better audit case selection through balance sheet … audit case selection through balance sheet...

An Oracle White Paper

October 2013

Better Audit Case Selection through Balance Sheet Forensics using Oracle Tax Analytics


Disclaimer

The following is intended to outline our general product direction. It is intended for information

purposes only, and may not be incorporated into any contract. It is not a commitment to deliver

any material, code, or functionality, and should not be relied upon in making purchasing

decisions. The development, release, and timing of any features or functionality described for

Oracle’s products remains at the sole discretion of Oracle.

Executive Overview ........................................................................... 3

Introduction ....................................................................................... 3

eXtensible Business Reporting Language - XBRL ............................. 4

Standardized Exchange of Data .................................................... 4

What is XBRL? .............................................................................. 5

Taxonomies ................................................................................... 5

Implementing XBRL in Europe ....................................................... 5

Annual Financial Statements in Germany .......................................... 6

XBRL Solution for Germany .......................................................... 7

Balance Sheet Management ............................................................. 7

Financial Forensics............................................................................ 9

Is there a case to audit? ................................................................ 9

Opportunity .................................................................................. 10

Mapping the XBRL Tags ................................................................. 11

Tax Analytics Implementation .......................................................... 12

Conclusion ...................................................................................... 13


3

Executive Overview

Some of the challenges Tax Authorities are facing encompass to deliver better services to

constituents and decrease administrative burden for constituents, realize higher compliance

and all against a lower cost and with less staff. Being efficient and selective in applying the

scarce resources will help to address the productivity challenge.

Financial Statements (Balance Sheets and Profit & Loss (P&L)) in a standardized and

electronic format offer possibilities for enhanced analytics. Using Analytics capabilities applying

an adequate financial forensics model, will help identify Red Flags which will help to narrow

down the selection of constituents for Audit. This will result in a higher probability of “catching”

fraudulent constituents whilst applying adequately the scarce resources.

Introduction

Tax Authorities (and governments in general) are facing a number of challenges where they

have to deliver better services, decrease the administrative burden for constituents on the one

hand and realize higher constituent compliance against a lower cost and with less staff. It is

therefore important to be selective in applying the scarce resources.

As a result, one of the trends is that governments are introducing and mandating constituents

and specifically taxpayers to do most tax related reporting in a standardized and electronic

way. This means that both governmental bodies and their constituents have to adapt their

processes and supporting applications to facilitate this new way of communications.

It also creates opportunity for Tax Authorities to improve on different aspects of their

“business”. Improvement on Audit Case Selection is one of those aspects where Tax

Authorities can find opportunities for cost reduction and increase of revenue.


4

In this white paper we will take a closer look at a standardized and electronic way of data

communication (XBRL1), already implemented at a vast number of Tax Authorities. When

Balance Sheets and P&L Statements are available in this format and a Tax Authority has these

over consecutive years, it will be possible to do comparative ratio analysis on the figures in the

Financial Statements. The model that we use in this paper, is the M-Score model created by

professor Beneish2. This model is a mathematical approach which evaluates eight ratios to

determine the likelihood of earnings manipulation and allows auditors to spot discrepancies

within the firm's financial statements. This will allow Tax Authorities to better select their audit

cases.

We will further use a case example called E-Bilanz3 that is relevant for German Tax

Authorities. E-Bilanz is the name of the project in Germany where businesses have to provide

their Financial Statements (Balance Sheet and P&L) to the German Tax Authorities. Although

the case example describes a closed loop process, the focus in this white paper will be on

Analytics (“Financial Forensics”) and the M-Score model.

eXtensible Business Reporting Language - XBRL

Standardized Exchange of Data

Each application stores its data in its own proprietary format (Balance Sheet, Profit & Loss Statement,

etc.). The data communication problem between different software applications used to be resolved by

exporting the data file in an ASCII format (e.g. CSV format) so that it is possible to retrieve the

original ASCII file into another application. Next to the fact this is very time-consuming, it may cause

the data to be skewed and placed in the wrong table, because the exporting application does not use

the same delimiting character as the importing application. Another well known problem is that most

1 XBRL (eXtensible Business Reporting Language), is developed by the XBRL.org, an international

consortium of companies and organizations, and is widely accepted by the international community.

