

301

219

2328

6.46

5.85

2006

2059

309

264

3639

9.53

8.84

2007

2127

319

295

4477

8.10

7.57

2008

2637

396

310

5221

7.75

7.29

Allahabad Bank

2009

341

Andhra Bank
2002

880

132

884

8.80

2003

1357

204

1116

8.19

2004

1588

238

1453

8.17

2005

1822

273

1837

8.03

7.19

191

2006

1561

234

238

2894

12.20

11.20

2007

1864

280

249

3156

9.98

9.19

2008

2001

300

262

3249

8.54

7.85

2009

271

Bank of Baroda
2002

2873

431

3827

7.56

2003

3366

505

4387

8.10

2004

4291

644

5131

8.47

2005

4284

643

5628

8.21

7.44
10.14

527

2006

4301

645

597

7845

10.98

2007

4959

744

644

8650

8.74

8.09

2008

5962

894

677

11044

7.63

7.16

2009

761

Bank of India
2002

2943

441

2845

6.37

2003

3678

552

3541

7.56

2004

3994

599

4010

7.47

2005

3393

509

531

4465

7.05

6.21

2006

3816

572

553

4984

6.75

6.00

2007

5003

750

560

5895

6.54

5.92

2008

6346

952

611

10589

7.70

7.26

2009

758
Table contd...

398 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
Bank of Maharashtra
2002

895

134

698

6.56

2003

1037

156

981

5.88

2004

1236

185

1436

7.03

2005

1267

190

158

1543

7.10

6.37

2006

1023

153

177

1573

7.47

6.63

2007

1359

204

176

1742

6.03

5.42

2008

1509

226

182

1782

5.13

4.60

2009

195

Canara Bank
2002

3250

488

3472

8.07

2003

3745

562

4149

7.85

2004

4755

713

5252

7.81

2005

4694

704

588

6109

7.29

2006

4898

735

660

7132

7.81

7.09

2007

5478

822

717

10354

7.17

6.67

2008

5751

863

754

10500

7.01

6.51

2009

6.59

806

Central Bank of India


2002

2135

320

1997

5.20

2003

2451

368

2424

5.66

2004

3086

463

2974

6.23

2005

3294

494

384

3265

6.08

5.37

2006

2911

437

442

3442

7.19

6.27

2007

2950

443

465

3790

6.32

5.55

2008

3014

452

458

5943

5.42

5.00

2009

444

Corporation Bank
2002

1008

151

2046

16.80

2003

1325

199

2370

17.30

2004

1481

222

2768

16.52

2005

1695

254

191

3054

13.55

12.70

2006

1700

255

225

3374

12.41

11.58

2007

1944

292

244

3765

11.30

10.57

2008

2144

322

267

4228

9.64

9.03

2009

289

Dena Bank
2002

795

119

977

4.36

2003

1005

151

999

5.31

2004

1209

181

1055

5.19

2005

997

150

150

1104

6.63

5.73

2006

1162

174

161

1339

5.96

5.25

2007

1248

187

168

1497

6.06

5.38

2008

1337

201

170

1801

6.75

6.11

2009

187

Annexure III

399

Indian Bank
2002

1033

155

4185

0.85

2003

1345

202

5130

7.51

2004

1864

280

5538

7.66

2005

1873

281

212

5936

7.60

7.33

2006

1974

296

254

2492

10.29

9.24

2007

2605

391

286

3841

12.28

11.37

2008

3060

459

323

5211

11.41

10.70

2009

382

Indian Overseas Bank


2002

1501

225

1133

6.17

2003

1742

261

1460

5.83

2004

2340

351

2081

6.74

2005

2654

398

279

2575

7.10

6.33

2006

2608

391

337

3178

8.54

7.63

2007

2948

442

380

3991

8.20

7.42

2008

3487

523

411

4857

7.86

7.20

1621

8.89

2009

452

Oriental Bank of Commerce


2002

1446

217

2003

1746

262

2110

10.72

2004

2178

327

2677

9.87

2005

2029

304

269

3327

5.42

4.98

2006

2158

324

298

5171

10.37

9.77

2007

2294

344

318

5601

10.05

9.48

2008

2299

345

324

5776

9.34

8.82

2009

338

Punjab & Sind Bank


2002

545

82

502

6.37

2003

693

104

489

6.11

2004

749

112

467

6.38

2005

829

124

440

5.26

4.07

99

2006

751

113

114

1222

10.05

9.12

2007

995

149

116

1406

9.58

8.79

2008

1103

165

129

2093

8.04

7.55

2009

142

Punjab National Bank


2002

3273

491

3381

6.34

2003

4374

656

4033

7.11

2004

5492

824

5012

7.01

2005

5683

852

657

8161

8.87

8.16

2006

5941

891

777

9376

10.06

9.23

2007

6556

983

856

10435

8.93

8.20

2008

7532

1130

909

12318

8.52

7.89

2009

1001

Table contd...

