The Product Development Section Presents
Annuity‐Based Solutions: Securing the “Golden Years” Seminar
May 9, 2018 | Baltimore Marriott Waterfront | Baltimore, MD
Presenters: Christopher Allen Bartak, FSA, MAAA
Nicholas E. Carbo, FSA, MAAA Jennifer L. Healy, FSA, MAAA
Justin Kahn Miranda McCarthy
Douglas L. Robbins, FSA, MAAA Peter Tian
SOA Antitrust Compliance Guidelines SOA Presentation Disclaimer
© Oliver Wyman
Fixed Annuity and Fixed Indexed Annuity Market and Product Trends Annuity-Based Solutions: Securing the “Golden Years” Seminar Baltimore, May 9, 2018
Nicholas Carbo, FSA, MAAA
1 © Oliver Wyman
Agenda
1 Backdrop for this seminar
2 Market and product perspectives
2 © Oliver Wyman
US population by age (MM of people) 2015 and 2030E
Backdrop for this seminar Need for guaranteed retirement income has never been greater and will continue to grow
50
10
0
20
40
30
80+
42 34
12
50 – 59 70 – 79 30 – 39
44 40 35
60 – 69
46 41
20
40 – 49
40
+14% 49
+64%
+72% -9%
20
+16% +13%
2030E 2015
Source: US Census Data
In-Retirement Average issue age of Fixed Annuities: 72
Pre-Retirement Average age: 30s to 40s Products offered: Group Retirement Plans, mutual funds platform, financial advice
Near/At-Retirement Average issue age of Variable and Index Annuities: 64
• Baby boomers will continue to enter retirement through 2030
• Average US life expectancy at age 65 increased by 3 years between 1975 and 2010, and is projected to continue to grow
• Explosion in population of retirees in income drawdown
3 © Oliver Wyman
US Personal Financial Assets – Landscape Overview; amounts in $BN
20%
100%
10%
0%
30%
40%
90%
60%
70%
50%
80%
DB
457
401k
DC
403b
2,032
Annuities
52,313
TSP
Non-qualified
IRA
Other
Retail
6,965
Private DB
8,249
State and Local
DB
Federal DB
Market
US personal financial assets Annuities are a minor fraction of US personal financial assets
Corporate Source: Investment Company Institute, US Retirement Market Statistics, Q3, 2016; Federal Reserve Statistical Release Z.1 Financial Accounts of the United States, Third Quarter 2016 Annuities include all fixed and variable annuity reserves at life insurance companies less annuities held by IRAs, 403(b) plans, 457 plans, and private pension
4 © Oliver Wyman
Retail annuity sales 2008 - 2017 RFIAs and fixed indexed annuities are the bright spots in the retail annuity market
Book value
SPIA
DIA
Mar
ket r
isk
for p
olic
yhol
der
Source: LIMRA
Low
er
Hig
her
2013 2012 2011 2010 2009 2008 2017 2014 2015 2016
Variable
Registered indexed
MVA
Fixed indexed
5 © Oliver Wyman
$238 $220 $230 $237 $236 $222 $204
0%
10%
20%
30%
40%
50%
60%
70%
$0
$50
$100
$150
$200
$250
$300
2011 2012 2013 2014 2015 2016 2017
Inco
me
elec
tion
(% o
f sal
es)
Sale
s ($
B)
Accumulation (FA/FIA/VA) VA GLWB FIA GLWB Income annuity % of income sales
Income vs. accumulation sales Income sales are decreasing!
VA GLWB sales declined sharply due to de-risking. Until 2016, this was partially compensated by rising FIA GLWB sales. The recent FIA GLWB slowdown is caused by DOL/distribution changes.
Source: LIMRA
6 © Oliver Wyman
Why are income annuity sales decreasing
The $MM question…
7 © Oliver Wyman
… This remains a puzzle for the industry!
Questions we will attempt to answer: • Why are FIAs successful? • Why have RFIAs seen so much growth? • Why are income annuity sales stagnant? • Will fee-based annuities break through?
8 © Oliver Wyman
Why are FIAs successful
9 © Oliver Wyman
Expanding FIA market The FIA market benefited from new entrants and wider acceptance with distributors
Rank Company Sales ($B) 1 Allianz Life 6.3 2 Aviva USA 4.5 3 American Equity 4.4 4 Great American 1.8 5 North American 1.7 6 Lincoln National 1.6 7 Midland National 1.5 8 Jackson National 1.5 9 ING 1.4 10 Security Benefit Life 0.9
Source: Wink’s Sales & Market Report
2011 FIA Market : $32.2B 2017 FIA Market: $54.0B
Rank Company Sales ($B) 1 Allianz Life 7.5 2 Athene USA 5.0 3 Nationwide 4.6 4 American Equity 4.1 5 Great American 3.5 6 AIG 3.4 7 Midland National 2.1 8 Global Atlantic 2.1 9 Symetra 2.0 10 North American 1.8
Top 3 market share 47% Top 10 market share 78%
Top 3 market share 29% Top 10 market share 65%
10 © Oliver Wyman
Competitive landscape (2005) The FIA market was previously heavily concentrated in “niche players” focusing on the independent channel
Banks and broker dealers Independent and banks Independent organizations
Rec
ent
Acqu
irers
U
S St
ock
Com
pani
es
Fore
ign
Subs
idia
ries
Mut
ual
Com
pani
es
Ow
ners
hip
Stru
ctur
e
Market share and dominant distribution channels by ownership structure and credit rating
Priv
atel
y H
eld
Size of bubbles represents 2005 market share
Primary distribution channels
Lower Higher Credit Rating
A++ A A- B++ B+ B A+
Source: Advantage Compendium and Oliver Wyman Research
Allianz Life
AmerUs (Athene/Aviva block)
Old Mutual (now Fidelity& Guaranty)
Midland National
Equi Trust
Jackson National
32.3%
9.9%
All other companies
8.8%
15.2%
American Equity ING (now VOYA)
Jefferson Pilot (now Lincoln)
Sun Life (now Delaware Life)
8.7% 7.5%
11 © Oliver Wyman
Competitive landscape (2013) In 2013, the market was already significantly broadened
Banks and broker dealers Independent and banks Independent organizations
Rec
ent
Acqu
irers
U
S St
ock
Com
pani
es
Fore
ign
Subs
idia
ries
Mut
ual
Com
pani
es
Ow
ners
hip
Stru
ctur
e
Market share and dominant distribution channels by ownership structure and credit rating
Priv
atel
y H
eld
Size of bubbles represents 2013 market share
Primary distribution channels
Lower Higher Credit Rating
A++ A A- B++ B+ B A+
Source: Wink’s Sales & Market Report and Oliver Wyman Research
Allianz Life
15.5%
Career
National Life
10.4% American Equity
Sagicor Life
CNO companies Phoenix
Midland National
North American
Pacific Life
11.7%
7.1%
7.9%
Security Benefit
Fidelity & Guaranty
Equi Trust Athene Lincoln Benefit Life
All other companies
Forethought
Genworth
Nationwide Western & Southern
OneAmerica
Jackson National
Great American
Lincoln
Protective Voya AIG
National Western Symetra
12 © Oliver Wyman
Competitive landscape (2017) Today’s FIA market has a broad range of players with substantial representation outside of the independent channel
Banks and broker dealers Independent and banks Independent organizations
Rec
ent
Acqu
irers
U
S St
ock
Com
pani
es
Fore
ign
Subs
idia
ries
Mut
ual
Com
pani
es
Ow
ners
hip
Stru
ctur
e
Market share and dominant distribution channels by ownership structure and credit rating
Priv
atel
y H
eld
Career
Size of bubbles represents 2017 market share
Primary distribution channels
Lower Higher Credit Rating
A++ A A- B++ B+ B A+
Source: LIMRA and Oliver Wyman Research
All distribution channels Allianz Life
Athene USA
Nationwide
American Equity
Midland National
Fidelity & Guaranty
North American Security Benefit
Equi Trust
National Life
Jackson National
Americo
Phoenix
Voya Great American Symetra
Pacific Life
Protective
AIG
Global Atlantic
Lincoln
Mass Mutual
13.9%
9.0%
8.5%
7.6% 6.5%
Small mutual companies Reliance Standard
Delaware Life
American National
Western & Southern
Bankers Life & Casualty
Ohio National
All other companies
9.2%
13 © Oliver Wyman
FIA distribution changes FIA sales have grown significantly due to penetration in institutional channels
Q4 2011 Q4 2017
Source: Wink’s Sales & Market Report
Independent agents, 89.4%
Bank, 5.1%
B/D, 2.6% Career, 2.9%
Independent agents, 55.9%
Bank, 16.9%
B/D, 20.2%
Career, 7.0%
14 © Oliver Wyman
The FIA market has grown partly due to large M&A capital inflows Acquirer/Seller (date)
Transaction type: IPO Japanese insurers Alternative buyers
2010 2011 2012 2013 2014 2015 2016 2017
15 © Oliver Wyman
Why have RFIAs seen so much growth
16 © Oliver Wyman
Registered fixed index annuity product launches Product launches have been accelerating recently with top competitors launching version 2.0 of their products
2010 2011 2012 2013 2014 2015 2016 2017
Oct 2010 • AXA introduces
Structured Capital Strategies
May 2013 • MetLife introduces
Shield Level Selector
August 2013 • CUNA introduces
Member’s Zone Annuity
September 2013 • Allianz Life
introduces Index Advantage
January 2015 • Voya introduces
Potential Plus
October 2015 • Voya closes
Potential Plus
August 2017 • Brighthouse
introduces Shield Level Select Access (Fee based)
August 2016 • CUNA introduces
