Research Project Report on

“Study of Probability of Defaulter”

Master in Business Administration- Digital Business (MBA-DB)

by

SUNDARAM MISHRA

20210305017

Under the guidance of

Dr. Venkata Mrudula

1
A STUDY ON THE PROBABILITY OF DEAUFAULTERS OF BANK IN
INDIA

SUNDARAM MISHRA

20210305017

Abstract

This paper explores the study of the probability of defaulters of banks in India. To find out
the desired result according to study being a sample of one public sector bank which is
Punjab National Bank and one private sector bank which is AXIS Bank is selected. Different
measurement tools are available to manage the different type of risks in banks, hence
operational risk measurement and management have been analyzed in this paper.

EXECUTIVE SUMMARY

Banking industry has gone through one of the worst crisis in recent times, and is still
recovering from the after-shocks. However, there were a lot of learnings that banks would
have taken away from this crisis. One of them is the need for a robust risk management
system. The crisis dealt a blow to the banking system, catching them off guard when it
came to foreseeing the risk. Banks, in the credit card business, face financial risk in the
form of both credit risk and fraud risk. This paper builds upon their technique to predict
the fraud risk posed by the merchants to the banks.

2
 Fraud risk is an important aspect of risk management systems, particularly in the credit
space. The uncertainty surrounding the receipt of paybacks calls for designing robust risk
prediction models. Fraud risk is very different from credit risk because fraud risk does not
follow a pattern. It happens suddenly, and may not always have a trend before it happens.
This creates a need for separate model for fraud risk prediction. This paper develops a
fraud risk prediction model that uses logistic regression technique, deployed using SAS.
The setup of the study is the merchant-bank relationship in the credit card industry.

The model developed in this paper triggers on a transaction level, and assigns a
‘probability score of default (PF) to each merchant for a possible fraud risk whenever a
transaction is done at the merchant. Such a score warns the management in advance of
probable future losses on merchant accounts. Banks can rank order merchants based on
their PF score, and instead of working on the entire merchant portfolio, they can focus on
the relatively riskier set of merchants. The PF model is validated by comparing the actual
defaults with those predicted by the model and a good alignment is found between the
two. The results show that the model can capture 62 percent frauds in the first decile when
the transactions are sorted by the probability of fraud computed by the model.

What is Credit Risk?

Credit risk arises when a corporate or individual borrower fails to meet their debt obligations.
It is the probability that the lender will not receive the principal and interest payments of a
debt required to service the debt extended to a borrower.

On the side of the lender, credit risk will disrupt its cash flows and also increase collection
costs, since the lender may be forced to hire a debt collection agency to enforce the
collection. The loss may be partial or complete, where the lender incurs a loss of part of the
loan or the entire loan extended to the borrower.

The interest rate charged on a loan serves as the lender’s reward for accepting to bear credit
risk. In an efficient market system, banks charge a high interest rate for high-risk loans as a
way of compensating for the high risk of default. For example, a corporate borrower with a
steady income and a good credit history can get credit at a lower interest rate than what high-
risk borrowers would be charged.

3
Conversely, when transacting with a corporate borrower with a poor credit history, the lender
can decide to charge a high interest rate for the loan or reject the loan application altogether.
Lenders can use different methods to assess the level of credit risk of a potential borrower in
order to mitigate losses and avoid delayed payments.

Types of Credit Risk

The following are the main types of credit risks:

1. Credit default risk

Credit default risk occurs when the borrower is unable to pay the loan obligation in full or
when the borrower is already 90 days past the due date of the loan repayment. The credit
default risk may affect all credit-sensitive financial transactions such as loans, bonds,
securities, and derivatives.

The level of default risk can change due to a broader economic change. It can also be due
because of a change in a borrower’s economic situation, such as increased competition or
recession, which can affect the company’s ability to set aside principal and interest payments
on the loan.

2. Concentration risk

Concentration risk is the level of risk that arises from exposure to a single counterparty or
sector, and it offers the potential to produce large amounts of losses that may threaten the
lender’s core operations. The risk results from the observation that more concentrated
portfolios lack diversification, and therefore, the returns on the underlying assets are more
correlated.

For example, a corporate borrower who relies on one major buyer for its main products has a
high level of concentration risk and has the potential to incur a large amount of losses if the
main buyer stops buying their products.

3. Country risk

Country risk is the risk that occurs when a country freezes foreign currency payments
obligations, resulting in a default on its obligations. The risk is associated with the country’s

4
political instability and macroeconomic performance, which may adversely affect the value
of its assets or operating profits. The changes in the business environment will affect all
companies operating within a particular country.

