Study Large Motor Third Party Claims the Netherlands NRV, 3th October 2013 Christof Van Der Aa
Technical studies @ QBE Re Our technical expertise is a key factor to provide excellent service to our clients Technical studies, analysis reinsurance structure and tailor-made covers for our clients. Motor large claims studies are made on a regular basis in order to have a clear insight in the Motor claims. Since 1992 we have been making large claims studies for the Dutch market. Our large MTPL claims Study 2012 is based on a 14-year observation period and represents a 88% market share. Individual benchmarks 2
Overview presentation Recent topics BI claims Changes in Social Security legislation : AWBZ Change in claims capitalisation rate («Rekenrente») Study Large Motor Third Party Liability Claims Data sample large claims study Indexation method Largest claims in last known situation Extrapolation method Conclusions IBN(E)R analysis Largest claims in ultimate situation Number of claims Burning Cost Payment pattern Road Safety and BI claims frequency International perspective Summary and conclusions 3
Recent topics BI claims (Re)insurers need to have a clear insight in the components of large MTPL claims, the majority of these claims being Bodily Injury claims. Loss of income Third Party Aid Long term care Medical Expenses Changes in the Social Security legislation l : AWBZ Change in law as from 01.01.2013 Higher personal AWBZ contribution («Vermogenstoets») Up to 2200/month personal AWBZ contribution! No retroactive impact on claims Capitalisation rate (= «Rekenrente») changes Article recently published in «Verkeersrecht»:from 3% to 2% Implementation depends on decision Supreme Court ( Hoge Raad ) Retroactive impact on claims 4
Capitalisation rate comparison 4.50% Capitalisation rate comparison 4.00% 3.50% 3.00% 2.50% 2.00% 1.50% 1.00% 0.50% 0.00% NL B FR D CH UK P Ireland 5
Capitalisation rate : impact on MTPL XL claims (ex 1) 8% 1% 10% 2% 3% 13% 6% 28% 27% 2% for 20 years, after 3% Accident 01 01 2008 Victim: male, born 01 01 1983, married Employee, net income of 25.000 EUR per year Tetraplegic Settlement 01 01 2013 100% disabled for work Retirement at age of 67 Placement institution 01 01 2013 (AWBZ) 1% 0% 3% 1% 7% 0% 10% 2% 12% 6% Medical costs Past loss of income 2% 7% 0% 9% 2% 11% 6% 29% 29% Future loss of income Past domestic help Future domestic help Past loss of self-activation 30% 30% 1% 0% 3% 1% Future loss of self-activation Personal contribution AWBZ Moral damage 1% 0% 3% 1% Material damage Costs Legal interest 6
Capitalisation rate : impact on MTPL XL claims (ex 1) Loss amount 3% 2% FGU Priority 1 Priority 2 + 12% + 16% + 24% Accident 01 01 2008 Victim: male, born 01 01 1983, married Employee, net income of 25.000 EUR per year Tetraplegic Settlement 01 01 2013 100% disabled for work Retirement at age of 67 Placement institution 01 01 2013 (AWBZ) Priority 2 1.000.000 New IBNR s Priority 1 500.000 1.968.895 2.202.810 7
Capitalisation rate : impact on MTPL XL claims (ex 2) 2% 0% 1% 0% 8% 1% 8% 3% 3% 8% 13% Accident 01 01 2008 Victim: male, born 01 01 1983, married Employee, net income of 25.000 EUR per year Several fractures, severely handicapped Settlement 01 01 2013 100% disabled for work without domestic help Retirement at age of 67 3% 0% 1% 0% 7% 1% 7% 3% 2% for 20 years, after 3% 56% 7% 3% 1% 7% 3% 7% 1% 1% 2% 7% 60% 11% 0% 58% 12% Medical costs Past loss of income Future loss of income Past domestic help Future domestic help Past loss of self-activation Future loss of self-activation Moral damage Material damage Costs Legal interest 8