400 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
Syndicate Bank
2002

1383

207

1683

8.47

2003

1705

256

1614

7.69

2004

2205

331

1905

6.75

2005

2284

343

265

2199

6.10

5.37

2006

2442

366

310

2834

7.40

6.59

2007

2768

415

347

3623

6.24

5.64

2008

2962

444

375

4291

6.62

6.04

2009

409

UCO Bank
2002

1313

197

2762

4.89

2003

1491

224

1214

5.19

2004

1819

273

1783

6.08

2005

1925

289

231

2069

5.75

5.11

2006

1940

291

262

2462

6.09

5.44

2007

2138

321

284

2662

5.78

5.16

2008

2260

339

300

2926

5.05

4.53

2009

317

Union Bank of India


2002

1836

275

2107

6.16

2003

2323

348

2607

6.86

2004

2567

385

3087

6.47

2005

2831

425

336

3614

6.07

5.51

2006

2869

430

386

4558

7.32

6.70

2007

3477

522

413

5190

7.79

7.17

2008

4173

626

459

7348

7.45

6.98

2009

526

United Bank of India


2002

996

149

1964

8.84

2003

1147

172

1962

12.63

2004

1286

193

1960

15.04

2005

1393

209

1957

14.15

12.91

171

2006

1457

219

191

1828

10.01

8.96

2007

1498

225

207

2415

7.72

7.06

2008

1370

206

217

2661

6.74

6.19

2009

216

Vijaya Bank
2002

674

101

781

8.86

2003

990

149

962

7.42

2004

1364

205

1336

8.37

2005

1338

201

151

1590

7.59

6.87

2006

1257

189

185

1670

9.26

8.24

2007

1347

202

198

1897

7.07

6.33

2008

1362

204

197

2460

5.73

5.27

2009

198

Annexure III

401

State Bank of India


2002

13255

1988

15224

9.22

2003

15718

2358

17203

8.81

2004

18798

2820

20231

8.34

2005

8515

1277

24072

8.04

2389

7.24

2006

8994

1349

2152

27644

9.36

8.63

2007

21823

3273

1815

31298

8.01

7.55

2008

25716

3857

1967

49032

8.48

8.14

2009

2827

State Bank of Bikaner & Jaipur


2002

791

119

752

9.22

2003

891

134

903

8.81

2004

1208

181

1149

8.34

2005

1352

203

145

1298

8.04

7.14

2006

1241

186

173

1406

9.36

8.21

2007

1432

215

190

1654

8.01

7.09

2008

1411

212

201

1713

6.95

6.13

2009

204

State Bank of Hyderabad


2002

1014

152

998

9.86

2003

1209

181

1251

9.84

2004

1548

232

1574

8.42

2005

1384

208

189

1765

7.58

6.77

2006

1553

233

207

2114

8.95

8.07

2007

1811

272

224

2541

8.25

7.52

2008

1790

269

2009

237

2694

7.24

6.60

258

1317

7.01

5.64

State Bank of Indore


2002

567

85

413

8.15

2003

669

100

584

9.40

2004

814

122

791

8.31

2005

680

102

103

904

6.67

5.91

2006

787

118

108

1018

7.55

6.75

2007

800

120

114

1177

6.74

6.09

2008

883

132

113

1317

7.01

6.41

2009

124

State Bank of Mysore


2002

548

82

352

6.70

2003

680

102

431

7.23

2004

794

119

582

7.18

2005

931

140

101

756

7.12

6.17

2006

948

142

120

935

7.44

6.48

2007

1034

155

134

1141

6.62

5.84

2008

1184

178

146

1378

6.54

5.85

2009

158

Table contd...

402 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
State Bank of Patiala
2002

921

138

1142

9.97

2003

1135

170

1412

10.39

2004

1452

218

1731

9.87

2005

1332

200

175

2045

11.05

10.10

2006

1344

202

196

2235

9.96

9.09

2007

1447

217

206

2488

8.36

7.67

2008

1486

223

206

2459

6.74

6.17

2009

214

State Bank of Saurashtra


2002

453

68

568

12.11

2003

534

80

625

11.66

2004

710

107

767

10.99

2005

641

96

85

794

8.68

7.75

2006

584

88

94

977

9.02

8.15

2007

575

86

97

1043

8.17

7.41

2008

571

86

90

1145

8.06

7.43

2009

87

State Bank of Travancore


2002

655

98

618

7.79

2003

822

123

731

6.80

2004

1153

173

925

6.23

2005

1305

196

132

1130

6.17

5.45

2006

1307

196

164

1332

7.24

6.35

2007

1356

203

188

1599

7.55

6.66

2008

1446

217

198

1718

6.94

6.14

2009

205

Axis Bank
2002

616

92

694

6.42

2003

733

110

1040

6.44

2004

1106

166

1138

6.44

2005

1147

172

123

2422

8.87

8.42

2006

1808

271

149

2886

7.26

6.88

2007

2577

387

203

3403

6.42

6.04

2008

4381

657

277

8769

10.17

9.85

2009

438

HDFC Bank
2002

962

144

2132

10.81

2003

1304

196

2608

9.49

2004

1818

273

2694

8.03

2005

2428

364

204

4520

9.60

9.17

2006

3669

550

278

5299

8.55

8.10

2007

5226

784

396

6433

8.57

8.04

2008

7511

1127

566

11497

10.30

9.79

2009

820

Annexure III

403

ICICI Bank
2002

1168

175

6618

7.47

2003

4583

687

7289

7.05

2004

4944

742

8360

6.09

2005

6255

938

535

12900

7.59

7.28

2006

8890

1334

789

22556

9.20

8.88

2007

12565

1885

1004

24663

7.42

7.12

2008

16115

2417

1386

46821

11.32

10.99

2009

Note: Net worth of a bank is computed as per the Reserve


Bank of Indias definition which is as follows: Net worth
would comprise of Paid-up capital plus Free Reserves
including Share Premium but excluding Revaluation
Reserves, plus Investment Fluctuation Reserve and credit
balance in Profit & Loss account, less debit balance in Profit
and Loss account, Accumulated Losses and Intangible Assets.
No general or specific provisions should be included in
computation of net worth. Infusion of capital through equity
shares, either through domestic issues or overseas floats after

1879

the published balance sheet date, may also be taken into


account for determining the ceiling on exposure to capital
market.
(i)

Gross Income = Net Interest Income + Non-Interest


Income
(ii) Impact on Tier I capital = capital charge/networth
(iii) Estimated Tier I = (1-impact on Tier I capital)* current
Tier I
These definitions are adopted in all relevant calculations.

ANNEXURE IV
Cost-Asset Ratio
Bank-Wise Operational Risk Capital Charge and Impact on Tier I Capital
(Nationalized Banks, State Bank Group & Axis Bank, HDFC Bank & ICICI Bank)

Year

Cost-Asset
Ratio (%)

Average
PSB (%)

Three
standard
deviations (%)

Total
Assets
(Rs. Crore)

Capital
Charge
(Rs. Crore)

Networth
(Rs. Crore)

Current
Tier I (%)

Estimated
Tier I
Capital (%)

