2006 Tillinghast Pricing Methodology Survey Results

Transcription

1 2006 Tillinghast Pricing Methodology Survey Results Actuaries Club of the Southwest Fall Meeting Dominique Lebel (415) November 15, 2007 Presentation overview Profit measures and objectives Stochastic analysis Profit targets versus achieved profitability Validation of pricing objectives 1

2 The Tillinghast business of Towers Perrin recently conducted the 2006 Pricing Methodology Survey The focus of the survey was on the profit objectives and assumptions used in pricing 2005 new issues of individual life, annuity and health products The survey covered the following eleven products: Term Insurance (TERM) Par Ordinary Life (PAR OL) Universal Life with no-lapse guarantee (UL NLG) Universal Life (UL) Variable Life/Variable Universal Life (VL) Fixed Deferred Annuity (FA) Variable Deferred Annuity (VA) Equity Indexed Deferred Annuity (EIA) Single Premium Immediate Annuity (SPIA) Long-Term Care (LTC) Individual Disability Income (IDI) 80 life insurance, annuity and health care writers participated (including some of the largest writers) Of these, 66 respondents were stock companies and 14 were mutual companies or fraternal organizations 2 Profit measures and objectives 3

3 Statutory return on investment (ROI) remains the most common primary profit measure Primary Profit Measures by Product (Percent of Responses) 12% 1 7% 4% 2% 12% 4% 22% 7% 1 8% 24% 12% 17% 1 9% 13% 14% 8% 15% 2% 64% 64% 73% 74% 77% 71% 58% TERM PAR OL UL UL NLG VL FA VA ROI ROE Profit Margin ROA Other ROI = statutory return on investment (internal rate of return). ROE = GAAP return on equity. Profit margin = present value of profits as a percent of present value of premium. ROA = present value of profits as a percent of present value of reserves/fund. 4 Product TERM PAR OL UL NLG UL VL FA VA EIA SPIA LTC IDI All The median target return on investment for all companies ranges from 10. to 14. for the various product types surveyed ROI/ROE Target Values* Mutual/Fraternal Stocks Minimum Maximum Median Minimum Maximum Median % % 8.5% % 12.5% 17.5% 12.5% % % 11. All Companies Minimum Maximum Median % 8.5% % *Results are based on primary profit measures and do not include the increase in equity returns that may be available through debt leverage. ROI = statutory return on investment (internal rate of return). ROE = GAAP return on equity. Note: Products with less than four responses were not reported 5

4 Biggest change in profit objective by product was for Par OL Median ROI/ROE Target Values 2004 Survey 2006 Survey Term Par OL UL VL FA VA 6 Nearly all companies that use ROI/ROE as the primary profit measure for life and annuities calculate profit targets after provision for taxes and target surplus Percent of Profit Targets After Provision for Tax and Target Surplus Primary Secondary Profit Measure Life Annuities Health Life Annuities Health ROI 98% 99% 10 91% 98% 5 ROE N/A 10 87% N/A ROA N/A 64% N/A 10 67% N/A Profit Margin 22% 46% 33% 8 63% 83% ROA and profit margin are less likely to be calculated after provision for target surplus 7

