SOA Risk Management Task Force

SOA Risk Management Task Force Update - Session 25 May, 2002 Dave Ingram Hubert Mueller Jim Reiskytl Darrin Zimmerman Risk Management Task Force Update Agenda Risk Management Section Formation CAS/SOA Joint ERM Seminar Sub Group Updates 2 1

This is an Open Forum Your Participation is Expected! 3 Risk Management Task Force Update - Agenda Risk Management Section Formation CAS/SOA Joint ERM Seminar Sub Group Updates 4 2

Risk Management Task Force Update Risk Management Section formation CAS/SOA Joint Seminar Task Force Sub Groups Equity Modeling Economic Capital Risk Covariance Risk Management Metrics Pricing for Risk Extreme Values 5 Risk Management Section Formation Risk Management Task Force was formed in 2001 By the Finance Practice Area Advancement Committee To hold Risk Management Seminars To identify and develop Risk Management materials for actuarial education Remainder of 2001 spent organizing February 2002 Call for Volunteers Most successful such call ever Now have over 250 interested parties And about 100 active volunteers 6 3

Risk Management Section Formation High Degree of member interest suggests that a Risk Management Section should be formed Organizing Committee formed in April, 2003 Solicitation of initial members May/June 2003 Need 200 to form section Petition to Board June/July 2003 Hope to Continue both RMTF and RM Section Exact division of roles to be determined Expecting to make this an addition SOA efforts in the area of developing Actuarial Risk Management Practices 7 Risk Management Section Formation Current Status: 8 4

Risk Management Task Force Update - Agenda Risk Management Section Formation CAS/SOA Joint ERM Seminar Sub Group Updates 9 CAS / SOA Joint Enterprise Risk Management Seminar Joint Project of the RMTF and the CAS ERM Committee July 29, 30 Washington, D.C. Follows immediately after CAS Risk and Capital Management Seminar 10 5

ERM Seminar - Concurrent Sessions: ERM in Banking Value of ERM How ERM is Consistent with Embedded Value Reporting Outside Perspective of ERM ERM Case Studies Sarbanes -Oxley Credit Risk Modeling and Management Market and Credit Risk Integration Reinsurance Counterparty Risk DFA and DFCA as ALM, ERM Tools Managing and Measuring Operational Risks IAA Progress on RBC Risk-adjusted Capital Allocation Risk and Capital Management Through ALM Risk Premium for Insurance Product Pricing Securitization & Other Instruments for Transferring Risk to the Capital Markets Modeling Extreme Market Movements Risk Metrics Risk Tolerances and Limits Measures of Correlation Measuring, Hedging, and Management of Correlation Risk 11 ERM Seminar - Speakers Include 12 6

Risk Management Task Force Update Agenda Risk Management Section Formation CAS/SOA Joint ERM Seminar Sub Group Updates 13 SOA RMTF Sub Groups: 1. Economic Capital Calculation and Allocation 2. Enterprise Risk Management 3. Equity Modeling 4. Extreme Value Models 5. Health Risk Management 6. Policyholder Behavior in the Tail 7. Pricing for Risk 8. Risk Based Capital Covariance 9. Risk Management Future and Strategy 10. Risk Management Metrics http://www.soa.org/sections/rmtf/rmtf.html 14 7

Risk Management Task Force Update Agenda Sub Group Updates Equity Modeling Policyholder Behavior Economic Capital Risk Metrics Risk Covariance Health Risk 15 Equity Modeling 1. Subgroup Focus: With the introduction of new product designs in recent years, equity risk is quickly outpacing interest rate risk as the dominant market risk in insurance company portfolios 1. Task 1 to assess the kinds of modeling tools and techniques available to measure equity risk 2. Task 2 - to provide a comprehensive description of such techniques, along with identification of parameters and assumptions 3. Task 3 To identify the pros and cons of each approach used to manage equity risk 2. Ultimate Product To develop a practical guide for modeling and managing equity risk geared towards actuaries 16 8