2 The Beneish M-Score Model (Beneish Model) can assist in evaluating the probability of earnings manipulation in a company, as well as identifying areas that may require greater scrutiny and has been created by Professor Messod Beneish. Professor Beneish got his PhD at the University of Chicago, Booth School of Business and is since 1996 a Professor of Accounting at the Indiana University, Kelley School of Business.

3 E-Bilanz is German and stands for Electronic Balance Sheet (“E-Balance”).


5

ASCII files4 do not contain any metadata (information on the date in the file) and therefore the

significance of that data is difficult to determine.

Since files can be exchanged in various formats (pdf, xls, html, doc, etc.), not limited to ASCII alone,

the need for a digital standard to exchange information among software applications emerged. Today,

this standard is known as XBRL (eXtensible Business Reporting Language, developed by XBRL.org,

an international consortium of companies and organizations), which is widely accepted by the

international community.

What is XBRL?

XBRL is a language for the electronic communication of business and financial data. XBRL provides

major benefits in the preparation, analysis, and communication of business information. It promises

cost savings, greater efficiency, and improved accuracy and reliability to all those involved in supplying

or using financial data.

The idea behind XBRL is simple. Instead of treating financial information as a block of text - as in a

standard internet page or a printed document - it provides an identifying tag for each individual item of

data. For example, company net profit has its own unique tag.

The introduction of XBRL tags enables automated processing of business information by computer

software, cutting out laborious and costly processes of manual re-entry and comparison. Computers

can treat XBRL data "intelligently": they can recognize the information in a XBRL document, select it,

analyze it, store it, exchange it with other computers and present it automatically in a variety of ways

for users. XBRL greatly increases the speed of handling of financial data, reduces the chance of error

and permits automatic checking of information.

Taxonomies

Taxonomies are the dictionaries used by XBRL. They define the specific tags for individual items of

data (such as company net profit). Different taxonomies are required for different financial reporting

purposes. National jurisdictions need their own financial reporting taxonomies to reflect their local

accounting regulations. Many different organizations, including regulators, specific industries or even

companies, require taxonomies to cover their own business reporting needs.

Implementing XBRL in Europe

4 An "ASCII file" is a text file that is readable by the naked eye. It only contains the letters a-z, numbers,

carriage returns, and punctuation marks.


6

The objective of the XBRL initiative in Europe is to significantly contribute to governmental targets to

reduce the administrative burden of businesses by 25%. At the same time it promises to improve and

simplify the electronic services of governments:

In France XBRL is seen as a “bridge” between tax reporting and standard EDIFACT

In the Netherlands XBRL is used for reports to banks, regulators, Tax Authorities and Chambers of

Commerce

In Germany the newly introduced E-Bilanz XBRL reporting needs to be uploaded directly to the

servers of Ministry of Finance.

Annual Financial Statements in Germany

All companies in Germany (with exceptions) are obliged to hand over their annual financial statements

in an electronic way, all in the context of SteuBAG5. This is the act to reduce tax bureaucracy.

The German Government chose XBRL as a method because it was already an existing solution for

disclosure of the trade balances. They based a new fiscal taxonomy on an existing trade taxonomy. The

Financial Statements (E-Bilanz6) created need to be uploaded to the (German) Elster7 Server.

For companies in Germany this newly introduced E-Bilanz is quite a challenge. The problem is that

there is no default Account Template given in Germany and therefore the newly introduced E-Bilanz

seems quite a challenge to implement. Lots of IT companies and tax consultants offer solutions to

their clients to check and adjust the used GL account charts, before the fiscal taxonomy elements can

be linked to them. Sometimes those links need to be done manually, one by one.

