Winning The Zero Sum Game: Leveraging Predictive Modeling for Commercial Insurance Underwriting

1
Winning The Zero Sum Game Leveraging Predictive
Modeling for Commercial Insurance Underwriting
0 or

• Presented by John Lucker Principal
• Deloitte Consulting, LLP
• March 20, 2007

2
Topics For Discussion

• Introduction
• The State of the PC Industry
• PC Industry External Market Assessment
• Winning The Zero Sum Game Leveraging Predictive
  Modeling
• Some Pragmatic Comments and Approaches
• The 3 Key Activities Implementation,
  Implementation, Implementation

3
Introduction
After suffering through one of the worst years
for catastrophic losses, the PC industry has
experienced one of the most profitable years in
recent history. Despite the positive results in
2006, factors like accumulating surplus, price
softening, declining opportunities for organic
growth, rising loss costs, changing demographics,
slumping housing prices, and slow net business
growth may indicate that significant challenges
lie ahead. With nearly 1000 competitors in a
highly regulated, capital intensive, price
sensitive mature industry, the ability to
differentiate oneself and emerge as a market
leader is a challenge faced by all competitors.
A few have emerged as market leaders through
innovative business strategies, improved business
operations and the leveraged and strategic use of
analytics and technology. Historically, PC
industry performance has been at the mercy of
catastrophic events and pricing cycles. Recent
trends show that rates have topped out and
organic growth has been replaced with MA
activity. Unless PC companies embrace analytic
and technology enabled strategies for growth and
profitability, they will continue to operate in
an environment where they are swapping risks in a
highly commoditized industry resulting in a Zero
Sum Game.
4
The State of the PC Industry

• Personal and Commercial Lines Industry Economics
  and Outlook

5
The State of the PC Industry Industry Economics
A steady general growth in surplus has not been
matched by returns on equity.
PC Insurance Industry Returns (1990 - 2005)
CR1 Avg. 106.4
Surplus2 CAGR 7.9
ROE Avg. 9.3
Source A.M. Best Deloitte Analysis The ROE for
2005 was sourced from Insurance Information
Institute The value of Surplus for 2005 is
estimated based on 2004 data 1. Combined ratio is
statutory trade combined ratio 2. Surplus as
defined by A.M. Best includes the total of
policyholders surplus which includes surplus
notes, capital and assigned surplus, and
unassigned surplus accounts on statutory basis
6
The State of the PC Industry Industry Economics
Low growth, low returns and high volatility has
been reflected in the markets treatment of the
stocks of publicly held insurers over this period
of time.
Equity Market Performance (1992 - 2005)
Dot Com market run-up
Northridge
Sept. 11
Lowest CAT losses in 15 years
Source Index data is the closing price for that
year from Dow Jones Reuters Business Interactive
(Factiva)
7
The State of the PC Industry Industry Economics
A highly fragmented industry with over 950
companies1 actively operating in the market
PC Insurance Industry Concentration
PC Insurers, Ranked and Tiered by Net Premiums
Written, 2005

Percentage of
overall industry Net Premiums Written

• 10 Largest companies (1-10)
• 209.2bn in Net Premiums Written
• Average Combined ratio -100.6
• Market share
• 1995 - 47.1
• 2000 - 45.7
• 2005 - 48.2
• List of Top 10 players
• State Farm
• AIG
• Allstate
• St. Paul Travelers
• Berkshire Hathaway
• Nationwide
• Progressive
• Liberty Mutual
• Farmers
• Hartford

• Smallest Companies (51-965)
• 98.5bn in Net Premiums Written
• Average Combined ratio - 100.9 (Estimated)
• Group is shrinking as more smaller companies are
  finding it difficult to compete

Smallest Companies (51 -965)
Largest 10 Companies (1-10)
22.8
48.2
Next 40 Companies (11-50)
29.0

• Next 40 companies (11-50)
• 125.9bn in Net Premiums Written
• Average Combined Ratio - 100.9
• Market share
• 1995 - 27.9
• 2000 - 29.7
• 2005 - 29.0