Equity Modeling - Update Created an annotated reading list and posted to the web site. Identify a list of about twelve gaps, and have addressed about half of them so far. Is there further materials that would address the gap, whether material should be commissioned to fill the gap or whether we should identify it as a topic for future consideration. We have a few hard core members who have been active throughout the piece, but also welcome new members to bring us new energy and new ideas. 17 Equity Modeling Gaps Topic 1: Equilibrium vs Non Arbitrage Models Topic 2: Markov Chain Topic 3: Parameter Estimation Topic 4: Correlations Topic 5: Random Number Generation/ Variance Reduction Techniques Topic 6: Monte Carlo Simulation Topic 7: Risk neutral (Q) vs. Real World (P) Valuations Topic 8: Stock Indices vs Actively Managed Funds Topic 9: Currency Risk Topic 10: Model Validation Topic 11: Historical Simulation Topic 12: State Price Deflators Topic 13: Practical Application 18 9

Update from the SOA Risk Management Task Force Economic Capital Sub Group SOA Spring Meeting Washington, DC Session 25 Hubert Mueller Tillinghast Towers Perrin, (860) 843-7079 EC Subgroup One of the 10 subgroups within the Risk Management Task Force (RMTF) Goals To research measurement of economic risk in life insurance companies, practiced in North America To document techniques and standards that quantify the material risks of life insurance companies To document techniques for allocating risk capital for use in pricing, budgeting and financial reporting Participants Over 100 people on the listserve Generally, 10-12 participants on regular conference calls Active participants provide a good mix of experience from insurers, consultants and rating agencies 20 10

Initial Working Products Industry survey on current practices related to EC Completed in the fall of 2002 Almost 500 participants Highlights discussed at the 2002 Boston Annual Meeting EC Specialty Guide Designed to be a source of information for practitioners Summarizes the results of the survey and provides additional insights by members of the subgroup Initial draft is now on website A work in progress 21 EC Specialty Guide Topics (Revised) Definition of Economic Capital Uses of EC in the current marketplace Tie-in of EC to regulatory/rating agency capital Current approaches to calculating EC Case studies (could use more) Regulatory developments (RBC C-3 II) Literature Review Ranked distribution of present values of future profits from each simulation $m + 0 Economic Capital Selected risk tolerance Cumulative probability 22 11

Risk Covariance Status of Research Project 23 Policyholder Behavior in the Tails Status of Chief Actuary survey 24 12

Risk Metrics Have produced a series of papers defining various risk metrics Risk Duration, Convexity Net amount at Risk Value at Risk Risk Based Capital Business Plan Risk Measures: Growth Rate, Unit Cost Earnings at Risk Option Adjusted Spread Embedded Value Greeks Latest paper Risk Exposure Limits In Process Risk Tolerance 25 Types of Exposure Limitations a. Event Specific b. vary by level of management c. can limit the exposure of a specific type of risk to a percentage of the total risk exposure. This would ensure total exposure is a combination of specific exposures that are not 100% correlated. d. limiting the exposure in any specific instance of a particular type of risk. e. can be used in contractual arrangements to describe how risk will be shared between two or more parties. f. can be used in combination with another risk management metric by placing a floor and or cap on that risk management metric. g. part of a flexible pricing process by adjusting prices up (or down) as the difference between the exposure limitation and the current exposure decreas es (increases). h. do not have to be monetary. Exposure limitations can also be used to limit operational risk or achieve operational objectives. 26 13

Health Risk Several projects underway: Risk Mapping Modeling Solvency 27 Risk Mapping Environmental Risk Financial Risk Operational Risk Pricing Risk Reputation Risk Strategic Risk Anti-selection Competition Data Financial Viability ofcapitated Entity Model Reinsurance Trend: Inflation Trend: Intensity/Severity Trend: Technology Trend: Utilization Underwriting Authority Morbidity/Mortality Regulatory/Legislative Definitions and identifying correllations 28 14