5 Targeted ROI/ROEs have trended downward for the past several years, perhaps driven by falling interest rates ROI/ROE Target Values* Mutual/Fraternal (PAR OL) Stock (UL) ROI/ROE <12. 83% 82% 10 67% 10 13% 2 28% 45% % 17% 12% 17% 45% 43% 43% 44% 33% % 17% 5 35% 37% 28% 12% >15. 5% 9% 1 Average 10-yr Treasuries** 6.4% 5.7% % 4.4% 6.4% 5.7% % 4.4% ROI/ROE targets for stock companies continue to trend downward, with a higher proportion targeting less than 12% as compared to the last survey *Results are based on primary profit measures and do not include the increase in equity returns that may be available through debt leverage. **The Average 10-year Treasuries are for the calendar year prior to the survey, as the survey is based on products priced in the previous calendar year. Source: Tillinghast Pricing Methodology Surveys; Federal Reserve. 8 The most common pricing horizon for UL, VL and PAR OL is 30 years, while for deferred annuity products it is 20/25 years Pricing Horizon by Product (Percent of Responses) % 4% 3% 4% 3% 4% 3% 3% 7% 13% 8 44% 32% 35% 27% 44% % % 44% 47% 42% 66% 57% 83% 2 89% 27% Other Level Term 40+ years, incl lifetime 30 years 20/25 years 10/15 years % 3% UL NLG and LTC products are more commonly priced over the lifetime of the business, while term products are most often priced over the level term period 31% 24% 9% 18% 19% 18% 22% 1 1 7% TERM PAR OL UL NLG UL VL FA VA EIA SPIA LTC IDI 6% 6% 23% 9

6 Profit targets and competitive position are most likely to be sacrificed when trade-offs are necessary to meet defined goals Sacrifices Made When Trade-offs are Necessary* Life Annuities Health Pricing assumptions Commission schedules Competitive position Profit targets 41% of respondents indicated competitive position as most likely to be sacrificed when trade-offs are necessary to meet defined goals, while 39% indicated profit targets as the first to be sacrificed * Values represent average of all responses, where respondents ranked relative importance on a scale of 1 to 4. 4 = least likely to be sacrificed, 1 = most likely to be sacrificed. 10 The majority of companies continue to use the NAIC RBC formula as the main driver of target surplus S&P Formula 15% Main Driver of Target Surplus (Percent of Respondents) Economic Capital 1 Other 2% The percentage of respondents using the NAIC RBC formula has increased from 63% in 2003 to 69% in 2005 The percentage of respondents using the S&P formula has decreased from 2 in 2003 to 15% in 2005 AM Best Formula 4% NAIC RBC Formula 69% 11

7 The number of companies setting target surplus above 25 has increased from 2003 % Respondents Using RBC Ratios Above For the 2006 survey, for companies with RBC ratios above 25: 5 indicated they used values in the range of of respondents used values above There is a fairly equal split among respondents using fully allocated and projected fully allocated expenses in calculating target profitability Type of Expense Assumption Used (All Products) (Percent of Responses) 25% 7% 5% 3% 8% 37% 47% 45% Other Marginal Projected Fully Allocated Fully Allocated 38% 43% 42% Mutual/Fraternal Stock All Companies Fully allocated = today s expense levels (includes overhead) Projected fully allocated = expense levels on a fully allocated basis expected to be achieved in future based on projected sales (includes overhead) Marginal = does not include overhead 13

8 The number of companies indicating that they projected future mortality improvements when calculating expected profitability has increased from 2003, with a few exceptions by product 14% 27% 8% 6% Projected Future Mortality Improvement (Percentage of Responses) 38% 29% 25% 19% 14% 15% 12% 5% Term PAR OL UL NLG UL VL LTC IDI Future expected population improvement was cited as the rationale for the projected improvements by the majority of respondents that project them Projected improvements were generally limited to the first 10 or 20 policy years This slide does not include immediate annuities because we presume that virtually all companies reflect future mortality improvements in pricing immediate annuities 14 Most companies did not include projected future improvements in LTC morbidity assumptions Projected Future Morbidity Improvement Yes 5% No 95% 15

9 Approximately half the respondents assume no variance in pricing assumptions by channel; for companies that have multiple distribution channels, nearly all vary pricing assumptions by distribution expenses 10 Pricing Assumptions Varied by Channel* (Percent of Respondents) 96% 92% 74% 64% 52% 62% Profit targets Mortality/morbidity Lapse Distribution expenses Maintenance expenses 38% % 4% 11% Life Annuity Health 81% of companies indicated that they use multiple distribution channels * Companies selected all choices that applied. 16 Stochastic analysis 17