Source A.M. Best, Deloitte Analysis 1. All PC
companies having positive net premiums written
for 2005 have been considered
8
The State of the PC Industry

• The Hunger for Growth

9
The State of the PC Industry Outlook for Growth
The desire to grow has been a cornerstone
strategy for PC companies.
Source State Farm 2005 Annual Report, Amica
2005 Annual Report
10
The State of the PC Industry Outlook for Growth
The desire to grow has been a cornerstone
strategy for PC companies.
Source Progressive 2005 Annual Report, Hanover
Insurance Companies 2005 Annual Report
11
The State of the PC Industry Outlook for Growth
The desire to grow has been a cornerstone
strategy for PC companies.
Source Safeco 2005 Annual Report
Source Safeco 2005 Annual Report
12
The State of the PC Industry Outlook for Growth
The desire to grow has been a cornerstone
strategy for PC companies.
I've often commented that financial results are a
trailing indicator of operational
accomplishments. And that certainly was true last
year as we completed several key initiatives that
drove our 2006 results and will be important
drivers of our future success. For example, we
implemented predictive modeling to help produce a
more consistent and efficient approach to
underwriting. We rolled out our agent portal to
make it easier for our agents to do business with
us. We fully implemented our regional field
structure and we have talented leadership teams
in each of our regions. In addition, in our field
underwriters we have a full complement of
top-quality decision-makers to go into our agency
partners offices every day, building and
strengthening relationships and competing
effectively for their best business.
Source Safeco 2005 Annual Report http//www.harl
eysville.com/fin/fin_4.html
13
The State of the PC Industry Outlook for Growth
Can PC companies achieve their desired growth
levels using traditional growth strategies?
As PC companies continue to strive for steady
annual growth, they must ask themselves questions
like

• Will the housing bubble burst? If so, what will
  be the impact on new homeowners policies?
• How will the stock market be impacted, if the
  bubble does burst?
• Can the record levels of housing appreciation be
  sustained? If not, what will the impact be on
  homeowners policy premium growth?
• Will the aging US population impact the number of
  new insured drivers and insured vehicles?
• With so many commercial product lines tied to
  business owner policies, what is the future
  outlook for small and mid-size business growth?
• How will the changing physician practice
  arrangements affect the number of insurable
  entities in the medical malpractice market?

14
PC Industry External Market Assessment

• Can the External Market Support PC Insurance
  Companies Desire for Organic Growth?

15
PC Industry External Market Assessment - Housing
The housing boom over the past 10 or so years has
fueled a steady growth of over 1 million new
housing units per year.
Annual Rate of Housing Unit Growth
Source US Economic Census
16
PC Industry External Market Assessment US
Demographics
The US population over the past 25 years
illustrates a demographic shift resulting in a
decline in the percentage of people under the age
of 18 and a rise in the number of people over 65.
In the years ahead, this will result in a
decline in the number of insurable drivers.
Source US Population Census
17
PC Industry External Market Assessment Motor
Vehicles
While there has been a major increase in the
number of motor vehicles over the past forty
years, the growth rate in the number of new
drivers and new motor vehicles registered has
declined in the past ten years.
Source Federal Highway Administration
18
PC Industry External Market Assessment US
Businesses
The number of new business openings are often
off-set by the number of new business closings,
especially for small and mid size businesses.
Source US Economic Census
19
PC Industry External Market Assessment Med Mal
Market
Despite a steady increase in the number of
physicians over the past 35 years, changing
physician practice arrangements will lead to a
fewer number of insurable physician entities.
Source American Medical Association
20
Winning the Zero Sum Game

• Strategies for Success in the Zero Sum Game
• Leveraging Predictive Modeling

21
Winning the Zero Sum Game
A Predictive modeling solution that is aligned
with key business objectives can support multiple
strategies.
Predictive Modeling Enabling Business Strategies
for Success

• Underwriting Excellence
• Improve pricing precision
• Increase objectivity throughout the underwriting
  process
• Enhance risk selection and risk avoidance
  capabilities
• Improve pricing competitiveness in profitable
  segments
• Improve underwriter negotiation capabilities

