Group Long Term Disability Claims Experience Study - An Update

Group Long Term Disability Claims Experience Study - An Update GUAA in Philadelphia June 2, 2009 Rick Leavitt, VP Pricing and Consulting Actuary, Smith Group

GLTD Termination Study - Agenda 1. Describe the Study 2. Claim Demographic Trends 3. Termination Trends 4. Early Duration Terminations 5. Impact of Cancer 6. Mortality Improvement 7. Change in Definition 8. Mental & Nervous Limit 9. Elimination Period 10. Composition of Terminations 11. Next Steps 2

Participating Companies Thanks for the participation! 3

97-06 GLTD Termination Study 21 participants: 72% of 2007 market share Top 15 carriers except #3, #5 and #8 1997 2006 claim termination experience Over 1 million LTD claims From fairly homogeneous products 120,000 deaths 560,000 recoveries 4

How big is that study? Term count: CGDT87-40k 97/06-680k Claim count: Table95a 0.25M 97/06 1.2M Life-Years: Group Life Waiver - 0.7M 97/06-2.4M Carrier count: 2000 GLTD -10 97/06 21 5

Controlling for Company s Exposure Top 2 = 50% of exposure Top 5 = 75% of exposure Table should reflect o the industry experience? o the average company s experience? Something in between: o Top 5: each one set to 12% o Remaining ones: 40% 6

Data Scrubbing and Company Variance Company data reviewed by data buddies Intercompany variance comparison Errors caught & resubmitted Limited trimming of exposure Remapping of Recovery to max-out Remapping of Recovery to limit 7

Claim Demographic Trends Percent of New Claims by Diagnosis Back Cancer Other Musculoskeletal Injury Circulatory M&N Maternity Nervous System 15% 13% 11% 9% 7% 5% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 8

Claim Demographic Trends Percent of New Claims by Diagnosis 3.0% Respiratory Digestive Ill-defined Diabetes 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 9

Claim Demographic Trends Average Age by Gender 51 50 Total Female Male 49 48 47 46 45 44 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 The average age of a new claimant has increased by 2 years since 1997 10

Termination Trends 130% 120% History of Recovery and Death A to E's Recoveries Death 110% 100% 90% 80% 70% 60% Table 95a 2008 Study 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 11

Termination Trends 2% Change in New Claim Reserve Factor 0% -2% -3% -5% -6% -8% All EP 90 EP180 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Estimated cost per new claim has dropped by close to 6% since 1997 12

Termination Trends Can the shift in Terminations be explained by a shift in Diagnosis? 140% 130% 120% 110% 100% 90% 80% 70% 60% Termination Actual to Expected Ratios Recov Raw Death Raw Recov Adj for Diagnosis Death Adj for Diagnosis 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 13

Termination Trends 10% 8% 6% 4% 2% 0% -2% -4% -6% Pct Change in Term Rates (97-00 to 03-06) Recovs Deaths 1 2 3 4 5 6 1: Back, Injury, Musculo 3: Mental & Nervous 5. Cancer 2: Digestive, Other 4: Circulatory/Phys Nervous 6: Respiratory/Diabetes 14

Termination Trends Pct Change in Term Rates (97-00 to 03-06) 15% 13% 11% 9% 7% 5% 3% 1% -1% -3% -5% Recovs Deaths F: Age < 40 F: 41-55 F: Age > 55 M: Age < 40 M: 41-55 M: Age > 55 Recovery Improvement increases with age: Death Rate Improvement hits males more than females 15

Early Duration Terminations Early Duration Recoveries per 1000 months exposed 2Q 3Q 4Q 5Q 6Q 7Q 8Q 9Q 10Q 11Q 12Q 120 100 80 60 40 20 0 1 2 3 4 5 6 1: Back, Injury, Musculo 2: Digestive, Other 3: Mental & Nervous 4: Circulatory/Phys Nervous 5. Cancer 6: Respiratory/Diabetes 16

Early Duration Terminations Early Duration Deaths per 1000 months exposed 40 35 30 25 20 15 10 5 0 2Q 3Q 4Q 5Q 6Q 7Q 8Q 9Q 10Q 11Q 12Q 1 2 3 4 5 6 1: Back, Injury, Musculo 3: Mental & Nervous 5. Cancer 2: Digestive, Other 4: Circulatory/Phys Nervous 6: Respiratory/Diabetes 17

Claim Survivorship Claim Survivorship by Duration and Diagnosis Group 100% 1 2 3 4 5 6 80% 60% 40% 20% 0% 1 Yr 2 Yr 3 Yr 5 Yr 10 Yr 1: Back, Injury, Musculo 3: Mental & Nervous 5. Cancer 2: Digestive, Other 4: Circulatory/Phys Nervous 6: Respiratory/Diabetes 18

Estimated Cost per New Claim Nervous System Diabetes Respiratory Circulatory Other Ill-defined and Misc Back Other Digestive Mental and Nervous Injury other than back Cancer Maternity EP 90 Nervous System Diabetes Respiratory Circulatory Other Ill-defined and Misc Back Other Digestive Mental and Nervous Injury other than back Cancer Maternity 0 10 20 30 40 50 60 EP 180 0 10 20 30 40 50 60 19

Impact of Cancer Claims Cancer Respiratory Digestive Diabetes Maternity Other Mental and Nervous Circulatory Ill-defined Nervous System Injury other than back Other Back Actual to Expected Deaths by Diagnosis 0% 50% 100% 150% 200% 250% 300% 350% Cancer Death Rate is 13.6 times as high as non-cancer death rate 20

Impact of Cancer Claims Actual to Expected Deaths 450% 400% 350% 300% 250% 200% 150% 100% 50% 0% Other Cancer Total 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Non-cancer claims show half the mortality improvement of cancer claims 21

