THIS SESSION WILL USE POLLING!

THIS SESSION WILL USE POLLING! (To access in an internet browser, go to vcia.cnf.io) Click on the Polling Icon on the VCIA app Click on your session Respond to the Polls

HOW TO USE SOCIAL Q&A! (To access in an internet browser, go to vcia.cnf.io) 2 1 Click on the Polling Icon on the VCIA app Click on your session 1. Vote for a Question 2. Ask a Question

Risk Analytics: Set the Scene for your Captive Moderator: Anne Marie Towle, Willis Towers Watson Speakers: Mark Tabler, IPS & Genesis Re Derrick Easton, Willis Towers Watson August 10, 2016 2016 VCIA; Speaker materials used by VCIA under license.

Welcome and Introductions 3

Welcome and Introductions Seminar Goals: Use actuarial reports creatively to efficiently structure risk transfer within a Captive. Interpret analytical actuarial summaries to help de-risk business, while improving profitability, financial stability and capital efficiency. Identify risks that do not work well in traditional insurance markets. 4

Agenda Risk Analytics Audience Q&A and Polling Case Study Audience Q&A and Polling Summary of Best Practices and Tools Audience Q&A and Polling Wrap-up and Audience Polling Q&A 5

Audience Polling 6

Risk Analytics 10

Risk Analytics Definition Interpreting the analysis Applying the outcomes of analytics to market purchasing Expanding captive utilization based upon risk analysis outcomes 11

Set the scene for your captive Reserve Studies Fit for purpose Typical Output Line by line analysis Apply various techniques Basis of output selection Point estimate of liability Breakdown Paid, Case Reserves, IBNR Use Draw down reserve balance during financial period Risk Reserves are not the whole story Understanding volatility Capital Adequacy Formulaic approach, typically maximum of Minimum Capital & Surplus Percentage of NWP Percentage of Reserves Or view of Regulator 12

Case Study: IPS, RRG 13

IPS Current State Established 2005 Gross written premium $3.0m < 400 insureds Medical professional liability coverage only Occurrence based policies No reinsurance required participate with the IN PCF caps loss at $250k Only write business in State of Indiana Domiciled in Vermont 14

IPS Future State SB 28 signed March 2016 into law 1 st increase to caps effective July 1, 2017 to $400,000/$1,200,000 2 nd increase to caps effective July 1, 2019 to $500,000/$1,500,000 Results in significant change to actuarial and risk analytics 15

Communicating Exposure TIV Weighted Hazard Rating Earthquake Tropical River Flood Tsunami Storm Surge 1.9 0.4 2.6 0.1 1.8 % of Locations Exposed* 1.0% 0.0% 10.8% 0.3% 2.9% % of TIV Exposed* 3.7% 0.0% 13.1% 0.0% 7.0% * Exposed to a 4 or 5 hazard Rating DATA QUALITY Very High High Medium Low Very Low Not Geocoded 16

Communicating Exposure cont. 17

Communicating Exposure cont. 18

Communicating Loss Trends 19

Outputs Activities Inputs Optimal Analytical Output For each line of business written by captive and in portfolio across all lines Historic loss and exposure data Projected exposure data Captive insurance program detail Captive premiums per risk/per layer Captive risk tolerance (per risk/overall) Review losses and exposure information Where no losses, develop scenarios Develop actuarial models Add stochastic (Monte Carlo) analysis Review existing insurance arrangements and capital allocations Graphical summaries of loss history Table of expected aggregate losses Table of percentiles per risk/portfolio Impact of expected losses on existing insurance and interpretation of results Program alignment to risk tolerance Example: Stochastic Output Simulation Results ('000) Ground-up Losses Losses modelled under Current Program Return Period Total Total Total (Years) Percentile Loss Retained Ceded 1 in 2 50.0% 4,539 4,501 0 1 in 4 75.0% 7,979 6,916 408 1 in 5 80.0% 9,200 7,566 1,462 1 in 10 90.0% 13,221 8,724 5,294 1 in 20 95.0% 18,227 8,944 9,856 1 in 50 98.0% 26,384 9,104 18,277 1 in 100 99.0% 34,124 9,227 25,678 1 in 200 99.5% 44,562 9,335 37,403 1 in 500 99.8% 57,687 9,460 48,880 1 in 1000 99.9% 72,153 9,540 65,912 Mean 6,487 4,715 1,772 Std Dev 7,380 2,649 6,015 20

Optimal Analytical Output Good Data Highly available historical exposure and loss information Forecast of future exposures Good Industry loss data & development factors Add stochastic element Clear understanding of correlation between risks written in captive 21