Capitalisation rate : impact on MTPL XL claims (ex 2) Loss amount 3% 2% FGU Priority 1 Priority 2 + 10% + 15% + 21% Accident 01 01 2008 Victim: male, born 01 01 1983, married Employee, net income of 25.000 EUR per year Several fractures, severely handicapped Settlement 01 01 2013 100% disabled for work without domestic help Retirement at age of 67 Priority 2 500.000 New IBNR s Priority 1 300.000 987.050 1.088.845 9
Data sample Our study is based on the large MTPL claims from accident years (AY) 1998-2010 up to and including development year (DY) 2011. Threshold depends on information provided by the participating companies MTPL premium market share max original thresholdh max indexed thresholdh group 1 2.015.698.697 87,8% 875.000 1.582.542 group 2 1.544.817.697 67,3% 500.000 904.310 market 2011 2.296.177.000 100,0% Analysis from ground-up and without index clause 10
Premium indexation Why Correct premiums for inflation and tariff adjustments How? Evolution of average MTPL premium (stable portfolio composition) Ideally specific company indexes are set up Market index used in our study 118 116 114 112 110 108 106 104 102 100 Premium Index MTPL 98 1998 2000 2002 2004 2006 2008 2010 2012 11
Claims indexation Claims indexation : two components Evolution wages index Super imposed inflation while the average MTPL claims cost increases faster than the wages index, this is due to: Changing claims culture Evolutions in medical technology Higher expenses for pain and suffering Better protection of weak road users Increasing life expectancy Evolution in «jurisprudence» 12
Claims indexation super imposed inflation 200 Average claim amount vs wages index 180 160 140 120 Index 100 80 60 40 20 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 index avg claim wages index Linear (index avg claim) 13
Index comparison Index Comparison premium index claims index frequency index 180 170 160 150 140 130 120 110 100 90 80 70 60 50 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Premium index Claims index Frequency index 14
Largest claims in last known situation (LKS) Original Indexed Year Claim amount Year Claim amount 2009 5.558.903 2009 6.416.928 2009 4.535.000 2009 5.260.403 2008 3.751.240 2000 4.639.896 2000 2.938.097 1998 4.602.911 2002 2.879.052 2008 4.537.351 2009 2.860.260 1999 4.473.103 2003 2.750.000 2002 4.118.806 2002 2709526 2.709.526 2001 4115413 4.115.413 2006 2.660.000 2002 4.084.928 2001 2.623.108 2000 4.023.428 1998 2.593.359 2000 4.008.858 1999 2.583.817 2003 4.006.847 2001 2.540.921 1998 3.870.799 2008 2.537.898 2001 3.853.037 2007 2.525.000 2000 3.805.979 2000 2.500.001 2000 3.735.786 2009 2.500.000 1998 3.727.499 2000 2.478.552 2000 3.722.338 2005 2.474.377 1998 3.668.969 2009 2.399.998 2000 3.656.578 Up to 2007 (before the 5th MTPL Directive) the old policy limits were still in force, which means that some claims were probably capped and hence we underestimate the claims burden. 15
Extrapolation : Chain Ladder methodology Loss amount IBNER 1,4 M 1M IBNR 1,2 M Threshold 09M 0,9 0,8 M 0,5 M Development year 1 2 3 4 5 We make a separate Chain Ladder calculation l for the triangles with claims amounts (IBNER) and number of claims (IBNR). 16
Conclusions IBN(E)R analysis IBNER analysis The (from a reinsurers point of view) known claims are slightly underreserved. Bigger claims are better reserved IBNR analysis Up to 1.5M : larger claims are subject to a bigger IBNR effect, as from 1.5M the IBNR effect stabilizes. Volatility for higher thresholds The IBNR has been analysed from a reinsurers point of view : possibly the IBNR effect originates from a reserve adjustment (IBNER) made by the insurer. 17