• Marketing and Retention
• Target the right risks for non-renewals
• Improve retention of profitable risks
• Increase cross-sell opportunities
• Identify geographic and product expansion
  opportunities
• Enhance recruiting of profitable producers

Analyze
Develop
Operations
Underwriting
Marketing

• Operational Efficiency
• Reduce transaction costs
• Straight through processing of select risk
  segments
• Improve ease of doing business with agents
• Improve claims management activities
• Improve customer service at all levels

• Enhanced Decision Making
• Increase fraud detection capabilities
• Improve monitoring of underwriting performance
• Enhance ability to react to market forces sooner
• Increase information processing capabilities and
  data governance

Select
IT
Deploy
22
Personal Lines Underwriting Predictive Modeling
Landscape
HIGH
Personal Insurance Underwriting Sophistication
LOW
LOW
HIGH
Personal Insurance Predictive Modeling
Sophistication
23
Commercial Lines Underwriting Predictive Modeling
Landscape
HIGH 100 of Potential Being Realized

• Small commercial insurance market - 100 billion
  in premium

Commercial Insurance Financial Benefits Being
Achieved Through Implementation
LOW 0 of Potential Being Realized
LOW Single Line, Few Variables Sources
HIGH Multiple Lines, New Renew, Many Variables
Sources
Commercial Insurance Predictive Modeling
Sophistication
24
Its No Longer Just About Class
Which is the better risk and will be more
profitable?
Roofers
Florists
www.oprf.com/tour/viewsDOP/08.jpg
www.wisconsun.org/images/solarshingles.jpg
www.elca.org/dcs/disaster/roofer-1a.jpg
www.letchworthgardencity.net/ shopping/florist.jpg

25
Traditional Risk Segmentation
Personal Automobile Homeowners Personal Umbrella
Workers Compensation Commercial Automobile CMP /
BOP Property General Liability DO / EO /
EPL Medical Malpractice Excess Casualty Umbrella
Roofers 18 Year Old Males
140
90
135
87
Overall Loss Ratio of 75
125
Florists 40 Year Old Females
115
82
78
110
75
100
72
90
68
80
65
70
63
60
Actual loss ratio
Below average
Internal data
Average
Above average
26
Deloitte Predictive Modeling Approach

• Building and deploying predictive models requires
  a specialized combination of skills covering data
  management, data cleansing, data mart
  construction, actuarial and statistical analysis,
  data mining and modeling, and insurance
  operational and business processing and
  technology.

Raw Model Score Raw Reason Codes Pass-through
Data
OUTPUTS
Data Sources
Scoring Engine Integration
Build And Run The Model
External Data
Data Aggregation Data Cleaning
Internal Data
Evaluate and Create Variables
Synthetic Data
Develop Loss Predictive Model
Operational Reporting
Business Rules Engine
Business Implementation

• Outputs
• Risk Selection
• Risk Pricing
• Acceptance/Declination
• Other Business Actions