CONFIDENCE LEVEL Optimal Analytical Output Limited Data - Getting to optimal Custom analytics process to identify risks and associated Frequency/Severity Develop plausible scenarios Create model with cost drivers and financial impact Mine expert organizational knowledge Review model output & calibrate with stakeholders Scenarios Selected in Workshop Model Output Scenario # Events Severity per Year Scenario 1 6.25 $876,244 Scenario 2 0.29 $719,937 Scenario 3 0.82 $7,654,847 Scenario 4 2.17 $1,080,735 Scenario 5 5.57 $951,872 Total 15.1 $11,283,635 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% LOSS DISTRIBUTION CURVE AT VARIOUS CONFIDENCE LEVELS BY LOSS SCENARIO & OVERALL RISK 0% $1,000 $10,000 $100,000 $1,000,000 $10,000,000 $100,000,000 $1,000,000,000 TOTAL EXPECTED AGGREGATE LOSSES (UNDISCOUNTED) - Logarithmic Scale Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5 Total Risk 22

Audience Polling 23

Case Study: Genesis Re 26

Genesis Re Current State Genesis Re Cell A - Large Employers Groups Cell B - SIHO self insured plans Cell C - TBD Established 2014 Segregated cell structure domiciled in Vermont Medical stop loss coverage 27

Genesis Re Cell A Large Employer Groups Large Employers Health Plans Stop-Loss Policies $500,000 SIHO Holdings, Inc. Fronting Insurer (Companion Life) Captive (Genesis) Excess Reinsurer (Zurich) 100% xs $500K MGU (Summit Re) Insured Excess (Zurich) Insured Captive $400,000 (Genesis Re) Self Insured (Large Employers) Captive Participation Current $500,000 $75,000 Employer Deductible 28

Genesis Re Future State Genesis Re Cell A Large Employer Groups Cell B SIHO fully insured plans TBD WC, Property, Cyber, Etc. Continue to Grow Risk Analytics Needed 29

Analytics to inform Risk Transfer Align analytical output to desired risk retention and reinsurance requirements Return Period Percentile Unlimited (Years) ($M) 1 in 2 50% 39.6 1 in 4 75% 47.0 1 in 5 80% 49.2 1 in 10 90% 54.1 1 in 20 95% 59.2 1 in 50 98% 66.6 1 in 100 99% 74.5 1 in 200 99.5% 78.0 1 in 500 99.8% 85.0 1 in 1000 99.9% 92.0 Mean 40.19 Std Dev 11.49 Calibrate program to captive s risk tolerance Understand reinsurance pricing model Expenses Risk Margin Expected Losses Recalculate captive s risk based capital requirements pre & post reinsurance to demonstrate adherence to risk tolerance guidelines Return Period Percentile Unlimited With Stop (Years) ($M) Loss ($M) 1 in 2 50% 39.6 40.2 1 in 4 75% 47.0 47.6 1 in 5 80% 49.2 49.8 1 in 10 90% 54.1 54.7 1 in 20 95% 59.2 59.6 1 in 50 98% 66.6 59.6 1 in 100 99% 74.5 59.6 1 in 200 99.5% 78.0 59.6 1 in 500 99.8% 85.0 65.6 1 in 1000 99.9% 92.0 72.6 Mean 40.19 40.39 Std Dev 11.49 10.49 Capital ($M) 15 15 Ratio to Premium 1:3 1:3 Risk Tolerance ($M) 60 60 Breach Point (Percentile) 95.9% 99.6% Return Period (in years) 24.4 250.0 30

Analytics to inform Risk Transfer cont. Calibrate program to captive s risk tolerance 31

Analytics to inform Risk Transfer cont. Recalculate captive s risk based capital requirements pre & post reinsurance to demonstrate adherence to risk tolerance guidelines 32

Analytics to inform Risk Transfer cont. Understand reinsurance pricing model 33

Risk Transfer Structures (examples) Prospective: Single-line or Multi-line Reinsurance Prospective: Structured Reinsurance Retrospective: Adverse Development 34

Risk Transfer Structures (examples) Prospective: Index based Reinsurance Prospective: Parametric Reinsurance Prospective: Catastrophe Bonds 35

Quantifying Captive Utilization Foundational Risks Complimentary Risks Traditionally Uninsured Risks Enterprise Total Cost of Risk Easy: WC GL AL E&O MPL Moderate: Terrorism EPL Equipment Breakdown Extended Warranty Difficult: Self-insured Medical Cyber Political Risk Credit D&O Reputational Brand Varies from Easy to Difficult: Group Life STD LTD Retiree benefits Third Party coverage Operational Risks 36

Summary of Best Practices and Tools 37

Know the risk Best Practices Ensure voracity and relevance of historic loss and exposure information Understand correlation between lines of coverage written Produce and scenario test analytical output By line of business and in portfolio Document assumptions and external sourced data (& weighting) Risk Appetite v. Risk Tolerance Reinsurance Seek strategic protection aligned to risk tolerance What/where/how much/benefit to captive Lessons from the field Know stakeholder views and requirements Pay careful consideration to volatility/correlation Your decisions, your legacy 38

Wrap-Up and Audience Q&A 39

Conclusion 41

PLEASE EVALUATE THIS SESSION Click on Evaluations Click on the session title Answer all the questions! Thank you!

Disclaimer This presentation contains general information only. VCIA and its guest speakers are not, by means of this presentation, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services. This presentation is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional advisor. Neither VCIA nor its guest speakers shall be responsible for any loss sustained by any person who relies on this presentation. 43