2017 05 Product Development New Valuation Manual Presentations

The Product Development Section Presents
The New Valuation Manual and the Life Product
Development Actuary Seminar
May 10, 2017 | Sheraton Seattle Hotel | Seattle, WA

Presenters:
Carrie Lee Kelley, FSA, MAAA
Leonard Mangini, FSA, FALU, FRM, MAAA
Sean M. Pena, ASA, MAAA
Douglas L. Robbins, FSA, MAAA
Kimberly M. Steiner, FSA, MAAA
Chris Whitney, FSA, MAAA
The New Valua+on Manual and the Life
Product Development Actuary Seminar:
SOCIETY OF ACTUARIES
An.trust No.ce for Mee.ngs

Ac+ve par+cipa+on in the Society of Actuaries is an important aspect of membership. However, any Society ac+vity that arguably could be
perceived as a restraint of trade exposes the SOA and its members to an+trust risk. Accordingly, mee+ng par+cipants should refrain from
any discussion which may provide the basis for an inference that they agreed to take any ac+on rela+ng to prices, services, produc+on,
alloca+on of markets or any other maIer having a market effect. These discussions should be avoided both at official SOA mee+ngs and
informal gatherings and ac+vi+es. In addi+on, mee+ng par+cipants should be sensi+ve to other maIers that may raise par+cular
an+trust concern: membership restric+ons, codes of ethics or other forms of self-regula+on, product standardiza+on or cer+fica+on. The
following are guidelines that should be followed at all SOA mee+ngs, informal gatherings and ac+vi+es:
•  DON’T discuss your own, your firm’s, or others’ prices or fees for service, or anything that might affect prices or fees, such as costs,
discounts, terms of sale, or profit margins.
•  DON’T stay at a mee+ng where any such price talk occurs.
•  DON’T make public announcements or statements about your own or your firm’s prices or fees, or those of compe+tors, at any SOA
mee+ng or ac+vity.
•  DON’T talk about what other en++es or their members or employees plan to do in par+cular geographic or product markets or with
par+cular customers.
•  DON’T speak or act on behalf of the SOA or any of its commiIees unless specifically authorized to do so.
•  DO alert SOA staff or legal counsel about any concerns regarding proposed statements to be made by the associa+on on behalf of a
commiIee or sec+on.
•  DO consult with your own legal counsel or the SOA before raising any maIer or making any statement that you think may involve
compe++vely sensi+ve informa+on.
•  DO be alert to improper ac+vi+es, and don’t par+cipate if you think something is improper.
•  If you have specific ques+ons, seek guidance from your own legal counsel or from the SOA’s Execu+ve Director or legal counsel.

2
Presenta.on Disclaimer

Presenta(ons are intended for educa(onal purposes only and do not
replace independent professional judgment. Statements of fact and
opinions expressed are those of the par(cipants individually and,
unless expressly stated to the contrary, are not the opinion or
posi(on of the Society of Actuaries, its cosponsors or its commi=ees.
The Society of Actuaries does not endorse or approve, and assumes
no responsibility for, the content, accuracy or completeness of the
informa(on presented. A=endees should note that the sessions are
audio-recorded and may be published in various media, including
print, audio and video formats without further no(ce.

3
The New Valua.on Manual and the
Life Product Development Actuary

Overview of Product Development Processes that Might Change
Wed, May 10, 2017
Leonard Mangini, FSA, FRM, FALU, CLU, MAAA

President and Managing Member
Mangini Actuarial and Risk Advisory LLC
Presenter Biography- Leonard Mangini
Leonard Mangini, FSA, FRM, FALU, CLU, MAAA
President and Managing Member, Mangini Actuarial and Risk Advisory LLC

Mr. Mangini brings clients over 27 years of life, annuity, and health insurance and reinsurance
exper+se, holding senior Financial, Risk Management, Reinsurance and Product-related roles at
Manulife, ACE, AXA, and USLIFE. Leonard has also assisted clients with Financial Repor+ng, Risk
Management, Underwri+ng, Product Development, Reinsurance, M&A, and Market Conduct
issues as a consultant with E&Y, Milliman, and now his own actuarial and risk advisory prac+ce.

Prior to forming Mangini Actuarial and Risk Advisory LLC, Leonard was SVP and Deputy Global
Corporate Chief Actuary at Manulife. This role included supervising Principle-Based valua+on
assump+on and margin se`ng, explaining Principle-Based earnings to stake-holders,
supervising Principle-Based valua+on governance in an ORSA environment for 19 business units
in the US, Canada, and Asia and serving on Global Product Risk and Global ALM CommiIees.

In previous reinsurance roles, Leonard has served as an internal Board member, President, Chief
Actuary, Chief Pricing Officer, and Chief Risk Officer, including green-fielding a US life reinsurer.

Mr. Mangini previously served as Chair of the SOA Financial Repor+ng Sec+on Council, is a
Member of the Academy’s PBR Life Reserve Work Group, a Member of the Academy’s
CommiIee on Professional Responsibility, and serves as Chair of the SOA Oversight Group on
PBR Earnings AIribu+on. Leonard is a Fellow of the Society of Actuaries (FSA), Cer+fied
Financial Risk Manager (FRM), Fellow of the Academy of Life Underwri+ng (FALU), and Member
of the Academy of Actuaries (MAAA), qualified to sign Statements of Actuarial Opinion (PSAOs).

New VM and the Life Product Product Development Actuary Seminar May 10, 2017
5
Liability Disclaimer, Copyright, Use of Slides
Although we’ve aIempted to capture the leIer and spirit of the Valua+on Manual and ASOPs,
faithfully- you have a personal professional duty to familiarize yourself with the original source
material and apply professional judgment as to its specific applica+on to your own work and those
working under your direc+on as you perform covered Actuarial Services. The nature of your work,
and other professional designa+ons you hold, may require you to be bound by addi+onal
professional requirements from other organiza+ons as well.

This material has been prepared for general informa+onal purposes only and is not intended to be
relied upon as accoun+ng, legal, tax, or other professional advice, nor is it an Actuarial Opinion by
Leonard Mangini, Arnold Dicke, Tim Cardinal, Steve Stockman, or their respec+ve firms Mangini
Actuarial and Risk Advisory LLC, AADicke LLC, or Actuarial Compass LLC. Please refer to your
advisors for specific advice. The views expressed by the presenter are not necessarily those of
Mangini Actuarial and Risk Advisory LLC, AADicke LLC, or Actuarial Compass LLC.

Much of the original source material on VM-20/PBR and Professionalism is copyrighted material of
the American Academy of Actuaries, Society of Actuaries, or Na+onal Associa+on of Insurance
Commissioners. This presenta+on paraphrases these for educa+onal purposes to capture the
intent of the regula+ons and standards of prac+ce or results of SOA research, and every aIempt
has been made to iden+fy and cite original sources.

These slides may NOT be copied, redistributed, or otherwise furnished to any party without prior
wriIen consent of Mangini Actuarial and Risk Advisory LLC, AADicke LLC, or Actuarial Compass LLC
other than as required to comply with an audit of the aIendee’s annual CPE compliance.
The New ValuaLon Manual and the Life PD Actuary May 10, 2017
6
Cynics View: Product Development in the “Good Old Days”

Pricing/Product Development Interact With MarkeLngà Brilliant Idea

Product Almost Ready for Marketà Hey We Need Some Reinsurance!

Modeling?:
•  AssumpLons: We have ours, ValuaLon has theirs, Capital worries about tail risk
•  Reinsurance is ProporLonal (Coinsurance) or Replaces AssumpLons (YRT)
•  Reserves pop out of our actuarial soWware system

Agent Roll-Out MeeLng in 10 Daysà Call Up ValuaLon to Kick the Tires

ValuaLon Does Whatever They Do

Roll-Out Product

Tell Admin and ERM You Have a New Product

Big Success- Pat Ourselves on the Back

Get Around to DocumenLng Stuff…When Things Are Slow

VM and the PD Actuary ©2015-7 Mangini Actuarial and Risk Advisory LLC May 10, 2017

PBR Valua.on Is Easy!