Having financial data available in an XBRL format makes it easy to derive key figures on the financial

position of a company like e.g. determining the Equity Ratio (in %):

5 In December 2008, a law was signed in Germany in order to reduce the administrative burden of

taxation (Steuerbürokratieabbaugesetz, or SteuBAG). It tells that all processes that are using paper must be replaced by electronic media. Germany’s Federal Tax Administration selected XBRL as the mandatory data standard for filing income statements and balance sheets to be in compliance with SteuBAG, It applies to all companies independent of size and industry. The exception of companies not required to file in XBRL are are companies that use cash-basis accounting – mostly small entities like restaurants, beauticians, etc.

6 E-Bilanz is a project where the German Federal Tax Administration worked closely with the German XBRL jurisdiction and has invited all relevant German associations to join the working groups to establish and develop an appropriate taxonomy, including several industry-specific modules.

7 Elster stands for Elektronische STeuerERklärung and means Electronic Tax Declaration and is a project of the German Tax Authorities for the safe electronic transmission of tax data (like tax returns).


7

Figure 1 – Equity Ratio calculation mapping on XBRL tags

XBRL Solution for Germany

For Tax Authorities analysis of Financial Statements, when not standardized, is a complex and time

consuming job, where, at best, data is entered in spreadsheets, processed and analyzed. Although very

important, analytics is just one step, part of a whole process, from intake and monitoring, analyzing,

identifying Red Flags, to processing the results and selecting only those cases for audit that make a

calculated sense.

The value of analysis will multiply when the outcomes or conclusions are actionable. This will be

further enhanced when the end-to-end process is supported with an IT-Solution. The architectured

solution for the E-Bilanz for Tax Authorities in Germany consists of a number of standard

components from the Oracle footprint for Tax and Revenue Authorities:

Self-Service Component (PSRM-SS – Public Sector Revenue Management – Self-Service)

Taxation Component (PSRM – Public Sector Revenue Management)

Policy Component (OPA – Oracle Policy Automation, part of PSRM)

Document Production Component (DocuMaker)

Tax Analytics Component (OTA – Oracle Tax Analytics on OBIEE – Oracle Business Intelligence

Enterprise Edition)

Fusion Middleware Component

Database Component

Balance Sheet Management

A typical architecture to support the end-to-end process for the current German situation, but equally

appropriate for any other Tax Authority or Revenue Management Agency / Administration /

Authority looks like the following picture:


8

Figure 2 – Proposed architecture for better audit case selection

The high level process is as follows:

1. Load Balance Sheet Data

The Balance Sheet data can be uploaded into the Taxation Component in bulk or individually e.g.

submission through a constituent self-service portal. The Taxation Component is able to handle

XBRL. It will process the data and create a readable version of the balance sheet that will be

validated (on errors or obvious mistakes using the Policy Component). Validated Balance Sheets can

be automatically posted. Balance Sheets that do not pass validation will stay in a pending state and a

To Do (workflow) with a link to the delinquent Balance Sheet will be created for a user to process. If

changes need to be made then a new version of the Balance Sheet will be created. All versions will be

kept and will be accessible. If a constituent decides to send in a changed Balance Sheet (either on

request by the Tax Authority or on own initiative) this will be treated as a later version of the

original, first submitted balance sheet.

2. Compliance Monitor and process Balance Sheets (including changes and versioning)

A compliance monitor monitors due dates for Tax Returns, Payments and Balance Sheets. It will

automatically create (and send) reminders (and summons) if constituent s do not submit or pay on

time.


9

3. ETL for Analytics including other sources

Data can be extracted from the Taxation Component, transformed and enhanced with 3rd party data

(external) or from own legacy systems and applications and loaded into the data warehouse, ready to

be analyzed.

4. Intelligent Analysis and Data Mining

With the Tax Analytics component it is possible to run different data mining scenarios or execute

specific detection models. Based on red flags identified by the Analytics component an Auditor can

select constituents for further investigation or audit. How these red flags can be identified will be

described in the next section (Financial Forensics).

5. Selection of constituent for audit

Selection of constituents for audit is as easy as “ticking the box” behind the name of the constituent,

exporting the relevant data.

6. Auto inject in operational process

An individual selected constituent for audit or a list of those constituents will be injected into the

operation process where audit cases are created and Tax Authority staff (auditors) is notified for

follow up.

