Thank you for joining us – the webinar will start shortly Thinking about a buy out in 2020? Make sure you capture the data dividend @ClubVita #datadividend January 30 th , 2020 linkedin.com/company/club-vita 12 noon ET 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 Douglas Anderson Matt McDaniel Bobby Gentry Nate Luepke @ClubVita #datadividend linkedin.com/company/club-vita 3
What do the results of an auction process look like? 4
Advisor’s perspective on variation in prices Premium relative to “market” liability* Deal pricing varies based on many factors, including: • Demographics • Transaction size • State of issue • Benefit size Two thirds • Plan design • Investment portfolio of deals between • Insurer capacity • Seasonality 98% and 103% But, insurer’s view of longevity is a key driver, and may help explain outlier results both on low (93%) and high (105%) end *Source; Mercer, US: Data based on 64 US retiree-only deals from May 2016 to October 2019. Market liability defined using Mercer Yield Curve and most recent SOA mortality tables with collar adjustment. 5
So, how do you increase the likelihood of getting a price at the bottom of that range? 6
Three sources of gains More data Longer back Cleaner data fields history 7
Deep cleansing: What about tracing missing participants? 8
Three reasons why 2. Shorter 3. More 1. Reduce life underwriting cash flow) + outgoings + expectancy confidence (improve (reduce (lower risk reserves) premium) = $$$m benefits to bottom line 9
How much can ZIP codes change pricing? 10
Zooming in on diversity Society of Club Vita Club Vita Club Vita Actuaries 5-digit ZIP 9-digit ZIP 2020+ tables code model code model model Range of healthy male 2.7 years 5.4 years* 7.6 years 9.6 years? life expectancy - Pri-2012 at age 65 • • • • Rating factors High/low pension 5-digit ZIP code 9-digit ZIP code Adds salary • • amount Pension amount Pension amount amount as • • • Blue/white collar Blue/white collar Blue/white collar affluence measure Factors used One All All All at one time 11
Impact of adopting a ZIP code longevity model Impact of moving from RP06 to 9 digit ZIP US VitaCurves (both MP18 improvements) Increasing plan size (benefits in payment) Source: Club Vita, Zooming in on ZIP codes 12
With ZIP codes …. 13
Quantifying the health bias in SBRAP* Individual assumptions vs 'average' assumptions 8% 6% 4% 2% 0% -2% -4% -6% -8% Split of liability by pension amount Smallest pensions Largest pensions * Small benefit retirement annuity purchase Source: Club Vita; for a sample large US plan, the effect on deciles of liabilities of moving from plan’s assumed accounting 14 assumption to individual VitaCurves for baseline longevity
What happens if we use salary instead of annual pension? 15
Adding final salary as a covariate Changes in pricing or reserves from adding salary as a rating factor (Based on 115 different pension scheme portfolios of pensioners and dependants) 3% Loss of 1/3 of profits? Loss of 1/3 of profits? £63m loss (0.6%) in £11bn business £63m loss (0.6%) in £11bn business Underpriced 2% written written Introducing final 1% salaries changed No 0% impact portfolio valuation by -1% +/-2%. -2% We now get Overpriced salaries in 70% of UK -3% <£200m records. £67bn missed business? £67bn missed business? -4% Portfolios which could be written at Portfolios which could be written at average reduced premium of 1.4% average reduced premium of 1.4% Increasing portfolio size Increasing portfolio size -5% 16
Do participant options tell us something about the type of people they are? 17
Proxies for marital status Retirees who opt into a joint-life (with a contingent survivor pension) live longer + 1.3 years + 1 year Mix varies considerably between plans Male Pensioners Female Pensioners 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% . . . . . 10% 10% 0% 0% Plan A Plan B Plan C . . . . . Plan 1 Plan 2 Plan 3 Joint life Single life Joint life Single life 18
Are people taking lump sums random? In UK, Club Vita’s non - pensioner (deferred vested) data has enabled more competitive pricing for deferred annuities. 19
More back history 20
We can quickly see how far back is reliable … 21
Finessing trend assumptions By end 2020, Club Vita will provide socio-economic improvements for US retirees Life Expectancy at age 65 (men) 22 21 20 19 ~ higher incomes 18 17 ~ everyone else 16 15 ~ live in most deprived areas 14 Narrowing of socio- Strong, stable Resilience of 13 economic gap improvements for all higher socio-economics 12 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Comfortable Making-Do Hard-Pressed 22
So, what’s the overall data dividend? 23
Good data delivers several benefits Convenience Cleanliness for insurers + + Competition (extra) Confidence Covariates Cost reduction for plan sponsors? Conservatively, 2% or $10m on $500m 24
Any questions? Douglas Anderson Matt McDaniel Bobby Gentry Nate Luepke @ClubVita #datadividend linkedin.com/company/club-vita 25
Recommend
More recommend