Member’s Horizon Annuity
Feb 2017 • AXA Structured
Capital Strategies Plus filed
In addition Lincoln plans to launch in May 2018
2018
February 2018 • Great West
introduces Capital Choice
March 2018 • Great American
introduces Index Frontier 7
17 © Oliver Wyman
Registered fixed index account options compared to VA and FIA RFIA products offer two types of index accounts which allow for a higher cap than an FIA capped account Variable annuity
AV
retu
rn
Index value Index value
Index value
• Provides direct exposure to equity markets – i.e., no capped upside or floored downside
• Effectively a long call spread on the index – i.e., − Long call to capture
upside − Short call to limit net
debit
• Effectively a long “seagull” option structure – i.e., − Call spread similar to FIA − Short put to finance a
more attractive call spread
AV
retu
rn
Fixed index annuity (cap design)
AV
retu
rn
Buffer design
Index value
• Effectively a ”collar” option structure – i.e., − Call spread similar to FIA − Short put spread to
finance a more attractive call spread
AV
retu
rn
Floor design
18 © Oliver Wyman
Motivations for introducing RFIAs
Capital requirements • Diversification with VA • Principles-based
Market demand • Growing sales • Accumulation story
Innovation • Higher cap rates • New product segment
Risk management • Diversification with VA • Low interest rates D
19 © Oliver Wyman
Why are income annuities sales stagnant
20 © Oliver Wyman
Income annuities FIA GLWB VA GLWB
Product features Income • Guaranteed income
• Various payout structures available
• Gender specific rates
• Guaranteed income
• Single and joint life options
• Unisex rates
• Guaranteed income
• Single and joint life options
• Unisex rates
Cash value • Cash value options limited • GMSV (Guaranteed minimum surrender value)
• Modest upside potential
• Full participation in market returns less fees
• Subject to downside risk
Death benefits • Various death benefit options or no death benefit
• Account value returned on death
• GMSV guaranteed upon death
• Account value returned on death
• Most VA GLWB have no guaranteed death benefits
• GMDB typically available
Pricing regime Primary pricing methodology
• Real world • Real world • Risk neutral
Customer options for guaranteed income There are three major guaranteed income products in the market
1 2 3
21 © Oliver Wyman
Product features analysis We will sequentially examine product features from a consumer view
Income guarantees
Cash value
Death benefits
1 2 3
22 © Oliver Wyman
300
400
500
600
700
800
900
1,000
0 1 2 3 4 5 6 7 8 9 10
VA GLWB Income annuity life only Income annuity cash refund FIA GLWB
Gua
rant
eed
mon
thly
inco
me
Deferral period
Guaranteed income rates by product and deferral
1.Income illustrated for a single premium of $100k
2. Rates based on common products in the marketplace
60 year old male guaranteed monthly income by deferral period FIA GLWB guaranteed
income exceeds income annuity life only
FIA GLWB guaranteed income exceeds income annuity cash refund
FIA GLWB income can exceed pure income annuities
23 © Oliver Wyman
10 years from issue
5 years from issue At issue
Cash value (immediate income scenario) FIA provide stronger cash value guarantees but with very limited upside relative to VA
$0
$50
$100
$150
$200
FIA GLWB VA GLWB Income annuity cash refund
$0
$50
$100
$150
$200
$250
FIA GLWB VA GLWB Income annuity cashrefund
$0
$25
$50
$75
$100
FIA GLWB VA GLWB Income annuity cash refund
Cas
h va
lue
($k)
Cas
h va
lue
rang
e ($
k)
Cas
h va
lue
rang
e ($
k)
1.Income illustrated for a single premium of $100k
2.Range based on S&P returns from 1/1/87- 1/1/2017, using the best and worst results for the number of years of experience
His
toric
al
expe
rienc
e H
isto
rical
ex
perie
nce
60 year old male with income starting at issue
1FIA GLWB has paid out $6k more than VA GLWB
1FIA GLWB has paid out $12k more than VA GLWB Product Cash value commentary
Income annuity cash refund No cash value offered, cash refund is only a death benefit
FIA GLWB Narrow range of possible cash values, with guarantee
VA GLWB Wide range of possible cash values
24 © Oliver Wyman
$0
$25
$50
$75
$100
FIA GLWB VA GLWB Income annuity cash refund
10 years from issue
5 years from issue At issue
Dea
th b
enef
it ($
k)
Death benefit (immediate income scenario) FIA offers strong guaranteed death benefit; assuming a GMDB is not present, VA have significant downside risk
$0
$50
$100
$150
$200
FIA GLWB VA GLWB Income annuity cash refund
$0
$50
$100
$150
$200
$250
FIA GLWB VA GLWB Income annuity cashrefund
Dea
th b
enef
it ra
nge
($k)
1.Income illustrated for a single premium of $100k
2.Range based on S&P returns from 1/1/87- 1/1/2017, using the best and worst results for the number of years of experience
60 year old male with income starting at issue
His
toric
al
expe
rienc
e H
isto
rical
ex
perie
nce
Product Death benefits commentary
Income annuity cash refund Death benefit slightly higher than FIA guarantees
FIA GLWB Narrow range of possible death benefits, with guarantee
VA GLWB Wide range of possible death benefits
Dea
th b
enef
it ra
nge
($k)
25 © Oliver Wyman
Income annuities FIA GLWB VA GLWB
Income 1 • Life only income guarantees are very strong 2 • The income guarantee is
very strong 3 • The income guarantee is weakest of the three options
Cash value (strong equity market) 3 • Limited or no cash value 2 • Modest upside potential 1 • The cash value is invested in
the market less fees
Cash value (weak equity market) 3 • Limited or no cash value 1 • Minimum of GMSV 2 • The cash value is invested in
the market less fees
Death benefits 2 • Various death benefit guarantees can be purchased 1 • GMSV paid on death
• Modest upside potential 3 • The death benefit value is
invested in the market less fees (assuming no GMDB rider)
Rankings of the three benefits FIA GLWB offers strong guaranteed income while preserving flexibility
Industry sales indicate that the flexibility of a GLWB is valued by consumers
1 2 3
26 © Oliver Wyman
60 year old male starting income at issue
1.0% 1.2%
3.7% 4.8%
9.7% 10.2%
1.0%
0%
2%
4%
6%
8%
10%
12%
Age 70 / Year 10 surrender Age 80 / Year 20 death Age 90 / Year 30 death Age 100 / Year 40 death
Client IRR Company IRR
Rat
e of
retu
rn
Time of policyholder exit
Client and company IRR for an FIA with GLWB starting income immediately Timing of surrender or death determines the rate of return achieved by the client and the company
1.Client IRR represents pre-tax return on invested cash flow (premium)
2.Company IRR represents the internal rate of return on after-tax distributable earnings
27 © Oliver Wyman
Will fee-based annuities finally break through
28 © Oliver Wyman
Q1 2016 Q2 2016 Q3 2016 Q4 2016 Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018
December 2015
• Midland National – MNL Prosper 5
Fee-based FIA product launches
August 2016
• Great American – Index Protector 7
February 2017
• Allianz – Retirement Foundation
ADV Annuity
February 2017
• Lincoln – Lincoln Core Capital
May 2017
• Pacific Life – Pacific Index Advisory
July 2017
• Symetra – Advisory Edge – Income Advisory Edge
July 2017
• Nationwide – Nationwide Summit
March 2018
• Jackson National – Market protector advisory
29 © Oliver Wyman
• The customer will likely receive a higher cap rate • The surrender charges will likely be lower or eliminated
• The company will not be paying an upfront commission, resulting in less strain to the insurance company
• Agents that only sell fee-based may be willing to look at FIA as a choice for their customers
• The uncertainty around the DOL fiduciary rule may increase demand for these products
Higher customer value
Less strain to insurer
Market opportunity
Benefits of fee-based FIAs Fee-based chassis allows for more competitive features
30 © Oliver Wyman
Key takeaways Broadening the reach of retail income annuities remains a difficult problem. The following successes give hints on what could turn income annuities around:
1 FIA sales expanded through broader consumer acceptance and distribution
2 FIA GLWB sale grew due to flexibility combined with strong guarantees (echoes earlier VA GLBs)
3 RFIA accumulation story appeal with distribution consumers
Potential success factors drawn from above: distribution capacity, consumer acceptance, flexibility and strong guarantees.