7. Run Audit Cases

After the audit cases have been created, the Taxation component can be used to run and track the

audit case.

Financial Forensics

Is there a case to audit?

Is unexpected and consistent sales growth while established competitors are experiencing periods of

weak performance the result of efficient business operations or an indicator / red flag for fraudulent

activity? Tax Authorities have specific techniques (most of the times undisclosed) of identifying red

flags which trigger possible audit cases. These techniques include, but are not limited to:

Through comparison with similar companies;

Through comparison over multiple years;

Comparative Ratio Analysis;

Experience with particular Industry Groups or Constituents;

Data Mining – Big Data.


10

Red flags are not necessarily an indication of an undoubted occurrence of financial statement fraud,

but merely provide a general overview of the warning signs. It signals that further in-depth research

must be conducted.

Spotting red flags can be extremely challenging, because companies that are engaged in fraudulent

activities will attempt to portray the image of financial stability and normal business operations.

Opportunity

One of the opportunities of financial data in XBRL format and especially the balance sheet of the E-

Bilanz is that it can be used to execute a number of financial forensic models. One of the models we

investigated, is the Beneish M-Score. The model is in fact a mathematical model that uses eight

financial ratios to identify manipulated earnings. The variables are constructed from the company's

financial statements and create a score to describe the degree to which the earnings have been

manipulated.

M-SCORE RATIOS

RATIO RATIO NAME FORMULA

DSRI Days' sales in receivable

index (Accounts Receivable

t / Sales

t) / (Accounts Receivable

t-1 / Sales

t-1 )

GMI Gross margin index ((Sales t-1

- Cost of Sales t-1

) / Sales t-1

) /

((Sales t - Cost of Sales

t) / Sales

t)

AQI Asset quality index (1 - (Current Assets t + PPE

t) / Total Assets

t) /

(1 - (Current Assets t-1

+ PPE t-1

) / Total Assets t-1

)

SGI Sales growth index (Sales t) / (Sales

t-1)

DEPI Depreciation index (DE t-1

/ (DE t-1

+ PPE t-1

)) / (DE t / (DE

t + PPE

t))

SGAI Sales and general and

administrative expenses

index (Sales, General and Administrative Expenses

t / Sales

t) /

(Sales, General and Administrative Expenses t-1

/ Sales t-1

)

LVGI Leverage index ((Long Term Debt t + Current Liabilities

t) / Total Assets

t) /

((Long Term Debt t-1

+ Current Liabilities t-1

) / Total Assets t-1

)

TATA Total accruals to total assets

index

((Working Capital t - Working Capital

t-1) - (Cash

t - Cash

t-1) +

(Income Tax Payable t - Income Tax Payable

t-1) + (Current Portion of Long Term

Debt t - Current Portion of Long Term Debt

t-1) - DE

t) /

(Total Assets t)

PPE = Plant, Property and Equipment t = current year

DE = Depreciation and Amortization Expense t-1 = previous year


11

Each individual index mentioned in the table above is an indicator in its own right whether or not a

company is less or more prone to manipulate earnings. Please take a look at the following 2 examples.

Gross Margin Index (GMI) – Measures the ratio of a company’s prior year’s gross margin to the

current year’s gross margin. When the results are greater than 1.0 the company’s gross margin has

deteriorated, which is a negative indicator of a company’s prospects, hence, making such companies

more prone to manipulate earnings.

Asset Quality Index (AQI) – Measures the quality of a company’s assets by calculating the ratio of

non-current assets (other than plant, property and equipment (PPE)) to total assets. It indicates the

amount of total assets that are less certain to be ultimately realized. When a company has potentially

increased its cost deferral or increased its intangible assets (in fact created earnings manipulation) you

will find an AQI greater than 1.0. The conclusion is that the greater the AQI, indicating a reduction in

asset quality, the greater the probability of earnings manipulation.

The strength of the model is the combination of all 8 ratios in an equation calculating the M-Score.

