Transcript Handout

2006 CAS RATEMAKING SEMINAR
CONSIDERATIONS FOR SMALL
BUSINESSOWNERS POLICIES
(COM-3)
Beth Fitzgerald, FCAS, MAAA
Agenda
• Definition of Risks
• Market Needs
• Use of Statistical Modeling
• Scoring Model Development
• Amount of Insurance Relativity Factors
Underwriting Small Commercial Risks
Eligible for Businessowners
• Size
– Area
– Gross sales
• Type of risk
– Office, apartments, retail, service
– Contractors, restaurants, motels, self-storage
facilities
– Light manufacturing
• Rating
– Class-rated
– Low average premium
Growth in Small Businesses
25,000,000
24,000,000
23,000,000
Establishments
with less than 10
Employees
22,000,000
21,000,000
20,000,000
19,000,000
18,000,000
1992
1997
2000
2003
Source: Office of Advocacy, U.S. Small Business Administration
Market Needs
• Efficient use of technology to allow for
faster, more consistent underwriting
decisions
• Add intelligence to the policywriting
process
• Low-cost solution due to low premium
size
What Makes Statistical Modeling
Possible?
• Advanced computer capabilities
• Advanced statistical data mining
tools
Uses of Statistical Modeling
• Scoring of small commerical risks
– Improve loss predictability of risks
– Increase accuracy of pricing decisions
– Cost effective, consistent underwriting
• Improve manual rating of risks
Development of Scoring Models
• Analyze historical policy and loss data
• Link policy and loss data with external
data:
– Business financial data
– Weather
– Demographics
• Use statistical data mining software and
techniques
Modeling Process
Data Linking
Data Gathering
Data Cleansing
Evaluation
Analyze
Variables
Business
Knowledge
Determine
Predictive Variables
Modeling
Statistical Modeling Techniques
Balance good fit with explanatory power
• Generalized Linear Models
• Classification Trees
• Regression Trees
• Multivariate Adaptive Regression
Splines
• Neural Networks
Benefits of Scoring Model
• Fast, cost-effective tool to help you
determine which risks to insure
• More accurate pricing decisions
• Reduce underwriting expense through
automated scoring process efficiencies
• Expand your markets
Risks of Not Scoring
• Lost market share
• Greater risk of adverse selection
Use of Statistical Modeling in
Manual Rating
• Improve rating relativities of current
rating factors
• Add new rating factor to manual using
a multi-variate statistical model
Amount of Insurance Relativities
• Amount of Insurance identified as important
•
•
variable in BOP Scoring analysis
Partially handled by insurers
Decision to include as variable in manual and
not in scoring model
Property Buildings
One Dimensional
Experience Ratio
2.5
2
1.5
1
0.5
0
7
10
13
16
19
30
45
70 100 175 300 700 000
10
Amount of Insurance in 000's
Current Rating for BOP
Property
• Base loss costs by state/territory for
buildings & personal property
• Multi-state Relativities
–
–
–
–
Rate number
Sprinkler
Protection
Construction
Current Rating for BOP
Liability
• Base loss costs by state/territory for
occupants & lessors
– Occupants vary by AOI, Payroll or Sales
exposure base
• Multi-state rating relativities
– Class group
Multivariate Analysis for Amount of
Insurance Relativities
• Variables used for Property
–
–
–
–
Rate number
Sprinkler
Protection
Construction
• Variables used for Liability
– Class group
BOP Implementation of AOI
Relativities
• Incorporation into manual
– Definition of base amount of insurance
– Building - vary by state/region
• Timeline
– 12 month lead time
– Interaction with other possible changes
– Filing late 2006