Risk Management Task Force Agenda has been entirely driven by participants Volunteers are needed to continue and conclude current projects Ideas and volunteers are needed for future projects 29 Additional Materials Pricing for Risk Survey Economic Capital Survey 30 15

Pricing for Risk 1. Objective: To evaluate the different pricing techniques as to their effectiveness in capturing and quantifying the varied risks associated with the sale and administration of life insurance and annuity products. 1. Task 1 Survey of current industry practice 2. Task 2 Development of a comprehensive analysis of existing methods, along with an analysis of advantages and shortcomings of each 2. Ultimate Product Practice Guide for Pricing for Risk 31 Pricing for Risk Terms Return on Investment Return on Equity Return on Liabilities Risk-Adjusted Return on Capital Premium Margin Break-even Year Internal Rate of Return Return on Assets Return on Capital Contribution to Surplus Revenue Margin Embedded Value (or Economic Value Added) Capital Allocations Risk-Adjusted Profit Target Stochastic Scenario Analysis Assumption PADS Assumption Stress Testing Regulatory Formula (RBC/MCCSR) Multiple Economic Capital Earnings (Value) at Risk Mean-Variance Analysis Efficient Frontier Conditional Tail Expectation Problem Scenario Analysis Reduction to Yield Covariance of Risk 32 16

Pricing For Risk Survey Conducted August & September 2002 RMTF listserve and Product Development Section Members were asked to participate 351 Responses 33 What Pricing Measure does your company use? Premium Margin 18% Return on Equity 14% Internal Rate of Return 13% Return on Investment 10% Break-even Year 9% Risk Adjusted Return on Capital 6% Embedded Value 6% Economic Value Added 5% Return on Capital 4% Contribution to Surplus 4% Return on Assets 4% Revenue Margin 3% Other 2% Return on Liabilities 1% 34 17

Pricing Measures Used Across All Products Internal Rate of Return 34% Premium Margin 21% Return on Equity 14% Embedded Value 11% Break-even Year 9% Return on Assets 4% CTS 4% Other 2% 35 Pricing Measure by Product Life Annuity Grp Life Grp A&H Grp Ann Ind Hth Premium Margin 19.23% 12.9% 32.1% 32.2% 14.0% 28.3% Return on Equity 12.72% 15.5% 20.5% 17.5% 18.4% 12.2% Internal Rate of Return 36.35% 37.7% 25.0% 24.0% 37.4% 33.5% Break-even Year 12.43% 8.5% 5.8% 7.6% 6.2% 8.3% Embedded Value 11.09% 11.1% 10.3% 9.9% 11.7% 10.0% Return on Assets 3.43% 9.6% 0.6% 0.6% 7.3% 0.2% Other 1.50% 2.6% 1.9% 1.2% 2.2% 2.3% CTS 3.24% 2.1% 3.9% 7.0% 2.8% 5.2% 36 18

Methods of Reflecting Risk Capital allocation pricing for capital as a cost Risk-adjusted profit target higher risk products have higher targets Stochastic scenario analysis Pricing centered on set of scenarios Assumption PADS Explicit or implicit conservatism Assumption stress testing Regular set of tests 37 How Do you Price for Risk? Capital allocation Risk-adjusted profit target Stochastic scenario analysis Assumption PADS Assumption stress testing ROI 25% 18% 19% 13% 25% IRR 26% 18% 17% 12% 26% ROE 30% 16% 21% 11% 20% CTS 21% 27% 17% 15% 19% Premium Margin 16% 23% 13% 18% 28% 38 19

What is the basis for Risk Adjustment? 3. If you use capital allocations for reflecting risk, how are these allocations determined? Regulatory formula multiple 147 55.26% Internal formula 69 25.94% Economic capital 44 16.54% Other 6 2.26% Total: 266 39 What is the basis for Risk Adjustment? 4. If you use assumption PADS, how are these PADS determined? Analysis of recent experience 66 50.00% Industry standard 36 27.27% Other 5 3.79% Stochastic scenario analysis 25 18.94% Total: 132 40 20