Session Q&A
© Oliver Wyman
DOL FIDUCIARY RULE IMPACTS
MAY 9, 2018
CONFIDENTIALITY Our clients’ industries are extremely competitive, and the maintenance of confidentiality with respect to our clients’ plans and data is critical. Oliver Wyman rigorously applies internal confidentiality practices to protect the confidentiality of all client information.
Similarly, our industry is very competitive. We view our approaches and insights as proprietary and therefore look to our clients to protect our interests in our proposals, presentations, methodologies and analytical techniques. Under no circumstances should this material be shared with any third party without the prior written consent of Oliver Wyman.
© Oliver Wyman
2© Oliver Wyman
• Motivation for regulatory change• Proposed changes under the DOL rule• DOL rule impacts• Summary of SEC Regulation Best Interest
Agenda
3© Oliver Wyman
DB
Non-qualified money
Qualified money Non-qualified money
IRAs
DC
Fiduciary standard currently applies
Institutional
Motivation for change Concern was that the advice standard for IRAs and non-qualified money was not high enough
Retail
No consistent standard
4© Oliver Wyman
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1974 1979 1984 1989 1994 1999 2004 2009 2014
% o
f ret
irem
ent a
sset
s
Year
IRAs
DC
Private DB
State & Local DB
Federal DB
Annuities
40-year 10-year
10% 4%
10% 5%
10% 4%
8% 4%
12% 7%
25% 8%
Motivation for changeIRAs have been the fastest-growing class of retirement assets with a majority of growth coming from rollovers
Annual growth rate
5© Oliver Wyman
DOL rule overviewThe DOL rule moved toward a fiduciary standard and the use of PTEs to continue receiving commission
A
B
Becoming an ERISA fiduciary would have required a “Best Interest” rather than a “Suitability” standard
Becoming a fiduciary would have prevented advisors from receiving payments from third parties unless they use a PTE
6© Oliver Wyman
2016 2017 2018
Apr May Jun Jul Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul
April 6, 2016DOL final ruling published
April 10, 2017Started implementing best interest standards
Jan. 1, 2018Original BICE implementation date
12-months for implementation
Transition period – BICE partially applies BICE would have fully applied
Fiduciary rule/best interest standard applies(including PTE 84-24)
DOL rule timelineThe rule had a staggered implementation, with the most significant provisions never being implemented
Nov 27, 2017DOL delays BICE requirement
Mar. 15, 2018DOL rule struck down by circuit court
7© Oliver Wyman
DOL rule impactsWe expected the rule to have broad impacts across the insurance industry, for both manufacturers and distributors
5. Product Impacts
Product and distribution
2. IMO 3. LevelizedComp1. Volumes 4. Robo-advisors
Compliance
6. Policyholder contact
7. Non commission compensation
8. Selling agreement
9. Orphans and grandfathering
8© Oliver Wyman
DOL rule impactsAlthough the DOL rule is not in effect, we have seen some industry impacts
2. Advice Model 3. Product Shelf 4. Compensation
5. Compliance & Controls
6. Technology & Operations
1. Business Model
More impact Less impact
9© Oliver Wyman
SEC rule elements
# Proposal Applicability Description
1 Proposed Rule Regulation Best Interest
Broker-dealers Best interests for recommendations of securities transaction or strategy involving securities for a retail customer
2 Proposed Form CRS Relationship Summary
Broker-dealers and IAs 1. For IAs, plain English disclosure to be included in ADV
2. New plain English disclosure requirement for BDs too, to be defined
3. Both would have to disclose registration status in certain customer communications
4. Broker-dealer registered representative cannot use the word adviser
3 Proposed Commission Guidance
IAs Clarifies fiduciary duty standards for IAs
10© Oliver Wyman
Insurance suitabilityFINRA suitability
Standards comparisonThe DOL rule and Reg. BI impact different segments due to jurisdictional differences
Employer-sponsored plans DC & DB
Individual accounts
Mutual funds Annuities
ERIS
A pl
ans
& IR
AsN
on-E
RIS
A pl
ans
&
non-
qual
ified
fund
s
ERISA fiduciary
DOL fiduciary
SEC best interest
Broker-dealers and investment advisors Insurance only agents
11© Oliver Wyman
Primary differences between DOL and SEC rule
3 No private right of action
4Appears to have limitedimpact on insurance-only agents
1 Best interest is not clearly defined
2Simplified disclosure requirements
12© Oliver Wyman
Closing remarks
• Future of the DOL rule
• Expectations for the SEC rule
• Industry and customer impacts
QUALIFICATIONS, ASSUMPTIONS AND
LIMITING CONDITIONS
This report is for the exclusive use of the Oliver Wyman client named herein. This report is not intended for general circulation or publication, nor is it to be reproduced, quoted or distributed for any purpose without the prior written permission of Oliver Wyman. There are no third party beneficiaries with respect to this report, and Oliver Wyman does not accept any liability to any third party.
Information furnished by others, upon which all or portions of this report are based, is believed to be reliable but has not been independently verified, unless otherwise expressly indicated. Public information and industry and statistical data are from sources we deem to be reliable; however, we make no representation as to the accuracy or completeness of such information. The findings contained in this report may contain predictions based on current data and historical trends. Any such predictions are subject to inherent risks and uncertainties. Oliver Wyman accepts no responsibility for actual results or future events.
The opinions expressed in this report are valid only for the purpose stated herein and as of the date of this report. No obligation is assumed to revise this report to reflect changes, events or conditions, which occur subsequent to the date hereof.
All decisions in connection with the implementation or use of advice or recommendations contained in this report are the soleresponsibility of the client. This report does not represent investment advice nor does it provide an opinion regarding the fairness of any transaction to any and all parties.
Income Annuity Trends
S O C I E T Y O F A C T U A R I E S
G O L D E N Y E A R S S E M I N A R
FOR FINANCIAL PROFESSIONAL USE ONLY. NOT FOR USE WITH THE PUBLIC.
CRN202004-229527
2FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Today’s discussion
• The New Retirement: According to the Boomer
• Costs of Living in Retirement– expectations versus experience– The Wants– The Needs– The Surprises
• Sources of Retirement Income – The emergence of DIY
• Living Longer
• Optimizing portfolio with guaranteed income
3
Retirement Snapshot (1,2)
Nearly 50 million retirees in the US
(2015)
$24 TRILLION in net worth
1 LIMRA, Secure Retirement Institute, “Sources of Retirement Income, A Study of Retirees,” 2017 2 LIMRA Secure Retirement Institute, “2016 Fact Book on Retirement Income”
$1.3 Trillion in retirement income
(2014)
Coming from Social Security and Pensions
4FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
The Retiree Market
The retirement market will grow for decades
2015Americans
Turning 65 per day 10,000
2015Number of
Retirees 49.5 million
2014Retirement
Income Market $13 trillion*
202111,000/day
202566 million
2025$25 trillion
2035 78 million
205012,000/day
*The retirement income market size estimate reflects consumers aged 25 or over, and households with assets between $50k and $4.9 million.
LIMRA Secure Retirement Institute analysis of U.S. Census Bureau’s Current Population Survey, March 2015 Supplement, National Population Projection, 2015, and 2013 Survey of Consumer Finance, Federal Reserve Board, 2014. As cited in LIMRA Fact Book on Retirement Income, 2016, A Review of Trends and Activity in the Retirement Income Market
5FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
A New Brand of Retirement
Breaking all the Rules
6FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Their plans for retiring
26%
66%54%
42%
28%38% 39%
Plan to stop workand retire at aspecific age orsavings amount
Plan to or alreadyare working pastage 65 or don't
plan to retire
Plan to keepworking after
retirement
Proactivelyupdating skills tokeep working if
needed
Have a backupplan for retirement
income if they can'twork
Expect SocialSecurity to be
primary source ofincome
Expect self-fundedaccounts such as401(k), IRAs and
other savings to beprimary source of
income
Transamerica Center for Retirement Studies, “Wishful Thinking or Within Reach? Three Generations Prepare for Retirement, 18th Annual Transamerica Retirement Survey of Workers,” December 2017.
“How do you envision retiring?”
7FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
A Boomer perspective on longevity
“What age do you expect to live to?”
45%
10%
21%
18%
5%
1%
0.50%
Not sure
100+
90-99
80-89
65-79
60-64
Less than 60AGES
Transamerica Center for Retirement Studies, “Wishful Thinking or Within Reach? Three Generations Prepare for Retirement, 18th Annual Transamerica Retirement Survey of Workers,” December 2017
Retirement ExpensesWants, Needs & Surprises
9FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Costs of Living in Retirement
•The WantsDiscretionary Expenses (dining out, travel for pleasure and entertainment, charities, legacies)
•The NeedsHousing, food, utilities, clothing, health care and long term care
•The SurprisesRespondents saved 11% of the income for emergencies
LIMRA, Secure Retirement Institute, “Sources of Retirement Income, A Study of Retirees,” 2017
The Needs 65%
The Wants 24%
Surprises11%
Results are based on a sampling of 2.025 retirees ages 55-79 who have retired for at least a year and have $35,000 or more in household income.
10FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
What Do you Think is the Number One Cost in Retirement?
Placeholder for Audience Poll Slide
11FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Housing costs
The biggest bite out of retiree budgets!
HOUSING
32.9%TRANSPORTATION
16.2%FOOD
12.3%HEALTHCARE
11.1%RECREATION
5.3%Ann C. Foster, “A closer look at spending patterns of older Americans ,” Beyond the Numbers: Prices & Spending, vol. 5, no. 4 (U.S. Bureau of Labor Statistics, March 2016), https://www.bls.gov/opub/btn/volume-5/spending-patterns-of-older-americans.htm
12FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Surprise!
The potential costs of“aging in place”
SOURCE: MIT Age Lab; Issue Brief Number 2015-12, Going Home: Senior Housing Options in Retirement Joseph F. Coughlin, Ignacio Chaparro, Olivia DaDalt; (Bureau of Labor Statistics, 2012); Home Advisor
New roof $6,582
Gutter Repair$328
New Furnace $3,875
Termites!
Wider Doors/Ramps
$500 –$1,000
New Windows
$6,582
13FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Surprises in Basic living expenses
6%
6%
15%
16%
26%
28%
50%
57%
73%
Other reasons
Financially supporting older family member
Family events (divorced, widowed, married)
Doing more activities than expected
Financially supporting younger family member
Income taxes higher than expected
Housing costs more than expected
Health care costs more than expected
Cost of goods/services generally more than expected
Why were expenses higher than expected?
Note: Based on 507 retirees aged 55 to 79 who have been retired for at least a year, have $35,000 or more in household income, and who experienced “somewhat” or “significantly” higher-than-expected basic living expenses. Multiple responses allowed.
LIMRA Secure Retirement Institute, Retirement Spending, Experience versus Expectations, a Study of Retirees, 2016
Sources of Retirement Income
15FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Social Security42%
Pension/Retirement30%
Employment Earnings11%
Interest & Dividends11%
Other6%
Individual retirement income by source
Social Security and pensions remain the two largest sources of retiree income – and both guarantee income for life.
LIMRA Secure Retirement Institute Analysis of U.S. Census Bureau’s Current Population Survey, March 2015 Supplement. Analysis based on fully retired households. Household income represents total income received in 2014. As cited in LIMRA Fact Book on Retirement Income, 2016, A Review of Trends and Activity in the Retirement Income Market.
16FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Pension plans are disappearing
55% 53%
40%
26%18%
12%
Age 75 andolder
65 to 74 55 to 64 45 to 64 35 to 44 Age 18 to 34
Percent of households with access to a defined benefit plan
LIMRA Secure Retirement Institute analysis of 2013 Survey of Consumer Finances, Federal Reserve Board, 2014. For married couples, age group assignment based on age of oldest spouse in 2013. Data represents only households with zero-plus financial assets.
Pensions as a source of income will plummet for future generations.
17FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
5%
11%
16%
4%
26%
26%
28%
29%
81%
80%
5%
11%
16%
25%
40%
45%
52%
89%
81%
95%
Income from business
Rental income
Part-time work
Roth IRAs
Annuity
Retirement savingsplans
IRAs or Rollover IRAs
Taxable assets
Pension benefits
Social Security
Income sources: available and used
(As a percentage of households)
LIMRA, Secure Retirement Institute, “Sources of Retirement Income, A Study of Retirees,” 2017
Available income sourcesUsed for income
18FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
53%
33% 34%
76%84%
33%
19% 17%
46%
63%
28%
0% 0%
56%65%
All Under age 65 Ages 65-69 Ages 70-74 Ages 75-79
Income from IRAs
Percent of retirees receiving income from traditional IRAs
Taking systematic withdrawalsTaking withdrawals only to meet RMDs
Taking withdrawals
LIMRA, Secure Retirement Institute, “Sources of Retirement Income, A Study of Retirees,” 2017
19FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Annuity ownership
21%
51%
27%
53%
39%
30%
33%
28%
33%
8%7%
8%
9%8%
6%8%
5%7%
Don't work with a financialprofessional
Work with a financialprofessional
No formal plan
Have a formal plan
Ages 75-79
Ages 70-74
Ages 65-69
Under Age 60
All
Percent of retiree households owning an annuity
Income AnnuityDeferred Annuity
LIMRA, Secure Retirement Institute, “Sources of Retirement Income, A Study of Retirees,” 2017
Income Annuity Trends in the Market“Longevity Insurance”
21FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Retail annuity sales by investment objective
$119.0 $115.4 $114.5 $107.7
$21.7 $27.5 $29.3 $30.7$16.3 $19.6 $23.1 $26.7$25.9 $33.8 $39.8 $41.3
2012 2013 2014 2015
Guaranteed Income Principal Protection Protection & Market Growth Market Growth
$183.0
Guaranteed income is the largest component of annual annuity sales.
Totals are for retail sales only and exclude sales in employer plans and structured settlements. Guaranteed income bucket includes VA sales with GLWB/GMIB/GMWB rider elected, fixed indexed sales with GLWB rider elected, immediate income annuity and deferred income annuity.
$196.3 $206.7 $206.5
U.S. Individual Annuity Yearbook —2015 and VA GLBElection Tracking Survey, LIMRASecure Retirement Institute, 2016.
22FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Average initial premium by annuity type
$69k
$95k $90k
$117k
$148k $146k $148k
Fixed Rate Indexed Variable Indexed withGLWB
VA with GLB DIA SPIA
Lifetime income annuities can attract big tickets
Deferred Annuity Lifetime Income
“U.S. Individual Annuity Yearbook — 2015; Profile of Individual Annuity Buyer Study and Income Annuity Buyer Study,” LIMRA Secure Retirement Institute, 2016.
23FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Putting it all together
•Boomer Market unlike any other with more assets than any other generation
•Retirement (and when) has not been clearly defined, thus they want flexibility
•Bottom line- retirees need to have some form of guaranteed income. There is no one right answer, no one size fits all. Specific to individual. Some need more liquidity, some need accumulation prospects too, some need tax benefits etc.
•But all need flexibility and products that can change with them as their life changes during the Golden Years
•This has led to DIY Retirement, and a slowing of DIA sales
•Opportunity for manufacturers to define retirement and build products that suit the changing Boomer landscape
24FOR PRODUCER USE ONLY. NOT FOR USE WITH THE PUBLIC.
Disclosures
Annuity products are issued by Massachusetts Mutual Life Insurance Company (MassMutual) and C.M. Life Insurance Company. C.M. Life Insurance Company, Enfield, CT 06082, is non-admitted in New York and is a subsidiary of MassMutual, Springfield, MA 01111-0001.
© 2018 Massachusetts Mutual Life Insurance Company (MassMutual), Springfield, MA 01111-0001. All rights reserved. www.massmutual.com
Douglas L. Robbins
May 9, 2018
The Science Behind Securing the “Golden Years”
2
Topics
▪A Reasonably Hot Research Topic
▪What Might a Typical Retiree do Wrong, and Why?
▪Changing the Odds
▪Educating “The People”
3
Topics
▪A Reasonably Hot Research Topic
4
What does our Gut Say, as Insurers?
▪ Retirees’ biggest financial risk is probably running out of money before they die
▪ This could happen for any number of reasons, but regardless, our intuition here makes perfect sense
▪ How can we as insurers act most effectively to help them?
5
A Clear Topic for Research
▪ Session 46 L from the 2016 L&A Symposium started my thinking along these lines, so credit is due there
▪ The SOA has recently published the “Managing Retirement Decision Series”▪ The article most similar to my topic today is called
“Designing a Monthly Paycheck in Retirement”
▪ It is more anecdotal, and less “techy” than my presentation today, but well worth a look
6
A Key Point of the DIA Session (46 L)
▪ The timing of guaranteed DIA payouts can have a dramatic effect on the probability a retiree outlives their money
▪ Why is that?▪ Each guaranteed payment directly impacts the “fund”
a retiree would otherwise use
▪ The timing and amount of that effect is material
▪ This all led, along with some other concepts, to the path I’ve taken here
7
Topics
▪What Might a Typical Retiree do Wrong, and Why?
8
▪A huge chunk of our current or soon-to-be retirees can tell you 2 Key Things about Investing
What and Why?
9
▪A huge chunk of our current or soon-to-be retirees can tell you 2 Key Things about Investing
▪A little knowledge is a what?
What and Why?
10
Two patterns of Fund Unit Values
11
Two patterns of Fund Unit Values
▪Which one is better? Why?
12
The answer is, it depends
▪ If I invest only upfront dollars, Fund1 is surely better
13
The answer is, it depends
▪What if I invest $10k a year?
14
The answer is, it depends
▪ I’d then end up with $233k from Fund1, and $258k from Fund2!!
15
“Dollar Cost Averaging”
▪ A standard investors’ goal should be: buy low, sell high
▪ Sounds good in theory
▪ Easier said than done
▪ “Market timing” can be treacherous
▪ But by investing equal amounts over time, you’ll tend to naturally get more units for which you’ve bought low and eventually sold high, than vice versa
16
“Diversification”
▪ Buy lots of arguably sound financial instruments!