VM and the PD Actuary ©AADicke LLC and Mangini Actuarial and Risk Advisory LLC May 10, 2017

PBR Valua.on Is Easy!
Just 28 Simple Steps…


PBR Valua.on Is Easy- Just 28 Simple Steps!...Slide 1
Determining Which Reserve Requirements Apply to Various Blocks of Policies
1.  Determine which policies are in scope of VM-20 (all but credit life, pre-need and industrial life)
2.  Determine if company qualifies (and wants to apply) for the company-wide exemp+on (Y or N)
3.  If YES, use VM-A/VM-C for all policies w/2017 CSO, (may use 2001 CSO un+l 2020)
4.  If NO, during 1st 3 years , determine which blocks to value under VM-20 (irreversible decision),
other blocks are valued under VM-A/VM-C (pre-PBR CRVM) , 2017 CSO may be Elected
Apply Exclusion Tests
5.  Calculate NPR for all VM-20 policies (NPR= VM-A/VM-C w/2017 CSO except for Term/ULSG)
6.  For ALL blocks of VM-20 policies, decide which SET test to use (SERT, Demo, Cer+fica+on) for each
block (any block passing NY 7 or VM-20 16 on a stand-alone basis can be cer+fied)*
7.  For any block for which SERT to be applied, build CF Model as per VM-20 Sec+on 7.B.2, with
an+cipated experience or using Cash Flows from Asset Adequacy Tes+ng
8.  IF SET is Failed or NOT applied, SR or DR MUST be calculated (if CDHS- then deemed to fail)
9.  For “Other” policies, if SET passed then apply DET
10.  For Term and ULSG and “Other” (where DET is Failed), DR must be calculated

* SET must be applied separately for Term, ULSG, “Other”. Cannot be grouped for SERT if “significantly different risk profile”
VM and the PD Actuary ©2016-17 AADicke LLC and Mangini Actuarial and Risk Advisory LLC May 10, 2017
Modeling policies requiring a DR or SR:
11.  Determine Model Segments- VM Sec+on 7A- align with insurer’s asset segmenta+on
plan, investment strategy, how allocate investment income for statutory report
12.  Build CF model- seria+m of in scope inforce or grouped cells as per VM Sec+on 7.B.2
13.  Determine an+cipated experience assump+ons for each non-stochas+c, non-
investment risk factor (if not already determined for SERT), unless its PBE is prescribed
14.  Determine PBE assump3ons- add margins to an3cipated experience assump3ons*
15.  Select star+ng assets for each segment
16.  Determine investment expense and default assump+ons, by model segment
17.  Calculate SR using VM-20 Sec+on 5 methodology (if required)
18.  Calculate DR using VM-20 Sec+on 4 methodology
19.  Calculate aggregate modeled reserve = ∑ max (SR, DR) over all model segments
20.  If aggregate star+ng assets are either < 98% of aggregate modeled reserve or
> max (102% of aggregate modeled reserve, aggregate NPR),
provide assurance that min reserve is not understated (or re-select star+ng assets)
* Mortality is a lible more complicated


Apply Sec.on 2 Aggrega.on
21.  (Aggregate) minimum reserve (MR) = (1) + (2) + (3)+(4)+(5)+(6)+(7):

(1) For group of Term policies, SET passed: MR= max (DR*,ANPR)
(2) For group of Term policies, SET failed: MR= max (SR*,DR*,ANPR)
(3) For group of ULSG policies, SET passed: MR= max (DR*,ANPR)
(4) For group of ULSG policies, SET failed: MR= max (SR*,DR*,ANPR)
(5) For group of “Other” policies, SET, DET pass: MR= ANPR
(6) For group of “Other” policies, pass SET, fail DET: MR= max (DR*,ANPR)
(7) For group of “Other” policies, fail SET, DET: MR= max (SR*,DR*,ANPR)
22.  Due and Deferred Premium Asset, DDPA: offse`ng statutory asset
where “surplus reduc+on” is max[ ANPR - DDPA, DR, SR]

ANPR= aggregate NPR= ∑ ( NPR) - ∑ ( SSAP61 reinsurance credits)
DR*= DR + DDPA (axer reinsurance) , SR*= SR +DDPA (axer reinsurance)

Apply Sec.on 2 Alloca.on
23.  Allocate within each of these 7 groups based on post-reinsurance NPR
24.  Min Reserve for group of policies = ∑ ( Allocated min reserve) for policies in group
25.  REPEAT ENTIRE PROCESS (starLng from Step 6) excluding ceded reinsurance to
determine the Pre-Reinsurance Ceded” minimum reserve
26.  Use allocated post- and pre-reinsurance-ceded reserves to develop minimum reserve
by product type reported in VM-20 Reserve Supplement
Perform Recalcula.ons for PBR Actuarial Report
27.  Repeat DR CalculaLon without material margins and without all margins to quanLfy
impact of material margins and of aggregate margins on MR for PBR Actuarial Report
28.  At least once every 3 years, (or if material change in risk profile) calculate stand-alone
SR for each product and compare to the minimum reserve for those products

Step 25: Reinsurance

VM-20 Sec.on 8 (CF Model for DR and SR) must reflect reinsurance cash flows
•  Based on Ceding Company assumpLons
•  AssumpLons for cedant and reinsurer different- unless companies agree to share
•  Reinsurer’s Reserve for Assumed Business—using Reinsurer’s assump3ons
ANPR is sum of individual policy NPRs less SSAP 61R credit
PBR Reserve is calculated before and aWer reinsurance
•  Pre- and post-reinsurance-ceded minimum reserves MAY be on different bases—
e.g. SR for post- and DR for pre-reinsurance, both SR but different scenarios in CTE 70

Reinsurance Takeawaysà:
•  PBR eliminates mirror reserving!
•  Early Product Development process CooperaLon with mul3ple external par3es
•  Need to examine underlying risk drivers in product designs for “quirky” behavior


14
Reinsurance Process Conclusion

I guess we’re calling our reinsurers earlier in the process!


Step 14: Prudent Es.mate (PBE) Assump.ons

Material Risks that Company has influence over

•  Mortality, Morbidity, Lapses, ParLal Withdrawals
•  Premium Persistency, Loans, Rider/OpLon ElecLon
•  NGE such as Interest CrediLng, Expenses
•  Pending PBR ASOP: reflect PH viewpoint of policy value and embedded op3ons
•  Pending PBR ASOP: reflect expected Management Ac3ons based on ERM in prac3ce
Actuary’s Prudent EsLmate (PBE) of anLcipated future experience, factoring in

credibility- periodically reviewed and updated
•  Margin for Adverse DeviaLon and EsLmaLon Error
•  Margin Magnitude reflects uncertainty in experience
•  Must include element of conservaLsm
•  Must include margin on each assumpLon


VM-G: Governance Guidelines…

•  PBR Assump+ons, Methods, Models must be consistent with company’s risk
management processes
•  Board must exercise oversight, review summary results, process documenta+on,
ac+ons to rely on processes
•  Management must provide info to Board, review results, implement necessary
controls, ensure adequate resources, competent staff, and maintain func+onality
•  Qualified actuary: legal requirement oversight over PBR calcula(on; reviews and
approves assump(ons, methods, models, controls; PBR Actuarial Report
•  Appointed Actuary legally opines on PBR, non-PBR reserves

àNeed to call in Valua.on and Risk Management earlier…!


Documenta.on of Assump.ons in Pricing

DocumentaLon for Reserves- that’s ValuaLon’s Problem- Right?...


Documenta.on of Assump.ons in Pricing

Documenta.on for Reserves- that’s Valua.on’s Problem- Right?...
•  Where did ValuaLon and Risk Management get their experience studies?
•  Who set the anLcipated experience assumpLons (without margins)?
•  Who decided on the material risks to use in pricing as inputs?
•  Who analyzed underwriLng and risk classificaLon? Who complied with ASOP 12?
•  Who applied credibility theory to input data? Who complied with ASOP 25?
•  Who spoke to reinsurers and decided on assumpLons gross and net?

It may have been YOU or your staff in Pricing and Product Development!

Actuarial Standard of PracLce 23
20
Revised Actuarial Standard of PracLce 23
Data Quality
Applies to any actuary using dataà clearly when preparing/reviewing PBR Report
ASOP recently revised:
•  Defines appropriate data as data suitable for intended purpose. Requires disclosure if
regulaLon/ other circumstances require use of “unsuitable data”
•  “Data” now defined to include data mathema.cally derived from other data
•  Actuarial Services provided on or aRer April 30, 2017 that “create data” relied on or
used by others MUST apply data quality ASOP 23 as if they were the end users!
•  Requires actuaries using data to document limitaLons and impact of uncertainty or bias
in data on actuarial work product

à QA legally charged with verifying appropriateness of assumpLons may rely on you
à ASOPs 21: Auditors and State Examiners are end users of data, may ulLmately rely on you
à VM requires uncertainty reflected in larger margins and NOT in assumpLons themselves
21
VM-30- New AOMR Requirements

AOMR…I’m a Pricing Actuary…Not the Appointed Actuary…

Stop Already!


VM-30- New AOMR Requirements- Reliance on Others
VM-30 SecLon 3.A.6:
“In forming my opinion on [specify types of reserves], I relied upon data, assumpLons,
projecLons, or analysis prepared by [name and Ltle each expert providing the data,
assumpLons, projecLons, or analysis] as cerLfied in the abached statements. I evaluated that
data, assump3ons, projec3ons, or analysis for reasonableness and consistency. I also
reconciled data to the extent applicable to [list applicable exhibits and schedules] of the
company’s current annual statement. In other respects, my examina3on included review of the
assump3ons projec3ons, and analysis used and tests of the assump3ons, projec3ons, and
analysis I considered necessary. I have received documenta3on from the experts listed above
that supports the data, assumpLons, projecLons, and analysis.”