2005

2.37

2.09

2.53

45145

1142

2006

1.87

2.05

1.36

55292

754

2328

6.46

3.29

3639

9.53

7.56

2007

1.52

1.77

0.88

67664

598

2008

1.4

1.54

0.72

82940

595

4477

8.10

7.02

5221

7.75

6.87

2005

2.53

2.09

1.13

3273

37

1837

8.03

7.87

2006

2.11

2.05

1.06

4067

43

2894

12.20

12.02

2007

1.96

1.77

1.02

47541

487

3156

9.98

8.44

2008

1.67

1.54

0.50

56592

286

3249

8.54

7.79

2005

2.09

2.09

0.27

94664

256

5628

8.21

7.84

2006

2.1

2.05

0.22

113393

248

7845

10.98

10.63

2007

1.78

1.77

0.11

143146

155

8650

8.74

8.58

2008

1.63

1.54

0.22

179600

394

11044

7.63

7.36

2005

2.03

2.09

0.39

94978

373

4465

7.05

6.46

2006

1.88

2.05

0.50

112274

559

4984

6.75

5.99

2007

1.84

1.77

0.41

141637

581

5895

6.54

5.90

2008

1.48

1.54

0.41

178830

734

10589

7.70

7.17

2005

2.19

2.09

1.09

32885

358

1543

7.10

5.45

2006

2.11

2.05

1.03

31215

321

1573

7.47

5.95

2007

1.91

1.77

0.39

39009

151

1742

6.03

5.51

2008

1.74

1.54

0.53

48151

257

1782

5.13

4.39

Allahabad Bank

Andhra Bank

Bank of Baroda

Bank of India

Bank of Maharashtra

Canara Bank
2005

1.91

2.09

0.78

110305

865

6109

7.29

6.26

2006

1.77

2.05

0.95

132822

1263

7132

7.81

6.43

2007

1.55

1.77

0.85

165961

1405

10354

7.17

6.20

2008

1.55

1.54

0.76

180529

1364

10500

7.01

6.10

Central Bank of India


2005

2.46

2.09

1.30

68596

892

3265

6.08

4.42

2006

2.3

2.05

1.09

74681

811

3442

7.19

5.50

2007

1.81

1.77

0.95

93008

885

3790

6.32

4.84

2008

1.41

1.54

0.60

123956

748

5943

5.42

4.74

Annexure IV

405

Corporation Bank
2005

1.97

2.09

1.13

33924

383

3054

13.55

11.85

2006

1.84

2.05

0.72

40507

293

3374

12.41

11.33

2007

1.52

1.77

0.74

52721

389

3765

11.30

10.13

2008

1.34

1.54

0.81

66598

541

4228

9.64

8.41

2005

2.56

2.09

1.17

24029

282

1104

6.63

4.93

2006

2.11

2.05

1.01

26545

268

1339

5.96

4.77

2007

1.94

1.77

1.07

31451

336

1497

6.06

4.70

2008

1.68

1.54

0.48

38642

187

1801

6.75

6.05

2005

2.08

2.09

1.09

43861

479

5936

7.60

6.99

2006

2.27

2.05

1.16

47635

552

2492

10.29

8.01

2007

2.22

1.77

1.06

56149

597

3841

12.28

10.37

2008

1.99

1.54

1.43

70508

1007

5211

11.41

9.20

Dena Bank

Indian Bank

Indian Overseas Bank


2005

2.28

2.09

0.44

50815

225

2575

7.10

6.48

2006

2.13

2.05

0.46

59358

274

3178

8.54

7.80

2007

1.69

1.77

0.47

82257

386

3991

8.20

7.41

2008

1.46

1.54

0.29

101860

300

4857

7.86

7.38

Oriental Bank of Commerce


2005

1.47

2.09

1.32

54069

711

3327

5.42

4.26

2006

1.64

2.05

0.87

58937

513

5171

10.37

9.34

2007

1.35

1.77

0.89

73936

659

5601

10.05

8.87

2008

1.19

1.54

0.74

90705

673

5776

9.34

8.25

3.63

2.09

5.15

15718

810

440

5.26

-4.42

Punjab & Sind Bank


2005
2006

2.54

2.05

5.10

19043

971

1222

10.05

2.06

2007

2.38

1.77

3.66

21963

805

1406

9.58

4.10

2008

1.81

1.54

1.76

30949

543

2093

8.04

5.95

2.09

1.15

126241

1447

8161

8.87

7.30

Punjab National Bank


2005

2.6

2006

2.08

2.05

1.11

145267

1610

9376

10.06

8.33

2007

2.05

1.77

1.24

162423

2007

10435

8.93

7.21

2008

1.77

1.54

0.77

199020

1535

12318

8.52

7.46

2005

2.43

2.09

2.14

52109

1114

2199

6.10

3.01

2006

2.35

2.05

1.15

61077

704

2834

7.40

5.56

2007

1.55

1.77

1.07

89277

954

3623

6.24

4.60

2008

1.4

1.54

0.84

107132

903

4291

6.62

5.23

2005

1.99

2.09

0.59

1085

2069

5.75

5.73

2006

1.9

2.05

0.46

1177

2462

6.09

6.08

2007

1.59

1.77

0.54

1193

2662

5.78

5.77

2008

1.45

1.54

0.53

1306

2926

5.05

5.04

Syndicate Bank

UCO Bank

Table contd...