Largest claims in ultimate situation Indexed LKS Indexed ultimate situation Year Claim amount Year Claim amount 2009 6.416.928 2009 6.964.416 2009 5.260.403 2009 5.709.218 2000 4.639.896 2008 4.830.178 1998 4.602.911 2000 4.639.896 2008 4.537.351 1998 4.602.911 1999 4.473.103 1999 4.473.103 2002 4.118.806 2001 4.248.514 2001 4.115.413 2002 4.137.169 2002 4.084.928 2002 4.118.806 2000 4.023.428 2003 4.111.745 2000 4.008.858 2000 4.015.799 2003 4.006.847 2000 4.001.257 1998 3.870.799 2001 3.977.653 2001 3.853.037 1998 3.870.799 2000 3.805.979 2000 3.805.979 2000 3.735.786 2000 3.728.703 1998 3.727.499 1998 3.727.499 2000 3.722.338 2000 3.722.338 1998 3.668.969 1998 3.668.969 2000 3.656.578 2000 3.656.578 18
Largest claims in ultimate situation Largest claims before and after extrapolation (IBNER) 100 90 80 70 60 50 40 30 20 10 0 2M-3M 3M-4M 4M-5M 5M-6M 6M-7M nb of original claims in LKS nb of indexed claims in LKS nb of indexed claims in ultimate situation 19
Largest claims in ultimate situation 5,000,000 Avg 5 largest indexed claims in ultimate situation 4,000,000 3,000,000 2,000,000 1,000,000 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Average 5 largest claims 20
Number of claims in ultimate situation per AY 50 Number of indexed claims per AY 40 30 20 10 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Threshold 1 M Threshold 1.2 M Threshold 1.5 M 15 Number of indexed claims per AY 10 5 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Threshold 1.8 M Threshold 2 M Threshold 2.5 M 21
Number of claims in ultimate situation per AY Contrary to what has been observed during our previous studies the number of large MTPL claims is no longer decreasing. This corresponds with the data published by SWOV (Stichting Wetenschappelijk Onderzoek Verkeersveiligheid). Road casualties 1,600 Number of road victims per year Severely injured 14,000 1,400 1,200 1,000 800 600 400 200 12,000 10,000 8,000 6,000 4,000 2,000 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 0 Road casualties Severely Injured 22
Number of claims in ultimate situation Number of MTPL claims in ultimate situation in excess of a number of thresholds is shown. Weighted average for the period 1998-2009 for a premium volume of 1000M. 40 Average number of claims (ultimate) 1998-2009 35 30 25 20 15 10 5 0 1M 11M 1.1 12M 1.2 13M 1.3 14M 1.4 15M 1.5 16M 1.6 17M 1.7 18M 1.8 19M 1.9 2M 25M 2.5 3M Number of claims Poly. (Number of claims) 23
Evolution number of claims in ultimate situation 40 Evolution number of claims 35 30 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Development year 10 Evolution number of claims Threshold h 1 M Threshold h ld12m 1.2 Threshold h ld15m 1.5 8 6 4 2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Development year Threshold 1.8M Threshold 2 M Threshold 2.5 M 24
Number of claims in ultimate situation per AY It takes a considerable number of years before the reinsurer has a good view on the expected number of claims. Development pattern number of claims 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Threshold 1 M Threshold 1.2 M Threshold 1.5 M Development year Threshold 1.8M Threshold 2 M Threshold 2.5 M 25
Burning Cost and number of claims The number of claims in ultimate situation (lambda) and the Burning Cost for a number of layers have been analysed. Premium volume = 1000M ; Index clause date of payment margin 10%. 1.6% Burning cost XOL layers with limit 5M 1.4% 1.2% 1.0% 0.8% 0.6% 0.4% 0.2% 00% 0.0% 1.0 M 1.2 M 1.4 M 1.6 M 1.8 M 2.0 M 2.2 M 2.4 M 2.6 M BC Expon. (BC) 26
