Thank you for joining us – the webinar will start shortly How to capture the pension risk transfer data dividend http://linkedin.com/company/club-vita/ February 26 th , 2020 12 noon EST @ClubVita #datadividend Club Vita LLP is an appointed representative of Hymans Robertson LLP, which is authorised and regulated by the Financial Conduct Authority and Licensed by the Institute and Faculty of Actuaries for a range of investment business activities.
Today’s aim A: Because cleaner, more complete data will save Q: Why are we bothering you money, creating a to create better data? high ROI. We would like to show you how … Plan Advisor sponsor 2
Today’s panel Richard Brown Benoît Labrosse Paul Forestell Tom Ault 3
What factors drive group annuity pricing? 4
Group annuity pricing is driven by several factors – many of which plan sponsors have control over Plenty of control Little control ▪ ▪ Transaction size Underlying population, demographics and location ▪ Timing of transaction ▪ Benefit provisions ▪ Tranching of transaction ▪ Insurers’ preferences ▪ Data cleanliness and ▪ completeness Insurers’ assets ▪ Assuris coverage Sheer luck ▪ Transaction day process ▪ Insurers’ confidence a ▪ Insurers not winning a specific transaction will occur deal ▪ Investment portfolio ▪ Insurers’ not meeting their ▪ Transparency of process annual objectives Source: Data extract from a sample of 30+ Morneau Shepell transactions over 2018 and 2019 5
So, how can better data help ensure plan sponsors receive the best possible price? 6
Three sources of gains More data Historical Cleaner data fields experience 7
Deep cleaning: Do you really know who your members are? 8
Key sources of membership uncertainty Unlocated deferred vested (DV) members • Basic data (e.g., date of birth) likely more unreliable than regular DVs Unnotified deaths • Reduced by comprehensive administrative practices, including existence checking programs Uncertainty regarding second lives • Unreliable form of pension data – could single lives actually be joint? • Have predeceased spouses been captured? 9
Three reasons why 1. Reduce 2. Shorter 3. More future life underwriting cash flow) + + payments expectancy confidence (improve (reduces (lower risk reserves) premium) = $$$m benefits to bottom line 10
What can your data tell you about the longevity of your members? 11
Tapping into the wealth of data in administrative records Age Gender Pension Job type Education Salary Union or non-union Industry Retiree vs. Marital Health at Postal code Survivor status retirement Province 12
What does data tell us about longevity expectations? Differences in period life expectancy at age 65 based on CV18 VitaCurves 9 Males 7.9 8 ears of Period Life Expectancy Females 7 6.5 6 5 4.6 3.7 4 3.6 3.4 3.4 3 Additional Y 2.0 2.0 1.9 2 1.4 1.3 1.3 1.2 1.2 1.0 1 0 Pensioner vs Joint life vs single Non-manual vs High vs low High vs low Longest vs Regular vs Most granular survivor life manual pension band salary band shortest lived disabled pensioner model longevity group pensioner Pen. Type Form Occupation Pension Salary Postal Code Ret. Health All Required data fields 13
Impact of capturing longevity factors 14
Lowering risk premium through better longevity insights Retirement Postal codes Occupation Salary health • Much better • Postal code has • High-level occupation • Disabled pensioners representation of become a standard information may be experience much affluence than PRT data field of limited value higher mortality pension • An address search can • Objective member- • Captured via plan- • Salary data isn’t increase coverage and specific data is the specific disability always readily accuracy most powerful pension benefits available but partial or and/or status (e.g., related data can still be helpful LTD) prior to retirement 15
What can historical experience tell us? 16
Assessing historical experience Projected vs. reported historical pensioner/survivor headcounts 4,500 4,000 3,500 Male Survivor 3,000 Female Survivor 2,500 Male Pensioner 2,000 Female Pensioner 1,500 Reported Counts 1,000 500 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 Year Ending December 31st Male mortality for your Plan compared to VitaBank 10.0% Crude mortality rate (log scale) 80-89 - VitaBank 80-89 - Your Plan 70-79 - VitaBank 70-79 - Your Plan 1.0% 60-69 - VitaBank 60-69 - Your Plan 0.1% 17 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Dos and don’ts of historical experience for PRT • Offer high quality historical experience data regardless of plan size • Verify that all historical deaths (and other exits) have been captured and Dos reconcile to past plan membership statistics • Provide as much data on longevity factors as possible • Rely heavily on valuation data for historical mortality experience • The quality of valuation-sourced mortality data is often poor Don’ts • Easy to miss deaths, particularly those soon after pension commencement and for original pensioner of survivors • Provide historical experience that hasn’t been fully validated/assessed 18
So, what’s the overall data dividend? 19
Good data delivers several benefits Convenience Cleanliness for insurers + + Competition (extra) Confidence Covariates Cost reduction for plan sponsors? Good data vs bad data could mean savings of 2% (or $10m on $500m) 20
Any questions? Richard Brown Benoît Labrosse Paul Forestell Tom Ault http://linkedin.com/company/club-vita/ @ClubVita #datadividend 21
Recommend
More recommend