406 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
Union Bank
2005

1.74

2.09

1.19

72413

859

3614

6.07

4.63

2006

1.57

2.05

1.46

89126

1304

4558

7.32

5.23

2007

1.44

1.77

1.44

102678

1480

5190

7.79

5.57

2008

1.28

1.54

1.35

124073

1679

7348

7.45

5.75

United Bank of India


2005

2.42

2.09

1.16

29098

336

1957

14.15

11.72

2006

2.45

2.05

1.38

33248

458

1828

10.01

7.50

2007

1.84

1.77

1.11

42310

470

2415

7.72

6.22

2008

1.66

1.54

0.90

54311

488

2661

6.74

5.50

2005

1.84

2.09

1.55

2934

45

1590

7.59

7.37

2006

1.98

2.05

0.63

3153

20

1670

9.26

9.15

2007

1.54

1.77

0.74

42357

312

1897

7.07

5.91

2008

1.25

1.54

0.80

56184

449

2460

5.73

4.68

2005

2.19

2.09

0.39

459883

1778

24072

8.04

7.45

2006

2.37

2.05

0.72

494029

3569

27644

9.36

8.15

2007

2.09

1.77

0.98

566565

5570

31298

8.01

6.58

2008

1.75

1.54

1.06

721526

7636

49032

8.48

7.16

Vijaya Bank

State Bank of India

State Bank of Bikaner & Jaipur


2005

2.66

2.09

1.58

23430

370

1298

8.04

5.75

2006

2.76

2.05

2.12

27514

583

1406

9.36

5.48

2007

2.18

1.77

2.12

34507

731

1654

8.01

4.47

2008

1.82

1.54

1.84

41154

756

1713

6.95

3.88

State Bank of Hyderabad


2005

1.92

2.09

1.53

34922

534

1765

7.58

5.29

2006

2.01

2.05

1.06

40630

432

2114

8.95

7.12

2007

1.65

1.77

0.45

49052

220

2541

8.25

7.53

2008

1.3

1.54

0.58

61620

355

2694

7.24

6.29

State Bank of Indore


2005

1.94

2.09

0.37

16898

62

904

6.67

6.21

2006

1.92

2.05

0.43

20711

90

1018

7.55

6.88

2007

1.67

1.77

0.47

24527

116

1177

6.74

6.08

2008

1.47

1.54

0.38

29275

111

1317

7.01

6.42

State Bank of Mysore


2005

2.89

2.09

2.40

16553

397

756

7.12

3.38

2006

2.63

2.05

2.33

19337

451

935

7.44

3.85

2007

2.09

1.77

2.20

26843

591

1141

6.62

3.19

2008

1.87

1.54

1.57

33070

519

1378

6.54

4.08

State Bank of Patiala


2005

1.52

2.09

1.86

31503

586

2045

11.05

7.88

2006

1.48

2.05

2.06

41233

849

2235

9.96

6.18

2007

1.39

1.77

1.89

47461

897

2488

8.36

5.35

2008

1.2

1.54

1.62

59060

958

2459

6.74

4.11

Annexure IV
State Bank of Saurashtra
2005

1.69

2.09

0.98

15053

148

794

8.68

7.06

2006

1.87

2.05

1.03

16530

171

977

9.02

7.45

2007

1.73

1.77

0.93

18847

176

1043

8.17

6.79

2008

1.84

1.54

0.75

21358

160

1145

8.06

6.94

State Bank of Travancore


2005

1.74

2.09

1.23

28875

354

1130

6.17

4.24

2006

1.98

2.05

1.03

31862

329

1332

7.24

5.45

2007

1.7

1.77

0.77

37993

293

1599

7.55

6.17

2008

1.56

1.54

0.21

44111

95

1718

6.94

6.56

NEW PRIVATE SECTOR BANKS


Axis Bank
2005

1.54

2.18

1.59

37744

377

2422

8.87

7.49

2006

1.64

2.41

1.94

49731

497

2886

7.26

6.01

2007

1.66

2.33

2.09

73257

733

3403

6.42

5.04

2008

1.97

1.97

1.77

109578

1096

8769

10.17

8.90

2005

2.11

2.18

0.81

51429

417

4520

9.60

8.71

2006

2.3

2.41

0.75

73506

553

5299

8.55

7.66

2007

2.65

2.33

0.73

91236

669

6433

8.57

7.68

2008

2.81

1.97

1.92

133177

2558

11497

10.30

8.01

2005

1.97

2.18

0.74

167659

1233

12900

7.59

6.86

2006

1.99

2.41

1.07

251389

2701

22556

9.20

8.10

2007

1.94

2.33

1.29

344658

4463

24663

7.42

6.08

2008

2.04

1.97

1.22

399795

4897

46821

11.32

10.14

HDFC Bank

ICICI Bank

407

ANNEXURE V
Cost-Income Ratio
Bank-Wise Operational Risk Capital Charge and Impact on Tier I Capital
(Nationalized Banks, State Bank Group & Axis Bank, HDFC Bank & ICICI Bank)

Year

Cost-income
Ratio (%)

Average
PSB (%)

Three
standard
deviations
(decimal)

Gross
Income
(Rs. Crore)

Capital
Charge
(Rs. Crore)

Networth
(Rs. Crore)

Current
Tier I (%)

Estimated
Tier I
Capital (%)