▪ Target-age funds, though, can help foster a long-term relationship with a tangible goal in mind
▪ Here’s the basic idea . . . .
17
“Diversification”
▪ An investor with 30 years to retirement might be shown this, regarding an Equity fund: *
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
3,000,000
3,500,000
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930
Retirement Fund Over 30 Years - $10k per Year
5th Median 95th
*Note: from here on, all examples assume 7% mu & 16% sigma for Equity. 4% each for Fixed.
18
“Diversification”
▪ Demonstrates the power of periodic Equity investing in the long term
▪ The median return looks great (not to even mention the 95th %-ile)
▪ Even the 5th %-ile shows at least some positive growth
▪ With a 30-year horizon, plenty of time to re-think, if early returns disappoint
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930
Retirement Fund Over 30 Years - $10k per Year
5th Median 95th
19
“Diversification”
▪ The situation is quite different in year 25, assuming median growth thus far
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930
Retirement Fund Over 30 Years - $10k per Year
5th Median 95th
20
“Diversification” – 50% Fixed
▪ For some, it might make great sense to narrow the ultimate outcome range
21
Lessons Learned . . . .
▪Planning for equal transactions over time = “Good”
22
Lessons Learned . . . .
▪Planning for equal transactions over time = “Good”
▪Diversification into Fixed = “Safety”
23
Lessons Learned . . . .
▪Planning for equal transactions over time = “Good”
▪Diversification into Fixed = “Safety”
▪What could possibly go wrong?
▪Which one is better? Why?
24
Two patterns of Fund Unit Values
▪ What if I have $200k, and plan to w/d $18k/yr (total $252k) for 14 yrs?
25
Two patterns of Fund Unit Values
▪ The blue fund (4%/yr) works perfectly; the red one (~6.5%/yr) runs out in yr 11
26
Two patterns of Fund Unit Values
27
Retirement Disaster
▪ My accumulation-phase red line was quirky – possibly misleadingly so
▪ The payout-phase one was just an S&P500 starting point of Jan 1, 2003 (which is by NO MEANS worst case!)▪ Not too quirky at all, really
▪ In fact, it could be reasonably viewed as an average to good-ish real-world scenario
▪ Yet it led to retirement disaster
28
Retirement Disaster
▪ My accumulation-phase red line was quirky – possibly misleadingly so
▪ The payout-phase one was just an S&P500 starting point of Jan 1, 2003 (which is by NO MEANS worst case!)▪ Not too quirky at all, really
▪ In fact, it could be reasonably viewed as an average to good-ish real-world scenario
▪ Yet it led to retirement disaster . . . b/c of timing
29
Topics
▪Changing the Odds
30
What is going on?
▪ “Reverse Dollar Cost Averaging”
▪ In a spiky market, if you are in the payout phase, taking annual income:
▪ You likely intend to withdraw equal amounts periodically – a standard retirement goal
▪ You may find it tough to adjust your income down by 40% if the market drops that much
▪ This will tend to cause you to sell the most units in market downturns, depleting your account rapidly in the worst possible scenarios
31
What can we do about it?
▪ Option 1: Meet our entire income need using a DIA/SPIA or a GLWB
▪ Positive: Fully guaranteed at amount planned
▪ Negatives: May result in very little left in accumulation accounts
Other risks may bite in this scenario, inflation being one key example
Nothing left for emergency needs
“Loss of control of funds,” aka liquidity risk
32
What can we do about it?
▪ Option 1: Meet our entire income need using a DIA/SPIA or a GLWB
▪ Positive: Fully guaranteed at amount planned
▪ Negatives: May result in very little left in accumulation accounts
Other risks may bite in this scenario, inflation being one key example
Nothing left for emergency needs
“Loss of control of funds,” aka liquidity risk
▪ The latter is very important to agents!!
33
Can we do better?
▪ I believe we can, via a more limited allocation to SPIA/DIA
▪ I will set up a case study to demonstrate:
▪ 30-year certain-period analysis (close to life expectancy of a 60-year old, but easier to grasp)
▪ Simulation based on independent lognormal annual returns, w/ parameters “mu” and “sigma”
Mu’s from 5-7%, net of expenses & fees
Sigma’s from 10-16%
▪ Varying planned withdrawal amounts
34
Can we do better?
▪ Assume that the policyholder has a planned withdrawal % of account value
▪ They will alter the $ amount as A/V varies
▪ (A “SWiP,” or Systematic Withdrawal Program)
▪ This changes the Lognormal net “Mu” only
▪ Cumulative returns then remain Lognormal
▪ Also assume that there is a minimum livable annual income of 70% of starting planned
▪ This gives us an array of “ruin” probabilities
▪ These can be used to compare strategies
35
Can we do better?
▪ For a 4.5% income strategy at Age 60, you get the following ruin probabilities:
36
Can we do better?
▪ A 5% income strategy at Age 60 produces results a bit more ominous:
37
Can we do better?
▪ Looking at those two parameters suggests something rather stunning, at least to me
▪ We’ve discussed the benefits of shifting fund mixes during the accumulation phase
▪ Does this work during retirement? No!
38
Can we do better?
▪ Looking at those two parameters suggests something rather stunning, at least to me
▪ We’ve discussed the benefits of shifting fund mixes during the accumulation phase
▪ Does this work during retirement? No!▪ E.g., with 16% sigma and 7% net mu, a 5% w/d here comes
with an estimated 14.5% failure rate in year 15
39
Can we do better?
▪ Looking at those two parameters suggests something rather stunning, at least to me
▪ We’ve discussed the benefits of shifting fund mixes during the accumulation phase
▪ Does this work during retirement? No!▪ E.g., with 16% sigma and 7% net mu, a 5% w/d here comes
with an estimated 14.5% failure rate in year 15
▪ By mixing in 33.3% fixed assets, I reduce sigma to 12% and mu to 6%, my failure rate drops to . . .
40
Can we do better?
▪ Looking at those two parameters suggests something rather stunning, at least to me
▪ We’ve discussed the benefits of shifting fund mixes during the accumulation phase
▪ Does this work during retirement? No!▪ E.g., with 16% sigma and 7% net mu, a 5% w/d here comes
with an estimated 14.5% failure rate in year 15
▪ By mixing in 33.3% fixed assets, I reduce sigma to 12% and mu to 6%, my failure rate drops to . . . 13.8%
41
Can we do better?
▪ Looking at those two parameters suggests something rather stunning, at least to me
▪ We’ve discussed the benefits of shifting fund mixes during the accumulation phase
▪ Does this work during retirement? No!▪ E.g., with 16% sigma and 7% net mu, a 5% w/d here comes
with an estimated 14.5% failure rate in year 15
▪ By mixing in 33.3% fixed assets, I reduce sigma to 12% and mu to 6%, my failure rate drops to . . . 13.8%
▪ Moral: We need to rethink “Safety”!
42
What do I do now, coach?
▪ This leads to an easy-to-grasp conception of the impact of adding a SPIA to an equity-based portfolio
▪ Say you have $100,000 invested in exactly the sort of fund projected in my scenarios, with the 30-year retirement horizon used
▪ You expect a net mu = 7%, Sigma = 16%
▪ Your desired income is $5k (w/ ruin at $3.5k)
▪ A retail SPIA is paying 6% per annum
▪ Clearly, using $83.3k to buy a SPIA would meet the entire desired income, with 0% risk
▪ But it would not leave much liquidity
43
And this matters because?
▪ But putting about $58.3k of your $100k in a SPIA has virtually the same risk impact!
▪ Why?
▪ It takes care of your “minimum income requirement,” leaving your liquid assets to hopefully cover the rest
▪ You have no chance of ever getting less than your $3,500 needed in any year – the SPIA ensures that!
▪ A 3.6% SWiP on your remaining liquid amount of almost $42k gets you to the desired $5,000 initially
▪ This deal makes it much more likely than “SWiP only” that the $ amount of the SWiP will grow over time
44
And this matters because?
▪ Generally, the best use under my assumption set (which is admittedly over-simplified) is when you have N years to go to retirement
▪ AT retirement, therefore a SPIA looks best
▪ But a DIA’s impact pre-retirement may be more pronounced than a SPIA at retirement:
▪ Because of the pre-retirement return volatility, a person retiring in 10 years has roughly double the ruin probability of a current retiree, if they stay in 100% Equity
▪ A 58% or so allocation to a DIA paying 12% of premium starting in year 10 still brings those to 0% 45
When Would a DIA be Better?
▪ In both cases, SPIA and DIA, the allocations can of course be lower:
▪ If the desired or minimum income levels are lower, or
▪ If the annuitant is willing to accept non-zero risk of ruin
▪ This lends itself to various sorts of tools that enterprising actuaries might develop for the field
▪ Such a tool would analyze allocations, using some optimization metric . . . .
