preventing tax evasion & benefits fraud through predictive analytics
TRANSCRIPT
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics Ian Pretty | Senior Vice President, Tax & Welfare, Capgemini June 4, 2014 | SAS Analytics Frankfurt
2
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Areas to be covered today
! Why should Tax & Welfare Agencies be concerned?
! The impact of technology on Fraud & Error
! How can Tax & Welfare Agencies respond?
! The Capgemini response
3
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Is there really a fraud and error problem?
It is estimated that approximately €100 billion in total is involved in the
wrongful non-payment of VAT within
the EU Member States each year
Source: EU MTIC Report
Shadow economies are estimated to have
accounted for £880 billion in lost tax in the EU
between 1999 and 2007
Source: tax justice network
It is estimated that MTIC VAT fraud contributed between £0.5 billion
and £1.0 billion to the UK VAT gap in 2010-11.
Source:
HMRC report (2012) Measuring tax gaps 2012; Tax gap estimates for
2010-11.
4
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Do Governments agree that there is a problem?
5
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
new modelling tools & techniques
So why does better Fraud Management matter?
So why does better Fraud Management
matter?
new & more data
growing demand for and expectations of public services
shorter reaction times
growing use of digital
identity theft Industrialization
of Fraud
growing complexity
growing fiscal deficits
reducing costly investigations
internal Fraud
6
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Typically we see 4 types of behaviour
Large Businesses & HNWI Relationship based monitoring to protect
Compliant Make it simple to get tax right
Casual avoiders Risk based campaigns to recover and deter
Deliberate evaders Full enquiries to recover & deter Criminals Investigate & prosecute or disrupt
Value at risk
Ris
k of
non
-com
plia
nce
Low
High
Low High
7
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
How Governments respond in a digital, data and analytics driven world will determine how they protect revenues
8
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Section 1 How will Technology impact the fight against Fraud & Error?
9
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
We are all aware of the rise of ‘Big Data’...
Many PBs of data
every day
25+ TBs of log data every day
12+ TBs of tweet
data every day
30 billion RFID tags
today (1.3bn in
2005)
100s of millions of
GPS enabled devices sold
annually
4.6 billion camera phones
world wide
76 million smart meters
in 2009… 200m by 2014
2+ billion people on the Web at end
2012
80% Of world’s data is unstructured
10
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
…but it is how you analyse that data that will be key to future success
Business
“Business” – it is the use of analytics to directly target a business issue or process and as such is sold to the Business. Examples are customer retention, increasing wallet share, fraud reduction…
Business Analytics is the uses of advanced analytical techniques to find trends and predict future outcomes which are used to optimize
business processes, customer interaction and manage risk and fraud.
Analytics
“Analytics” – it makes extensive use of data, statistical and quantitative analysis, explanatory & predictive modeling, and fact-based management to drive decision making.
Governments will have to become data-driven, analytics-enabled organisations
11
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Moving faster to an analytics enabled world means a shift in our Big Data thinking
Each business area can have their own analytics on the same data
Each area can get their own insights
The Business Data Lake provides a place to land the big data
Big data is driven by business use cases
Business Data Lakes
Insights can then be shared across the business
12
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
So we will need data lakes to support this new world of analytics
Store everything
Govern only the common
Encourage local Treat global as a local view
21
34
Business Data Lake
It’s all about insight at the point of action
13
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
But Governments will also have to operate in a digital world with increased risks for fraud and error….