2005

53.39

48.46

0.37

2004

735

2328

6.46

4.42

2006

50.30

51.94

0.19

2059

382

3639

9.53

8.53

2007

48.30

51.08

0.13

2127

266

4477

8.10

7.62

2008

43.90

48.32

0.12

2637

306

5221

7.75

7.30

2005

45.50

48.46

0.14

1822

259

1837

8.03

6.90

2006

54.96

51.94

0.12

1561

187

2894

12.20

11.41

2007

50.05

51.08

0.09

1864

172

3156

9.98

9.44

2008

47.18

48.32

0.07

2001

144

3249

8.54

8.16

2005

46.27

48.46

0.08

4284

347

5628

8.21

7.70

2006

55.45

51.94

0.11

4301

473

7845

10.98

10.32

2007

51.30

51.08

0.09

4959

436

8650

8.74

8.30

2008

49.21

48.32

0.08

5962

459

11044

7.63

7.31

2005

56.94

48.46

0.20

3393

689

4465

7.05

5.96

2006

55.42

51.94

0.20

3816

750

4984

6.75

5.73

2007

52.13

51.08

0.20

5003

980

5895

6.54

5.45

2008

41.68

48.32

0.16

6346

1019

10589

7.70

6.96

Allahabad Bank

Andhra Bank

Bank of Baroda

Bank of India

Bank of Maharashtra
2005

56.83

48.46

0.18

1267

226

1543

7.10

6.06

2006

64.42

51.94

0.32

1023

326

1573

7.47

5.92

2007

54.89

51.08

0.33

1359

447

1742

6.03

4.48

2008

55.40

48.32

0.31

1509

475

1782

5.13

3.76

2005

44.93

48.46

0.14

4694

677

6109

7.29

6.48

2006

47.92

51.94

0.16

4898

782

7132

7.81

6.95

2007

46.82

51.08

0.15

5478

794

10354

7.17

6.62

2008

48.53

48.32

0.12

5751

715

10500

7.01

6.53

Canara Bank

Central Bank of India


2005

14.57

48.46

0.78

3294

2569

3265

6.08

1.30

2006

22.32

51.94

0.96

2911

2799

3442

7.19

1.34

2007

57.08

51.08

0.96

2950

2842

3790

6.32

1.58

2008

57.93

48.32

0.67

3014

2028

5943

5.42

3.57

Annexure V

409

Corporation Bank
2005

39.35

48.46

0.37

1695

630

3054

13.55

10.75

2006

43.94

51.94

0.29

1700

495

3374

12.41

10.59

2007

41.36

51.08

0.33

1944

641

3765

11.30

9.38

2008

41.60

48.32

0.30

2144

649

4228

9.64

8.16

Dena Bank
2005

61.79

48.46

0.30

997

296

1104

6.63

4.85

2006

48.28

51.94

0.30

1162

354

1339

5.96

4.38

2007

49.04

51.08

0.30

1248

370

1497

6.06

4.56

2008

48.62

48.32

0.09

1337

119

1801

6.75

6.30

2005

48.80

48.46

0.29

1873

546

5936

7.60

6.90

2006

54.71

51.94

0.26

1974

507

2492

10.29

8.20

2007

47.87

51.08

0.09

2605

235

3841

12.28

11.53

2008

45.75

48.32

0.11

3060

322

5211

11.41

10.71

Indian Bank

Indian Overseas Bank


2005

43.65

48.46

0.16

2654

415

2575

7.10

5.95

2006

48.39

51.94

0.13

2608

345

3178

8.54

7.61

2007

47.08

51.08

0.15

2948

449

3991

8.20

7.28

2008

42.59

48.32

0.17

3487

580

4857

7.86

6.92

Oriental Bank of Commerce


2005

39.23

48.46

0.51

2029

1032

3327

5.42

3.74

2006

44.76

51.94

0.41

2158

893

5171

10.37

8.58

2007

43.50

51.08

0.30

2294

678

5601

10.05

8.83

2008

46.98

48.32

0.22

2299

513

5776

9.34

8.51

Punjab & Sind Bank


2005

68.88

48.46

0.89

829

735

440

5.26

-3.53

2006

64.31

51.94

0.90

751

674

1222

10.05

4.51

2007

52.56

51.08

0.51

995

505

1406

9.58

6.14

2008

50.86

48.32

0.27

1103

298

2093

8.04

6.90

Punjab National Bank


2005

57.68

48.46

0.20

5683

1165

8161

8.87

7.60

2006

50.88

51.94

0.20

5941

1197

9376

10.06

8.78

2007

50.73

51.08

0.20

6556

1292

10435

8.93

7.82

2008

46.81

48.32

0.04

7532

299

12318

8.52

8.31

2005

55.34

48.46

0.37

2284

854

2199

6.10

3.73

2006

58.76

51.94

0.25

2442

620

2834

7.40

5.78

2007

50.07

51.08

0.21

2768

572

3623

6.24

5.25

2008

50.47

48.32

0.15

2962

454

4291

6.62

5.92

2005

56.36

48.46

0.26

1925

504

2069

5.75

4.35

2006

60.67

51.94

0.26

1940

497

2462

6.09

4.86

2007

55.80

51.08

0.27

2138

576

2662

5.78

4.53

2008

57.79

48.32

0.29

2260

658

2926

5.05

3.91

Syndicate Bank

UCO Bank

Table contd...

410 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Table contd...
Union Bank
2005

44.40

48.46

0.15

2831

435

3614

6.07

5.34

2006

48.87

51.94

0.12

2869

357

4558

7.32

6.75

2007

42.45

51.08

0.21

3477

739

5190

7.79

6.68

2008

38.17

48.32

0.29

4173

1210

7348

7.45

6.22

United Bank of India


2005

50.54

48.46

0.17

1393

232

1957

14.15

12.47

2006

55.87

51.94

0.18

1457

260

1828

10.01

8.59

2007

51.94

51.08

0.10

1498

144

2415

7.72

7.26

2008

65.91

48.32

0.38

1370

525

2661

6.74

5.41

2005

40.21

48.46

0.30

1338

396

1590

7.59

5.70

2006

49.64

51.94

0.26

1257

325

1670

9.26

7.46

2007

48.33

51.08

0.19

1347

257

1897

7.07

6.11

2008

51.47

48.32

0.10

1362

138

2460

5.73

5.41

2005

26.72

48.46

0.47

8515

4000

24072

8.04

6.70

2006

28.34

51.94

0.69

8994

6169

27644

9.36

7.27

2007

54.18

51.08

0.68

21823

14923

31298

8.01

4.19

2008

49.03

48.32

0.51

25716

12992

49032

8.48

6.23

Vijaya Bank

State Bank of India

State Bank of Bikaner & Jaipur


2005

46.01

48.46

0.07

1352

93

1298

8.04

7.46

2006

61.24

51.94

0.21

1241

257

1406

9.36

7.65

2007

52.58

51.08

0.21

1432

296

1654

8.01

6.58

2008

53.15

48.32

0.22

1411

317

1713

6.95

5.66

State Bank of Hyderabad


2005

48.48

48.46

0.34

1384

465

1765

7.58

5.58

2006

52.54

51.94

0.23

1553

351

2114

8.95

7.47

2007

44.62

51.08

0.14

1811

249

2541

8.25

7.44

2008

44.64

48.32

0.16

1790

283

2694

7.24

6.48

State Bank of Indore


2005

48.24

48.46

0.34

680

230

904

6.67

4.97

2006

50.44

51.94

0.23

787

178

1018

7.55

6.23

2007

51.25

51.08

0.03

800

26

1177

6.74

6.59

2008

48.81

48.32

0.03

883

30

1317

7.01

6.85

State Bank of Mysore


2005

51.45

48.46

0.07

931

67

756

7.12

6.49

2006

53.69

51.94

0.08

948

75

935

7.44

6.84

2007

54.35

51.08

0.10

1034

105

1141

6.62

6.01

2008

52.11

48.32

0.11

1184

133

1378

6.54

5.91

State Bank of Patiala


2005

35.96

48.46

0.50

1332

671

2045

11.05

7.43

2006

45.54

51.94

0.42

1344

571

2235

9.96

7.42

2007

45.54

51.08

0.32

1447

463

2488

8.36

6.80

2008

47.58

48.32

0.18

1486

268

2459

6.74

6.01

Annexure V
State Bank of Saurashtra
2005

39.63

48.46

0.27

641

175

794

8.68

6.77

2006

52.74

51.94

0.27

584

156

977

9.02

7.58

2007

56.70

51.08

0.22

575

128

1043

8.17

7.17

2008

69.00

48.32

0.45

571

260

1145

8.06

6.23

State Bank of Travancore


2005

38.54

48.46

0.26

1305

342

1130

6.17

4.30

2006

48.36

51.94

0.26

1307

336

1332

7.24

5.41

2007

47.57

51.08

0.24

1356

320

1599

7.55

6.04

2008

47.51

48.32

0.11

1446

156

1718

6.94

6.31

NEW PRIVATE SECTOR BANKS


Axis Bank
2005

50.65

48.46

0.21

1147

241

2422

8.87

7.99

2006

45.02

51.94

0.30

1808

544

2886

7.26

5.89

2007

47.15

51.08

0.25

2577

634

3403

6.42

5.22

2008

49.19

48.32

0.24

4381

1073

8769

10.17

8.93

2005

44.69

48.46

0.19

2428

455

4520

9.60

8.63

2006

46.09

51.94

0.27

3669

988

5299

8.55

6.96

2007

46.33

51.08

0.29

5226

1515

6433

8.57

6.55

2008

49.87

48.32

0.23

7511

1740

11497

10.30

8.74

2005

52.74

48.46

0.10

6255

623

12900

7.59

7.22

2006

56.25

51.94

0.10

8890

881

22556

9.20

8.84

2007

53.25

51.08

0.03

12565

409

24663

7.42

7.30

2008

50.60

48.32

0.04

16115

571

46821

11.32

11.18

HDFC Bank

ICICI Bank

411

ANNEXURE VI
Graphical presentation of the Operational Risk Capital Charge
under the Basic Indicators Approach for Public Sector Banks
(2005-2009)

Graphical presentation of the Operational Risk Capital Charge


under the Basic Indicators Approach for Public Sector Banks
(2005-2009)

