SOA Life & Annuity Symposium May 16-17, Session 31 PD, Does Anyone Else Want to be Illustration Actuary this Year?

Transcription

1 SOA Life & Annuity Symposium May 16-17, 2011 Session 31 PD, Does Anyone Else Want to be Illustration Actuary this Year? Moderator: Donna Christine Megregian, FSA, MAAA Presenters: Gayle L. Donato, FSA, MAAA Donna Christine Megregian, FSA, MAAA

2 SOA May 16 17, 2011 Life & Annuity Symposium Session 31 - PD Does Anyone Else Want to be an Illustration Actuary this Year? Presenter: Gayle L. Donato, FSA, MAAA donatog@nationwide.com Moderator: Donna Christine Megregian, FSA, MAAA Presenter Disclaimer Presentation not endorsed by SOA or AAA Presentation - solely the opinions of the presenter FSA / MAAA Illustration Actuary since 2007 Prior to 2007 assisted Illustration Actuaries 1

3 Agenda History Regulatory Requirements Challenges Solutions Best Practices Conclusion and Questions History Pre Life Insurance Illustrations Model Regulation Need for regulation: Lack of consumer understanding No consistent content / format for illustrations Misleading illustrations Vanishing premiums Aggressive non-guaranteed interest rates, dividends, etc. Terms undefined or not clear 2

4 History continued Life Insurance Illustrations Model Regulation Primary purpose consumer protection and education Effective date covered policies issued after 1/1/1997 Covered policies all individual and group life new and inforce illustrations except: Variable Life Individual and Group Annuities Credit Life Death Benefit less than $10,000 Regulatory Requirements Regulations and guidance Illustration Actuary Responsible Officer 3

5 Regulatory Requirements regulations and guidance Life Insurance Illustrations Model Regulation Actuarial Standard of Practice #24 Practice Note ASOP No. 24 Compliance with the NAIC Life Insurance Illustrations Model Regulation Illustration Work Group updating Other related ASOP s Non-guaranteed charges #1 Risk classification #12 Dividends #15 Data quality #23 Actuarial communication #41 Regulatory Requirements Illustration Actuary Certification with disclosures -> Board of Directors and State s Insurance Commissioner Annually certify Familiar with Actuarial Standard of Practice #24 Disciplined i current scales are conformity with ASOP Illustrated scales meet Model Regulation requirements Includes not using or describing non-guaranteed elements in a manner that is misleading or has the capacity or tendency to mislead 4

6 Regulatory Requirements Illustration Actuary If payable scale has been reduced for reasons outside of experience factors underlying the disciplined current scale If non-guaranteed elements for new and inforce differ e.g., Re-priced current COI s for new business, new money interest rate for several years then portfolio rate If non-guaranteed elements illustrated differ from those being paid e.g., Charges lower than illustrated charges Method used to allocate overhead expenses fully allocated, marginal, GRET Polling question Are you currently an Illustration Actuary or do you currently support an Illustration Actuary (includes consultants)? 1. Yes 2. No 61% 39% 1 2 5

7 Current Requirements Responsible Officer Not the Illustration Actuary May or may not be an Actuary Guidance only from: Life Insurance Illustrations Model Regulation Certify annually: Illustration format meets Model Scales illustrated are those certified by Illustration Actuary Agents were provided with expense allocation method Fully allocated, marginal or GRET Polling question Are you currently a Responsible Officer or do you support your Responsible Officer (includes consultants)? 94% 1. Yes 2. No 6% 1 2 6

8 Timeline of Challenges Newborn Toddler Pretender School-age Teenager Timeline of Challenges Newborn Testing new business policy form must be selfsupporting and cannot be lapse supported Self-supporting At every illustrated point in time starting with the fifteenth policy anniversary (with the twentieth for second-or-laterto-die policies), the accumulated value of all policy cash flows, when using experience assumptions underlying the disciplined current scale, should be equal to or greater than the illustrated policyholder value, i.e., the cash surrender values and any other illustrated benefit amounts available at the policyholders' election Prevent lapse-supported policy form must be self-supporting under the same assumptions and with the same level of aggregation as described above, changing only the persistency assumption to 100% persistency after year 5 7

