Reinsurance Symposium 2016

Transcription

1 Reinsurance Symposium 2016 MAY 10 12, 2016 GEN RE HOME OFFICE, STAMFORD, CT A Berkshire Hathaway Company

2 Reinsurance Symposium 2016 MAY 10 12, 2016 GEN RE HOME OFFICE, STAMFORD, CT What Your Data Can Do For You? Presented by Lori Walters and Karen Morello Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 2

3 Agenda Analytics at Gen Re Your Data and the Reinsurance Proposal Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 3

4 Competitive Landscape The digital and technology revolution is driving change in all aspects of our business. Telematics, driverless cars, Smart Homes, etc. Agent and consumer expectations Global marketplace and interconnectivity of risk New and changing distribution channels Explosion of rich data and analytics To successfully navigate these changes, the industry will need to embrace technology and advanced analytics. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 4

5 Competitive Landscape DEFENSIVE Will competitors use analytics to develop more accurate pricing? OFFENSIVE Can we use analytics to help identify and close more profitable business? Will competitors use analytics to select the best risks, leaving us the worse risks? Are there operational decisions we could improve through analytics? Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 5

6 There are many different types of Analytics Business Intelligence / Reporting Advanced Analytics What has happened? Why did it happen? What will happen next? What should we do next? Competitive Advantage Statistical Analytics Predictive Analytics Forecasting Predictive models Prescriptive Analytics Optimization Decision Modeling Descriptive Analytics Trend detection Segmentation analysis Standard/Ad-hoc reports Dashboards Degree of Complexity Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 6

7 Analytics Can be Applied across the Insurance Lifecycle Fraud Detection Claim Resource Allocation Customer Retention Premium Audit Customer Segmentation Lead Generation Internet and Direct Mail Targeting Campaign Design Claims & Service Marketing Pricing & Underwriting Distribution & Sales Pricing Model Underwriting Rules Price Optimization Loss Trend Monitoring Agency Segmentation New Agency Appointment Submission Prioritization Cross Sell / Upsell Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 7

8 Predictive Analytics in Insurance Predictive Analytics is widely used in Pricing, but is under-utilized in other business functions. Predictive Modeling Usage 1,2 by Business Function Reserving 18% Claims 21% Marketing 27% Underwriting 52% Pricing/Rating 81% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 1. Source: ISO and Earnix 2013 Insurance Predictive Modeling Survey; 269 respondents from U.S., Canada P&C carriers. 2. Frequently or always using Predictive Modeling. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 8

9 Challenges The most significant challenge in developing predictive models is a lack of sufficient data. Primary challenge for large carriers is data quality (57%) 1 Small carriers struggle with not having enough data (42%) 1 User adoption is the biggest concern in implementing Analytics 2. According to a Valen Analytics survey 2, 82% of respondents cited underwriter adoption as significant or high concern when implementing predictive analytics. 1. Source: ISO and Earnix 2013 Insurance Predictive Modeling Survey; 269 respondents from U.S., Canada P&C carriers. 2. Source: 2015 Valen P/C Analytics Summit. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 9

10 Combining Analytics and Expertise The best results often combine human expertise and analytics Source: Valen Analytics, 2016 Outlook Report: Property & Casualty Insurance Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 10

11 Analytics at Gen Re Challenges and Learnings Challenges Data, Data, Data Accuracy and consistency Limited granularity to increase differentiation Enterprise Data Strategy Learnings Consistent, quality data capture Strategic evaluation of supplemental external data Reinsurance vs. Insurance Often driven by rare events and emerging issues More heterogeneity and more complex purchasing behaviors Build early wins Start with feasible use cases Demonstrate success and build core capabilities Business Readiness Initiatives without strong business ownership and crossfunctional support are less likely to succeed Focus on change management Strong business support is essential Change management and involvement of IT and business stakeholders throughout increase the chance of success Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 11

12 Takeaways Analytics are critical to navigate the digital and technology revolution. New opportunities to compete can be found through creative use of data and analytics. Analytics are a tool to aid decision making; the best results often combine human expertise and analytics 1. High quality data and user adoption are essential for success analytics, but are often the most challenging. Gen Re is invested in analytics we look forward to partnering with you to navigate the change. 1. Source: Valen Analytics, 2016 Outlook Report: Property & Casualty Insurance Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 12

13 Your Data and the Proposal Your Proposal Booklet contains May 2016 Your Quote Information and Exhibits which support your quote 2016 Reinsurance Opportunity Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 13

14 Casualty Proposal Exhibits Automobile Liability Personal Commercial General Liability Workers Compensation Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 14

15 Loss Graph For Retention Setting Personal Automobile Graphing Untrended Losses and Expense $600,000 $500,000 $400,000 $300,000 $200,000 $100,000 $ Untrended BI/PD Loss Untrended UM/UIM Loss Expense $125,000 Current Retention Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 15

16 Loss Exhibit For Retention Setting Total Layer Incurred xs $125,000 Claim Count and Untrended Losses and Expense Total Year All Years BI/PD Claim Count Excess Loss $ 652,025 $ 63,657 $ 266,386 $ 497,269 $ 228,967 $ 1,708,304 Average Loss $ 170,830 UM/UIM Claim Count Excess Loss 126, , , , ,831 $ 701,372 Average Loss $ 70,137 Total All Lines Claim Count Excess Loss $ 778,108 $ 195,801 $ 377,834 $ 630,135 $ 427,798 $ 2,409,676 Average Loss $ 240,968 Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 16

17 Casualty Personal Automobile Personal Auto providing detailed information by claimant allows us to provide exposure analysis by Age of driver Number of passengers in vehicle Line of business (BI, UM/UIM, etc...) NUMBER OF CLAIMANTS INVOLVED BY CLAIM COUNT Total Total Multiple Claimants Number Claims With One Two Three Four of Claims Multiple Claimants Total All Years % to Total 55% 59% Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 17