The empirical equation that employs the eight financial ratios to detect earnings manipulation is as

follows:

M = -4.84 + 0.92 DSRI + 0.528 GMI + 0.404 AQI + 0.892 SGI + 0.115 DEPI –

0.172 SGAI + 4.679 TATA – 0.327 LVGI

An M-Score (M in the equation) value greater than -2.22 warrants further investigation as a value less

than -2.22 suggests that the company is not a manipulator. In his research Beneish used companies in

the Compustat database between 1982-1992 and found that he could correctly identify 76% of

manipulators, whilst only incorrectly identifying 17.5% of non-manipulators.

In his studies, Beneish excluded financial institutions when calculating the M-Score because their

business models are very different from the manufacturing and other service firms. Therefore, one

should take extreme care when looking at M-Scores of financial firms.

Mapping the XBRL Tags

The individual elements in the formulas used in the M-Score equation can be mapped on the XBRL

tags. In the next figure an example is given how the elements necessary for calculating the Days Sales

in Receivables Index (DSRI) can be mapped.


12

Figure 3 – Mapping of XBRL Tags

Tax Analytics Implementation

With the data in XBRL format in the database and having mapped the individual elements in the

formulas of the M-Score model on the XBRL tags, a dashboard with 2 Tabs have been created.

The first Tab consist of an overview, with reports on the percentages of expected manipulators and

non-manipulators, a top 10 of Industry Groups with highest number of potential manipulators.

The second Tab is the Financial Forensics dashboard and contains a report where a top 25 (50-1000

selectable through filter setting) of potentially delinquent constituents. It is possible to select a

constituent from the list and look into the details of a selected constituent, highlighting the M-Scores

over several years, looking at the details of the individual ratios (indexes) and having them compared

against benchmarks. More detail can be found in the parameters used where d detailed list is given of

elements like Cost of Goods, Current Assets and Current Liabilities, amongst others.

Filters can be applied in order to look at a certain Industry Group or Tax Year. The default is the

current year minus 1 (in 2013 the default will be 2012) and the M-Score report will display year 2012

with 4 previous years if and when data is available. Please note that in order to calculate the M-Score

data is needed from 2 consecutive years.


13

Figure 4 – Example of a Financial Forensics dashboard

Conclusion

Challenges of Tax Authorities (and governments in general) such as the need to deliver better services

and decrease the constituent burden on the one hand and realize higher constituent compliance against

a lower cost and with less staff create opportunity for Tax Authorities to improve on different aspects

of their “business”.

One aspect discussed in this white paper is improvement on Audit Case Selection where Tax

Authorities can find opportunities for reduced costs and increased revenue.

Balance Sheets and P&L in standardized XBRL format offer possibilities for enhanced analytics.

Applying the M-Score model helps identifying Red Flags, which help to narrow down selection of

constituents for Audit Cases. Hence, this is resulting in a higher probability of “catching” fraudulent

constituents whilst applying adequately the scarce resources.

Better Audit Case Selection through Balance

Sheet Forensics using Oracle Tax Analytics

October 2013

Author: Jan H. Beck

Oracle Corporation

World Headquarters

500 Oracle Parkway

Redwood Shores, CA 94065

U.S.A.

Worldwide Inquiries:

Phone: +1.650.506.7000

Fax: +1.650.506.7200

oracle.com

Copyright © 2013, Oracle and/or its affiliates. All rights reserved.

This document is provided for information purposes only, and the contents hereof are subject to change without notice. This

document is not warranted to be error-free, nor subject to any other warranties or conditions, whether expressed orally or implied in

law, including implied warranties and conditions of merchantability or fitness for a particular purpose. We specifically disclaim any

liability with respect to this document, and no contractual obligations are formed either directly or indirectly by this document. This

document may not be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose, without our

prior written permission.

Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.

Intel and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license and

are trademarks or registered trademarks of SPARC International, Inc. AMD, Opteron, the AMD logo, and the AMD Opteron logo are

trademarks or registered trademarks of Advanced Micro Devices. UNIX is a registered trademark of The Open Group. 1013

better audit case selection through balance sheet … audit case selection through balance sheet...

Documents