9 Self Supporting Accumulated cash flows > Cash surrender value Year 15 and beyond Accumulated cash flows year t= Prior accumulated cash flows year t-1 + Premium + Investment income - Benefits (deaths, surrenders, dividends, etc.) - Expenses (commissions, admin, sales, overhead, etc.) - Taxes Timeline of Challenges Newborn Focused on basic needs: How to run the self-support and lapse-support tests Model setup through maturity, report output, etc. Assumptions setting through maturity Expense decision fully allocated, GRET or marginal Documentation model, results, assumptions, etc. Certification process File states & company board 8

10 Timeline of Challenges Toddler Focused policies inforce greater than a year Per ASOP 24 section The DSC, for a policy in force one year or more, continues to be in compliance with the Model and this standard, if any of the following apply: A. The currently payable scale has not been changed since the last certification and the illustration actuary determines that experience since the last certification does not warrant changes in the disciplined current scale that would make it significantly less favorable to the policyholder; or B. The currently payable scale has been changed since the development e e of the disciplined current scale most recently certified only to the extent that changes are reasonably consistent with changes in experience assumptions underlying the disciplined current scale; or C. The currently payable scale has been made less favorable to the policyholder since the last certification and the change is more than the change in the current experience would dictate Timeline of Challenges Toddler Key challenges in meeting A, B or C A. No change in scale and experience lines up with model Experience studies verses assumptions modeled often not credible B. Scale changed and are consistent with experience Straightforward for maintaining interest spread Relies some on actuarial judgment C. Scale made less favorable and it is more than experience would dictate Need to consider ASOP s like #1 when determining scale Document rational, experience studies, etc. for meeting A, B or C 9

11 Timeline of Challenges Pretender Increased focus on policies inforce over a year Harder to comply with section 3.7 of ASOP 24 as experience changes e.g., distribution, lapse, expenses, earned rates, mortality, policyholder behavior such as payments, etc. Focus shifted to these challenges: Modeling inforce Experience studies Staffing Timeline of Challenges Pretender: Modeling inforce Just Do It! Inforce models: Model each issue year and project forward Aggregate issue years when possible, and project forward Current inforce model 10

12 Modeling inforce changes / challenges Product knowledge Separating results by policy form vs. aggregating results e.g., NY vs. Non-NY NY version -combine if similar il Testing only policies issued > 1/1/1997 Assumption changes e.g., Mortality without improvement, fully allocated expenses, investment returns based on experience, etc. Policyholder behavior e.g., Premium deterioration, dynamic lapses Starting surplus Cash value Reserves Reserves and Target Surplus Timeline of Challenges Pretender Experience Study Considerations Data grouped by product & aggregated by similar product Credibility Actual to expected expected from model Mortality Late durations Underwriting changes / audit results / x-factor changes Lapse focus on product / policyholder trends No-lapse products low in later durations & zero once paid-up Dynamic lapses based on policy behavior Shock lapses Term 11

13 Timeline of Challenges Pretender Experience Study Considerations Continued Investments earned rates, spread & credited rate Illustrated interest rate shown < earned rate per Model Regulation Earned rate and spread by product Reflect deterioration that has occurred Do not reflect future improvements Consider likely future deterioration Test credited rate or spread that varies by duration Sensitivity test assumptions with low credibility, range of earned rates, etc. Timeline of Challenges School-age Reinsurance Modeling third party and or captive reinsurer Charges guaranteed? Product shifts Distribution sweet spots Underwriting Commission changes Maturity age 120 Model and test through age 120 Late duration mortality vs. late duration COI s Grade late duration mortality into VBT 12