Operational Risk Capital Charge: Comparative Performance


for Public Sector Banks in 2005

Operational Risk Capital Charge: Comparative Performance of


Public Sector Banks in 2007

12000
12000
11345

9000

9000

Rs. Crore

Rs. Crore

8926

6000

6000

3000

3000
2389

1815

158

588

644 560

657
384
191 150 212

279 269

99

265 231 336


171 151

Graphical presentation of the Operational Risk Capital Charge


under the Basic Indicators Approach for Public Sector Banks
(2005-2009)

717

176

465
286380318
244
168
116

413
347
284
207198

190

Alla FSB
hab
a
An d
dhr
a
Bo
B
Bo
Bo I
Ma
Cs h
nar
a
Co Centr
rpo al
ratio
n
De
n
Ind a
ian
IoB
O
Pu rienta
n&
l
Sin
d
PN
B
Sy
n
UC
O
Un
io
Un n
ite
Vija d
ya
SB
_B SBI
ik_
J
SB ai
_H
SB yd
_
SB Ind
_M
y
SB s
_
SB Pat
_S
SB au
_Tr
av

295 249

145 189 103101 175 85 132

224
206 97
188
114 134

Alla FSB
hab
a
An d
dhr
a
Bo
B
Bo
Bo I
Ma
Cs h
nar
a
C
e
Co
rpo ntral
ratio
n
De
n
Ind a
ian
IoB
O
Pu rienta
n&
l
Sin
d
PN
B
Sy
n
UC
O
Un
ion
Un
ite
Vija d
ya
SB
S
_B BI
ik_
SB Jai
_H
SB yd
_
SB Ind
_M
y
SB s
_P
SB at
_S
a
u
SB
_Tr
av

527 531
191 219

856

Graphical presentation of the Operational Risk Capital Charge


under the Basic Indicators Approach for Public Sector Banks
(2005-2009)

Operational Risk Capital Charge: Comparative Performance of


Public Sector Banks in 2006

Operational Risk Capital Charge: Comparative Performance of


Public Sector Banks in 2008

12000

12000

12049

10312

9000

Rs. Crore

6000

3000

6000

3000
2152

1967

777
660
553
442
386
254337298
310 262
225
177
191185
161
114

173

207
196
108 120
94 164

677611
310 262
182

754
458
323411324
267
170
129

909
375

300

459
217197

237
201 113146 206 90 198

FSB
Allahabad
Andhra
BoB
BoI
BoMah
Csnara
Central
Corporation
Dena
Indian
IoB
Oriental
Pun & Sind
PNB
Syn
UCO
Union
United
Vijaya
SBI
SB_Bik_Jai
SB_Hyd
SB_Ind
SB_Mys
SB_Pat
SB_Sau
SB_Trav

597
264 238

Alla FSB
hab
a
An d
dhr
a
Bo
B
B
Bo oI
Ma
Cs h
nar
a
Co Centr
rpo al
ratio
n
De
n
Ind a
ian
IoB
O
Pu rienta
n&
l
Sin
d
PN
B
Sy
n
UC
O
Un
ion
Un
ite
Vija d
ya
SB
_B SBI
ik_
SB Jai
_H
SB yd
_
SB Ind
_M
y
SB s
_
SB Pat
_S
SB au
_Tr
av

Rs. Crore

9000

Annexure VI

Graphical presentation of the Operational Risk Capital Charge


under the Basic Indicators Approach for Public Sector Banks
(2005-2009)
Operational Risk Capital Charge: Comparative Performance of Public
Sector Banks in 2009
15000

13111

12000

Rs. Crore

9000

6000

2827

3000

761 758

195

806

1001
526
452
444 289
409
338
216
317
187 382
198
142

204

258
158 214 87 205
124

Alla FSB
hab
a
An d
dhr
a
Bo
B
B
Bo oI
Ma
Cs h
nar
a
Co Centr
rpo al
rati
on
De
n
Ind a
ian
IoB
O
Pu rienta
n&
l
Sin
d
PN
B
Sy
n
UC
O
Un
ion
Un
ite
Vija d
ya
SB
_B SBI
ik_
SB Jai
_H
SB yd
_
SB Ind
_M
y
SB s
_
SB Pat
_S
SB au
_Tr
av

341 271

413

ANNEXURE VII
Graphical presentation of the operational Risk Capital Charge
under the Basic Indicators Approach for New Private Sector
Banks (2005-2009)
Rs. Crore

2500

Com parative Pe rform ance of Operational Ris k Capital Charge in 2005

1925

1500

1004

1000
500

1121

1200

Rs. Crore

2000

Com parative Pe rform ance of Ope rational Ris k Capital Charge


in 2007

203

396

0
PVT

1000

Ax is

HDFC

ICICI

800
535

600
400
123

200

204

0
PVT

Axis

HDFC

ICICI

Graphical presentation of the operational Risk Capital Charge


under the Basic Indicators Approach for New Private Sector
Banks (2005-2009)

Com parative Pe rform ance of Operatinal Risk Capital Charge in


2009

Com parative Perform ance of Operational Risk Capital Charge in 2006

1494
1500
789

1000
500

149

278

0
PVT

Rs. Crore

2500
2000

Axis

HDFC

ICICI

Com parative Pe rform ance of Ope rational Ris k Capital Charge


in 2007
1925

1500

1004

1000
500

203

396

0
PVT

Ax is

HDFC

ICICI

Rs. Crore

Rs. Crore

2000

4000
3500
3000
2500
2000
1500
1000
500
0

3627
1879
438
PVT

Axis

820

HDFC

ICICI

ANNEXURE VIII
Operational Risk Capital Charge (BIA) Bank of Baroda
700
Rs. Crore

Graphical presentation of the operational risk capital charge


that individual nationalized banks have to keep aside on account
of operational risk and its impact on their Tier I capital for
2005 to 2009 under the Basic Indicators Approach