46
Further Thoughts
▪ There are various other weapons in the annuity arsenal, which you’ve heard about, that could be applied
▪ Generally, GLWBs mimic either SPIA or DIA to some extent, depending on when they are first utilized by a client
▪ Are they better or worse than an Account Value / SPIA or DIA combo?
47
Further Thoughts
48
Topics
▪Educating “The People”
▪ Populace 1: Retirees
▪ Want to maximize value of accumulated assets
▪ Want to not outlive assets in decumulation
▪ Populace 2: Sales Force
▪ Want to maximize value of assets
▪ Want to maintain control of assets
49
First Question: Who is That?
▪ Populace 1: Retirees
▪ Want to maximize value of accumulated assets
▪ Want to not outlive assets in decumulation
▪ Populace 2: Sales Force
▪ Want to maximize value of assets
▪ Want to maintain control of assets
▪ Want to maximize value of assets controlled
50
First Question: Who is That?
▪ A mix of account value (“AV”) driven assets of some type with a SPIA/DIA is typically less risky than going full AV
▪ (The exception would be when the income % desired is higher than SPIA/DIA would provide)
▪ Outside of that exception, the client’s desire to not be starving at age 90 could make this an easy sell
▪ It should simply be a matter of assessing the client’s preferred risk (of AV depletion) level, and showing them how to achieve it
51
Retirees s/b easy to convince
▪ Payout (non-AV-driven) Annuities have long been equated with loss of asset control
▪ Ever hear of the term “annuicide”?
▪ Prospect of one last upfront commission at annuity date, has not been sufficiently motivational
▪ Any educational effort for the Sales Force has to focus at least as much on asset preservation as it does on retirement security
▪ Could be a misconception based on a “myth”
52
Sales Force may be less so
▪ What “myth” do we need to educate away to be successful in getting SPIA/DIA utilized?
▪ I hypothesize that it’s related to general human risk tolerance, but in an unusual way
▪ Let me ask you a few financial questions – Scenario 1:
▪ You get a $50 gift card from a friend
▪ When you go to the store to use it, the manager offers you a coin flip bet –tails you lose the card, heads he replaces it with a $100 one
▪ Do you take the deal or keep the original card?
▪ Second scenario:
• At a cousin’s birthday party, you find a $50 gift card on the floor
• Do you keep it, or tell the guest of honor? 53
Busting the “Myth”?
▪ Now how about this?
▪ You get a $150 red light camera ticket, which you know is bogus – there was a cop who waved you through, but he’s not in the photo
▪ The state allows you to settle things for $75 if you plead guilty & mail it in
▪ If you fight it, you estimate a 50% chance the judge will buy your story and let you off, but a 50% chance that you’ll now owe $150
▪ Do you take the $75 mail-in discount, or go fight the ticket in court?
54
Busting the “Myth”?
▪ Now how about this?
▪ You get a $150 red light camera ticket, which you know is bogus – there was a cop who waved you through, but he’s not in the photo
▪ The state allows you to settle things for $75 if you plead guilty & mail it in
▪ If you fight it, you estimate a 50% chance the judge will buy your story and let you off, but a 50% chance that you’ll now owe $150
▪ Do you take the $75 mail-in discount, or go fight the ticket in court?
▪ Here’s the thing: the first and third scenarios are stochastically identical
• Most of the time, response %s are different because in the third one, something is being “taken” from you and you become LESS risk averse in response, trying to fight to keep what’s yours
• The 2nd scenario was just so you wouldn’t notice, but if you turned in the gift card, take 5 minutes of professionalism credit
55
Busting the “Myth”?
▪ Now how about this?
▪ You get a $150 red light camera ticket, which you know is bogus – there was a cop who waved you through, but he’s not in the photo
▪ The state allows you to settle things for $75 if you plead guilty & mail it in
▪ If you fight it, you estimate a 50% chance the judge will buy your story and let you off, but a 50% chance that you’ll now owe $150
▪ Do you take the $75 mail-in discount, or go fight the ticket in court?
▪ Here’s the thing: the first and third scenarios are stochastically identical
• Most of the time, response %s are different because in the third one, something is being “taken” from you and you become LESS risk averse in response, trying to fight to keep what’s yours
• The 2nd scenario was just so you wouldn’t notice, but if you turned in the gift card, take 5 minutes of professionalism credit (kidding!)
56
Busting the “Myth”?
▪ What “myth” do we need to educate away to be successful in getting SPIA/DIA utilized?
▪ I hypothesize that it’s related to risk tolerance
▪ “If I put 20% of my assets into a SPIA now, I’ll have only 80% left under my control.”
▪ “Something I had is being taken away from me.”
▪ “I’m willing to incur risk* to avoid that.”
▪ Is this analysis of the situation correct?
• In the very short term, yes, of course!!
• But it ignores the cash-flow picture of a 100% A/V strategy
*For my client, but also for myself, as we shall see! 57
Busting the “Myth”?
▪ The following illustration assumes that the starting position is in a balanced (1/3 Equity, 2/3 Fixed) fund for “safety,” assuming 5% net AV growth w/ 8% vol, and the client wants a very typical 4% W/D
▪ Let’s assume that the agent is in “full control” of these funds – no CDSC or MVA, etc
▪ And let’s say his concern is this:
▪ He’ll get an upfront commission, roughly the same, on either a move into a final deferred annuity (remember DOL!) or a SPIA
▪ But on the deferred annuity, he’d get a 1% trail after year 7 – any SPIA money “loses out” on this
58
What is the “Myth,” Exactly?
▪ Here’s an illustration of where things might end up
59
Myth Busting
▪ I’ll note here that although I didn’t make a graph, the ruin odds for, say, 2/3 Equity aren’t as bad as you’d think, though in the tails, it accelerates a bit
▪ But your illustration system notes that with 20% in a SPIA paying 6%, one can invest the rest in 2/3 Equity, with better ruin probabilities throughout
▪ The SPIA takes care of 1.20% of the required w/d leaving only 2.8% to come from the fund
▪ The improved growth and lower w/d from the fund-value component illustrates very well!
60
Myth Busting
▪ Here is the revised illustration:
61
Myth Busting
▪ And here is what you may have missed . . . .
62
Myth Busting
▪ And here is what you may have missed . . . .
63
Myth Busting
▪ Can a Lifetime W/B benefit not accomplish the same thing, though?
▪ What if I illustrate a VAGLB that allows 4% W/Ds at our given issue age of 60?
▪ Presumably (in most cases, anyway), you still can’t be in 100% Equity
▪ Assume 2/3-1/3 again, and a fee of 100bp
64
Myth Busting
▪ Here is the VAGLB fund-value illustration:
65
Myth Busting
▪ The median return is worse off than the 20% SPIA/80% VA combo in years 12+
▪ All tail fund-value measures are much worse
▪ The tails are also, obviously, even more badly off over time than with the basic 4% w/d option
▪ It should be our job to get this word out to sellers
66
Myth Busting
▪ The median return is worse off than the 20% SPIA/80% VA combo in years 12+
▪ All tail fund-value measures are much worse
▪ The tails are also, obviously, even more badly off over time than with the basic 4% w/d option
▪ It should be our job to get this word out to sellers
▪ The policyholder ruin situation, though, is clearly superior with the VAGLB . . .
67
Myth Busting
▪ The median return is worse off than the 20% SPIA/80% VA combo in years 12+
▪ All tail fund-value measures are much worse
▪ The tails are also, obviously, even more badly off over time than with the basic 4% w/d option
▪ It should be our job to get this word out to sellers
▪ The policyholder ruin situation, though, is clearly superior with the VAGLB . . . or IS IT?
68
Myth Busting
▪ Why, come to think of it, are companies able to offer (despite the big tail risk and/or hedge costs, AND all that client liquidity) an income guarantee on a VA even remotely comparable to a SPIA?
▪ The only reasonable answer, I believe, is policyholder behavior
▪ The client and agent really should consider that risk aspect, too (and/or, what will happen to these benefits if policyholder behavior becomes broadly more efficient!)
69
Myth Busting
▪ FIA WB may be the worst case of all
70
Myth Busting
▪ You can improve your client’s ruin situation, AND your own long-term funds under control, by including some SPIA/DIA
▪ VAGLB and/or FIAWB show mixed results
▪ The client’s ruin risk can go to 0% IFF they use the benefit completely optimally; if not, who knows?
▪ Long-term fund values are generally worse; does client truly hope to live on their guarantee only?
▪ Moral: It makes tons of sense to mix some form of SWiP w/ SPIA/DIA!!
71
Myth Busting – Summary to Sellers
▪ I’ve ignored mortality today, because this session is mostly about financial risk
▪ However, it is conceivable that companies could assume higher mortality for lifetime WBs than for SPIAs and/or DIAs
▪ This, in combination with the behavior assumptions noted earlier, could create occasional situations where lifetime WBs pay more than SPIA/DIA
▪ This is worth considering, but the adverse side of behavior risk should nevertheless be part of the analysis
72
A final caveat
© Oliver Wyman
NAIC VA RESERVE AND CAPITAL REFORMANNUITY-BASED SOLUTIONS SEMINARMAY 9, 2018
PETER TIAN
CONFIDENTIALITY Our clients’ industries are extremely competitive, and the maintenance of confidentiality with respect to our clients’ plans and data is critical. Oliver Wyman rigorously applies internal confidentiality practices to protect the confidentiality of all client information.