Types of Risks in Banks

Broadly speaking, Risks in the Banking sector are of two types namely Systematic Risks and
Unsystematic Risks. Lets us define these two types of risks in Banks and understand the
concept behind them.

1. Systematic Risks:

 It is the risk inherent to the entire market or a market segment, and it can affect a large
number of assets.
 Systematic risk is also known as Undiversifiable Risk or Volatility and market risk.
 Systematic risk affects the overall market and not just a stock or industry in particular.
 This type of risk is both unpredictable and impossible to avoid completely.
 Examples of it include interest rate changes, inflation, recessions, and wars.

2. Unsystematic Risks:

 It is the risk that affects a very small number of assets.

 It is also called Non-systematic Risk, Specific Risk, Diversifiable Risk, and Residual
Risk.
 This type of risk refers to the uncertainty inherent to a company or industry investment
in particular.
 Examples include a change in management, a product recall, a regulatory change that
could drive down company sales, and a new competitor in the marketplace with the
potential to take away market share from a company in which you’ve invested.
 It is possible to avoid Unsystematic Risks through diversification.

5
In addition to these broad categories of Risks, there are several other specific risks that the
banking sector faces.

3. Credit of Default Risk:

It is the risk in which a borrower is unable to pay the interest or principal on its debt
obligations. The Basel Committee on Banking Supervision defines credit risk as the potential
that a bank borrower, or counter-party, will fail to meet its payment obligations regarding the
terms agreed with the bank. It includes both uncertainties involved in repayment of the
bank’s dues and repayment of dues on time.

4. Market Risk:

The Basel Committee on Banking Supervision defines market risk as the risk of losses in on-
balance or off-balance sheet positions that arise from movement in market prices. Market risk
is the most prominent for banks present in investment banking. 

The four components of market risk:

 Interest Risk: It causes potential losses due to movements in interest rates. This risk
arises because a bank’s assets usually have a significantly longer maturity than its
liabilities. In the banking language, management of interest rate risk is also called
asset-liability management (ALM).

 Equity Risk: It causes potential losses due to changes in stock prices as banks accept
equity against disbursing loans. Banks can accept equity as collateral for loans and
purchase ownership stakes in other companies as investments from their free or
investible cash. Any negative change in stock price either leads to a loss or diminution
in investments’ value.

 Commodity Risk: It causes potential losses due to changes in

commodity (agricultural, industrial, energy) prices. Massive fluctuations occur in these
prices due to continuous variations in demand and supply. Banks may hold them as
part of their investments, and hence face losses. The commodities’ values fluctuate a

6
 great deal due to changes in demand and supply. Any bank holding them as part of an
investment is exposed to commodity risk.

Foreign Exchange Risk:

 It causes potential loss due to changes in the value of the bank’s assets or liabilities
resulting from exchange rate fluctuations as banks transact with their customers or
other stakeholders in multiple currencies. Banks transact in foreign exchange for their
customers or the banks’ accounts. Any adverse movement can diminish the value of
the foreign currency and cause a loss to the bank.

5. Liquidity Risk:

It can be defined as the risk of a bank not being able to finance its day-to-day operations.
Failure to manage this risk could lead to severe consequences for the bank’s reputation as
well as the bond pricing and ratings of the bank in the money market.

6. Country Risk:

Country risk refers to the risk that a country won’t be able to honour its financial
commitments. When a country defaults on its obligations, it can harm the performance of all
other financial instruments in that country as well as other countries it has relations with.
Country risk applies to stocks, bonds, mutual funds, options, and futures that are issued
within a particular country. 

7. Operational Risk:

The Basel Committee on Banking Supervision defines operational risk as to the risk of loss
resulting from inadequate or failed internal processes, people, and systems or external events.

7
All banks face operational risks in their day-to-day operations across all their departments
including treasury, credit, investment, and information technology.

There are three main causes of this risk:

 Human Intervention & Error

 Failure of the IT/internal software & systems.
 Failure of Internal Processes to transmit data & information accurately

8. Reputational Risk:

Reputational risk implies the public’s loss of confidence in a bank due to a negative

perception or image that could be created with/without any evidence of wrongdoing by the
bank. Reputational value is often measured in terms of brand value. Advertisements play a
significant role in forming & maintaining public perception, which is the key reason for
banks spending millions in content marketing dollars.

The reputational risk could stem from:

 The inability of the bank to honour government/regulatory commitments

 Nonobservance of the code of conduct under corporate governance
 Mismanagement/Manipulation of customer records
 Ineffective customer service/after-sales services

9. Systemic Risk:

This risk includes the possibility of bringing the entire financial system to a standstill. This is
caused due to a domino effect where the failure of one bank could ripple down the failure of
its counterparties/other stakeholders, which could, in turn, threaten the entire financial
services industry.