Beginning of Web Session
Login Transaction and Logout
Pre-Authentication Threats Post-Authentication Threats
DDOS Attacks Phishing Attacks Parameter Injection Man in the Browser New Account Registration Fraud
Account Takeover Fraudulent Reclaims Vulnerability Probing Risking Intelligence Gathering Password Guessing
Disruption and/or Intelligence Gathering
Theft of information and/ or Money
Nation States – Hacktavists – Organised Criminals
News > UK > Crime
Source: http://www.independent.co.uk/news/uk/crime/cybercrime-boss-offers-a-ferrari-for-hacker-who-dreams-up-the-biggest-scam-9349931.html
Cybercrime boss offers a Ferrari for hacker who dreams up the biggest scam
14
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
…meaning they will need to think data, analytics and digital
Source: Capgemini Consulting-MIT Analysis – Digital Transformation: A roadmap for billion-dollar organisations (c) 2011
Iterative Transformation Roadmap
Dig
ital E
ngag
emen
t Digital G
overnance
Digital Building Blocks Customer
Insight Operational
Process New Business
Model
Customer understanding
Customer touch points
Improved compliance Worker enablement
Performance management
Process digitisation
Global collaboration
New outsourcing/ partner models
Digitally modified business
Digital Capabilities
Tax Investigators
Channels
Tax Policy
Process Innovation
Customer Knowledge
Culture
Partnership Network
Brand
Strategic Assets
Digital Investment Skills Initiatives
Transformative Digital Vision
Use of new analytical capabilities & tools
Using cross-government &
third party data sources
Real time identify verification and data
validation
Digital by default – intervention by
exception
Bilateral and multilateral exchange
of data
Mobile access to data & tools
Near real time dashboards
15
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Digital will fundamentally change the tax administration model
16
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Section 2 How should Tax & Welfare Agencies respond?
17
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Tax and Welfare Agencies will need to move from ‘checking’ to ‘risk based’ analytics....
Up-front data matching accuracy
and eligibility checks
Pre-emptive and initial risking
Synthesis of risk and case
prioritisation
Sophisticated, algorithm-based
response
Compliance rules Risk rules Risk score Risk-based treatment
Individual reports income ‘A’ and
compliance rule is used to compare it to known
income value ‘B’ reported by employer
Individual reports income ‘A’, risk rule is
used to assess the propensity to risk, e.g.
by comparing income to possession of assets
Individual triggers multiple (risk) rules
which are combined into single risk score that
enables the Agency to differentiate between
the level of risk between individuals
Individual triggers multiple risk factors and based on predictive risk score, this individual is
treated differently
Cha
ract
eris
tics
Exam
ple
18
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
...which means risking using advanced analytics to link multiple data sets and generate a risk score.....
Historic Approach Looking for data matches to prove
fraud and error
Leading practice model Spotting likelihood of an event through multivariable analysis
Outlier analysis Entity Network Analysis
Hybrid risk modelling approach
Location Demographics & behaviour
Income Assets Funds
Multiple data sources brought together
X
X X
X X
X
Data set 1 Data set 2
Data match/ mismatch triggers risk rule
19
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
...and an analytics–based risk methodology
Compliance activity
WHAT is happening?
WHO is doing it?
WHY are they doing it?
HOW to respond?
Understanding the type of non-compliance
(simple error; evasion; avoidance;
underreporting income)
Understanding the characteristics of the taxpayer or benefits
group (segment)
Understanding the reasons (low level of services; complicated legislation; criminal
attack)
Understanding the best option (targeted
compliance campaign; preventive action; better information/ service; penalties)
Analytical insight
Client Segmentation
Behavioral Analysis
Predictive Modelling
Campaign Design & Mgt
Risk Rule Design & Mgt
Anomaly Detection
20
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Section 3 Trouve: The Capgemini answer in partnership with SAS
21
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Back to Mr. Hyde
22
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Trouve applies new sources of data and advanced analytics to create an end to end risking & interventions process....