Similarly, our industry is very competitive. We view our approaches and insights as proprietary and therefore look to our clients to protect our interests in our proposals, presentations, methodologies and analytical techniques. Under no circumstances should this material be shared with any third party without the prior written consent of Oliver Wyman.
© Oliver Wyman
2© Oliver Wyman
Agenda
• Provide background of the NAIC VA reserve and capital reform initiative• Recap current state of the VA statutory balance sheet• Discuss recommendations for revisions to the current framework• Outline expected implications of revisions to VA industry
3© Oliver Wyman
Recent history of VA statutory reserve and capital standards
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Enactment of C3 Phase II for VA TAR
“Rise of VA captives”
• Poor alignment of statutory risk factors with economic – and GAAP – risk profile
• Increasing variation in company hedging objectives
• (Nearly) continuously falling interest rates
Timeline2006-2019
Enactment of AG 43 for VA reserves
NAIC commissions VA reform initiative
OW presents initial recommendations
QIS I QIS II
OW presents final recommendations
4© Oliver Wyman
Overview of the current VA statutory reserve and capital framework
Total Funding Requirement
Total Asset Req.(C3P2)
Reserves(AG43)
C3P2 Standard Scenario C3P2 Stochastic AG43 Stochastic AG43 Standard Scenario
CTE 90 “Best-efforts” run
CTE 90 “Adjusted” run
CTE 70 “Best-efforts” run
CTE 70 “Adjusted run
MAX
MAX MAX
Weighted avg. Weighted avg.
Difference between TAR and reserves, if positive, is the
RBC C3 charge
5© Oliver Wyman
The statutory balance sheet is characterized by asymmetric accounting, non-harmonized assumptions, and unstable reserve vs. capital boundary
Total Asset Requirement
(“TAR”)
Separate account value
Fair value of hedge assets
Assets Liabilities
Other GA assets
Statutory reserve
General account assets• Derivatives held at fair value
• Other assets mostly amortized cost
Required capital for
market risk
1
2
3
1
2 Statutory reserves• Entirely not fair value
– 50-70% of FV equity sensitivity– 10-20% of FV interest rate sensitivity
• Not comparable across companies
3 Required capital for market risk• Arithmetic difference between TAR and reserves, which have
dissimilar levels of market sensitivity
• Highly volatile, but reduced by voluntary reserves dollar-for-dollar
Volatile boundary
Surplus
Assumed long-term 20-yr. UST in industry
2 3 5 2 3
<4.0%4.00-4.49%4.50-4.99%5.00-5.49%≥5.50%
6© Oliver Wyman
To address these shortcomings in the current VA statutory framework, Oliver Wyman recommended five types of framework revisions
Shortcomings in the current framework Framework revisions
Economic-based hedging can adversely impact statutory financials Align economically-focused hedge assets with liability valuations
Volatile interactions between stochastic and Standard Scenario calculations
Reform Standard Scenario calculationsStandard Scenario is excessively conservative in some areas and aggressive in others
Volatile interactions between reserves and TAR, incentivizing voluntary reserves Align TAR and reserves
Derivative admissibility limits prevent hedging benefit from being fully recognizedRevise asset admissibility for derivatives and DTAs
DTA admissibility limits exacerbate mismatches between statutory and tax accounting
Companies use different capital markets assumptions for scenario generation Standardize capital markets assumptions
1
2
3
4
5
7© Oliver Wyman
The revisions align asset-liability valuations, stabilizes the boundary between reserves and capital, and removes incentives for using voluntary reserves
CTE 98
Separate account value
Fair value of hedge assets
Assets Liabilities
Other GA assets
Statutory reserve
Required capital
1
2
Surplus 3
¼ × (1 − Tax Rate) ×(CTE 98 − Reserves)
General account assets• Hedge assets carried on a similar valuation basis as liabilities
• Other assets mostly amortized cost
1
2 Statutory reserves• If hedging fully: market-sensitivity aligned with fair value
• If not hedging: similar market-sensitivity as current framework
Revised Standard Scenario reserve floor
• Essentially CTE reserve with prescribed actuarial assumptions, calibrated based on industry data
3 Required capital for market risk• More stable, as CTE 98 and CTE 70 reserve have similar levels
of market sensitivity
• Requires $4 of voluntary reserves to reduce each $1 of C3
8© Oliver Wyman
What a “400% RBC” VA balance sheet looks like under today’s IR conditions
Total Asset Requirement (100% RBC)
Separate account value
Fair value of hedge assets
Assets Liabilities
Other GA assets
Statutory reserve
(CTE 70)
CTE 98
Separate account value
Fair value of hedge assets
Assets Liabilities
Other GA assets
Statutory reserve
(CTE 70)
Current framework Revised framework
Surplus or voluntary reserve
Hedging increases TAR – and therefore assets required for 400% RBC
Voluntary reserve
Hedging reduces CTE 98 – and therefore assets required for
400% RBC
Total Asset Requirement (100% RBC)
9© Oliver Wyman
Expected implications of the framework revisions on the VA industry
Implication Description
1 Increase in minimum funding required to meet target RBC ratio
• Harmonized interest rate scenarios are more adverse than many companies’ current assumptions
• Reduced capital benefit from voluntary reserves
• Partially offset by removal of Working Reserve, hedge run-off, and current Standard Scenarios
2 Greater incentives for more extensive economic hedging
• Current framework penalizes full economic hedging; thus, optimal strategy is to hedge partially
• Under revised framework, full economic hedging minimizes balance sheet volatility and target funding for a typical target RBC ratio
3 Reduced need for large capital buffer to maintain target RBC ratio, if hedging
• Total balance sheet volatility reduced for companies with extensive hedging
• Full economic hedging eliminates need for capital buffer, as hedge gains would fully offset liability movements in stress
4 Reduced proposition of VA as a capital-efficient vehicle for market risk-taking
• Current framework allows capitalization for unhedged VA market risk to be below fair value of guarantees in low interest rate environments
• Revised framework requires capitalization to be near fair value of guarantees in most interest rate conditions to achieve a typical target RBC ratio
5 More level playing field across carriers • Reserve and TAR floored at levels calculated using prescribed market and actuarial assumptions
Summary of recommendationsAppendix
11© Oliver Wyman
Summary of recommendations1 of 2
Framework Recommendation
CTE Amount Use VM-20 scenario generator for interest rate scenarios
Use VM-20 scenario generator for separate account returns, but recalibrated based on data from 1926 to 2016
Allow companies to use proprietary scenario generators if – and only if – they do not reduce Total Asset Requirement
Introduce principles to govern implied volatility scenario generation, with a “safe harbor” approach provided
Remove the Working Reserve when calculating scenario GPVAD
Discount deficiencies at the Net Asset Earned Rate on Additional Assets
Follow VM-20 guidance on general account asset projections, with additional constraint on borrowing cost
Permit immediate liquidation of currently-held hedges and non-reflection of mark-to-market hedge gains and losses
Reduce minimum allowable CDHS “error factor”, but require back-testing disclosure to support chosen “error factor”
Differentiate treatment of non-guaranteed revenue sharing income by affiliated funds vs. non-affiliated funds
Standard Scenario Amount
Align AG43 Standard Scenario calculations with CTE (“adjusted”)
Remove the C3 Phase II Standard Scenario
Project Standard Scenario on an aggregated basis, but with disclosure of aggregation benefit observed
Refresh prescribed policyholder behavior assumptions to align with industry experience
Use the Standard Scenario construct to govern model choices and actuarial assumptions only, via a reserve “add-on”
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
12© Oliver Wyman
Summary of recommendations2 of 2
Framework Recommendation
Standard Scenario
Calculate Standard Scenario based on company-specific market paths (selected from a panel of standardized paths)
Allow the Standard Scenario Amount to be calculated as a CTE Amount with prescribed assumptions
C3 charge Calculate C3 as the difference between total statutory reserve and CTE 95 on same distribution
Permit smoothing to be conducted on the C3 charge, but not on the Total Asset Requirement
Disclosure requirement
Disclose Sharpe ratio and correlations for all funds not generated by mapping to the VM-20 scenario generator
Disclose modeled vs. actual hedge performance over the past 12 to 36 months for explicit CHDS reflection
Disclose historical Greek coverage over the past 12 to 36 months for implicit CDHS reflection
Disclose positioning of the dollar amount of CTE (“best-efforts”) relative to the unhedged CTE and fair value
Disclose a “cumulative decrement” analysis under companies’ own and prescribed Standard Scenario assumptions
Other topics Increase admissibility limit for designated VA hedges
Increase admissibility limit for VA-related DTAs
Endorse hedge accounting for interest rate derivatives that are part of VA hedge programs
Allocate aggregate reserve to seriatim level based on Present Value of Accumulated Product Cash Flows
16
17
18
19
20
21
22
23
24
25
26
27
28
QUALIFICATIONS, ASSUMPTIONS AND
LIMITING CONDITIONS
This report is for the exclusive use of the Oliver Wyman client named herein. This report is not intended for general circulation or publication, nor is it to be reproduced, quoted or distributed for any purpose without the prior written permission of Oliver Wyman. There are no third party beneficiaries with respect to this report, and Oliver Wyman does not accept any liability to any third party.