14 Timeline of Challenges Teenager Index Universal Life (IUL) Illustrated interest credited rate blend of fixed and index Fixed rate set based on assets backing product Index 7 8% at max based on 25-year look-back is common Rate often applies to non-loaned and loaned value Include hedge costs in testing Timeline of Challenges Teenager Index Universal Life (IUL) Variable loan illustrated interest rate Not permitted by all states Loan charged rate e.g., 5.22% (Moody s 2/2011) Loan credited rate often the index return between 7 8% Illustrating credited > charged in all durations Supportable? Spirit of the Model Regulation? Moody s 25-year average 7.15% (2/2011) Regulatory requirements -protect & educate consumers /not mislead Illustrate current, min and max loan charged rate? 13

15 Solutions to failing self and lapse support tests Refine and enhance model Peer review when close to passing OR failing Decrease commissions Stop selling e.g., low band, higher issue ages Change illustrated charges COIs, premium load, interest credited rate, etc. Change on admin system also work through ASOP 1 Illustrate only guaranteed elements No Lapse Guarantee UL s recently some have done Distribution of surplus Best Practices Review Model Regulation, ASOP 24 & Practice Notes Start early and plan key steps Utilize existing experience studies, models & product information Stress or sensitivity test Shows product risk & can helps avoid re-testing during year Document assumptions, testing method, model and results methodology behind assumptions File on time with states and Board of Directors 14

16 Conclusion Aware of the latest illustration actuary requirements Understand issues facing illustration actuaries Questions Additional questions submit to IWG We will consider as we update the Practice Notes Thank you for attending! 15

17 Does anyone else want to be an illustration actuary this year? Presented by Donna Megregian, FSA, MAAA Actuary May 18, 2011 Illustration Survey Milliman conducted a survey of illustration actuaries and pricing actuaries that do illustration testing Topics included: General demographics New business issues Inforce issues Assumption Development Product Specific questions 21 companies responded 2 May 18,

18 Audience Question Did you participate in the Milliman survey? 1. Yes, I got the survey and participated i t 2. No. I got the but I missed the deadline 3. No. I didn t get the so I couldn t participate 4. I m not sure if we participated or not 35% 63% 0% 2% Audience Question How much time do you spend on Illustration testing for your annual certification (outside of the pricing process)? 1. 1 month months months months 5. It s a year round process 38% 59% 3% 0% 0%

19 How much time does your illustration testing for your annual certification take you (outside of the pricing process)? It's a year round process 10% 6-11 months 10% 4-6 months 0% 2-3 months 43% 1 month 38% 5 May 18, 2011 Audience Question How many new business policy forms do you certify? % 2. 1 to to to % 3% 6% 6% 0% 6%

20 How many new business policy forms do you certify? 26+ 0% % % 11 to 15 14% 6 to 10 19% 1 to 5 43% 0 5% 7 May 18, 2011 Audience Question Are you finding illustration testing requirements are restricting product design? 1. Yes 2. No 3. Don t know 4. Not issuing new products 33% 55% 10% 2%

21 Are you finding illustration testing requirements are restricting product design? No new products 5% Don't know 0% No 38% Yes 57% 9 May 18, 2011 Audience Question For the lapse support test on new business, how do you accommodate for premium persistency especially with regard to after year 5 when lapses are set to zero? 1. We allow policies to lapse due to 28% inadequate premium 2. We allow policies to catch up on premiums 3. We allow policies to pay the minimum 19% 19% premium to stay inforce 4. We do not use a premium persistency assumption 6% 5. It depends on the type of product 3% 6. Premium modeled is more than sufficient to carry the policy 7. Other 13% 13%