à Opinions for 12/31/17 Appointed Actuary can’t “merely rely” on others in AOMR,
regardless of whether valuaLon is being performed using PBR or “Old CRVM”:
Probably will be coming to YOU for documenta3on and support of assump3on se\ng,
projec3ons and analysis and will be reviewing and tes3ng Pricing’s work to the extent
that it was used in helping set PBR valua3on assump3ons!


Pricing Sohware and Modeling Implica.ons

Margins, the VM-31 PBR Actuarial Report, and Internal Stakeholders
•  Intent of Exhibits in PBR Actuarial Report is to show “dollar amount” of margins so relaLve
size compared to minimum PBR Reserve is transparent to various stakeholdersà can
judge degree of conservaLsm embedded in PBR reserves
•  Opera9onally, must repeat DR calculaLon without material margins and without all
margins to quanLfy impact of material margins, aggregate margin in reserves
•  Combined with need to have Pre- and Post- Reinsurance model runs to quanLfy
reinsurance reserve credit highlights importance of model run-Lme issues

Takeawaysà
•  Simple models to streamline mulLple pricing and valua3on iteraLons, VM-31
•  ValuaLon must be conversant, aligned on reinsurance, involved early in PD, may need to
train your colleagues on the intricacies of your treaLes and involve in negoLaLon
•  QA, ERM must be involved closely and early in model risk management


24
Who should Maintain Models?

Each FuncLonal Unit?

Centralized Teams with Pricing, ValuaLon and ERM clients?


25

Internal Hedging: Company Asset Segmenta.on
And ERM Plans Impact PBR Reserves


26

Strategic Decision: “Internal Hedging” and Risk Offsets
Company Asset Segmenta.on and ERM Plans Impact PBR Reserves
•  VM S7.A: Requires aligning model segments for DR/SR with insurer’s asset
segmentaLon plan and how investment income allocated in statutory financials
•  VM S5: Governs permibed aggregaLon of SR for Risk Offsets
•  Term
•  ULSG
•  Other
•  VM S7.B.3: Requires AggregaLon Sub-Groups ~ Integrated Risk Management

Risk Offset Takeaways à
•  Evaluate which Segments Provide Opportunity for Self-Hedging
•  Consider Revising Asset SegmentaLon and ERM to OpLmize Risk Offsets
•  ERM must be involved closely and early


27

Strategic Decision- Should We Delay Implementa.on?
A Quick Comment on Taxa.on

28
Final Product Development Process in a PBR World?
Pricing/Product Development Interact with ValuaLon and RM from the Start
Keep in Contact with Tax Department, not just worried about 7702 issues…
Document Data, Credibility, AssumpLons, Inputs, Decisions, Models as You Go
Reinsurers called in Early: Provide you with AssumpLons? Reserve Credit Info?
Investment Department, CFO, and ERM to OpLmize AggregaLon Offsets?
Centralized Internal Modeling Teams that have Business Unit/FuncLonal Clients?

“One Version of The Truth”- Best EsLmate AssumpLons agreed upon by Pricing,
ValuaLon, ERM, Capital/Solvency a a Team?
Different Margins around these Best EsLmates for Different Purposes?
•  US GAAP, PBR Statutory, ERM and Solvency Capital
Feedback Loops for Unlocking AssumpLons

Data Repository for Storing AssumpLons, SensiLviLes, Experience Study Freshness?
Explaining Profits- Release of Margins and Actual versus Best EsLmate AssumpLons

Contact Informa+on
Leonard Mangini, FSA, FRM, CLU, FALU, MAAA
President , Mangini Actuarial and Risk Advisory LLC
E-mail: leonard@manginiactuarial.com
Web: www.manginiactuarial.com
Mobile: (516) 418-2549

30
The New Valua+on Manual and the Life
Product Development Actuary Seminar:
The Role of the Company, Management, Qualified Actuary,
Appointed Actuary in PBR Governance under VM-G
President, Mangini Actuarial and Risk Advisory LLC
May 10,2017
Liability Disclaimer, Copyright, Use of Slides
Although we’ve aIempted to capture the leIer and spirit of the Valua+on Manual and ASOPs,
faithfully- you have a personal professional duty to familiarize yourself with the original source
material and apply professional judgment as to its specific applica+on to your own work and those
working under your direc+on as you perform covered Actuarial Services. The nature of your work,
and other professional designa+ons you hold, may require you to be bound by addi+onal
professional requirements from other organiza+ons as well.

This material has been prepared for general informa+onal purposes only and is not intended to be
relied upon as accoun+ng, legal, tax, or other professional advice, nor is it an Actuarial Opinion by
Leonard Mangini, Arnold Dicke, Tim Cardinal, Steve Stockman, or their respec+ve firms Mangini
Actuarial and Risk Advisory LLC, AADicke LLC, or Actuarial Compass LLC. Please refer to your
advisors for specific advice. The views expressed by the presenter are not necessarily those of
Mangini Actuarial and Risk Advisory LLC, AADicke LLC, or Actuarial Compass LLC.

Much of the original source material on VM-20/PBR and Professionalism is copyrighted material of
the American Academy of Actuaries, Society of Actuaries, or Na+onal Associa+on of Insurance
Commissioners. This presenta+on paraphrases these for educa+onal purposes to capture the
intent of the regula+ons and standards of prac+ce or results of SOA research, and every aIempt
has been made to iden+fy and cite original sources.

These slides may NOT be copied, redistributed, or otherwise furnished to any party without prior
wriIen consent of Mangini Actuarial and Risk Advisory LLC, AADicke LLC, or Actuarial Compass LLC
other than as required to comply with an audit of the aIendee’s annual CPE compliance.
The New VM and the PD Actuary © Mangini Actuarial and Risk Advisory LLC May 10, 2017
32
Agenda: VM-G and PBR Governance
Why am I hearing this first before today’s other sessions?

Governance Provides Context for Roles and Responsibili.es
•  Who is Required to Do What?- Some are guidance some legal requirement
•  Where do you have leeway?
•  What must be disclosed?
The Players and Their Roles:
•  Regulators vs. Company- and the duLes of actuaries
•  Board- VM-G SecLon
•  Senior Management- VM-G SecLon
•  One or more Qualified Actuaries- VM-G SecLon
•  Role of Qualified Actuary vs. Appointed Actuary
•  Other Actuaries and Employees, Intended Users and Stakeholders
33

The Roles of the Regulator and Company
Under the Revised SVL

34
Background: Standard ValuaLon Law

“The commissioner shall annually value, or cause to be valued, the reserve liabiliLes
(hereinaWer called reserves) for all outstanding life insurance contracts, annuity and
pure endowment contracts, accident & health contracts, and deposit-type
contracts of every company issued on or aWer the operaLve date of the…manual.”

à  Except for insolvency or authorized control, usually the Commissioner causes
the Company to value reserves. The Company has the responsibility for selecLng
assumpLons, methods and models used in a PBR ValuaLon.
à  Commissioner can hire independent actuary to do valuaLon if not saLsfied with
the Company’s work in establishing reserves
VM and the PD Actuary ©2016-17 AADicke LLC- used with permission May 10, 2017
2016 Valua+on Actuary Symposium August 2016
VM-G: Company Assigns ResponsibiliLes to
One or More Qualified Actuaries

In carrying out the responsibility for each group of policies and contracts
subject to a principle-based valuaLon, the company shall assign to one
or more Qualified Actuaries certain responsibiliLes listed later in VM-G

VM and the PD Actuary ©2016-17 AADicke LLC-

2016used withActuary
Valua+on permission
Symposium May 10, 2017
August 2016
Summer 2016: LATF Request for Comment

LATF requested comments on whether Qualified Actuaries assisLng or
advising the Company regarding its responsibility to establish principle-
based reserves under VM-20 represent the interest of the Company or
the interest of the Commissioner.

Under VM-G, one or more Qualified Actuaries are essenLally overseeing
work that has been assigned to the Company by the Commissioner.

So whose interest does the Qualified Actuary represent?

ACLI Response to LATF Request
Points to legal status of employeesà have duty to their employer
Also the Actuarial Code of Conduct- “Principal” is employer or client

Code of Conduct outlines responsibiliLes of Actuary to his/her Principal
•  Precept 5: Must idenLfy Principal in communicaLons
•  Precept 7: Must disclose conflicts of interest
•  Precept 10: Must cooperate with others (including Commissioner) in Principal’s interest

AAA Life PracLce Council Response
Interest NOT clearly definedà response doesn’t speak directly to LATF’s request
Points to Code of Conduct and ASOPs
Code of Professional Conduct:
•  Precept 1: must fulfill profession’s responsibility to public, uphold profession’s reputaLon
and must NOT provide services that violate or evade the law
•  Precept 3: must saLsfy applicable ASOPs
•  Precept 7: “Actuarial CommunicaLon” must be clear and appropriate to circumstances
and intended audience (interests of Company, Commissioner could be served differently)
ASOP 41 (CommunicaLons): defines “Intended User”: Any person whom actuary
idenLfies as able to rely on the actuarial findings (clearly includes Commissioner)
If assumpLons or methods chosen by someone other than the actuary, 3 choices:
•  If actuary finds them “reasonable,” no disclosure is required
•  If “significantly conflict” with what actuary finds reasonable, disclose fact and who set them
•  If the actuary is unable or not qualified to judge reasonableness, disclose that fact

VM and the PD Actuary ©2016-17 AADicke LLC-

2016used with
Valua+on permission
Actuary Symposium May 10, 2017 August 2016

So Whose Interest Does the QA Represent?