Prioritized (risk based)
flow
Large scale Data
Networking & Network Analysis
High Analytical Performance
Data Visualization
Applying insight
across the value chain
Measurement and Continuous
Improvement
Applying analytics internally
(workforce, case
management)
Building a citizen
centric view
Hybrid Analytics Models
Advanced Campaign
Management
Receive Understand Interact Review
23
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
...and enterprise capabilities to improve compliance outcomes
Filing / (Re) Payment request
Calculation Assessment
Payments in/out Pre-reg Registration /
Application Reconciliation Compliance & Debt Enforcement
Bus
ines
s op
erat
ing
mod
el
Solu
tion
arch
itect
ure
laye
rs
Downstream risking
Debt Management
Internal Fraud
Upstream risking
ID assurance
Voluntary compliance
Tailored solution to deliver new capabilities and maximize value
Design Develop Deploy
Organization
People & Skills
Processes
Technology
Business Services
Information Systems
Capabilities For more information about TROUVE visit: www.capgemini.com/trouve
Debt Management
Information Mgt
Work and Workforce management
Investigation and Audit Campaigns Profiling and Risking
Performance Management
Strategy and Policy
Set Risk Policy / Strategy
Set Service and Channel Strategy
Simulate Policy / Strategy
Develop Policy / Strategy
Monitor Legal Compliance
Manage Customer Service
Manage Yield Effectiveness
Manage Resources and Workforce Efficiency
Profile Citizens
Prioritize Risk
Validate Citizen Identity
Identity Registration
Risk
Identity Returns
Risk
Identity Repayment Risk
Identity Compliance
Risk
Identity Debt Risk
Set Channel Selection Rules
Design Campaign
Execute Campaign Case
Record / Verify Response
Profile Citizens Investigate
Non-Compliance
Find ‘Ghosts’
Investigate False Passes
Detect Internal Fraud
Pursue Compliance
Case
Pursue Internal Case
Assess debt risk
Set work priorities
and allocate
Manage Case
Worklist
Manage Contact Worklist
Set Resource / Skills strategy and capacity
Set Data Acquisition and Mgt
Policy
Select/Model new information sources
Monitor Data Quality
Import and Check External Data
Prioritize debt
Create inventory of
debt
Monitor Insolvencies
Pursue Debt Case with the
Citizen
Administer Insolvencies
Set workflow
rules
24
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
The functionality of TROUVE addresses the requirements of the future compliance model
Downstream Risking & Case Management Operates outside of operational processing. Extends the capabilities of post processing compliance to maximise money yield and money recovery based on optimising available resources.
Upstream Risking & Case Management Uses predictive models to identity high risk transactions to withhold services such as payments or repayments and initiates interventions.
Protecting Online channels from ID Theft Using transaction monitoring and the application of identity assurance within the transaction to prevent ID Theft
Uncovering Internal Fraud & Collusion Applies the analytics techniques on internal operational and customer data and to identify anomalies in behaviours that signal fraud, either individual working alone or collusion with external fraudsters.
Improving Debt Management by understanding customers attitudes and behaviours we can determine the optimal treatment strategy balancing cost and business results. An integrated feedback mechanism leads to a continuous improvement.
Supporting Voluntary Compliance maximizing the use of digital communication channels, methods and campaigns to drive up voluntary compliance via targeted & tailored service, eliminating the need for compliance activity
25
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Client Concerns Method
But we also know that each client has a different starting point
Client Situation
! Requirement identified but there is no clearly articulated vision or high level design
! Vision and High Level design exist – unsure of where to start and in what order
! Concerns remain about clarity and progress
! Will it work at all and if so will it be scalable
! Desire to start to build initial components quickly
! Value Discovery
! Target Operating Model and detailed Roadmap
! Business Assurance
! Proof of Concept/Pilot
! Design & Build Fraud Management System
I need to do something
I have a Vision
Show me it works
I get it. When can we start?
Am I on the right track?