Risks such as systemic risk, which the banks have little or no control over, can only be
mitigated if banks have a strong capital base, to ensure a sound infrastructure.

8
Factors Affecting Credit Risk Modelling

In order to minimize the level of credit risk, lenders should forecast credit risk with greater
accuracy. Listed below are some of the factors that lenders should consider when assessing
the level of credit risk:

1. Probability of Default (POD)

The probability of default, sometimes abbreviated as POD, is the likelihood that a borrower
will default on their loan obligations. For individual borrowers, POD is based on a
combination of two factors, i.e., credit score and debt-to-income ratio.

The POD for corporate borrowers is obtained from credit rating agencies. If the lender
determines that a potential borrower demonstrates a lower probability of default, the loan will
come with a low interest rate and low or no down payment on the loan. The risk is partly
managed by pledging collateral against the loan.

9
2. Loss Given Default (LGD)

Loss given default (LGD) refers to the amount of loss that a lender will suffer in case a
borrower defaults on the loan. For example, assume that two borrowers, A and B, with the
same debt-to-income ratio and an identical credit score. Borrower A takes a loan of $10,000
while B takes a loan of $200,000.

The two borrowers present with different credit profiles, and the lender stands to suffer a
greater loss when Borrower B defaults since the latter owes a larger amount. Although there
is no standard practice of calculating LGD, lenders consider an entire portfolio of loans to
determine the total exposure to loss.

3. Exposure at Default (EAD)

Exposure at Default (EAD) evaluates the amount of loss exposure that a lender is exposed to
at any particular time, and it is an indicator of the risk appetite of the lender. EAD is an
important concept that references both individual and corporate borrowers. It is calculated by
multiplying each loan obligation by a specific percentage that is adjusted based on the
particulars of the loan.

LITERATURE REVIEW

Cressey (1953), a criminologist, first studied the motivation behind committing fraud. He
formulated the well-known “fraud triangle”, which has its three vertices as ‘Pressure’,
‘Opportunity’, and ‘Rationalization’. Cressey concluded from his research that “Trust
violators when they conceive of themselves as having a financial problem which is non-
shareable, have knowledge or awareness that this problem can be secretly resolved by
violation of the position of financial trust, and are able to apply to their own conduct in that
situation verbalizations which enable them to adjust their conceptions of themselves as
trusted persons with their conceptions of themselves as users of the entrusted funds or

10
property” (Cressey, 1953, p. 742). Credit card fraud risk prediction and modelling has been a
research interest of many banks around the world and a number of techniques, with special
emphasis on regression techniques. Molyneaux (2007) and George (1992) also discussed the
ethics of banking and Clarke (1994), in his work, mentioned about the moral complexity of
fraudulent behaviour. Ghosh and Reilly (1994) proposed a fraud detection system that used
past fraud cases due to lost cards, stolen cards, and application frauds. Syeda (2002) used
parallel granular neural networks (PGNNs) to improve the knowledge discovery process in
credit card fraud detection. Fan, Prodromidis, and Stolfo (1999) suggested the application of
distributed data mining in credit card fraud detection. Brause, Langsdorf, and Hepp (1999)
developed an approach that involved advanced data mining techniques and neural network
algorithms to obtain high fraud coverage. Chiu and Tsai (2004) proposed web services and
data mining techniques to establish a collaborative scheme for fraud detection in the banking
industry. Not much literature is available on addressing the risk faced by a bank from a
merchant. This study aims to fill this existing gap

Fraud Risk

Fraud Risk is the possibility that a counter party engages in an activity which entails earning
revenue through the means that act as a breach of contract with the signatory party. The
authors discuss the cause of fraud risk in the merchant-acquirer relationship. Fraud risk is a
challenge to the acquiring banks because the merchants always have a motivation to earn
money through fraudulent ways, primarily because merchants look upon the discount rate
charged by the banks as a cost, and they engage in activities in order to compensate for this
perceived cost. These activities invariably breach the contract between the merchant and the