Responses were reviewed by LATFà no acLon was taken

Two PracLcal Ways to View PBR AssumpLon Seyng!
1) Company sets assumpLons
•  Do the actuaries (QAs, AAs, others) agree? If not, document and disclose per ASOP 41
•  QA under VM-G ensures complies with VM
•  AA under VM-30 tests and opines on whether ensuing reserves are truly adequate
à Range: Low and High?? à Actuaries pick what they feel comfortable with
2) An actuary sets assumpLons on behalf of Company

•  Self-police that comply with Code of Conduct, ASOPs, laws, regulaLons- document per ASOP 41
•  QA under VM-G ensures complies with VM-20 or VM-21
•  AA under VM-30 tests and opines on whether ensuing reserves are truly adequate
à  Range: Low and High?? à Company picks what they feel comfortable with
Both are defensible implementaLons since ulLmately the Company has chosen the assumpLons,
models and methods; a QA and AA have fulfilled their legal roles and ASOP disclosure duLes; and if
the two parLes don’t agree there is documentaLon about how and why for intended users!

The Board’s Role
Under the Revised SVL

42
VM-G DefiniLon of the Board
VM-G Sec.on 1.C.2: The term “board” and “board of directors” means:

(a) the board of an insurance company that has not been designated to be part of a
group of insurance companies, or
(b) the board of a single company within a group of insurance companies that is
designated by the senior management of any holding company of all the insurance
companies in such group of insurance companies, or a commibee of such board,
consisLng of members of such board, duly appointed by such board and authorized
by such board to perform funcLons substanLally similar to those described in this
secLon
The New VM and the PD Actuary May 10, 2017
43
VM-G: Guidance for the Board
VM-G Sec.on 2: Commensurate with materiality of PBR reserves in to overall
company risk and consistent with oversight role, Board is responsible for:
•  Overseeing processes of Senior Management to idenLfy/correct material weaknesses
•  Overseeing infrastructure (policies, procedures, controls, resources) to implement PBR
•  Receiving/Reviewing Qualified Actuaries Reports/CerLficaLons under VM-G SecLon 3.A.6
•  InteracLng with Senior Management to resolve quesLons
•  DocumenLng Board acLons related to PBR in Board Minutes
à ImplementaLon Plan: will go the Board

à Processes/Controls: Model Audit Rule (MAR)/Sarbanes (SOX) due to “material weakness”
languageà Stakeholders: ERM/CRO, Controller/Internal Audit, External Audit, Regulator

44

Senior Management’s Role
Under VM-G

45
VM-G DefiniLon of Senior Management
VM-G Sec.on 1.C.3:

“includes the highest ranking officers of an insurance company or group of
insurance companies with responsibiliLes for operaLng results, risk assessment,
and financial reporLng (e.g., the chief execuLve officer, chief financial officer, chief
actuary, and chief risk officer) and such other senior officers as may be designated
by the insurance company or group of insurance companies”
The New VM and the PD Actuary May 10, 2017
46
VM-G: Guidance for Senior Management
VM-G Sec.on 3: Responsible for direcLng implementaLon and ongoing operaLon
of PBR valuaLon funcLon:
•  Ensuring infrastructure (policies, procedures, controls, resources) to implement PBR
•  Reviewing PBR AssumpLons, Methods, Models for consistency with other risk processes
•  Reviewing and assessing significant unusual issues and findings of PBR valuaLon and
sensiLvity tests
•  Ensuring adopLon of internal controls that all material risks accounted for, comply with
VM and actuarial standards; annual evaluaLon of controls à report to Board
•  Adequate resources for modeling with skill and competence, process to ensure SVL
compliance, process to validate data for non-prescribed assumpLons, process to ensure
appropriate model input assumpLons, process to find and limit material errors/weakness,
and adequate reporLng on controls
•  Annual report to Board on infrastructure adequacy, PBR vs. ERM, knowledge and
experience of Senior Management that monitor and control PBR
à Puts Senior Management into ORSA/Solvency II role as Pillar 2 “guardians”


47

Qualified Actuary’s Role
Under VM-G

48
“Qualified Actuary” vs. Appointed Actuary

VM-01 definiLons:
The term “qualified actuary” means an individual who is qualified to sign the
applicable statement of actuarial opinion in accordance with the American
Academy of Actuaries qualificaLon standards for actuaries signing such
statements and who meets the requirements specified in the ValuaLon Manual.
The term “appointed actuary” means a qualified actuary who is appointed or
retained in accordance with the ValuaLon Manual to prepare the actuarial
opinion required in SecLon 3A of the Standard ValuaLon Law (VM-05).
Qualified Actuary is assigned responsibility for a group of policies
•  VM-Gà Legal responsibiliLes, not merely guidance


Qualified Actuary Legal DuLes
VM-G SecLon 4:
Oversee calculaLon of PBR reserves for that group of policies
Verifying compliance with VM
•  AssumpLons, methods, models
•  Company documented internal standards for processes and
•  Controls over those processes
CerLfy that AssumpLons are Prudent EsLmates with appropriate margins
Provide Summary Report to Board and Senior Management
•  ValuaLon Process
•  ValuaLon Results
•  General Level of ConservaLsm
•  Materiality of PBR to total reserves of company
•  Significant or unusual findings
Prepare PBR Actuarial Report under VM-31

50
QA InteracLons with External Audit and Regulators
Prepare the VM-31 PBR Report

Disclose significant unresolved issues
NOT cerLfying adequacy of aggregate reserves for group of policies, the
company’s surplus, or Company’s financial condiLon
à that is sLll the role of the ONE Appointed Actuary

51
Model Audit Rule- ReporLng Requirements
QA thrown into “Controls World” fortunately enters into an Exis.ng Framework
•  Model Audit Rule (MAR) LegislaLon- eff. 1/1/2010 (or when domicile adopts)
•  Detailed Requirements- Controls, ReporLng on Controls for Statutory AccounLng
•  SecLon 17: $500 Million Direct/Assumed Premium Threshold on Report on Controls
•  SecLon 17.D.5: Defines two terms- “Significant” and “Material Weakness”
•  Former- warrant abenLon of governance, Laber- reasonable probability of material misstatements
•  Insurer MUST report “Material Weakness” to domicile within 60 days Audited Financials
•  Model Law DOES NOT prescribe parLcular framework for review/evaluaLon of controls
•  MAR Guide indicates most SEC registrants adopt COSO Internal Control-Integrated framework and
that COSO ERM-Integrated framework and PCAOB Guidance for Smaller Companies are relevant
Available Resources:
•  NAIC Audit Rule Implementa.on Guide (MAR Guide)- clarifies MAR (without changing contents)
•  Academy MAR Prac.ce Note (Nov 2010)- guidance preparing documentaLon/controls for reserves

52
Contact Informa+on
President , Mangini Actuarial and Risk Advisory LLC

E-mail: leonard@manginiactuarial.com
Web: www.manginiactuarial.com
Mobile: (516) 418-2549

53
VM-20 Mortality Assumption Example
The New Valuation Manual and
Life Product Development Actuary Seminar
Carrie Kelley, FSA, MAAA
May 10, 2017
willistowerswatson.com
© 2017 Willis Towers Watson. All rights reserved.
Agenda
 Prescribed Mortality
 Prudent Estimate Mortality Assumptions
 Potential Challenges
willistowerswatson.com 2
© 2017 Willis Towers Watson. All rights reserved. Proprietary and Confidential. For Willis Towers Watson and Willis Towers Watson client use only.
Introduction to VM-20 Mortality
Introduction to VM-20 mortality
2 mortality assumption sets: Net Premium Reserve’s prescribed assumptions

and the modeled reserves prudent best estimate
 2017 CSO and 2017 CSO Preferred Structure

Prescribed Mortality (2001 CSO tables are available during transition)
 GI/SI: 2017 table in development for use
 Company experience
Prudent Estimate  Margin
Mortality  Industry tables
Mortality for VM-20 Products
VM-20 § Application
 3.C.1 - Mortality for Ordinary Life policies to be used in the Net Premium Reserve
calculation
Net
Premium  Company may elect either the 2001 CSO or 2017 CSO for issues on or after 1/1/2017 but before
1/1/2020
Reserve
 2017 CSO for issues on or after 1/1/2020
 Conditions for use of the 2017 CSO Preferred Structure Table
Non-  Mortality defined in VM-02

forfeiture  Company may use 2001 CSO or 2017 CSO following the above transition rules
Exclusion  Anticipated experience assumption (a component of prudent estimate assumptions)