10 Most companies do not make an adjustment for the cost of interest rate embedded options in assets and liabilities for UL and UL NLG Methodology for Reflecting Interest Rate Embedded Options in Assets and Liabilities (Percent of Responses)* UL NLG 61% 34% 5% UL 74% 21% 6% FA 44% 41% 14% EIA 2 69% 11% None. Single Interest Mean of Scenarios "Haircut" to Interest Margin The results of this slide are surprising to us. There is not as much provision for interest rate risk as we would have thought. 18 Practices vary by company for pricing guarantees on VAs 26% 23% Method Used to Determine Cost of Guarantees on VAs (e.g., GMDB, GMIB) (Percent of Respondents) 19% 3 6% 12% 24% 21% 26% 14% 2006 Survey 2004 Survey Mean cost based on stochastic scenarios Fiftieth or other percentile based on stochastic scenarios Set equal to reinsurance cost Set equal to expected hedge costs Other The 85th percentile was the median response for companies that use the fiftieth or other percentile For Other, the responses varied from company to company 19

11 Most companies use risk neutral stochastic scenarios to determine the cost of guarantees on VAs Type of Scenario (Percent of Respondents) Other 13% The most common responses for Other were: Consultant data Multiple types of scenarios Real World 34% Risk Neutral 53% 20 Assumed separate account growth rates have decreased since Gross Separate Account Earned Rate* (Number of Respondents) < % % % % % > Survey 2004 Survey *Gross separate account earned rate assumed in calculating expected profitability (before deductions for mortality and expense risk charge, investment advisory fee or other fund expenses). 21

12 Profit targets versus achieved profitability 22 Median GAAP ROEs achieved for life and annuity business in-force in 2005 are generally lower than targeted median ROI/ROEs for 2005 new issues GAAP ROE 35% 3 Max Median Min 2005 Actual GAAP ROE vs. Targeted ROI/ROE (Unleveraged) % 2 15% 1 5% 12.5% % % % % Actual Target Actual Target Actual Target Actual Target 0.9% Life Annuity Life Annuity Mutual/Fraternal Stock Note: Actual and targeted minimum and maximum results are not necessarily for the same company. 23

13 Almost one-fifth of respondents indicated that product sales in 2005 are not likely to achieve aggregate profit targets, with some variation by product Likelihood of 2005 Issues Meeting Aggregate Profit Targets (Percent of Responses) 3% 27% 25% 43% 14% 15% 21% 5% 46% 54% 47% 58% 49% 53% 43% 23% 19% 15% 39% 31% 32% 36% TERM PAR OL UL NLG UL VL FA VA Very likely Somewhat likely Not likely Don't know 24 The likelihood of achieving aggregate profit targets varies by the type of expense assumption used Likelihood of 2005 Issues Meeting Aggregate Profit Targets Fully Allocated Expenses Don t know 1% Projected Fully Allocated Expenses Don t know 1% Marginal Very likely 48% Not likely 17% Somewhat likely 34% Very likely 16% Not likely 23% Somewhat likely 6 Very likely 18% Somewhat likely 5 Not likely 32% Respondents who used Fully Allocated Expenses are more confident that they will achieve aggregate profit targets The proportion of very likely for Fully Allocated Expenses has increased from 43% in 2003 to 48% in 2005 Fully allocated = today s expense levels. Projected fully allocated = expense levels on a fully allocated basis expected to be achieved in future based on projected sales. 25

14 Validation of pricing objectives 26 55% of the participants measure actual vs. targeted profitability by monitoring GAAP ROE 55% Method Used to Measure Actual vs. Expected Profitability (Percent of Respondents) 4 44% 29% 23% GAAP ROE Embedded value calculations Run pricing model using actual experience Gain by source analysis Other Monitoring actual to expected was the most common answer for Other Note: Companies selected all the methods that applied. 27