11
bank, and are termed as fraud. To understand this, an example of a merchant selling high
value product, with irregular business cycle is considered. Such a merchant could have a
small customer base. This concept is termed as ‘card member concentration’, which means
that most of the transactions at the merchant are initiated by a small group of customers and
therefore the transactions are ‘concentrated’ around a few customers. In such a scenario, the
possibility of fraud increases tremendously, because the merchant and customer can tie up
together for committing the fraud. Another scenario where fraud is possible is when a
merchant uses his personal credit card, issued by the same bank as the acquiring bank, and
swipes the card on his POS device. In this way, the merchant has not sold anything. He has
just swiped his personal card, and submits the slip generated by the POS device to the
acquiring bank for claiming the money. Typically, this money is credited in two to three days,
after deducting 2-3 percent as the surcharge, also known as the discount rate. This amount
will appear on the statement of the personal card of the merchant, which would need to be
paid after 50 days approximately. Upon looking at this transaction from another perspective,
it was realized that this amounted to taking a loan from the bank for 50 days, at a rate of 2-3
percent. This is a breach of contract by the merchant, and should be termed as fraud. The last
scenario that is considered for fraud is the possibility of a merchant who has a low business
volume under usual circumstances, but has some big spikes in the volume in the recent past.
In such a scenario, the mathematical slope of the spike in volume is calculated, and if that
slope is more than a certain cut-off value, then that merchant is tagged as a possible fraud
case.

FRAUD RISK MODELING IN MERCHANTACQUIRER

RELATIONSHIP

The Need An acquiring bank can have millions of merchants in its global portfolio.
While all merchants would be risky at some level, all of them do not pose the same credit
risk. The management of the bank cannot build risk management strategy for the entire
merchant portfolio. Hence, it would be helpful for the bank if they can assign a
‘probability of fraud’ (PF) score to the merchants based on the level of risk posed by

12
 them, and then sort the list of merchants in decreasing order of this probability score. The
authors propose to build a statistical model, which would help the acquiring bank in the
following ways: • Generate PF scores for all merchants in the portfolio of the bank, which
would help the bank in identifying the high risk merchants, and plan accordingly • Act as
an “Early Warning System”, which would give the acquiring bank some visibility into the
potentially “invisible” fraud risk in the future The above-mentioned points enable the
banks to create sufficient reserves in advance, in view of the fraud risk they foresee using
the model. Also, the banks can do further analysis on the kind of industries, geographies,
etc. that are posing more risk, and plan accordingly.

Methodology:

As discussed earlier, the number of merchants in the global portfolio of the bank can be
in millions. This creates a need for a statistical model. To make an effective model, the
authors propose to first create segments of merchants, based on their perceived risk
profile, and then create the required models for each of the segments.

13
 1
Create Merchant Segments

2
Identify the type of statistical
model to be built

3
Credit
Risk Identify the variables suitable for
Modelling the model for each segment

4
Test the statistical significance
of each variable

5
Build models of each segment
and validate the results using
“out of time-out of sample” data

PROFILE OF PUNJAB NATIONAL BANK (Based on September’22

Results)

Punjab National Bank (PNB), India’s first Swadeshi Bank, commenced its operations on
April 12, 1895 from Lahore, with an authorized capital of ₹ 2 lac and working capital of

14
 ₹ 20,000. The Bank was established by the spirit of nationalism and was the first bank
purely managed by Indians with Indian Capital. During the long history of the Bank, 9
banks have been merged/ amalgamated with PNB.

BUSINESS PERFORMANCE
PNB is the second largest Public Sector Bank (PSB) in the country with Global Gross
Business at ₹ 20,23,713 Crore. The Bank continues to maintain its forte in low cost CASA
deposits with a share of 44.91%. Bank’s focus has been on qualitative business growth,
recovery and arresting fresh slippages.

The Business figures of the Bank as on September’2022 is given as under: (₹

Crore)
YoY variation
Sl. Parameters Sep’21 Mar’22 Sep’22
Amt. %
1 Domestic Deposits 1098117 1125049 1167872 69755 6.4%
2 Global Deposits 1115373 1146218 1193501 78128 7.0%
3 Gross Domestic Advances 715799 759214 796839 81040 11.3%
4 Gross Global Advances 735724 785104 830212 94488 12.8%
5 Gross Domestic Business 1813916 1884263 1964711 150795 8.3%
6 Gross Overseas Business 37181 47060 59002 21821 58.7%
7 Gross Global Business 1851097 1931322 2023713 172616 9.3%
8 CASA Deposits 498757 533654 524448 25691 5.2%
9 Current Deposits 71958 81974 72741 783 1.1%
10 Saving Deposits 426799 451680 451707 24908 5.8%
11 CASA Share Domestic (%) 45.42% 47.43% 44.91%

PROFITABILITY

In terms of bottom-line parameters, the Bank’s Operating Profit increased by 38.5% on

Y-OY basis to ₹ 5567 Crore in Q2 FY’23 from ₹ 4021 Crore in Q2 FY’22. Net Interest

15
Income (NII) increased by 30.2% on Y-o-Y basis to ₹ 8271 Crore in Q2 FY’23 from ₹
6353 Crore in Q2 FY’22. Global NIM improved by 21 bps to 3.00% in Q2 FY’23 from
2.79% in Q1 FY’22.
Cost to income ratio for Q2’FY23 at 49.93% improved by 834 bps on Y-o-Y basis.

