Discussion Paper

Fraud Detection Using Data Analytics

in the Banking Industry
Table of Contents
What is Fraud?.................................................................................................................................................................................. 1

Who is Responsible for Fraud Detection? ........................................................................................................................................ 1

Why Use Data Analysis for Fraud Detection? ................................................................................................................................... 2

Analytical Techniques for Fraud Detection ....................................................................................................................................... 3

Fraud Detection Program Strategies ................................................................................................................................................ 4

Banking ............................................................................................................................................................................................. 5

Other Resources ............................................................................................................................................................................... 7

Fraud Detection Using Data Analytics in the Banking Industry 1

What is Fraud?
Fraud encompasses a wide range of illicit practices and illegal acts involving
The International Professional Practices Framework
intentional deception or misrepresentation. The Institute of Internal Auditors’ (IPPF) contains the following Standards on fraud and
International Professional Practices Framework (IPPF) defines fraud as: internal audit’s role:
1200 – Proficiency and Due Professional Care
"… any illegal act characterized by deceit, concealment, or violation of trust.
These acts are not dependent upon the threat of violence or physical force. 1210-A2 – Internal auditors must have sufficient
knowledge to evaluate the risk of fraud and the manner
Frauds are perpetrated by parties and organizations to obtain money, property, or in which it is managed by the organization, but are not
services; to avoid payment or loss of services; or to secure personal or business expected to have the expertise of a person whose
advantage." primary responsibility is detecting and investigating
fraud.
Fraud impacts organizations in several areas including financial, operational, and 1220 – Due Professional Care
psychological. While the monetary loss owing to fraud is significant, the full impact 1220.A1 – Internal auditors must exercise due
of fraud on an organization can be staggering. The losses to reputation, goodwill, professional care by considering the following:
and customer relations can be devastating. As fraud can be perpetrated by any Extent of work needed to achieve the engagement’s
employee within an organization or by those from the outside, it is important to objectives;
have an effective fraud management program in place to safeguard your Related complexity, materiality, or significance of
matters to which assurance procedures are applied;
organization’s assets and reputation.
Adequacy and effectiveness of governance, risk
management, and control processes;
Probability of significant errors, fraud, or
noncompliance; and
Who is Responsible for Fraud Detection? Cost of assurance in relation to potential benefits.
While senior management and the board are ultimately responsible for a fraud 2060 – Reporting to Senior Management and the
management program, internal audit can be a key player in helping address fraud. Board
By providing an evaluation on the potential for the occurrence of fraud, internal The chief audit executive must report periodically to
audit can show an organization how it is prepared for and is managing these fraud senior management and the board on the internal audit
risks. activity’s purpose, authority, responsibility, and
performance relative to its plan. Reporting must also
In today’s automated world, many business processes depend on the use of include significant risk exposures and control issues,
including fraud risks, governance issues, and other
technology. This allows for people committing fraud to exploit weaknesses in matters needed or requested by senior management
security, controls or oversight in business applications to perpetrate their crimes. and the board.
However, the good news is that technology can also be a means of combating
2120 – Risk Management
fraud. Internal audit needs to view technology as a necessary part of their toolkit
that can help prevent and detect fraud. Leveraging technology to implement 2120.A2 – The internal audit activity must evaluate the
potential for the occurrence of fraud and how the
continuous fraud prevention programs helps safeguard organizations from the risk organization manages fraud risks.
of fraud and reduce the time it takes to uncover fraudulent activity. This helps both
2210 – Engagement Objectives
catch it faster and reduce the impact it can have on organizations.
2210.A2 – Internal auditors must consider the
probability of significant errors, fraud, noncompliance,
and other exposures when developing the engagement
objectives.
Fraud Detection Using Data Analytics in the Banking Industry 2

Why Use Data Analysis for Fraud Detection?

Data analysis technology enables auditors and fraud examiners to analyze an The Association of Certified
organization’s business data to gain insight into how well internal controls are Fraud Examiners’ 2010 Global
operating and to identify transactions that indicate fraudulent activity or the
Fraud Study found that the
heightened risk of fraud. Data analysis can be applied to just about anywhere in
an organization where electronic transactions are recorded and stored. banking and financial services
industry had the most cases
Data analysis also provides an effective way to be more proactive in the fight
across all industries –
against fraud. Whistleblower hotlines provide the means for people to report
suspected fraudulent behavior but hotlines alone are not enough. Why be only
accounting for more than 16%
reactive and wait for a whistleblower to finally come forward? Why not seek out of frauds.
indicators of fraud in the data? That way, organizations can detect indicators of www.acfe.org
fraudulent activity much sooner and stop it before it becomes material and creates
financial damage.