Testing  Assumption used for asset adequacy testing
Modeled
Reserves
 Section 9.C - Prudent estimate mortality
Prudent Estimate Mortality
Assumptions
Prudent estimate mortality assumptions
Overview of basic components
 Prudent estimate assumptions are company experience + margin

 Mortality prudent estimate assumptions have additional requirements
 Prescribed steps for development of company experience, margins, and use of industry
tables
 Grade company experience with industry experience based on credibility of company

data
Prudent estimate mortality assumptions
Step by step process
Step 1: Step 2: Step 3:

Determine Determine Determine
Company Own margins
Mortality Company based on
Segments Experience credibility Loaded Step 5:
company table Grade to
industry table Prudent
based upon Estimate
credibility & Mortality
Step 4: Loaded sufficient
Applicable industry table data period
industry
mortality table
 Add margins
for industry
table
VM-20 prescribed approach to set mortality assumptions
Breaking down by step
1 Mortality segment
2 Company experience rates
3 Credibility and experience margins
4 Industry mortality table and margins
5 Blending company and industry
Step 1: Mortality segment
Getting started
Expect mortality to vary by several factors, possibly including:

 Risk class
 Smoking status
 Face amount band
 Gender
 Distribution channel
 Underwriting type
Example: Risk class and gender are key variables
Step 2: Company experience rates
Tools and data necessary
 Determined from company’s own experience where available

 Actual company experience for books of business in the segment
 Experience from other books of business with similar underwriting
 Experience from other sources, if available and appropriate
 Company experience data:

 Must be at least three exposure years and no more than ten
 Must be updated every three years
 Must be adjusted to reflect significant changes that are expected to persist
Example
Consistent with segment definitions, split experience by smoking status and

gender
% A / E Based on 2015 VBT (SM / NS)

Smoking Female – Male –
Status Female – Raw Male – Raw After HMI* After HMI*
Non-smoker 80% 85% 78% 82%
Smoker 100% 110% 97% 107%

Figures are illustrative, not based on any company experience
*1% improvement for 3 years
Example
Split FNS across risk classes
% A / E Based on 2015 VBT
Relative Experience
Risk Class Raw A/E Weight Factor Factor
Best NS 59% 45% 80% 62%*
2nd Best NS 83% 35% 100% 78%
Residual NS 111% 20% 145% 113%
Total NS 78% 100% 100% 78%

Figures are illustrative, not based on any company experience
*62% = 80% * 78%
Note: Make sure weighted average of experience factor is at least as great as

aggregate factor
Step 3: Level of credibility
Selecting method
 Credibility impacts margins and the industry grading period

 Credibility may be determined on an aggregate level
 Limited Fluctuation or Bühlmann Empirical Bayesian Method by amount
 Parameters of 95% minimum probability and 5% maximum error margin
 Parameters and method mandated for valuations using the 2015 VBT
Bühlmann Credibility Results

Smoking Credibility – Credibility – Male
Status Female
Non-Smoker 72% 100%
Smoker 33% 49%

Note: Figures are illustrative, not based on any company experience
Step 3: Prescribed mortality margin
Example
Attained Credibility: Credibility: Credibility: Credibility: Credibility:

… …
Age 0 - 7% 63 - 67% 68- 72% 73 - 77% 99+%
<45 20.4% … 13.7% 12.7% 11.6% … 2.3%
46 - 47 20.2% … 13.7% 12.7% 11.6% … 2.3%
48 - 49 20.0% … 13.5% 12.5% 11.4% … 2.3%
… … … … … … … ….
Step 4: Applicable industry table
Selecting the right one
 Company experience grades into the applicable industry table over time
 The industry table is the 2015 VBT for issues on or after 1/1/2017
 A modified industry table may be permitted for limited situations where an industry basic
table is not appropriate: simplified underwriting or substandard lives
 The underwriting criteria scoring procedure may be used to select the appropriate
industry table for mortality segments
 The SOA’s RRTool can be used to determine the appropriate RR tables
 Actuaries may apply judgement to adjust up or down two tables to account for factors not
recognized in the tool
 Historical mortality improvement is allowed based on the prescribed mortality
improvement table
Step 4: Applicable industry table
Example
 Used 2015 VBT RR Tables

Risk Class RR Table
Best NS 70
2nd Best NS 90
Residual NS 120
 2015 VBT Margins

Attained Age Load
≤45 20.4%
46 - 47 20.2%
48 - 49 20.0%
… …
Step 5: Prudent assumption
Sufficient data period
 Sufficient data period is the last duration of experience with more than 50 claims
 The sufficient data period impacts how long company experience can be used
before grading to the industry table
Duration Number Deaths Duration Number Deaths
1 43 6 132
2 57 7 114
3 75 8 84
4 118 9 52
5 147 10 31
Credibility over exposure period = 72%
Determine grading parameters
Credibility of Maximum # of Maximum # of Maximum # of years in which the

company data years for data years in which assumption must grade to 100%
over sufficient to be considered to begin grading of an applicable industry table
data period sufficient after sufficient (from duration where sufficient
data no longer data no longer exists)*
exists
20-39% 10 2 8*
40-59% 20 4 12*
60-79% 35 7 17*
80-100% 50 10 25*
* The maximum # of years in which the assumption must grade to 100% of an applicable industry table shall be the lesser
of
(a) the appropriate number of years stated in the chart above, or
(b) the number of years of sufficient data + 15 times the credibility percentage applicable to column (1) in the above chart
Note: Different grading parameters specified in VM-20 for valuations prior to 1/1/2017
Summary of parameters
Item Value
 Credibility 72%
 Last Duration with Sufficient Data 9
 Max Yrs to be Sufficient 35
 Max Yrs to Begin Grading after last 7

significant duration
 Max Yrs to Grade to 100% of Industry Min (17, 9 + 15 * 72% = 19.8) = 17

 Duration to Begin Grading 16
 First Duration at 100% of Industry 26
Summary of rates FNS, Issue Age 32, Preferred NS
Experience Experience Experience Experience Industry Industry Industry Final

Duration
Factor* (% of 70 RR) Margin Weight Factor*** Margin Weight Rate
1-14 62% 62% * X** 12.70% 100% 100% 20.4% 0% 70% * X
15 62% 82% 12.70% 100% 100% 20.2% 0% 92%****

16 62% 82% 12.70% 100% 100% 20.2% 0% 92%
17 62% 82% 12.50% 90% 100% 20.0% 10% 95%
18 62% 81% 12.50% 80% 100% 20.0% 20% 97%
19 62% 81% 12.30% 70% 100% 19.8% 30% 100%
20 62% 81% 12.30% 60% 100% 19.8% 40% 102%
21 62% 81% 12.10% 50% 100% 19.6% 50% 105%
22 62% 81% 12.10% 40% 100% 19.6% 60% 108%
23 62% 80% 11.90% 30% 100% 19.2% 70% 110%
24 62% 80% 11.90% 20% 100% 19.2% 80% 113%
* Based on 2015 VBT 100 RR Table

** X = (100 RRx / 70 RRx)
*** % of the 70 RR Table
**** 92% = 82% x (1 + 12.7%) x100% + 100% x (1 + 20.2%) x 0%
Summary of rates FNS, Issue Age 32, Preferred NS
Experience Experience Experience Experience Industry Industry Industry Final

Duration
Factor* (% of 70 RR) Margin Weight Factor*** Margin Weight Rate
1-14 62% 62% * X** 12.70% 100% 100% 20.4% 0% 70% * X
15 62% 82% 12.70% 100% 100% 20.2% 0% 92%****

16 62% 82% 12.70% 100% 100% 20.2% 0% 92%
17 62% 82% 12.50% 90% 100% 20.0% 10% 95%

18 62% 81% 12.50% 80% 100% 20.0% 20% 97%
19 62% 81% 12.30% 70% 100% 19.8% 30% 100%
20 62% 81% 12.30% 60% 100% 19.8% 40% 102%
21 62% 81% 12.10% 50% 100% 19.6% 50% 105%
22 62% 81% 12.10% 40% 100% 19.6% 60% 108%
23 62% 80% 11.90% 30% 100% 19.2% 70% 110%
24 62% 80% 11.90% 20% 100% 19.2% 80% 113%
* Based on 2015 VBT 100 RR Table

** X = (100 RRx / 70 RRx)
*** % of the 70 RR Table
**** 92% = 82% x (1 + 12.7%) x100% + 100% x (1 + 20.2%) x 0%
Potential Challenges
Potential challenges
Form of company experience table
Mortality Experience A/E on Industry Table

An aggregate % of (e.g., 2015 VBT)
160%
an underlying table
may not properly
140%
capture emerging
experience.
120%
Our recommendation: 100%
Where available consider

using a company specific 80%
table, with adjustments,

instead of an industry 60%
table.
40%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
IA <45 IA 45-65 IA 65+
Note: Figures are illustrative, not based on any company experience
Aggregation for credibility
What the manual states