18 Casualty Commercial Automobile Commercial Auto providing class codes enables us to use your account specific data to allocate the risk (light, medium, heavy, extra heavy, all other), and not rely on industry allocation. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 18

19 Commercial Auto By Classification Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 19

20 Casualty General Liability CLEAR Can a fresh look at your exposure help manage your book? CLEAR is a Gen Re proprietary model developed confidentially for our Casualty clients. The provision of GL class codes can assist you in: Your Risk Distribution (PremOps/Products Table Assignment) Exclusion Detail Construction Job Site Ranking Detail Mold Ranking Detail Year-over-Year Risk Analysis Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 20

21 What CLEAR Helps You See Your Risk Distribution CLEAR delivers a comprehensive distribution of your book using premium data on a subline and class-level basis. You can see what types of risks generate what premium amounts, and how much of that distribution has changed year after year. The reports analyze business using ISO-GL and BOP, AAIS-GL and BOP, or SIC codes.) Market Share Analysis CLEAR helps you compare your risk profiles and experience with that of other ISO reporting companies in your line of business. You are able to identify which classes of business have been profitable for others, and total available premium, so you can evaluate your strategy and pricing. As a result, you can learn how much of your book is a high-, medium- or low-hazard for each exposure on a state-by-state basis. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 21

22 Casualty Workers Compensation The provision of premium by NCCI (or bureau) class code by state including the statistical class codes can provide: A full analysis using Gen Re s WC Rating Proprietary Model Excess Loss Ratio by Attachment Point (ELR). Exposure Relativity Factor (ERF); these are factors applied to state rates, to adjust for class of business mix. The class and loss ratio have direct impact on the final rate. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 22

23 WC Exposure Relativity Distribution Reinsurance Symposium Percent of Class Code Premium ERF Comparison 2015 and Submission 2014 Submission Account Account Account Industry Group Description Premium Distribution Distribution Agriculture $ 351, % 3.6% Contracting Light $ 3,823, % 20.8% Contracting Medium $ 2,369, % 8.5% Contracting Heavy $ 729, % 2.6% Forestry $ - 0.0% 0.1% Healthcare Hospitals, Nursing Homes $ 3,147, % 10.9% Healthcare All Other $ 259, % 0.9% Manufacturing Chemical NOC $ 162, % 0.8% Manufacturing Chemical Refineries $ - 0.0% 0.0% Manufacturing Heavy $ 311, % 1.5% Manufacturing All Other $ 822, % 5.1% Mercantile Convenience Stores $ 57, % 0.1% Mercantile Food & Provision Dealers $ 542, % 3.0% Mercantile Gas Stations $ 496, % 3.3% Mercantile Hotels, Clubs & Restaurants $ 2,740, % 11.5% Mercantile All Other $ 3,854, % 15.3% Mining $ - 0.0% 0.1% Public Entity Police, Firemen $ 64, % 0.4% Public Entity All Other $ 685, % 4.5% Schools, Colleges & Churches $ 150, % 0.4% Transportation Trucking $ 98, % 1.0% Transportation Warehousing, Dist. Centers $ 120, % 0.6% Transportation All Other $ 476, % 3.0% Utilities Power Generation $ 15, % 0.0% Utilities All Other $ 304, % 0.8% Aircraft $ 225, % 1.2% Total $ 21,810, % 100% Subtotal of Top 7 Groups $ 17,487, % 76.6% All Other Groups $ 4,322, % 23.4% Total Weighted Average ERF Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 23

24 Property Proposal Exhibits Commercial Property Personal Property Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 24

25 Commercial Property PEER PEER is a client confidential Gen Re proprietary model developed for our property Treaty clients, which provides a unique way to view your business in the following ways: Optimize the mix of commercial property business you write With the use of Hazard Grade Assessment Tools, identify characteristics associated with each hazard grade (low, medium, high, very high) Detailing agency books of business that contain higher than average hazard composition Net retained versus Ceded Exposure Analysis Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 25

26 PEER Example by Hazard Group Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 26

27 PEER Exposure by Hazard Risk, Premium, TIV, count Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 27

28 Personal Property Policy Level Detail Policy level detail file including the following Protection Code Year Built Policy Form The following slides are exhibit examples of above Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 28

29 Property Exposure by Protection Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 29

30 Property Exposure by Year Built Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 30

31 Property Exposure by Policy Form Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 31

32 Personal Property Average Rate Analysis Reinsurance Symposium Insurance Group PEERTM Dashboard 2015 Homeowners Property Data TIV Limit Bands 0-300, ,001 to 500, ,001 to 700, ,001 to 1,000,000 1,000,000 > Grand Total 2015 Policy Count 5,552 3,530 5,800 4, , Premium 388,725 2,437,666 1,805, ,661 43,231 5,265, Total Insured Va 250,522,000 1,650,000,321 1,380,065,220 1,525,500, ,300,252 5,144,388,028 Total Avg. Rate Average Rate Average Rate Denotes material change Note: The data sources were the 2015 and 2014 company data submissions; Property policy level detail. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 32

33 Your Data and the Proposal Summary Takeaways The more detail you provide, the more we can do with your data! Class Codes are key! For any further detail on Gen Re Proprietary Models discussed, please ask for the line of business contact information. Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 33

34 Reinsurance Symposium 2016 MAY 10 12, 2016 GEN RE HOME OFFICE, STAMFORD, CT Thank You! Lori Walters / / lori.walters@genre.com Karen Morello / / kmorell@genre.com Reinsurance Symposium 2016 Client Data Needs Lori Walters and Karen Morello 34