To effectively test for fraud, all relevant transactions must be tested across all applicable business systems and applications.
Analyzing business transactions at the source level helps auditors provide better insight and a more complete view as to the
likelihood of fraud occurring. It helps focus investigative action to those transactions that are suspicious or illustrate control
weaknesses that could be exploited by fraudsters. Follow-on tests should be performed to further that auditor’s understanding of
the data and to search for symptoms of fraud in the data.1

There is a spectrum of analysis that can be deployed to detect fraud. It ranges from point-in-time analysis conducted in an ad
hoc context for one-off fraud investigation or exploration, through to repetitive analysis of business processes where fraudulent
activity is likely to more likely to occur. Ultimately, where the risk of fraud is high and the likelihood is as well, organizations can
employ an “always on” or continuous approach to fraud detection – especially in those areas where preventative controls are not
possible or effective.

Once an organization gets started with data analysis, they usually find that they want to do more and dig deeper into the data.
Modern organizations have increased management demands for information and the audit paradigm is shifting from the
traditional cyclical approach to a continuous and risk-based model. Technology therefore offers a range of solutions, varying by
the size and sophistication of the audit organization. From ad hoc analysis, through to repeatable automated procedures, and
continuous auditing and monitoring, analytics provide insight into the integrity of financial and business operations through
transactional analysis. Technology provides more accurate audit reports and better insight into the internal controls framework,
and improves the ability to access and manage business risk.

1 Coderre, David G., Fraud Analysis Techniques Using ACL, John Wiley & Sons, 2009.
Fraud Detection Using Data Analytics in the Banking Industry 3

Analytical Techniques for Fraud Detection

Getting started requires an understanding of: “As a large multinational
 The areas in which fraud can occur organisation, our internal audit
 What fraudulent activity would look like in the data process is under scrutiny from
 What data sources are required to test for indicators of fraud multiple global and local
authorities. It was essential for
The following techniques are effective in detecting fraud. Auditors should ensure
us to find a solution which
they use these, where appropriate. They include the following: allowed fast processing of
 Calculation of statistical parameters (e.g., averages, standard deviations,
financial data, was scalable to
high/low values) – to identify outliers that could indicate fraud. our needs and streamlined the
 Classification – to find patterns amongst data elements. entire process to increase
 Stratification of numbers – to identify unusual overall efficiency. ACL met
(i.e., excessively high or low) entries. every one of our needs.”
 Digital analysis using Benford’s Law – to identify unexpected occurrences Cyrille Oudard
of digits in naturally occurring data sets. Head of Internal Audit Tools, BNP Paribas
 Joining different diverse sources – to identify matching values (such as
names, addresses, and account numbers) where they shouldn’t exist.
 Duplicate testing – to identify duplicate transactions such as payments, claims, or expense report items.
 Gap testing – to identify missing values in sequential data where there should be none.
 Summing of numeric values – to identify control totals that may have been falsified.
 Validating entry dates – to identify suspicious or inappropriate times for postings or data entry.2

Please note that random sampling is not listed as an effective fraud detection technique. While sampling is an effective data
analysis technique for analyzing data values that are consistent throughout the data population, the very nature of fraud is
different as it tends not to occur randomly.

2 Global Technology Audit Guide: Fraud Prevention and Detection in an Automated World. The Institute of Internal Auditors, 2009.
Fraud Detection Using Data Analytics in the Banking Industry 4

Fraud Detection Program Strategies

Instead of relying on reactive measures like whistleblowers, organizations can and
should take a more hands-on approach to fraud detection. A fraud detection and “ACL Analytics Exchange
prevention program should include a range of approaches – from point-in-time to leverages ACL's proven
recurring and, ultimately, continually for those areas where the risk of fraud
analytical strengths to provide
warrants. Based on key risk indicators, point-in-time (or ad hoc) testing will help
identify transactions to be investigated. If that testing reveals indicators of fraud,
auditors with a means to
recurring testing or continuous analysis should be considered. perform data analyses and
help effectively detect and
A KPMG Forensics’ Fraud Risk Management report states, “unlike retrospective
analyses, continuous transaction monitoring allows an organization to identify prevent fraud."
potentially fraudulent transactions on, for example, a daily, weekly, or monthly basis. David Coderre
leading author of books such as “Computer-Aided Fraud
Organizations frequently use continuous monitoring efforts to focus on narrow bands Detection & Prevention: A Step-by-Step Guide”
of transactions or areas that pose particularly strong risks.”3

By leveraging the power of data analysis technology organizations can detect fraud
sooner and reduce the negative impact of significant losses owing to fraud.