Impact of Cancer Claims Death rates by duration relative to first year rates 140% 120% 100% 80% 60% Other Cancer 40% 20% 0% Yr 2 Yr 2 Yr 4 Yr 5 Yr 6 Yr 7 Yr 8 Yr 9 Yr 10 Yr 11+ Cancer rates decline by duration while the others increase 22

Impact of Cancer Claims 900% 800% 700% 600% 500% 400% 300% 200% 100% 0% Death rates by age relative to 42 yr-old rates Other Cancer 22 27 32 37 42 47 52 57 62 67 72 77 82 Other rates increase with age while cancer rates do not 23

Impact of Cancer Claims 16 14 Ratio of Death Rates by age and gender vs Avg 42 yr old rate F-Other F-Cancer M-Other M-Cancer 12 10 8 6 4 2 0 22 27 32 37 42 47 52 57 62 67 72 77 82 24

Mortality Improvement 115% Non-Cancer: A to E Deaths by Calendar Year 110% 105% 100% 95% 90% 85% Raw: Normalized to total A/E Adjusted for change in Mix 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Roughly half of non-cancer mortality improvement is explained by claim mix 25

Mortality Improvement 115% 110% 105% Cancer: A to E Deaths by Calendar Year Raw: Normalized to total A/E Adjusted for change in Mix 100% 95% 90% 85% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Cancer mortality improvement is not affect by claim mix 26

Mortality Improvement 20% 15% Percent Corrected Death Rate Improvement since 1997 Cancer Other Total 10% 5% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 10% improvement in death rates equals about 1.5% lower total claim cost 27

Change in Definition 0.06 0.05 0.04 Recovery Rates vs Months since Change in Definition 0.03 0.02 0.01 0-3 -2-1 0 1 2 3 Some Change in Definition Recoveries spill over into the next two months Approximately 7% of claims recover = 8 times the normal rate 28

Change in Definition Pct of Claims Recov Bump Other Musculo 10.4% 11.0 Injury 15.2% 9.8 Back 11.8% 9.7 Respiratory 3.5% 8.9 Nervous Syst 4.8% 7.7 Ill-defined 6.6% 7.6 Diabetes 3.0% 7.1 Circulatory 3.5% 6.7 Maternity 16.9% 5.8 Other 3.9% 5.4 Digestive 4.1% 4.2 M&N 3.5% 3.8 Cancer 2.7% 3.6 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% Oth er Musculo Injury Pct of Claims Ill-defined Nervous Syst Respiratory Back Diabetes Circulato ry Recov Bump M&N Digestive Other Maternity Cancer 12 10 8 6 4 2 0 29

After the Change in Definition 6 6 5 5 4 4 3 3 2 2 1 1 0 0 Monthly Recovery Rates (per 1000) Monthly Recovery Rates (per 1000) Any Occ Any Occ Own Occ Own Occ Yr 4 Yr 5 Yr 6 Yr 7 Yr 8 Yr 9 Yr 10 Yr 11+ Yr 4 Yr 5 Yr 6 Yr 7 Yr 8 Yr 9 Yr 10 Yr 11+ Post 36 Month A to E s: Any Occ: 127% Own Occ: 91% 30

After the Change in Definition 3.5 Monthly Death Rates (per 1000) 3 2.5 2 1.5 1 0.5 0 Any Occ Own Occ Yr 4 Yr 5 Yr 6 Yr 7 Yr 8 Yr 9 Yr 10 Yr 11+ Post 36 Month A to E s: Any Occ: 94% Own Occ: 85% 31

Mental & Nervous Limits 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% M&N Closures at the Limit Date by Age and Gender Limit Maxouts F: Age>40 M: Age> 40 F: Age< 40 M: Age< 40 Total Percent Closed F: Age>40 79.2% M: Age> 40 76.6% F: Age< 40 86.6% M: Age< 40 85.1% 32

Impact of Elimination Period 180 160 140 120 100 80 60 40 20 Recovery Rates by Duration and EP EP-30 EP-90 EP-180 0 2Q 4Q 6Q 8Q 10Q 12Q Yr 5 Yr 7 Yr 9 Yr11+ 33

Impact of Elimination Period 40 35 30 25 20 15 10 5 Recovery Rates by Duration and EP EP-30 EP-90 EP-180 0 2Q 4Q 6Q 8Q 10Q 12Q Yr 5 Yr 7 Yr 9 Yr11+ 34

Impact of Elimination Period 6 5 4 Recovery Rates by Duration and EP EP-30 EP-90 EP-180 3 2 1 0 Yr 4 Yr 5 Yr 6 Yr 7 Yr 8 Yr 9 Yr10 Yr11+ EP Effects have worn off by the 4 th year of duration 35

Composition of Terminations 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Percent of Closures by Type and Duration Yr 1-3 Yr 4-6 Yr 7-10 Yr 11+ Settled Maxouts Limits Deaths Recov Recovery + Settlements = 94% of Deaths after year 6 36

Composition of Terminations 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Pct of Closures after 5 Years by Type and Attained Age <40 40-55 55-65 >65 Settled Maxouts Limits Deaths Recov 37

97-06 GLTD STUDY Observations There have been significant recovery improvements in the last few years There are significant variations by claim diagnosis Cancer claims have a very different pattern than non-cancer claims There has been some mortality improvement in recent years, with a greater improvement in cancer versus non-cancer claims Own occupation claims have a different long-term recovery pattern than any occ claims A significant percent of M&N claims survive past their limit date 38

97-06 GLTD STUDY Next Steps 1. Report writing 2. Table Vendor Selection 3. Table Development 4. New Valuation Table 2010? 39