RoA improved by 3 bps in Q2’FY23 to 0.12% from 0.09% in Q1’FY23. Roe improved
by 55 bps in Q2’FY23 to 2.56% from 2.01% in Q1’FY23. Net Profit of the Bank
increased to ₹ 411 Crore during Q2’FY23, with Q-o-Q growth of 33.4%.

CAPITAL ADEQUACY RATIO

CRAR was 14.74% for September’22. Tier-I is at 12.20% (CET-1 was at 10.88%, AT1
was at 1.32%) and Tier-II is at 2.54% as at September’22.

ASSET QUALITY
Management of stressed assets continues to be one of the top priorities for the Bank. As a
result of focused efforts in this direction, Gross Non-Performing Assets (GNPA) declined to
at
₹87035 Crore as on September’22 as against ₹100291 Crore as on September’21 declined by
13.21% on YoY basis.

Net Non-Performing Assets (NNPA) were at ₹29348 Crore as on September’22 as against

₹36934 Crore in as on September’21 declined by 20.53% YoY basis.

In terms of ratios, GNPA ratio improved by315 bps to 10.48% in September’22 from
13.63% in September’21. There was an improvement of 79 bps on QoQ basis and NNPA
ratio improved by 169 bps to 3.80% in September’22 from 5.49% in September’21.
There was an improvement of 48 bps on QoQ basis. PCR including TWO improved by
319 bps to 83.96% in September’22 from 80.77% in September’21.

PRIORITY SECTOR LENDING

PNB has constantly been achieving National Goals and Targets allocated under flagship
schemes of the nation for upliftment and employment of targeted groups. The outstanding
under Priority Sector as on September’22 stood at ₹ 2,82,256 Crore. National Goal
achievement of Priority Sector Advances was 43.54% of ANBC against the National Goal of
40% as at the end of

16
September’22. Agriculture advances stood at ₹123332 Crore, exceeding the National Goal of
18 % and was at 19.03 % of ANBC as at the end of September’22. Credit to Small and
Marginal farmers stood at ₹65865 Crore in September’22. National Goal achievement is
10.16% of ANBC, exceeding the target of 9.5%. Credit to Weaker Sections stood at ₹91294
Crore in September’22. National Goal achievement is 14.08% of ANBC, exceeding the target
of 11.5%. Credit to Micro
Enterprises stood at ₹54142 Crore as on September’22. The Bank has achieved National
Goal at 8.35% of ANBC as against the target of 7.5 %.

ENHANCING DIGITALISATION
Digital Banking Initiatives: Bank has focused on creation of digital canvas for customers by
allowing all tools for customers like debit card, PNB One, BHIM UPI etc. Digital
Transactions improved to 83% in September’22.The progress under digitalization is outline
below:

 Internet Banking Services (IBS) users increased to 370 Lakhs as at September’22

from 284 lakhs in September’21, registering 31% Y-o-Y growth.
 PNB One Mobile Banking Services (MBS) users increased 37% YoY to 373 Lakhs as
at September’22 from 272 Lakhs in September’21.
 UPI transactions increased by 85% on YoY basis to 85 Crores as at September’22
from

46 crores in September’21.

FINANCIAL INCLUSION
The Bank has been a pioneer in taking initiatives in the area of Financial Inclusion. Under
PMJDY, 441 Lakh accounts were opened. As on 30.09.2022, progress under the social
security schemes is as under:-

 Pradhan Mantri Suraksha Bima Yojana [PMSBY]: 194.6 lakh customers enrolled
under PMSBY.
 Pradhan Mantri Jeevan Jyoti Bima Yojana [PMJJBY]: 48.46 lakh customers enrolled
under PMJJBY.
 Atal Pension Yojana [APY]: 23.76 lakh customers enrolled under APY.

17
NEW INITIATIVES UNDERTAKEN
Liability Product

 PNB “AROGYA” Saving Scheme (Age group 18-Upto 60 years) with Health Care
Insurance.

 Saving Elite Scheme for HNI Customers

 Term Deposits Scheme for 405 days, 600 days with attractive rate of interest.

 Integration for CERSAI (CKYC) for paperless KYC.

 TAB Banking in all branches.

 New Tie-Ups/ MoUs for Rakshak Plus, Salary Accounts with Indian Army, Assam
Rifles & Indian Air force.