3 “Fraud Risk Management: Developing a Strategy for Prevention, Detection and Response.” KPMG International, 2006
Fraud Detection Using Data Analytics in the Banking Industry 5

Banking
Fraud detection in banking is a critical activity that can span a series of Banking/Financial Services – 298 Cases
fraud schemes and fraudulent activity from bank employees and Number Percent
Scheme of Cases of Cases
customers alike. Since banking is a relatively highly regulated industry,
Corruption 101 33.9%
there are also a number of external compliance requirements that
Cash on Hand 64 21.5%
banks must adhere to in the combat against fraudulent and criminal
Billing 37 12.4%
activity. Check Tampering 35 11.7%
Non-Cash 33 11.1%
Skimming 32 10.7%
Larceny 29 9.7%
Banking Related Fraud Schemes: Expense Reimbursements 20 6.7%
Financial Statement Fraud 16 5.4%
Here are a few typical fraud schemes encountered in banking and
Payroll 9 3.0%
some examples of the way data analysis can be applied to detect and
Register Disbursements 8 2.7%
prevent them:
Distribution of Fraud Schemes in Banking/Financial Services 4

Corruption
 Find customers who appear on the US Treasury Department Office of Foreign Asset Control (OFAC) list.
 Ensure Financial Action Taskforce on Money Laundering (FATF) compliance.
 Produce listing of transactions with organizations on the list of non-cooperative countries and territories.

Cash
 Identify cash transactions just below regulatory reporting thresholds.
 Identify a series of cash disbursements by customer number that together exceed regulatory reporting thresholds.
 Identify statistically unusual numbers of cash transfers by customer or by bank account.

Billing
 Identify unusually large number of waived fees by branch or by employee.

Check Tampering
 Identify missing, duplicate, void or out of sequence check numbers.
 Identify checks paid that do not match checks issued, by bank, by check.
 Locate check forgery or falsification of loan applications.

4 2010 Global Fraud Study: Report to the Nations on Occupational Fraud and Abuse, Association of Certified Fraud Examiners
Fraud Detection Using Data Analytics in the Banking Industry 6

Skimming
 Highlight very short time deposit and withdrawal on the same account.
 Find indicators of kiting checks.
 Highlight duplication of credit card transactions and skimming.

Larceny
 Identify customer account takeover.
 Identify co-opted customer account information.
 Locate number of loans by customer or bank employee without repayments.
 Find loan amounts greater than the value of specified item or collateral.
 Highlight sudden activity in dormant customer accounts – identify who is processing transactions against these accounts.
 Isolate mortgage fraud schemes – identify “straw buyer” scheme indicators.

Financial Statement Fraud

 Monitor dormant and suspense General Ledger accounts.
 Identify Journal Entries at suspicious times.
 Fraud Detection Using Data Analytics in the Banking Industry 7

Other Resources
 Association of Certified Fraud Examiners (ACFE): the world's largest anti-fraud
organization and premier provider of anti-fraud training and education. www.acfe.org
To find out how ACL can help
 The Institute of Internal Auditors (The IIA): www.theiia.org your organization combat fraud,
» International Standards for the Professional Practice of Internal Auditing (Standards): contact us at +1-604-669-4225
The Standards are mandatory requirements – principle-focused and providing a or info@acl.com to arrange for
framework for performing and promoting internal auditing. a free consultation.
» Internal Auditing and Fraud Practice Guide: guidance on how to comply with the
International Standards for the Professional Practice of Internal Auditing.
» GTAG 13: Fraud Prevention and Detection Techniques in an Automated World: step-by-step process guide for auditing
a fraud prevention program, including an explanation of the various types of data analysis to use in detecting fraud,
and a technology fraud risk assessment template.
 ACL Detecting Fraud resource page: anti-fraud materials including industry reports, case studies and on-demand
webinars. www.acl.com/bankingfraud

About ACL
ACL delivers technology solutions that are transforming audit and risk management to give
organizations unprecedented control over their business.

Our integrated family of products—including our cloud based audit and compliance management
solution and flagship data analytics products—are used at all levels of the enterprise to help maximize
growth opportunities by identifying and mitigating risk, protecting profits, and accelerating
performance.

Thanks to 25 years of experience and our consultative approach, we implement flawlessly so

© 2010 ACL Services Ltd.
DP/BNK/021113 customers realize concrete business results fast at low risk. Our actively engaged community of more
ACL and the ACL logo are trademarks or registered than 14,000 customers around the globe—including 89% of the Fortune 500 and hundreds of
trademarks of ACL Services Ltd. All other trademarks
are the property of their respective owners. governments—tells our story best. Visit us online at .