What is the basis for Risk Adjustment? 5. If you risk-adjusted profit objective, how is it determined? Judgment 81 61.83% Formula 44 33.59% Other 6 4.58% Total: 131 41 What is the basis for Risk Adjustment? 6. If you use assumption stress testing, how are the parameters determined? Judgment 137 59.05% Confidence limits 48 20.69% Worst case historical experience 42 18.10% Other 5 2.16% Total: 232 42 21

What is the basis for Risk Adjustment? 7. If you use stochastic scenario analysis, how is the distribution of results analyzed? Percentiles 83 30.51% Mean-variance analysis 44 16.18% Conditional tail expectation (CTE) 44 16.18% Problem scenario analysis 38 13.97% Value at risk 24 8.82% Efficient frontier 23 8.46% Earnings at risk 14 5.15% Other 2 0.74% Total: 272 43 How are different risks handled? 8. A. How do you capture the risk associated with asset default in pricing? Reduction to yield 196 59.57% Capital allocation 66 20.06% Assumption stress testing 31 9.42% Stochastic scenario analysis 28 8.51% Other 8 2.43% Total: 329 44 22

How are different risks handled? 8. B. Who determines parameters and magnitude of this provision? Both areas 132 49.81% Investment area 62 23.40% Actuarial area 61 23.02% Other 10 3.77% Total: 265 45 How are different risks handled? 8. C. How is the amount of the adjustment determined? Internal model 146 51.23% Rating agency factors (e.g. Moody's) 97 34.04% Other 25 8.77% Commercial software 17 5.96% Total: 285 46 23

How are different risks handled? 9. How do you capture the risk associated with interest rate changes in pricing? Stochastic scenario analysis 103 27.03% Assumption stress testing 96 25.20% Capital allocation 58 15.22% Reduction to yield 57 14.96% Duration analysis 57 14.96% Other 10 2.62% Total: 381 47 How are different risks handled? 10. A. How do you capture the risk associated with the volatility of equity returns? Stochastic scenario analysis 94 34.31% Assumption stress testing 78 28.47% Capital allocation 41 14.96% Historical tests (back testing) 28 10.22% Capital markets price analysis 19 6.93% Other 14 5.11% Total: 274 48 24

How are different risks handled? 10. B. If you use stochastic scenario analysis, what manner of assumptions do you use? Historical 81 32.79% Mean reversion 61 24.70% Capital markets 37 14.98% Arbitrage free 35 14.17% Risk neutral 27 10.93% Other 6 2.43% Total: 247 49 How are different risks handled? 11. How do you capture the risk associated with adverse claim deviation in pricing? Assumption stress testing 134 36.31% Assumption PAD 84 22.76% Capital allocation 47 12.74% Dynamic lapse formula 40 10.84% Stochastic scenario analysis 37 10.03% Reinsurance price 22 5.96% Other 5 1.36% Total: 369 50 25

How are different risks handled? 12. How do you capture the risk associated with shortterm claim fluctuation in pricing? Assumption stress testing 94 40.00% Assumption PAD 66 28.09% Capital allocation 42 17.87% Stochastic scenario analysis 27 11.49% Other 6 2.55% Total: 235 51 How are different risks handled? 13. How do you capture the risk associated with modeled customer and agent behavior in pricing? Assumption stress testing 109 38.52% Dynamic lapse formula 74 26.15% Assumption PAD 48 16.96% Stochastic scenario analysis 41 14.49% Other 11 3.89% Total: 283 52 26