Asset Product
 Revamped Pre-Approved Personal Loan and Pensioner Loan.

 Insta (in-built) sanction of Car loan facility for existing MSME/ Corporate borrowers.

 Modification of Gold Loan Scheme for best interest rate advantages.

 Introduction of Krishi Swarna Yojana.

 Tie ups with

 National Small Industries Corporation Ltd. For financing MSMEs.

 Schwing Stetter India Pvt Ltd for Financing Construction Equipment.

18
AWARDS AND RECOGNITIONS
 PNB becomes one of the first Bank to launch UPI on Rupay Credit Cards.

 Punjab National Bank won Second prize in “Rajbhasha Kirti Puraskar” for FY 2021-22

 Punjab National Bank was declared winner under 2 Themes and EASE 4.0 awards ü Tech
Enabled Banking – 1st Runner up.
ü Governance and HR – 1st Runner up.

19
 AXIS BANK PROFILE

Corporate Profile

Axis Bank is the third largest private sector bank in India. The Bank offers the entire
spectrum of financial services to customer segments covering Large and Mid-Corporates,
MSME, Agriculture and Retail Businesses.
The Bank has a large footprint of 4,758 domestic branches (including extension counters)
with 10,990 ATMs & 5,972 cash recyclers spread across the country as of 31st March 2022.
The Bank has 6 Axis Virtual Centers with over 1,500 Virtual Relationship Managers as of
31st March 2022. The Overseas operations of the Bank are spread over eight international
offices with branches in Singapore, Dubai (at DIFC), and Gift City-IBU; representative
offices in Dhaka, Dubai, Abu Dhabi, Sharjah and an overseas subsidiary in London, UK. The
international offices focus on Corporate Lending, Trade Finance, Syndication, Investment
Banking, Liability Businesses, and Private Banking/Wealth Management offerings.
Axis Bank is one of the first new generation private sector banks to have begun operations in
1994. The Bank was promoted in 1993, jointly by Specified Undertaking of Unit Trust of
India (SUUTI) (then known as Unit Trust of India), Life Insurance Corporation of India
(LIC), General Insurance Corporation of India (GIC), National Insurance Company Ltd., The
New India Assurance Company Ltd., The Oriental Insurance Company Ltd., and United India
Insurance Company Ltd. The shareholding of Unit Trust of India was subsequently
transferred to SUUTI, an entity established in 2003.
With a balance sheet size of Rs. 11,75,178 crores as on 31st March 2022, Axis Bank has
achieved consistent growth and with a 5-year CAGR (2016-17 to 2021-22) of 14% each in
Total Assets & Advances and 15% in Deposits.

MARKET RISK IN TRADING BOOK

Market risk is the risk of loss to the Bank’s earnings and capital due to changes in
the market level of interest rates, price of securities, foreign exchange rates and
equities’ price, as well as volatility risk in the option book. The Bank is exposed
to market risk through its investment activities and also trading activities, which

20
are undertaken for customers as well as on a proprietary basis. The Bank adopts a
comprehensive approach to market risk management for its trading, investment
and asset/liability portfolios. For market risk management, the Bank has:
 Board approved market risk policies and guidelines which are aligned to the
regulatory guidelines and based on experiences gained over the years. The policies
are reviewed periodically keeping in view regulatory changes, business
requirements and market developments.
 Process manual which are updated regularly to incorporate and document the
best practices.
 Market risk identification through elaborate mapping of the Bank’s main
businesses to various market risks.
 Statistical measures like Value at Risk (VaR), supplemented by stress tests,
back tests and scenario analysis.
 Put in place non-statistical measures/limits on positions, gaps, stop loss,
duration and option Greeks etc.
 Management Information System (MIS) for timely market risk reporting to
senior management functionaries. Key risk metrics are presented to the Risk
Management Committee of the Board through Risk Dash-Boards. Risk limits such
as position limits, stop-loss limits, alarm limits, gaps and sensitivities (duration,
PVBP, option Greeks) are set up and reviewed periodically, based on a number of
criteria including regulatory guidelines, relevant market analysis, business
strategy, size of the investment and trading portfolio, management experience and
the Bank’s risk appetite. These limits are monitored on an intra-day/daily basis by
the Treasury Mid-office and the exceptions are put up to ALCO and Risk
Management Committee of the Board. The Bank uses Historical Simulation and
its variants for computing VaR for its trading portfolio. VaR is calculated and
reported on a daily basis for the trading portfolios at a 99% confidence level for a
one-day holding period, using 250 days of historical data or one year of relative
changes in historical rates and prices. The model assumes that the risk factor
changes observed in the past are a good estimate of those likely to occur in the
future and is, therefore, limited by the relevance of the historical data used. The
method, however, does not make any assumption about the nature or type of the
loss distribution. The VaR models for different portfolios are back-tested at