 The regulations allow for the setting of assumptions for mortality by segments; this
does not necessarily require setting of credibility on a segment level
 9.C.4.b - “Credibility may be determined at either the mortality segment level or at a
more aggregate level if the mortality for the sub-classes (mortality segments) was
determined using an aggregate level of mortality”
How to interpret
 Weighted average segment level assumptions should not result in an overall
assumption less than would have been assumed on the aggregate data; the
relationship between developed tables should make sense (NS/SM)
Credibility impacts
margins and sufficient data period
Appropriate amount and type of experience
Balance between
 Using all available experience (to increase credibility and sufficient data period)
 Using experience that is relevant to the particular block of business
Questions to determine the relevance of available experience

 Similar risk class structure?
 Similar underwriting era?
 Any significant underwriting concessions during data period?
 Similar product design?
 Similar distribution channel?
 Should conversions, PLTP, and/or other special business be included?
Projecting Mortality
The assumption will change as your company collects experience

 Credibility will improve as the company collects more claims data
 Sufficient data period will also be increasing
 Additional years of historical mortality improvement will be allowed
Considerations in how to project the assumption

 What is a reasonable estimate of future claims experience? Best estimate mortality?
 What changes will be necessary to the inner and outer loop projection?
 How material are these changes to my profit metrics?
 What simplifications can be made?
Summary
 Prescribed mortality for the NPR is very similar to other formulaic reserve regimes
 Prudent estimate assumptions are principles based, with prescribed methods and
margins
 There are many key considerations in developing your prudent estimate assumptions
Contact
Carrie Kelley, FSA, MAAA

Consultant
One Alliance Center, 3500 Lenox Road, Suite 900, Atlanta, GA 30326-4238
T +1 404 365 1595

E Carrie.Kelley@willistowerswatson.com
Reinsurance Considerations for Product Development
under PBR
The New Valuation Manual and Product Development Seminar
Seattle – May 10, 2017
Chris Whitney, FSA, MAAA
© 2017 Oliver Wyman

Agenda
1 Background
2 Case study
3 Discussion to date
4 Impacts on product development and pricing
5 Conclusion
© 2017 Oliver Wyman 1

Background
Section 8 of VM-20 pertains to the impact of reinsurance on the components

of reserves under PBR
Component Considerations in determining reinsurance impact
 No change as compared to pre-PBR

Net premium  Coinsurance: The NPR is reduced by the percentage coinsured
reserve
Maximum
 Yearly Renewable Term (YRT): The NPR is reduced by the unearned

cost of insurance that is reinsured
 Requires two separate calculations, pre- and post-reinsurance

Deterministic and
stochastic reserve  Exclusion testing, if elected, must be performed on a pre- and post-
reinsurance basis
 The starting asset collar does not apply to pre-reinsurance reserves

Final PBR reserve
 Credit = 𝑀𝑀𝑀𝑀𝑀𝑀 𝑁𝑁𝑁𝑁𝑁𝑁𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 , 𝐷𝐷𝐷𝐷𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 , 𝑆𝑆𝑆𝑆𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 𝑀𝑀𝑀𝑀𝑀𝑀 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 , 𝐷𝐷𝐷𝐷𝑁𝑁𝑁𝑁𝑁𝑁 , 𝑆𝑆𝑆𝑆𝑁𝑁𝑁𝑁𝑁𝑁
The reserve credit for reinsurance under PBR is significantly different from the
formulaic approach that insurers have become accustomed to
Background
Several sources of guidance exist for the modeling of reinsurance cash

flows. The guidance is non-prescriptive and takes the form of considerations
and required disclosures.
Source Guidance
 The actuary should assume that the counterparty is likely to act efficiently
 The assumptions used may differ between the ceding and assuming company
VM-20  Additional (outside the cash flow model) stochastic analysis may be required for
certain types of reinsurance (i.e. stop-loss)
 Considerations are similar to those for liability modeling
 Requires a description of assumptions and methodology used to model

VM-31
reinsurance cash flows
 Recommends consistency between reinsurance assumptions and other
assumptions
PBR ASOP  Margins should consider the guarantees in the arrangements, past practices of the
reinsurer and how the company might respond to different actions the reinsurer
could take
 States that “some actuaries will assume less than 100% selection against the
company”
AAA Practice note
 Recommends analyzing the financial impact on the reinsurer and assuming more
selection if the financial impact is significant
Case study
A cohort of new business with $50MM of first year premium consisting of 10-,
20- and 30-year term products was projected for 30 years
 30 year projection horizon

Model
 Reserve revalued annually
 Mortality follows 100% of 2015 VBT

Best estimate
 Mortality experience is 30% credible with 10 years of sufficient data
assumptions
 Expenses, commissions and lapses set at industry averages
Prudent estimate  Mortality is improved up to each valuation date at 1% per year

assumptions  100% shock lapse at end of level term period
 The NPR uses the 2017 CSO and a valuation interest rate of 4.5%
Reserve  DR scenarios are re-generated at each valuation date
assumptions  Starting assets at each valuation date use the ‘direct iteration’ approach
 The cohort is assumed to pass the Stochastic Exclusion Test (SET)
Assumptions used and products modeled are for an illustrative term portfolio
intended to be reasonably representative of products offered in the market today
Case study
The gross NPR and DR for this cohort of new business are shown below
200
Gross NPR
Gross DR
150
Reserve ($MM)
100
50
0
0 10 20 30
Duration (Years)
The DR starts much higher than the NPR, but the gap closes over time, partially
because mortality improvement to date is reflected at future valuation dates
Case study
Three 50 percent first dollar coinsurance agreements were modeled. The

coinsurance allowances were assumed to be guaranteed.
200
Net DR: Coinsurance 1
Net DR: Coinsurance 2
150 Net DR: Coinsurance 3
Net NPR
Reserve ($MM)
100
50
0
0 10 20 30
Duration (Years)
Coinsurance 1: Reimburse proportion of VM-20 prudent expenses and commissions
Coinsurance 2: Reimburse proportion of best estimate expenses and commissions
Coinsurance 3: Reimburse best estimate expenses and commissions, level % of premium

Case study
A 50 percent first dollar YRT reinsurance arrangement with the current

premium scale set equal to 100 percent of the best estimate mortality
assumption was modeled
200
Net DR: YRT Scenario 1
Net DR: YRT Scenario 2
150 Net DR: YRT Scenario 3
Net NPR
Reserve ($MM)
100
50
0
0 10 Duration (Years) 20 30
YRT Scenario 1: No change in rates
YRT Scenario 2: Change rates to eliminate any gain/loss from reinsurance
YRT Scenario 3: Increase rates by 15%

Case study
The difference in net reserves under the YRT scenarios modeled is driven by
the level of margin in the VM-20 mortality assumption
The result below is for 35-year-old male, preferred non-tobacco, time 1 valuation
9 100%
Excess of VM-20 over Best Estimate (%)

80%
7
Mortality Rate per thousand
6
60%
5
4
40%
3
2
20%
0 0%
0 10 20 30
Duration (Years)
VM-20/Best Est Best Est Mortality VM-20 Mortality
The mortality assumption under VM-20 contains no future mortality improvement

and is based on a company-specific prudent assumption grading to a prudent
industry table when sufficient data no longer exists
Case study
The gross and net reserves resulting from a hypothetical internal YRT
arrangement are shown below for illustrative purposes
Direct Writer Internal Reinsurer
6.0 6.0 6
6 5.8
5.0 (2.0)
(2.8) (2.3)
(1.0)
4.0 4.0
4 4
3.5
Reserve ($MM)
Reserve ($MM)
3.2
2.8 2.8
2.3
2 2
1.0
0 0
NPR DR SR Final PBR NPR DR SR Final PBR
The internal reinsurance arrangement increases net reserves by 13%. Total

reserves increased from $6MM to $6.8MM because of the arrangement.
Discussion to date
The results of the NAIC’s PBR company pilot project showed that seven
companies reported a reduction in post-reinsurance reserves, one reported
an increase and three only reported pre-reinsurance reserves
Gross and net of reinsurance reserves

led to many discussions and NAIC
staff will research the impact of PBR
reserves on accounting for
reinsurance.
– Mike Boerner, Chair of the NAIC’s Principle-Based Review (EX) Working Group

Discussion to date
The reduction in primary security requirement under AG 48 due to

reinsurance was limited to ½ Cx. Comments were received in support of and
against this change.
A ceding insurer might use one set of

assumptions to manufacture a large reserve
credit, while the reinsurer uses a different set
of assumptions to calculate a much smaller
reserve. A company potentially could exploit
this type of gap. We recommend that LATF
explore improvements to the Valuation Manual
that could mitigate the risk of this type of
gaming.
– John Finston, Chair, NAIC Reinsurance (E) Task Force

Impacts on product development and pricing
The following impacts are expected because of PBR and the associated
reinsurance considerations
Reinsurance type Impact
 Less attractive due to lower tax reserves and uncertainty in AG 48 financing

Financial solutions
 Carriers realizing a tax benefits may defer PBR election
 Stronger guarantees on YRT rates (i.e. lower cap, fully guaranteed)

 Additional disclosure on when/how reinsurance rates will be reviewed and
YRT
changed
 Industry convergence on level of prudence to assume
 Restructuring allowances or expense assumptions