Information furnished by others, upon which all or portions of this report are based, is believed to be reliable but has not been independently verified, unless otherwise expressly indicated. Public information and industry and statistical data are from sources we deem to be reliable; however, we make no representation as to the accuracy or completeness of such information. The findings contained in this report may contain predictions based on current data and historical trends. Any such predictions are subject to inherent risks and uncertainties. Oliver Wyman accepts no responsibility for actual results or future events.
The opinions expressed in this report are valid only for the purpose stated herein and as of the date of this report. No obligation is assumed to revise this report to reflect changes, events or conditions, which occur subsequent to the date hereof.
All decisions in connection with the implementation or use of advice or recommendations contained in this report are the soleresponsibility of the client. This report does not represent investment advice nor does it provide an opinion regarding the fairness of any transaction to any and all parties.
How Can We Win Over The Sales Force?
Chris Bartak
• Independent annuity agent channel & business model
• Product selection from an agent point-of-view
• Trends and (limited!) predictions for the future
Topics
• View is particular to my experience in this channel at this point in time
• Tilted towards Fixed Indexed Annuities, planning goes beyond that
• Specific examples hypothetical / representative
Disclaimers & Caveats
Parties & Terms
Client Agent FMO Carrier
CustomerConsumerRetiree
ProducerAdviser(or)
Field Marketing OrgIndependent Marketing Org(IMO)Wholesaler
Insurance CompanyUnderwriterProduct Manufacturer
1. Client
• In or near retirement, Age: 55-85, median ~63
• 500k - $2M in investable assets
• Current Adviser & Portfolio
Typical Profile
1. Retirement income planning2. Market volatility3. Social Security planning4. Longevity5. Legacy6. Impact of Long Term Care expenses7. Tax planning (RMDs)8. Accumulation of assets9. Inflation
Client ConcernsOne or more of:
2. Agent
• Prior experience
• Business Structure– Solo– Development of team
• Specialization
Profile
Business Flow
• Most difficult & expensive part of business
• Cost of customer acquisition drives the entire model
• Effective methods evolve over time, successful ideas can hit “capacity”
Direct Consumer Marketing
Direct Mailer to prequalified prospects– <1% response rate; several thousand / event– Invite to dinner event & presentation on topic of
interest to retirees– Clear language that annuities and other insurance
may be discussed, but no sales will be made at event– RSVP process
Ex. Funnel – Seminar Marketing
Event– Mid to high end restaurant– 15-30 attendees– Educational presentation; typically focused on one-or-
more of previous “client concerns”– Entirely group based– One-One appointment follow up process
Seminar Marketing part 2
Other marketing funnels1. Radio - shows / ads2. Television3. Book4. Referrals
– Much more than “have any friends?”– Proactive campaigns, events, community building
Client-Agent Meeting / Planning Process
Less More
Planning Process1. Typically multiple appointments
2. Most securities licensed & do comprehensive review of entire financial plan
– Investments– Income– Insurance
Example Planning Outcomes1. Not a fit, go separate ways2. Income
– FIA with GLWB to provide core pension-like income3. Accumulation
– No fee FIA as safe money bond alternative4. Legacy
– Leverage unneeded RMDs into legacy
Post Sale Service1. Annual reviews
2. Portfolio management
3. Policy service – withdrawals, rider utilization, claims, etc
3. FMO
Core Purpose #1Help agent support & grow their business
– Product access & education– Operations back office– Compliance / suitability– Materials, websites, brochures, brand– Coaching, development– Technology (electronic application, statement
aggregation, etc)
Core Purpose #2Deliver quality business to Insurance Carrier
– Efficiently market on a variable cost basis– Pre-screen and deliver apps IGO– Compliance / Suitability – Point of contact for agent
4. Carrier
First, do no harm
– Marketing materials accurately & simply describe products
– Website works, phone gets answered– New business ops & inforce service– Renewals & other non-guaranteed elements– Credible management– Vision for future
Second, win on customer value
– Though actual client facing sale may or may actually not be competitive, agents act as if it is• Agent typically has access to large number of carrier• Product flows & behavior chase strongest benefits
– Compensation?
Future
• Development from “annuity sales” into full service planners– RIA Platforms
• DOL• Technology
– 2018 large shift towards e-app– Integrated planning, CRM, reporting software
Trends
Douglas L. Robbins
May 9, 2018
The Science Behind Securing the “Golden Years” –Case Study
2
Let’s use a more realistic look at “Ruin”
▪ No more “30-year certain” look
▪ We will instead use 70% of a2000 mortality, for simplicity (rather than messing with improvement factors)
▪ If the fund runs out, ruin now occurs with probability tPx
▪ Same fund characteristics, in terms of balancing Equity and Fixed
3
Assume an Agent has a Potential Client
▪ Post-DOL world▪ Presume that one goal is to give advice that is unlikely to
be said to have failed in retrospect
▪ Agent and client basically agree on <2.5% P(Ruin)
▪ Client has $100k to invest
▪ Client is age 60 and plans to begin taking Social Security at 67
▪ Non-SS Income need is stated as $7.2k per year until 67, and $3.2k thereafter
4
Chart of Income-Need Problem
5
The Agent has 4 Relevant Key Tools
▪ 1. A Variable Annuity with a 7-yr CDSC, Comp of 5% now and 1% trail yrs 8+
▪ 2. Same, w/ 4.25% W/B rider for 150bp
▪ 3. A SPIA that pays 5% per year, and has 5% comp upfront, only
▪ 4. A DIA that pays 8.5% per year starting in year 8, also with 5% comp upfront
6
The Agent’s Prior Practice was SWiP
▪ Would have used 20% Equity, 80% Fixed
▪ By using the Monte Carlo tool you’ve designed, they find that the probability estimate of ruin for that is about 8.3%
▪ The tool allows them to monkey with the fund mix, to little avail▪ Best case is 90% Fixed, still with P(Ruin) over 8%
▪ Reasonably close at 80% Fixed, so that is “Base”
▪ Median Base PV of Trail comp @3% = $11.0k
7
What are some other options?
▪ Use some % of SPIA to underlie the entire series of cash flows
▪ Use some % of DIA to underlie the post-Age67 cash flows
▪ Use the VA, and include the VAGLB Rider:▪ Presume efficient policyholder behavior
▪ This rules out use for the pre-SSA excess amount, since this payout must be level
8
An Effective Tool would Have a “Solve”
▪ I won’t go into all the detail today of what that might look like
▪ Essentially, though, this study will look by hand at some options that attempt to meet the client’s agreed upon Ruin need
▪ Of those options, we will attempt to maximize the PV of trail comp
▪ You can quickly see a couple general trends▪ More SPIA at the same Equity Mix reduces both Ruin% and PV (Comp)
▪ More Equity at the same SPIA mix does the opposite
▪ Ruin, once you’re at least partly into a SPIA, doesn’t mean the same thing as it does in a pure SWiP
▪ In Case 6*, Ruin ~= running out of “extra money”9
Here is a Chart w/ Inclusion of some SPIA
▪ DIA seems to be more helpful▪ The risk of Ruin in years 1-7 isn’t great if you don’t overspend on DIA
▪ The DIA structure, in this case, provides cheaper out-year protection
▪ In this case the Post-Ruin Outcome* is only true at ages 67+ -- this should be carefully conveyed
▪ Case 6** Ruin is similarly described as with SPIA
10
Here is a Similar Chart for DIA
▪ I’ve expressed the fund-value component in terms of selling agent’s perspective, but it also measures likely “extra money” that a given policyholder will have around
▪ Something like this would be pretty easy to automate, and solve for a best possible outcome, assuming chosen conditions
▪ All outcomes are, of course, dependent on input market assumptions and SPIA/DIA rates; in my example, DIA was the “best deal”
11
Important Follow-On Considerations
▪ The non-level income desired, and desire to use any VAGLB efficiently, rule out keeping 100% of funds in a VA
▪ Assume the extra $28k of withdrawals needed pre-age 67 can be bought for ~$25k at issue
▪ The remaining premium can fund the $3.2k per year as an efficient withdrawal stream
▪ The PV(Comp), with 0% chance of “ruin” in that case is $10.0k, better than SPIA options, but below best-case DIA
▪ Again, “ruin-free” assumes perfect utilization!
▪ And the probability of “extra money” going to $0 is 26%!!
12
What About the VAGLB Option?
▪ Although it may seem like it, the goal of this study wasn’t to extol one form of guaranteed income over another
▪ It’s to argue that we can better educate, and design better tools for, the field, in this general area of practice
▪ I believe this is how we will ultimately best serve our true end clients
13
Conclusion