21
 regular intervals and the results are used to maintain and improve the efficacy of
the model. The VaR measure is supplemented by a series of stress tests and
sensitivity analysis that estimates the likely behaviour of a portfolio under extreme
but plausible conditions and its impact on earnings and capital. The Bank
undertakes stress tests for market risks for its trading book, IRS, forex open
position and forex gaps on a monthly basis as well as for liquidity risk at the end
of each quarter. The Bank has built its capabilities to migrate to advanced
approach i.e. Internal Models Approach for assessment of market risk capital.

Market Risk & Liquidity Risk of PNB

The investment policy covering various aspects of market risk attempts to assess and
minimize risks inherent in treasury operations through various risk management tools.
Broadly, it incorporates policy prescriptions for measuring, monitoring and managing
systemic risk, credit risk, market risk, operational risk and liquidity risk in treasury
operations.

Besides regulatory limits, the Bank has put in place internal limits and ensures adherence
thereof on continuous basis for managing market risk in trading book of the bank and its
4 business operations. Bank has prescribed entry level barriers, exposure limits, stop loss
limits, VaR limits, Duration limits and Risk Tolerance limit for trading book investments.
Bank is keeping constant track on Migration of Credit Ratings of investment portfolio.
Limits for exposures to Counterparties, Industry Segments and Countries are monitored.
The risks under Forex operations are monitored and controlled through Stop Loss Limits,
Overnight limit, Daylight limit, Aggregate Gap limit, Individual Gap limit, Value at Risk
(VaR) limit, Inter-Bank dealing and investment limits etc.

22
 For the Market Risk Management of the Bank, Mid-Office with separate Desks for
Treasury & Asset Liability Management (ALM) has been established.

Market Risk Management Committee (MRMC) and Asset Liability Management

Committee (ALCO) are primarily responsible for establishing the market risk management
and asset liability management of the Bank, procedures thereof, implementing risk
management guidelines issued by regulator, best risk management practices followed
globally and ensuring that internal parameters, procedures, practices/policies and risk
management prudential limits are adhered to. ALCO is also entrusted with the job of Base
rate / MCLR/ advances linked to external benchmark and pricing of advances & deposit
products and suggesting revision of MCLR/Base Rate/ BPLR/ external benchmark linked
advances to Board.

The policies for hedging and/or mitigating risk and strategies & processes for monitoring the
continuing effectiveness of hedges/ mitigants are discussed in MRMC and based on views
taken by /mandates of MRMC, hedge deals are undertaken.

Liquidity risk of the Bank is assessed through gap analysis for maturity mismatch based on
residual maturity in different time buckets as well as various liquidity ratios and management
of the same is done within the prudential limits fixed thereon. Advance techniques such as
Stress testing, simulation, sensitivity analysis etc. are use

Identify the type of statistical model to be built

In the fraud risk model built here, the dependent variable is “whether a merchant will go into
a fraudulent activity or not”. This is a binary variable, with the value “1” representing the
possibility that the merchant will do a fraud, and the value “0” representing the possibility
that the merchant will not do a fraud. Hence, logistic regression was used to model the risk.

Another choice that was made was based on the nature of model required. Broadly speaking,
there are two types of models: Transaction level and Account level. The transaction level
model gets triggered when transaction is recorded at a merchant and transaction variables are
used. The drawback of this model is that if a merchant does not have too many transactions
(As per industry standards, the merchant should have at least 100 transactions per day on an
average in the last one year), or has seasonal transactions, then this model may not be useful
because there will not be enough transactions to get a reliable PF score. To overcome this,

23
 one can use account level model which uses variables which are merchant specific,
irrespective of whether a transaction is being recorded at the merchant or not. However, the
drawback of this model is that the variables are not updated because they are based on
merchant history rather than on fresh transactions. This paper builds a transaction level model
to demonstrate its usefulness

Merchant_Id Trans_Dt Sub_Dt trans_amt slop of spi no of customer last 5 days

CM_CONC

9100011002 13/3/2011 13/2/2011 234234 4 3567 3

9100011003 12/5/2011 13/5/2011 556435 2.4 3875 2
9100011004 24/6/2011 25/6/2011 13401361.33 8776 5
9100011003 25/1/2011 27/1/2011 406124410.9 12376 2
9100011002 27/7/2011 28/7/2011 6672323200 9876 3
9100011001 31/8/2011 1/9/2011 30042371 12345 5
9100011004 10/9/2011 11/9/2011 22442670.8 4678 5