Coinsurance
 Higher coinsurance allowances reflective of lower reserve requirements
 An important consideration will be the aggregate reserve impact

Internal
 Use depends on the aggregate reserve and capital impact

Conclusion
Key takeaways
1 Reinsurance has a more significant impact on pricing
Pricing models must be granular enough to properly account for the

2 asymmetry arising from reinsurance structures
It is critical for carriers to understand the pricing implications of

3 reinsurance early on in the pricing process
4 Implications of potential regulatory changes should be considered

Other Product Assumptions
The New Valuation Manual and Life Product
Kim Steiner, FSA, MAAA

May 10, 2017
Agenda
 NPR prescribed assumptions

 Prudent estimate assumptions
 Pricing specific considerations
Introduction to PBR assumptions
Focus on two assumption sets
Prescribed assumptions
 Methods consistent with prior formulaic
include mortality, lapse, reserve calculations
and discount rates
Prudent estimate  Assumptions are company experience

plus a margin
assumption methodology
 Some aspects of the mortality assumption
provides some guidance include a fair amount of prescription
but leaves a lot of room
 Other liability assumptions are less
for judgment prescribed
Prescribed Assumptions
The Details
Prescribed components – Interest rates
 Section 3.C.2 of VM-20 details required interest rates for the NPR calculation
 Interest rates are fixed based on calendar year of issue
 Same formula as current statutory reserves, but for products with no non-forfeiture
value, the rate is increased by a minimum of 1.5% or a factor of 125%
 I = .03 + W * (R1 - .03) + (W/2) * (R2-.09)

 Where: R1 is the lesser of R and .09
 R2 is the greater of R and .09
 R is the lesser of the average 36 month or 12 month monthly average of the composite yield
on seasoned corporate bonds as published by Moody’s
 W is a weighting factor for the policy based on guarantee duration
 I = interest rate
 Interest rates remain unchanged if they do not differ from prior year calculated
interest rates by 0.50%
Prescribed components – Lapse rates
 ULSG lapse rates are determined based the expected funding level of secondary
guarantees
 Lapse rates are 0% in the calculation of the ULSG floor (calculated as UL w/o SG)
 Term product lapse rates are prescribed and vary by duration and size of the
premium jump (for post level term period lapse rates)
Prudent Estimate Assumptions
Guidance
General guidance on prudent estimate assumptions
Section 9 of VM-20 – Anticipated Experience
1 Identify risk factors not prescribed or stochastically modeled
2 Anticipated experience
Use relevant and credible company experience; credibility blend

3
elsewhere
Where credibility theory is of limited use rely on other accepted actuarial

4
practices
5 Sensitivity testing
6 Annual review
General guidance on prudent estimate assumptions
Section 9 of VM-20 – Margin
 Set a margin for each assumption independently
 Should increase reserves
 Assume independent variables
 Level of uncertainty should impact margin
 Not required if variations in the assumption do not have a material impact
Specific liability assumptions
Category Assumption Types
Policyholder behavior Lapse, premiums, withdrawals/surrenders
Corporate Expenses and taxes
Asset Spreads, defaults, investment expenses
Product features COIs, crediting spreads, reinsurance etc.
Term Lapses, post level profits, conversions
ULSG Lapse, premiums, withdrawals/surrenders
Policyholder behavior assumption guidance
 Generic
 Vary by characteristics that will have a material impact on the modeled reserves
 Appropriate for scenarios that drive the modeled reserve
 Based on experience for block of business or similar
 Reflect historical experience only where experience is relevant to risk being modeled
 Reflect impact of policyholder options
 Appropriate level of granularity
 Use dynamic modeling based on scenario, where appropriate
Policyholder behavior assumption guidance
 Margins should consider  Necessary sensitivity tests

 Credibility of data  Minimum premium scenario
 Policyholder efficiency over time  No further premium payment scenario
 Whether experience is from modeled block  Pre-payment of premiums –
single premium scenario
 Marketing or administrative practices
 Pre-payment of premiums –
level premium scenario
Corporate – Expenses and taxes
 Selection of expense driver

 Certain IT and capital expenditures can be spread over a reasonable number of years,
 Use fully allocated expenses, assuming no expense improvements
 Experience should be used to determine appropriate assumptions
 For a new block of business use similar mature business, where available
 Include inflation using national inflation data
 Premium tax – State premium tax applicable to product
 Do not include income taxes
Assets
 VM-20 requires the use of prescribed asset default costs and spreads
 Company assets are mapped to PBR credit rating and assigned a weighted average life
(WAL)
 Annual default costs are determined as NAIC published default costs with spread
adjustments
 Long-term spreads are determined using an NAIC published spread table
 Additional adjustments are made for investment expenses and option adjusted spreads
 Embedded options should be properly modeled
 Revenue sharing may be reflected if certain stipulations are met
Assumption related to product features
 Non-Guaranteed Elements (NGEs)  Reinsurance

 How would these be impacted by prudent  Assumes that the counterparty is a reasonable
estimate assumptions? (e.g., COIs) entity
 Would the company make changes in  What contractual tools does the reinsurer
response to the modeled prudent estimate have to adjust for poor performance?
assumptions?
 When would these actions occur?
 When would the change be made?  What actions would the ceding company take?
 Can the change to NGE be justified?  Can they be justified?
 Should it be modeled?  Should they be modeled
Term specific considerations
 Level period
 Assume policyholder efficiency improves over time
 Company experience
 Analyze later durations in the level term period to ensure consistency with low lapse rates
observed in industry data
 A simple increase may not be appropriate for margins
 Shock Lapse
 Company and industry experience
 Capture appropriate characteristics such as premium jump
 Appropriate margin will be determined by whether the company has profits or losses in post level
term period
 Post level term profits are limited
 Term Conversions
 Companies should rely on their own experience where credible
 When using industry experience company should make adjustments for product differences and
company experience
 Model the liability of the conversion policy, considering materiality
Example step 1: Risk characteristics
Determining appropriate factors
Lapse Experience by Product 20 Year Term

Product Lapse Rate Lapse Experience by FA Band
(A/E by Count)
10 year 8.9%
Face Amount Lapse Rate
20 year 5.8%
<$200K 7.2%
$200K-$500K 6.0%
20 Year Term
Lapse Experience by Risk Class $500K+ 6.0%
Risk Class Lapse Rate
Preferred 5.0%
Experience seems to vary by level
Standard 6.7% term period, face amount, and risk
class suggesting that lapse rates
Substandard 9.0% should vary by these factors.
Example Step 1: Risk Characteristics (cont.)
Determining Appropriate Factors
20 Year Term 20 Year Term

Lapse Experience by Issue Age Lapse Experience by Gender
Issue Age Lapse Rate Gender Lapse Rate
20-29 9.4% Male 5.6%
30-39 5.6% Female 6.0%
40-49 5.0%
50-59 5.2%
60-69 5.8%
Experience seems to be similar for all issue ages except 20-29. Judgement
must be used to determine if sufficient credibility exists to suggest that the
assumption should differentiate.
Experience doesn’t suggest material differences by gender.
Example step 2: Company experience assumption
Assumption by duration for 20-year term
Company Experience
Policy Year
Company Industry
Credibility
CW Lapse Smoothed The primary
Experience Experience Rate Lapse Rate
drivers of lapse
1 10.3% 6.3% 99.3% 10.3% 10.5% rates are level
2 8.2% 6.3% 96.0% 8.1% 8.0% term period and
3 5.4% 5.6% 84.0% 5.4% 5.5% duration. Likely
4 4.3% 4.8% 75.3% 4.4% 4.5% most companies
5 4.7% 4.3% 71.3% 4.6% 4.5%
will only have
… … … … … …
credibility by
duration and
11 4.3% 2.9% 16.7% 3.1% 3.0%
product in
12 4.6% 2.8% 10.7% 3.0% 3.0%
aggregate. One
13 4.1% 2.8% 5.3% 2.9% 3.0%
approach is to
14 3.8% 2.7% 2.0% 2.7% 3.0% credibility weight
15 5.8% 3.0% 0.7% 3.0% 3.0% duration based
… … … … … … on experience.
Total 5.8% 4.8%
Example step 2: Company experience assumption
20-year term assumption for other risk factors
Lapse Experience by Risk Class

Risk Class Lapse Rate Differential Assumption
Preferred 5.0% 86% 85%
Standard 6.7% 116% 115%
Substandard 9.0% 155% 150%
Total 5.8% 100%
Lapse Experience by Face Amount

Face Amount Lapse Rate* Differential Assumption
<$200K 7.2% 113% 110%
$200K+ 6.0% 94% 95%
Total 6.4% 100%
*Lapse rate shown is by count, where others are by amount
Step 3: Margin
Approach
 One approach is to use confidence

intervals to develop margins Volatility of Lapse Experience
 The method is actuarially sound and Calendar Year Lapse Experience

justifiable
2016 4.4%
 The mortality margins are developed 2015 5.3%
assuming a 90% confidence interval
2014 4.5%
and may be reasonable to assume the
same for other assumptions 2013 7.0%
2012 5.1%
2011 6.7%
2010 6.5%
2009 5.1%
2008 6.6%
2007 5.7%
Average 5.7%
Std Dev. 0.9%
Step 3: Margin
Level of Credibility
Margin Based on Company SD
Probability Confidence Interval
75% +/- 6.8%