Technical Reporting Monitoring
27
Varied Data Sources Improved Segmentation
Customer Data
Coverage Information
Policy Records Correspondence
Product Coverage Options
3rd Party Databases
Policyholder Info
Business Credit Personal Credit Crime
Statistics Traffic Patterns / Stats Economic CLUE
/ MVR Check Cashing Sub-Prime Lending Credit
Bureaus Real Estate Geographic/Geocode Demographic
Psychographic Bureau Data Sources Consumer /
Lifestyle Enhanced Census Behavioral
Experience Data Policyholders Insured
Weather
3rd Party Data
Agency Information
Customer Data
Billing Data
Marketing and Sales
Weather
Heat / Cold Extremes Precipitation
Extremes Hail Wind / Storms Event Extremes
Policy Information
Coverage Information
Claims Data
Claims Data
Agency Information
Billing Data
Losses Frequency Timing / Patters Loss Control
Data Fraud / Lawsuit
Retention Recruiting Profitability Adjusted
Premium Ratio New Business Volume Continuing
Education
Marketing / Sales
Billing / Payment Hist Accepted
Applications Rejected Applications
Campaign, Promotion Cust Response Scores Cust
Segmentation
28
Sample Representation of a Commercial Lines Model
The model produces a score of 1 100 that
indicates the relative risk of a policy on the
basis of predicted loss ratio
Predictive Model Equation
a (HolClaims) b (LR3 Year) c (Agent Distance)
d (BillHistory) e (BusinessAge) f
(FinScore)
55
Weights/ Coefficients
Predicted Loss Ratio Score
Policies that have higher than average predicted
loss ratio
Average Score
Policies that have lower than average predicted
loss ratio
29
Improved Class Segmentation
Business Segment
Roofers (or) 18 year old Males
12
32
56
Florists (or) 40 Year Old Females
46
35
19
30
Decile Management The Holes Revealed Quickly
Plugged
Model benefits are achieved by acting on various
underwriting directives that are dictated by the
predicted profitability of individual risks -
actions are driven by the decile management
concept.
Bobs Flower Shop 821
Traditional Non-Renewal Area
Janes Flower Shop 324
31
Significant Benefits Have Been Realized
Combined Ratio Improvements
Business Actions
Company
Targeted Non-Renewals Focused Renewal
Pricing Retention Programs New Business Growth
6
1

• National commercial insurer
• 2B commercial book 70 LR
• BOP, GL, Property, WC, Auto

2
NB Growth
1
Renewal Pricing
2
Retention
Non-Renewals
Total
32
What it IS What it IS NOT

• What it IS NOT
• A Black Box approach
• Stork delivery
• Replacement for underwriters
• Score used to communicate decision
• Score drives results
• A single variable magic bullet
• Actuarial and/or systems project
• Class plan underwriting

• What it IS with Deloitte
• Scoring drivers are known / understood
• Collaborative with knowledge transfer
• Additional underwriting toolset
• U/W reason messages are developed
• Implementation drives results
• Relationship among variables is power
• Business initiative
• Efficient segmentation of policyholders

33
Some Pragmatic Comments, Topics and Approaches

• The debate of underwriting as Science v. Art
• Typically, predictive modeling is not a dramatic
  departure from what and how you do things today
  but its more consistent and objective
• The quality of data question can predictive
  modeling be done with the data available?
• How can rapid benefits be achieved to help fund
  more extensive predictive modeling projects?
• Dealing with data size and credibility for
  smaller companies
• How wide should the data net be cast to get
  value from internal/external data?
• What new data can be gathered for the future?
• What lines of business are mature for predictive
  modeling and which are evolving?
• The comfort of investing in technology v. the
  discomfort of investing in operations
• Is the status quo really an option?

34
End-To-End Implementation Making Models Come
Alive

• Predictive Models must be effectively implemented
  to derive their benefit potential
• The financial benefits can be so significant that
  urgency should drive the pace of the project
• Create a benefit analysis and use the benefits to
  drive the project a complex process (PIF
  counts, LR management, retention, not written,
  etc)
• Competitive jockeying should also drive project
  pace first adopter advantages
• A best practice is to create a continuum of
  implementation solutions and phases
• Initial implementation should focus on extracting
  value from models before automation
• Tactical implementation can be achieved in 2-4
  months
• Planning, planning, and then some more planning

35
End-To-End Implementation Making Models Come
Alive

• Steering Committee and Project Committee
  Structure
• Phased structure and focus on 8020 Rule
• Development of End-State-Vision Project
  Planning Document some key questions are
• How will predictive modeling guide decision
  making, pricing, and tier placement?
• How will predictive models impact existing
  business processes (e.g. by line / account)?
• How will predictive models be blended into the
  field and agency management process?
• What key performance measures must be achieved?
• How will underwriters/raters/other personnels
  compliance be measured?
• What level of automation is desired for various
  business processes?