22 For the lapse support test, how do you accommodate for premium persistency especially with regard to after year 5 when lapses are set to zero? Other, please specify 10% We do not use a premium persistancy assumption 24% We allow policies to pay the min to stay inforce 14% We allow policies to catch up on premiums 10% We allow policies to lapse due inadequate premium 43% 11 May 18, 2011 Audience Question Which of the following do you include in your illustration testing of new business? 1. Substandard policies 2. Riders 3. Policy Loans 4. No Lapse Guarantees 5. Some of the above 50% 6. All of the above on some products 7. None of the above since excluding them is conservative 6% 18% 9% 15% 3% 0%

23 Which of the following do you iclude in your illustration testing? (mark all that apply) Substandard policies 33% Riders 52% Policy loans 24% No-Lapse Guarantees 62% All of the above on some products 14% None of the above 5% Other, please specify 24% 13 May 18, 2011 Audience Question During the pricing/illustration testing process, do you stress test your DCS to gain insight into when it might fail? 1. Yes, we stress testing certain DCS assumptions during the pricing process but not outside of the pricing process 2. Yes, we stress test the DCS during new business illustration testing, but not during the pricing process 3. Yes, we stress test the DCS during the pricing and new business illustration testing process 4. No, we do not stress test the DCS during the pricing or new business illustration testing process 36% 27% 24% 12%

24 During the pricing/illustration testing process, do you stress test your DCS to gain insight into when the DCS might fail? Other, please specify 10% No, we do not stress test the DCS in the pricing or new business illustration testing process 14% Yes, we stress test various DCS assumptions during both the pricing and new business illustration testing process 57% Yes, we stress test various DCS assumptions during the new business illustration testing, but not during the pricing process 5% Yes, we stress test various DCS assumptions during the pricing process, but not outside of the pricing process 14% 15 May 18, 2011 Audience Question How many inforce products do you certify? to to to to to % 18% 24% 9% 18% 12% 3%

25 How many inforce products do you certify? % % % 11 to 15 19% 6 to 10 5% 1 to 5 5% 0 5% 17 May 18, 2011 Audience Question Have you used distributions of surplus or prior gain in inforce models in order to support your DCS? 1. No, we do not assume distributions because our DCS is supportable? 2. No, we have adjusted our DCS and not assumed any distributions of surplus 3. Yes, we have done so in the past to support a DCS, but have not required an additional distribution assumption 4. Yes, we have done so and still do use accumulated surplus distributions to support inforce DCS 5. Not sure how this would work in inforce testing 6. Don t know 26% 26% 17% 17% 9% 6%

26 Have you used distributions of accumulated surplus or prior gain in inforce models in order to support your DCS? Other, please specify 10% Don't know 0% Not sure how this would work in inforce testing 19% Yes, we have done so and still do use accumulated surplus distributions to support inforce DCS 19% Yes, we have done so in the past to support a DCS, but have not required an additional distribution assumption 0% No, we have adjusted our DCS and not assumed distributions of surplus 0% No, did not assume distributions because our DCS is supportable 52% 19 May 18, 2011 Audience Question Have you been challenged by a state department of insurance regarding a proposed non-guaranteed element change? 1. Not applicable 2. No 3. Yes, by a few states 4. Yes, by several states 5. Don t know 66% 14% 14% 6% 0%

27 Have you been challenged by a state department of insurance regarding a proposed Non-guaranteed element change? No 52% Yes, by a few states 10% Yes, by several states 5% Not applicable 33% 21 May 18, 2011 Audience Question What type of expenses did you use in your most recent illustration testing? 1. GRET 2. Marginally allocated 3. Fully Allocated 35% 57% 8%

28 What type of expenses did you use in your most recent illustration testing? GRET 33% Marginally allocated 0% Fully allocated 67% 23 May 18, 2011 Audience Question What is your biggest assumption concern with regard to illustration testing? 1. Mortality 2. Expenses 3. Persistency 4. Investment Return 5. Distribution 6. Reinsurance 13% 28% 19% 38% 3% 0%

29 What is the biggest assumption concern when doing illustration testing? Reinsurance 1% Distribution 3% Investment Return 46% Persistency 10% Expenses 10% Mortality 29% 25 May 18,