Credit History is Predictive of Loss Ratio Relativities

2000 CAS Ratemaking Seminar The Use of Credit History And Other Demographic Data in Underwriting PA-35 Credit History is Predictive of Loss Ratio Relativities By Eddy Lo 2000 CAS Ratemaking Seminar The Use of Credit History And Other Demographic Data in Underwriting PA-35 Credit History is Predictive of Loss Ratio Relativities Agenda Traditional Insurance Loss Experience Credit and Demographic Data Scoring Definitions Scorecard Example Results Summary Traditional Insurance Loss Experience Age of dwelling vs loss ratio relativities Use of vehicle vs loss ratio relativities Vehicle performance vs loss ratio relativities Natural Phenomenon Illustration Age of Dwelling 1200 1000 Loss Rario Relativity 811 844 876 910 941 962 1000 800 600 400 200 0 6+ Natural Phenomenon Illustration Use of Vehicle 1200 1000 Loss Ratio Relativity 1000 800 921 760 665 600 400 200 pleasure use drive to work business use farm use Natural Phenomenon Illustration Vehicle Performance 1600 1365 1400 1185 Loss Ratio Relativity 1200 1185 1000 1000 800 600 400 200 standard intermediate high sports Credit and Loss Experience Credit characteristics correlate with loss ratio relativities Credit and Loss Experience (cont’d) Personal property 230,000 policies with claims 1,000,000 policies without claims 11 archives Univariate Analysis HO-3 Number of Adverse Public Records 1600 1540 1400 1200 Loss Ratio Relativity 1000 1000 800 600 400 200 zero 96% one or more Univariate Analysis HO-3 Months Since Most Recent Adverse Public Record 1800 1678 1600 1400 1226 Loss Ratio Relativity 1200 1000 1000 800 600 400 200 no public record 96% less than 48 48 or more Univariate Analysis Standard Auto 1400 Number of Trade Lines Opened in the Last 12 Months 1270 1200 1083 Loss Ratio Relativity 1000 1000 800 600 400 200 zero or one 82% two or three four or more Demographics and Loss Experience Length of residence Homeowner Auto Univariate Analysis Homeowner Length of Residence 1.20 1.118 Loss Ratio Relativity 1.00 0.883 0.80 0.60 0.40 0.20 0.00 < 15 years 15+ years Length of Residence Univariate Analysis Auto Length of Residence 1.20 1.161 Loss Ratio Relativity 1.15 1.10 1.054 1.05 1.019 1.00 0.976 0.95 0.90 0.85 this year one year two-nine years Length of Residence ten or more years Scoring Definitions A score for an insurance risk is a numeric summary of the impact on loss ratio relativity based on a certain set of predictive characteristics of the risk A scorecard is an algorithm, a table, or a piece of computer software that will calculate a score based on a certain set of characteristics provided for a risk Scorecard Example Simple homeowner scorecard Simple Homeowner Scorecard Number Adverse Public Records zero one or more 30 Months Since Most Recent Adverse Public Record no public record less than 48 30 48 or more 10 Number of Trade Lines 60+Days Delinquent in Last 24 Months zero one 25 10 two or more Number of Collections zero one or more 20 Number of Trade Lines Opened on the Last 12 Months zero one 20 10 two three four or more 5 Results Low scores correlate with high loss ratio relativities High scores correlate with low loss ratio relativities Homeowner Loss Ratio Relativities 1.6 1.4 1.2 1.0 0.8 0.6 0.4 low Score Range high Loss Ratio Relativities Personal Auto 1.6 1.4 1.2 1.0 0.8 0.6 0.4 low Score Range high Summary Use of Credit History Leads to Precision Underwriting Facilitates Consistent Underwriting Does Not Make Decisions, People Do Provides Input to Refine Decisions Use of Credit History Provides More Objectivity and Accuracy Summary (cont’d) Use of Credit History Helps Underwriters Focus on Risks Needing Attention Most Use of Credit History Helps to Reduce Premium Subsidies/Inequity Use of Credit History Helps to Open Up Markets Biography Name: Eddy Lo Product Manager, Insurance SBU, Fair, Isaac and Company, Inc Fair, Isaac work experience: Joined Insurance SBU as Senior Project Manager in January of 1996 Worked on new product development, regulatory and legislative issues, USER scorecard development Provided training on actuarial concepts and ratemaking principles Prior work experience: From 1994 to 1995, worked as a research and development program manager for personal lines at Fireman's Fund Insurance Company; built and implemented Geographic Information System in the home office and regions Also worked on developing insurance products for affinity groups and catastrophe management Over the prior 17 years, managed the pricing and filing of rates for personal and commercial lines of insurance for CNA, United Pacific/Reliance, PMI Mortgage Insurance, and Fireman's Fund Actuarial examinations passed part through 5, earned M S in Actuarial Science in 1977 and B B A in Finance and Business administration in 1976 from University of Wisconsin-Madison ... Seminar The Use of Credit History And Other Demographic Data in Underwriting PA-35 Credit History is Predictive of Loss Ratio Relativities Agenda Traditional Insurance Loss Experience Credit and Demographic... Insurance Loss Experience Age of dwelling vs loss ratio relativities Use of vehicle vs loss ratio relativities Vehicle performance vs loss ratio relativities Natural Phenomenon Illustration Age of Dwelling... Summary (cont’d) Use of Credit History Helps Underwriters Focus on Risks Needing Attention Most Use of Credit History Helps to Reduce Premium Subsidies/Inequity Use of Credit History Helps to Open