500
300
100
-100

Operational Risk Capital Charge (BIA) Allahabad Bank

2005

2006

Rs. Crore

295

264
219

100

Impact on tier I Capital (BIA) Bank of India

10
2006

2007

2008
Capital Charge

Year

Percent

2005

9.53
6.46

8.84

8.1

7.75

7.57

7.29

6.75

7.7

6.54

7.26
6.21

6.00

5.92

2005

2006

2007
Year
Current Tier I

2
2008
Estimated Tier I Capital

Operational Risk Capital Charge (BIA) Bank of India


2006

2007
Year
Current Tier I

2008
Estimated Tier I Capital

Im pact on Tier I Capital (BIA) Andhra Bank

800

758

600
400

531

553

560

611

200
0

15
12.2

Percent

7.05

5.85

2005

6
4
0

Rs. Crore

Percent

Impact on Tier I Capital (BIA) Allahabad Bank

10

2008
Capital Charge

310

300

12
10
8
6
4
2
0

2007
Year

400

200

677

644

597

527

8.03

2005

9.98

11.20

9.19

7.19

2006

8.54

2007

2008

Year

2009
Capital Charge

7.85

0
2006

Year

2007

Current Tier I

Estimated Tier I Capital

Percent

Im pact on Tier I Capital (BIA) Bank of Baroda


12
10
8
6
4
2
0

10.98
8.21

2005

8.09

2006

Year

300
200
100

195
158

177

2005

2006

176

182

2007

2008

8.74

10.14

7.44

Operational Risk Capital Charge (BIA) Bank of Maharashtra

2008

Rs. Crore

2005

2007

Current Tier I

7.63
7.16

2008
Estimated Tier I Capital

Year

2009
Capital Charge

416 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Operational Risk Capital Charge (BIA) Corporation Bank

Impact on Tier I Capital (BIA) Bank of Maharashtra


7.47

6.03

400

6.63

6.37

5.13
4.60

5.42

Rs. Crore

7.1

300
200
0

2005

2006

2007

Year

2005

2008

Current Tier I

806
588

Percent

400

2008

2009
Capital Charge

Impact on Tier I Capital (BIA) Corporation Bank


15

754

717

660

2007

Estimated Tier I Capital

800
600

2006

Year

Operational Risk Capital Charge (BIA) Canara Bank

Rs. Crore

289

267

244

225

191

100

200

10

13.55

12.41

12.70

11.3

11.58

9.64

10.57

9.03

0
2005

2006

2007

2008

2009

Year

0
2005

Capital Charge

Impact on Tier I Capital (BIA) Canara Bank


7.29

7.81
7.09

6.59

6.67

6.51

100

187
150

161

2005

2006

2006

Year

2007

2008

Current Tier I

Estimated Tier I Capital

170

2007

2008

Impact on Tier I Capital (BIA) Dena Bank

Operational Risk Capital Charge (BIA) Central Bank of India

9
384

465

442

458

444

200

6.63
Percent

400

2009
Capital Charge

Year

600

Rs. Crore

168

0
2005

5.96
6.06

6
5.73

5.38

5.25

0
2005

2006

2007

2008

5.37

500

7.19

6.32
5.42

6.27

5.55

5.00

0
2005

2006

2007
Estimated Tier I Capital

Operational Risk Capital Charge (BIA) Indian Bank

Year

2007

Current Tier I

2008
Estimated Tier I Capital

Rs. Crore

6.08

2006
Current Tier I

Impact on Tier I Capital (BIA) Central Bank of India


9

Year

2005

2009
Capital Charge

Year

2008
Estimated Tier I Capital

7.01

2007
Year
Current Tier I

200

7.17
Rs. Crore

Percent

2006

Operational Risk Capital Charge (BIA) Dena Bank

10

Percent

Percent

400

382

300
200
100

212

254

286

323

0
2005

2006

2007
Year

2008

2009
Capital Charge

417

Annexure VIII
Operational Risk Capital Charge (BIA) Punjab & Sind Bank
150

12.28

10.29
7.6

11.37

10.70

9.24

7.33

100

2006

2007

Year

50

2005

2008

Current Tier I

452

400
380

411

337

279

12
10
8

0
2006

2007

2008

10.05

7.55

2009

2005

Percent

7.86

7.42

6.33

7.20

3
0
2005

2006

Year

2007

1200
1000
800
600
400
200
0

2008

Current Tier I

777

2005

2006

2007

10.05

269

Percent

Rs. Crore

200

2008

2009
Capital Charge

9.58

10
324

318

909

856

Im pact on Tier I Capital (BIA) Punjab National Bank


12

300

8.04

9.12

5.26

8.79

7.55

4.07

100

2
0

0
2005

2006

2007
Year

2008

2005

2009
Capital Charge

10.37

9.77

6
3

10.05
9.48

2008
Estimated Tier I Capital

Operational Risk Capital Charge (BIA) Syndicate Bank

9.34
5.42

2007
Current Tier I

500

8.82

4.98

409

400
Rs. Crore

2006
Year

Im pact on Tier I Capital (BIA) Oriental Bank of Com m erce


12

Percent

657

Year

338
298

2008
Estimated Tier I Capital

1001

Estimated Tier I Capital

Operational Risk Capital Charge (BIA) Oriental Bank of


Com m erce

400

2007
Current Tier I

Operational Risk Capital Charge (BIA) Punjab National Bank

8.2

7.63

2006
Year

Rs. Crore

7.1

8.79

4.07

Im pact on Tier I Capital (BIA) Indian Overseas Bank


8.54

2009
Capital Charge

9.58

9.12

Capital Charge

Year

2008

8.04

5.26

6
4
2
0

100
2005

2007
Year

Impact on Tier I Capital (BIA) Punjab & Sind Bank

Percent

Rs. Crore

500
300

2006

Estimated Tier I Capital

Operational Risk Capital Charge (BIA) Indian Overseas Bank

129

116

114

99

2005

200

142

11.41
Rs. Crore

Percent

Impact on Tier I Capital (BIA) Indian Bank


14
12
10
8
6
4
2
0

300
200

265

310

347

375

100
0
2005

2006
Year

2007
Current Tier I

2008
Estimated Tier I Capital

0
2005

2006

2007
Year

2008

2009
Capital Charge

418 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Im pact on Tier I Capital (BIA) Syndicate Bank
7.4