Description of Variables used in PF Model

Variable Variable Description Variable

Merchant_Id A unique identifier for the Type
merchant Text
Trans_dt Date of transaction between merchant and the customer Date
Sub_dt Date on which the merchant submits the transaction slip to the acquiring bank Date
Trans_amt Transaction amount Numeric
CM_conc A measure of the concentration of card members at the merchant. Numeric
Calculated as the ratio of number of transactions in the last 5
days at the merchant to the number of card members initiating
those transactions

24
Slope spikeThe slope of the spike in current month’s business volume over last month’s Numeric
volume
No_of_cust_last 5 daysNumber of customers transacting at the merchant in the last 5 days Numeric

Percentile Distribution of Defaults captured by PF Model

Percentile No. of Cumulative Percent
Cumulative
Defaults No. of No.of
Defaults Defaults
10 689 689 62
20 201 890 80
30 67 957 86
40 46 1003 90
50 37 1040 94
60 24 1064 96
70 17 1081 97
80 14 1095 98
90 10 1105 99
100 7 1112 100

ROC (Receiver Operating Characteristics) Curve

the graph, known as ROC (Receiver Operating Characteristics) curve shows that the model
captures 63 percent defaults in the top 10 percent of the transaction data when the data is
sorted in decreasing order of
the PF score generated by the model. A high percentage of default capture in the top
percentile of ROC curve indicates that the model is working as expected, and has been able to
fit the data well.

25
The steepness in ROC Curve indicates the amount by which using a model is better than not
using a model at all.

degree line, indicates the incremental benefit emanating from the use of PF model.

Once the probability score is obtained from each transaction, the dataset is sorted in
descending order of probability score and then consolidated into deciles and the actual and
predicted default rate calculated for each decile as shown in Table 4.

Actual default rates are calculated as the ratio between the number of defaults in that decile
and the total number of merchants in that decile. Predicted default rates are calculated in a
similar way, but the numerator in this case represents the total number of merchants tagged as
defaulters by the PF model.

26
Comparison of Actual Default Rates and Default
Rates Predicted by the PF Model
Percentil No. of No. of Actua Predicte
e l d
Default Transactio Defau Default
s ns lt
Rate Rate
(%) (%)
10 689 8613 8.00 7.7
20 201 8613 2.33 2.0
30 67 8613 0.78 1.1
40 46 8613 0.53 0.5
50 37 8613 0.43 0.5
60 24 8613 0.28 0.3
70 17 8613 0.20 0.2
80 14 8613 0.16 0.2
90 10 8613 0.12 0.1
100 7 8613 0.08 0.1

Comparison of Actual vs Predicted Default Rate for the PF Model

27
 Figure shows the comparison of actual de

fault rate and the default rate predicted by the model for each decile and helps in
understanding the model performance in real time.

CONCLUSION
The management of risk in global business scenario has taken more importance than ever. In
the financial services sector, fraud risk is inherent in the nature of business. The aim of this
research was to create an Early Warning System for banks to effectively manage the risk
arising out of their merchant-acquirer relationship. To facilitate this, the PF (Probability of
Fraud) Model was proposed. The PF model assigns a PF score to the entire merchant
portfolio of the acquiring bank, which can be used by the banks to assess their merchant
portfolio more effectively.

In this paper, the business framework of the merchant acquirer relationship was explained,
along with the ways in which fraud risk can arise from this association. The PF model was
created in SAS and applied to the transactional data of a bank. It was observed that the model
captured 63 percent defaults in the top decile of transactions, when sorted in decreasing order
of PF scores. The PF model was also validated by comparing the actual defaults with those
predicted by the model and it was found that there was a very good alignment between the
two. This indicated that the model rank ordered the defaults quite close to reality.

28
REFERENCES

Agarwal N., & Sharma M. (2013). Default risk modelling in credit cards: A study of
merchant and acquiring bank relationship. IIMS Journal of Management Science, 4(1),
4048.
Brause, R., Langsdorf, T., & Hepp, M. (1999). Neural data mining for credit card fraud
detection. Proceedings of IEEE International Conference. Tools with Artificial
Intelligence, pp. 103-106.

Chiu, C., & Tsai, C. (2004). A web services-based collaborative scheme for credit card fraud
detection. Proceedings of IEEE International Conference. e-Technology, e-Commerce
and e Service, pp. 177-181.

Clarke, M. (1994).Fraud and the politics of morality. Business Ethics: A European Review,
3(2), 117-122.

Cressey, D. R. (1953). Other people’s money. Montclair, NJ: Patterson Smith.

29
30