26
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
Case study: Implementing the strategic risking solution for HM Revenue & Customs
! Capgemini supported HMRC to design, build, deploy and run their strategic risking tool – Connect
! Takes information from 28 different data sources ! Cross-matches one billion internal and third party data items ! Uncover hidden relationships across organizations, customers and their
associated data links (bank interest, lifestyle indicators and stated tax liability) ! Connect uses analytical and ‘spider diagram’ visualization tools ! HMRC analysts produce target profiles and models to risk assess
transactions and generate campaigns and cases for investigation ! Automated feeds into HMRC’s case management system ! Streamlined risk and intelligence operations are delivered by with
40% fewer staff. Connect produces in minutes what previously took months of research, or was simply not possible to do manually or on a volume basis
! Skilled staff concentrate on tackling aggressive evasion rather than correcting errors, which historically took much time and which is now tackled in other ways.
£
In total HMRC has recovered £2.6bn additional tax yield to date, through the use of Connect
The project has won several awards:
27
BIM
Copyright © 2014 Capgemini. All rights reserved.
Preventing Tax Evasion & Benefits Fraud through Predictive Analytics | May 2014
! Proven success stories in UK, Netherlands and in the Financial Services Sector
! £2.6bn additional tax yield to date for HM Revenue & Customs.
! Partner to 35 Tax & Welfare globally
! Understanding of the Tax business process
! Compliance Framework ! End to End solution ! World’s foremost provider
of Business Information Management (BIM) services.
Our capabilities
Register/Change of
Details
ProcessApplication
/Return
EstablishLiability/Benefit
ManagePayments
In/outReconcile
Investigation/Audit/
Enforcement
ReceiveCustomer
Submission
Enforcement/Debt
collection/criminal
proceedings
Prevent Protect Uncover Resolve
Feedback
Prove
An integrated approach takes a holistic approach on which to base a business strategy that develops and deploys common capabilities actively managed to deliver the best business outcome.
Prevent transmission of incorrect information – either error or fraud
Protect against incorrect /repayments/repayments through the identification and management of risks
Identify that fraudulent or non-compliant activity has taken place
Provide evidence to prove the case so that the authority can take remedial action
Successful resolution through recovering the monies or securing criminal prosecution
Infrastructure
Data Sources
Data Preparation
Data Linking/Networking Creation
Analytical Environment
Network Visualisation
Risk Model Management
Analytical Capability
Execution Ability
Investigative Capability
Case ManagementEnterprise Compliance Capabilities
! Strategic global partnership with SAS on Fraud management solutions
! BIM Centre of Excellence in India
! Business & Solution Architects
! Local footprint.
Delivery capability Domain expertise
SAS CoE
! Dedicated lab for all SAS products
! High performance servers installed
! Hands on experience for building proof of concepts
! Build better knowledge infrastructure to share and learn SAS
! Premium partnership agreement with SAS
Report generation & delivery
Predictive models, scorecards,
segmentation, decision trees, web analytics
Forecasting optimization, social media,
solutions
Value Proposition
Skills
! Analytical consultants
! Business analysts! Statisticians! Tools experts! BI architects! Data architects! MDM experts! Change experts! Quality experts! Process leads! Domains
Analytics Maturity Assessment
Specialized skill pool
Cloud based offering
Analytics CoE to support the known requirements
of today and the unanticipated needs of
the future
Easy to use and relevant scorecards and reports
that enable greater visibility into operating
and financial metrics
Ad-hoc sales, marketing and functional reporting
for a streamlined, integrated and automated
operation
Solution
Social Media Analytics
Marketing Campaign Analytics
Big Data Analytics
Kno
wle
dge
Inte
nsiv
e Resource Intensive
Proven value
The information contained in this presentation is proprietary. Copyright © 2014 Capgemini. All rights reserved.
Rightshore® is a trademark belonging to Capgemini.
www.capgemini.com/bim
About Capgemini With more than 130,000 people in over 40 countries, Capgemini is one of the world's foremost providers of consulting, technology and outsourcing services. The Group reported 2013 global revenues of EUR 10.1 billion. Together with its clients, Capgemini creates and delivers business and technology solutions that fit their needs and drive the results they want. A deeply multicultural organization, Capgemini has developed its own way of working, the Collaborative Business Experience™, and draws on Rightshore®, its worldwide delivery model.