How are different risks handled? 14. How do you capture the risk associated with the expense assumptions used in pricing? Assumption stress testing 126 43.75% Assumption PAD 67 23.26% Inflation model 58 20.14% Capital allocation 16 5.56% Stochastic scenario analysis 15 5.21% Other 6 2.08% Total: 288 53 How are different risks handled? 15. What additional explicit work is done to capture the covariance of risks? No covariance assumed 118 50.00% Capital allocation 38 16.94% Interest rates and dynamic behavior 28 15.32% Multi-risk stochastic analysis 28 14.52% Reduced assumption PAD 9 1.61% Other 8 1.61% Total: 229 54 27

Who Responded? 1. What is your primary area of practice? Pricing 144 41.02% Risk Management/Corporate 74 21.08% Valuation 70 19.94% Other 32 9.12% Investments 24 6.84% Marketing 7 1.99% Total: 351 55 Who Responded? 2. What is your type of employment? Insurance 201 71.28% Consulting 46 16.31% Reinsurance 16 5.67% Investment Company 9 3.19% Other 6 2.13% Software Developer 2 0.71% Marketing Organization 1 0.35% University or College 1 0.35% Total: 282 56 28

Who Responded? 3. For which product lines are you mainly responsible for? Individual Life 124 26.84% Individual Annuity 116 25.11% Individual Health (including critical illne 50 10.82% Group Health 41 8.87% Group Life 40 8.66% Group Annuity 35 7.58% Other 30 6.49% Reinsurance 26 5.63% Total: 462 57 Who Responded? 4. Which is your country of primary practice? United States of America 235 83.63% Canada 26 9.25% Other 20 7.12% Total: 281 58 29

Who Answered? 5. As a proxy for the size of your company, indicate the approximate number of qualified actuaries employed at your company worldwide. Less than 10 80 29.09% More than 150 55 20.00% 30 to 80 51 18.55% 80 to 150 45 16.36% 10 to 30 44 16.00% Total: 275 59 Highlights of the EC Survey Results Conducted over a six week period mid-july through early September, 2002 Blast e-mail was sent to members of International Section Financial Reporting Section Investment Section 491 survey participants 44% multinationals 32% US-based 8% North American 4% Canadian 57% had assets greater than $20 billion 68% were stock companies 24% of respondents were consultants 60 30

Over 80% of participants agreed to a strawmandefinition of EC At the enterprise level, Economic Capital is typically defined as sufficient surplus capital to meet negative cash flows at a given risk tolerance level. 81% agreed (strongly or somewhat) 9% disagreed 10% didn t know 61 Most respondents agreed that EC should cover various types of risks Interest Rate Risk 96% Pricing Risk 93% Credit Risk 92% Equity Market Risk 91% Liquidity Risk 86% Operational Risk 79% 62 31

Almost half of the respondents have been using the concept of EC at their company/in their work Have used EC 45% Have not used EC 55% 63 Determining economic capital for selected risk tolerance Ranked distribution of present values of future profits from each simulation $m + 0 Economic Capital Selected risk tolerance Cumulative probability 64 32

When determining EC, various risk tolerance measures are currently in use Specified Percentile 60% Multiple of Standard Deviation 17% CTE 15% Other 9% 65 The main reasons for calculating EC include risk and performance measurement 1. 44% of respondents cited risk management as a key reason 2. 32% cited performance measurement 3. 59% have established a formal framework for the calculation of EC 66 33

For measurement of EC, less than half of respondents currently use stochastic methods How has EC been measured for various lines of business? Stochastic model 43% Formulaic approach 31% Deterministic 28% Mean Variance Covariance 18% 67 While only a minority are aware of the new RBC C-3 Phase II requirements,... Aware of RBC C-3 Phase II 38% Not aware 62% 68 34

, a majority of companies expect EC to have greater significance going forward What are your plans/expectations for future developments of EC at your company? Greater significance 60% No change 14% Lower significance 2% Don't know/other 24% 69 70 35

David Ingram, FSA, PRM (212) 279 7166 david.ingram@milliman.com 36