36
End-To-End Implementation Making Models Come
Alive

• Extract, Transform, Load process for Internal,
  External, and Synthetic Data
• Management of external data vendors and external
  data acquisition
• Data quality and cleansing issues
• Construction of Scoring Engine(s) (real time,
  batch, manual, etc)
• Construction of Business Rules Engine or similar
  process
• Construction of operational data mart
• Design of technical architecture, data flow,
  messaging, data management, etc.
• Model maintenance

37
End-To-End Implementation Making Models Come
Alive

• Underwriting workflow (renewal business, new
  business, touch level)
• How model scores will be used in the U/W process
  (scores, reason codes, action thresholds, risk
  avoidance, risk acquisition, retention
  management, account vs. monoline, etc)
• How will models be used as risks proceed from new
  to renewals (disruption issues)?
• Use of model scores for downstream processes
• Relationship of model usage to field and producer
  management
• Business rule creation, optimization and
  maintenance

38
End-To-End Implementation Making Models Come
Alive

• What systems modifications are required to
  accommodate the process?
• What will different people in different roles see
  throughout the process?
• Focus on tactical and strategic to avoid sole
  focus on expansive systems development
• Consider ease-of-management since models will
  change periodically
• Loosely integrate key components to allow for
  flexibility (ETL, scoring engine, rules engine,
  reporting engine)

39
End-To-End Implementation Making Models Come
Alive

• Identification of all new process stakeholders
  (Underwriting, actuarial, systems, executive,
  legal/regulatory, claims, field, agency,
  training, project management, etc)
• How will predictive modeling be described
  internally and externally to all stakeholders?
• Manage any legal/regulatory issues and concerns
• Once communicated, how to deal with questions,
  concerns, issues, etc. from Underwriters, field
  personnel, agents, market analysts, etc.
• Development of change management, communication,
  and training activities and materials
• Development of necessary implementation materials
  for all stakeholders
• Development of feedback mechanisms using
  objective and subjective criteria

40
End-To-End Implementation Making Models Come
Alive

• Creation of management reports and metrics
  measurement processes including dashboards
• Communication of model usage, results tracking,
  and management metrics at all process points to
  all constituencies
• Loop back processes to manage compliance or
  deviation of model usage business plan

41
Contact Information
42
About Deloitte Deloitte refers to one or more of
Deloitte Touche Tohmatsu, a Swiss Verein, its
member firms and their respective subsidiaries
and affiliates. Deloitte Touche Tohmatsu is an
organization of member firms around the world
devoted to excellence in providing professional
services and advice, focused on client service
through a global strategy executed locally in
nearly 150 countries. With access to the deep
intellectual capital of 120,000 people worldwide,
Deloitte delivers services in four professional
areas, audit, tax, consulting and financial
advisory services, and serves more than one-half
of the worlds largest companies, as well as
large national enterprises, public institutions,
locally important clients, and successful,
fast-growing global growth companies. Services
are not provided by the Deloitte Touche Tohmatsu
Verein and, for regulatory and other reasons,
certain member firms do not provide services in
all four professional areas. As a Swiss Verein
(association), neither Deloitte Touche Tohmatsu
nor any of its member firms has any liability for
each others acts or omissions. Each of the
member firms is a separate and independent legal
entity operating under the names Deloitte,
Deloitte Touche, Deloitte Touche Tohmatsu,
or other related names. In the US, Deloitte
Touche USA LLP is the US member firm of Deloitte
Touche Tohmatsu and services are provided by the
subsidiaries of Deloitte Touche USA LLP
(Deloitte Touche LLP, Deloitte Consulting LLP,
Deloitte Financial Advisory Services LLP,
Deloitte Tax LLP and their subsidiaries), and not
by Deloitte Touche USA LLP. The subsidiaries of
the US member firm are among the nation's leading
professional services firms, providing audit,
tax, consulting and financial advisory services
through nearly 30,000 people in more than 80
cities. Known as employers of choice for
innovative human resources programs, they are
dedicated to helping their clients and their
people excel. For more information, please visit
the US member firms web site at
www.deloitte.com/us.