Comparison Burning Costs We compared the Burning Cost (BC) with the BC of previous studies for a number of layers with limit 2.5M Upward trend confirmed by our latest t study Priority Limit BC 2004 BC 2006 BC 2008 BC 2010 BC 2012 1.000.000 2.500.000 0,530% 0,836% 1,197% 1,192% 1,320% 1.100.000 2.500.000 0,437% 0,698% 0,967% 0,944% 1,138% 1.200.000 2.500.000 0,365% 0,588% 0,798% 0,755% 0,938% 1.300.000 2.500.000 0,307% 0,498% 0,667% 0,608% 0,778% 1.400.000 2.500.000 0,257% 0,422% 0,558% 0,492% 0,590% 1.500.000 2.500.000 0,212% 0,353% 0,465% 0,398% 0,472% 1.600.000 2.500.000 0,176% 0,300% 0,385% 0,322% 0,476% 1.700.000 2.500.000 0,145% 0,252% 0,317% 0,259% 0,400% 1.800.000 2.500.000 0,118% 0,209% 0,257% 0,207% 0,287% 1.900.000 000 2.500.000 000 0,092% 092% 0,169% 0,207% 0,163% 0,252% 2.000.000 2.500.000 0,070% 0,133% 0,162% 0,125% 0,174% 27
Payment pattern from ground up Payments related to the ultimate claims amount (from ground up) per DY. Average payment pattern: After 5 years 25% is paid After 10 years 55-60% is paid After 14 years 75-80% is paid Final claims settlement after 18-20 years Claims size has no material impact on settlement pace Payment pattern 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Threshold 1 M Threshold 1,2 M Development year Threshold 1,5 M Threshold 1,8 M Threshold 2 M Threshold 2.5 M 28
Payment pattern for the reinsurer The reinsurers payments start later but increase gradually while the payment pattern FGU is linear. The claims size has a material impact on the settlement pace of the reinsurer : while for a 1M claim the reinsurer paid 15% after ten years, he only paid 4,1% after 10 years in case of 2M claim Payment pattern (reinsurer) 60% 55% 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Threshold 1 M Threshold 1,2 M Threshold 1,5 M Threshold 1,8 M Threshold 2 M Threshold 2.5 M 29
Road Safety figures and MTPL quotations Claims frequency index MTPL XL = number of BI claims / number of Motor vehicles Based on official i statistics ti ti we calculate l the BI claims evolution : as we don t dispose for the Netherlands of the total number of BI claims, we made an approximation by using the evolution of the road casualties. In most of our markets we observe a decreasing BI claims frequency. index 110 100 90 80 70 60 50 40 30 20 10 0 Comparison frequency indexes 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Duitsland België Nederland VK Ierland Spanje 30
BC Comparison priority limit NL B FR D SP UK P I 1.000.000 5.000.000 2,19% 6,78% 0,51% 4,75% 0,54% 2.000.000 5.000.000 0,57% 3,61% 3,83% 1,92% 0,06% 2,70% 0,13% 0,46% 4.50% BC Comparison 4.00% 3.50% 3.00% 2.50% 2.00% 1.50% 1.00% 0.50% 0.00% 00% NL B FR D SP UK P I Layer 3M xs 2M 31
Summary and conclusions Recent topics BI claims : material impact on XL claims! AWBZ : changes in law as from 01.01.2013, no retroactive impact on claims Decrease capitalisation rate : decision Supreme Court, retroactive impact Largest MTPL claims : 3.5M - 7.0M, development > 15 years Claims reservation (from a reinsurers point of view): IBNER effect is moderate Bigger claims are better reserved IBNR (number of claims) remains key factor Largest claims tend to increase as from 2007? Impact 5th Directive MTPL Before 2007 : breach of policy limits? Increased burning cost Premium index Frequency index Claims index Payment pattern FGU almost independent of claims size 32
Any questions? I will be happy to answer them! 33