6.24

6.1
Percent

6
4

Operational Risk Capital Charge (BIA) United Bank


250

6.62

6.59

6.04

5.37

5.64

216

200

Rs. Crore

150
100

171

50
0

0
2005

2006
Year

2007
Current Tier I

2005

2008
Estimated Tier I Capital

15

300

284

262

231

Percent

Rs. Crore

317

300
200

2005

2006

2007

2008

7.72
6.74

8.96

5.44

5.11

2006
Year

6.19

2007
Current Tier I

2008
Estimated Tier I Capital

Operational Risk Capital Charge (BIA) Vijaya Bank

5.16
4.53

200
Rs. Crore

Percent

198

150
100

185

198

197

151

50
0
2006

2007
Current Tier I

2005

2008
Estimated Tier I Capital

10

386

336

459

413

2006

2007
Year

2008

2009
Capital Charge

Impact on Tier I Capital (BIA) Vijaya Bank


9.26

7.59

7.07

526
Percent

Rs. Crore

10.01

12.91

2005

Operational Risk Capital Charge(BIA) Union Bank

8.24
6.87

5.73
6.33

5.27

2
0
2005

2006

2007

2008

2009
Capital Charge

Year

Im pact on Tier I Capital (BIA) Union Bank


10

Percent

10

250
5.05

Year

7.32

7.79

6.07

6
4

2009
Capital Charge

2009

Im pact on Tier I Capital (BIA) UCO Bank


5.75
6.09
5.78

2005

2008

14.15

Capital Charge

Year

600
500
400
300
200
100
0

2007
Year

7.06

7
6
5
4
3
2
1
0

2006

Impact on Tier I Capital (BIA) United Bank

Operatioal Risk Capital Charge (BIA) UCO Bank


400

100

217

207

191

6.70

5.51

7.17

7.45
6.98

2
0
2005

2006
Year

2007
Current Tier I

2008
Estimated Tier I Capital

2005

2006
Year

2007
Current Tier I

2008
Estimated Tier I Capital

ANNEXURE IX
Im pact on Tier I Capital Bank of Maharashtra
8

6
Percent

A graphical comparison of the impact of Operational Risk


Capital Charge on Tier I Capital on Public Sector Banks and
Major New Private Sector Banks (Bank Wise) under the
Basic Indicators Approach and the alternative measures of
Operational Risk

Im pact on Tier I Capital Allahabad Bank

Percent

10
9
8
7
6
5
4
3
2
1
0

0
2005

2006

Current Tier I

Year

BIA

2007

2008

Cost-Income Ratio

Cost-Asset Ratio

Im pact on Tier I Capital Canara Bank

2005

2006
Current Tier I

Year
BIA

2007
Cost-Income Ratio

2008
Cost-Asset Ratio

8
6

Percent

Im pact on Tier I Capital Andhra Bank


4

12
10
Percent

8
6

0
2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

0
2005

2006
Current Tier I

Year
BIA

2007

2008
Im pact on Tier I Capital Central Bank of India

Cost-Income Ratio

Cost-Asset Ratio
8

Im pact on Tier I Capital Bank of Baroda

Percent

12
10

Percent

6
4

0
2005

2006

Current Tier I

0
2005

2006
Current Tier I

Year

BIA

2007
Cost-Income Ratio

Year

2007

2008

BIA

Cost-Income Ratio

Cost-Asset Ratio

2008
Cost-Asset Ratio

Im pact on Tier I Capital Corporation Bank

14

Im pact on Tier I Capital Bank of India


8

12
10

Percent

Percent

8
6
4

2
0

0
2005

2006
Current Tier I

Year
BIA

2007
Cost-Income Ratio

2008
Cost-Asset Ratio

2005

2006

Current Tier I

2007
Year
BIA
Cost-Income Ratio

2008

Cost-Asset Ratio

420 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Im pact on Tier I Capital Dena Bank

Im pact on Tier I Capital Punjab & Sind Bank


12

10
8

Percent

6
Percent

4
2
0
-2

2006

2007

2008

-4

0
2005

2006

Current Tier I

Year

2007

BIA

Cost-Income Ratio

-6

2008

Current Tier I

Cost-Asset Ratio

Im pact on Tier I Capital Indian Bank

BIA

Cost-Income Ratio

Cost-Asset Ratio

Impact on Tier I Capital Punjab National Bank

12

10

10

8
Percent

Percent

Year
2005

8
6

6
4

4
2

0
2005

2006
Current Tier I

Year
BIA

2007

2008

Cost-Income Ratio

Cost-Asset Ratio

2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

Impact on Tier I Capital Syndicate Bank

Im pact on Tier I Capital Indian Overseas Bank


8

8
6

Percent

Percent

6
4

4
2

0
2005

2006
Current Tier I

Year
BIA

2007
Cost-Income Ratio

2005

2008

2006

Current Tier I

Cost-Asset Ratio

Im pact on Tier I Capital Oriental Bank of Com m erce

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

Impact on Tier I Capital UCO Bank


7

10

6
5
Percent

Percent

8
6
4

4
3
2

1
0

0
2005

2006
Current Tier I

Year
BIA

2007
Cost-Income Ratio

2008
Cost-Asset Ratio

2005

2006

Current Tier I

Year

BIA

2007

2008

Cost-Income Ratio

Cost-Asset Ratio

Annexure IX
Impact on Tier I Capital United Bank of India

421

Impact on Tier I Capital State Bank of Hyderabad

16

10

14

12

Percent

Percent

10
8
6

2
0

0
2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

2005

Current Tier I

Cost-Asset Ratio

Im pact on Tier I Capital Vijaya Bank

Cost-Income Ratio

2008
Cost-Asset Ratio

Percent

Percent

2007

BIA

Impact on Tier I Capital State Bank of Indore

4
2

2
0

2005

2006
Current Tier I

2007

Year

BIA

Cost-Income Ratio

2008

2005

Cost-Asset Ratio

2006

Current Tier I

Impact on Tier I Capital State Bank of India

Year

BIA

2007

2008

Cost-Income Ratio

Cost-Asset Ratio

Impact on Tier I Capital State Bank of Mysore


8

10
9
8
7
6
5
4
3
2
1
0

6
Percent

Percent

Year

10

4
2
0

2005

2006

Current Tier I

Year

BIA

2005

2007

Cost-Income Ratio

2006

Current Tier I

Cost-Asset Ratio

Impact on Tier I Capital State Bank of Bikaner & Jaipur

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

Impact on Tier I Capital State Bank of Patiala

10

12

10
8

Percent

Percent

2006

4
2

6
4
2
0

0
2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

2005

2006

Current Tier I

Year

BIA

2007

2008

Cost-Income Ratio

Cost-Asset Ratio

422 Operational Risk Measurement for the Indian Banking Sector: Alternative Measures
Impact on Tier I Capital State Bank of Saurashtra

Impact on Tier I Capital HDFC Bank


10

Percent

10

Percent

6
4
2

4
2

2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

2005

Cost-Asset Ratio

2006

Current Tier I

Year

BIA

2007

2008

Cost-Income Ratio

Cost-Asset Ratio

Impact on Tier I Capital ICICI Bank

Impact on Tier I Capital State Bank of Travancore

12

10
8

Percent

6
Percent

6
4
2

0
2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

Impact on Tier I Capital Axis Bank


10

Percent

8
6
4
2
0
2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

2005

2006

Current Tier I

Year

BIA

2007

Cost-Income Ratio

2008

Cost-Asset Ratio

You might also like