85% +/- 7.0%
90% +/- 7.2%
95% +/- 7.5%
Work must be done to determine the direction of the margin.
ULSG specific considerations – Policyholder behavior
Lapse Premiums Other assumptions :
 Consider the value of  Level and persistency  Withdrawals

guarantees
 Robust premium  Surrenders
 Dynamic policyholder experience studies
 Loan utilization rates
behavior
 Run required sensitivities  Allocation between
 Robust experience  Funding scenarios investment and crediting
studies
options
 Canadian Term to 100
experience
Pricing Specific Considerations
Assumption unlocking
 Assumptions for the modeled reserve may change at each valuation date
 This increases the volatility of earnings for products
 Pricing sensitivities will be more complex to model
 Changes in the best estimate assumption likely mean changes in your prudent estimate
assumptions
 Pricing actuaries will need to be comfortable with wider ranges of pricing results
 May initially need more sensitivities to understand how reserves change
 Need to understand sensitivities and outer loop projections
Implementation
 Coordinate with the valuation team

 Use simplifications, where appropriate
 Documentation will be important to understand any differences that arise with actual
reserve
Questions
Contact
Kim Steiner
Director
175 Powder Forest Drive, Weatogue, CT 06089
T +1 860 843 7068

E Kim.Steiner@willistowerswatson.com
Documentation
The New Valuation Manual and Life Product
Kim Steiner, FSA, MAAA

May 10, 2017
Agenda
 Importance of Documentation
 VM-31: The Actuarial Report
 VM-50/51: Experience Reporting
 Pricing Considerations
The need for documentation in a PBR world
 Unlike CARVM, principles-based reserves leave a lot of room for judgment
 Assumptions for the modeled reserves are not fully prescribed
 Projection models will be used to determine some of the reserve components
 Companies are free to use reasonable judgment in many areas to properly

capture their business
 How do you prove the reasonableness of these decisions to regulators and

auditors?
Documentation will be important for internal and external purposes
The importance of documentation
Internal
General  Ownership of assumptions and models

 Change protocols
 Judgment and simplification used in assumptions and
models
 VM-20’s applicable regulations and guidance and the
methods used to comply
 Compliance with internal governance policies
Pricing-specific  How are the VM-20 validation models different from the
pricing models for VM-20?
 Are their pricing specific models or simplifications to models
that need to be documented?
 Are there adjustments to the VM-20 assumptions that are
pricing specific?
The importance of documentation
External
 Documentation will also be necessary to satisfy auditors and regulators
 The VM includes mandatory documentation that will be submitted to regulators

 VM-31: The Actuarial Report
 VM-50/51: Experience Reporting
 VM-G: Corporate Governance Guidance
 These documentation requirements should be considered throughout the PBR

process
 Consistent and timely documentation updates will ensure that judgement decisions
are not overlooked in documentation
 Aligning internal documentation with VM requirements will save time and effort
throughout the process
VM-31: The Actuarial Report
The Requirements
Background
 The PBR Actuarial Report must be submitted if deterministic or stochastic

reserves were calculated
 Certain documentation on the exclusion tests is still required if modeled
deterministic and stochastic reserves were not calculated
 Documentation and disclosure sufficient for another qualified actuary in the
same practice area to evaluate the work
 Descriptions of all material decisions made and information used by the
company
 Sub-report prepared for each group of policies that comprise a model
segment (e.g., product)
Requirements – Overview Section
 Key risks
 Reliances on data and other individuals
 Description of methods for assumptions, margins, and significant changes
from prior year
 Asset modeling methodology
 Description of risk management approach
 Description of rationale for determining if a decision was material
 Certification
Requirements – Full Report
 A discussion of modeling systems and methods

 Assumption details:
 Mortality assumption methodology
 Policyholder behavior assumptions
 Expense assumptions
 Asset assumptions
 Revenue sharing
 Reinsurance modeling methodology
 Non-guaranteed elements determination
 Exclusion testing
 Impact of margins
Assumption and Margin Specific Requirements
 Any significant changes from prior year
 List of key risks and experience reporting elements the company is using
to monitor changes in experience from year to year
 Methodology used to develop the margins
 Disclosure of key valuation assumptions that are materially inconsistent

with the company’s risk assessment process
 How the materiality of assumptions was determined
Mortality Specific Requirements
 Mortality assumption segment determination and rationale

 Description of methods used to select industry relative risk tables
 Details of any external data used to supplement company experience
assumption
 Rationale behind any adjustments to company experience
 Credibility method
 Details of period used to grade from company experience to industry
tables
Other Assumption Requirements
 Sources of data
 Determination of assumption, including methods used for less than fully
credible data
 Description of anticipated experience assumptions
 Margin methods and determination
 How NGEs are impacted by policyholder behavior
 How lapse and mortality assumptions are adjusted for anti-selection
 Expense assumption allocation methods
 Asset assumption methodology
Margin Impact
 The impact of margins on reserve

 Calculated by subtracting the deterministic reserve calculated with
anticipated experience by the deterministic reserve as reported
 This calculation is done for each risk factor and for each model segment
 If the company believes the deterministic reserve includes an implicit
margin (e.g., no future mortality improvement) anticipated experience can
be adjusted to reflect this
 For prescribed assumptions (e.g., asset) that are prudent estimate
assumptions, the company may choose to use other anticipated
experience assumptions for determining margin impact
Best Practices
 Robust document
 Material judgements
 Level of detail
 Include the why and not just the what
 Independent peer review
 Timely documentation
VM-50 / 51: Experience Reporting
The Requirements
VM-50: Experience reporting requirements
 The purpose of the collected experience includes:

 Establishment of industry tables
 Assist regulators and other parties in reviewing PBR Actuarial Reports
 Create an independent check on the accuracy and completeness of company
experience studies
 Inform determination of prescribed assumptions and margins
 Provide a common benchmark for regulators to use when reviewing company
experience
 Companies with more than $50M direct premiums are required to report
experience
 Reported experience should exclude reinsurance assumed
 Data must conform to common data definitions
 Must assist in rectifying critical data errors
VM-51: Experience report formats
Mortality
 Company Level Information

 Preferred Class Structure Questionnaire
 Mortality Claims Questionnaire
 Additional Plan Code Form
 Annual submission
 Gross of reinsurance
 Ordinary business, which excludes:
 COLI / BOLI
 SI / GI
 Worksite
 Individually solicited group life
 Direct response
 Final expense
 Pre-need
 Home service
 Conversion identified separately with conversion issue date
 Riders are a separate record
 Potential exclusion due to data issues
Pricing Considerations
Communication with other areas
 Pricing area may be first implementer
 Assumptions for new products
 Learnings for modeling implementation
 Methodology chosen could impact future profitability
 Role in completing formal documentation requirements
How pricing memos will change
 No change for states that do not require reserve demonstration
 Unclear exactly what is required for reserve demonstration
 Likely to evolve over time, which may result in more disclosure
Questions
Contact
Kim Steiner
Director
175 Powder Forest Drive, Weatogue, CT 06089
T +1 860 843 7068

E Kim.Steiner@willistowerswatson.com

2017 05 Product Development New Valuation Manual Presentations

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

2017 05 Product Development New Valuation Manual Presentations

Caricato da

Copyright:

Formati disponibili

Leonard Mangini, FSA, FRM, FALU, CLU, MAAA

21. (Aggregate) minimum reserve (MR) = (1) + (2) + (3)+(4)+(5)+(6)+(7):

Material Risks that Company has inﬂuence over

Actuary’s Prudent EsLmate (PBE) of anLcipated future experience, factoring in

Actuarial Standard of PracLce 23

• VM S7.B.3: Requires AggregaLon Sub-Groups ~ Integrated Risk Management

Centralized Internal Modeling Teams that have Business Unit/FuncLonal Clients?

Feedback Loops for Unlocking AssumpLons

Why am I hearing this ﬁrst before today’s other sessions?

VM and the PD Actuary ©2016-17 AADicke LLC-

VM and the PD Actuary ©2016-17 AADicke LLC-

2) An actuary sets assumpLons on behalf of Company

The New VM and the PD Actuary May 10, 2017

à ImplementaLon Plan: will go the Board

VM-G Sec.on 1.C.3:

The New VM and the PD Actuary May 10, 2017

à Puts Senior Management into ORSA/Solvency II role as Pillar 2 “guardians”

Prepare the VM-31 PBR Report

Carrie Kelley, FSA, MAAA

May 10, 2017

 Prudent Estimate Mortality Assumptions

2 mortality assumption sets: Net Premium Reserve’s prescribed assumptions

 2017 CSO and 2017 CSO Preferred Structure

Non-  Mortality defined in VM-02

Exclusion  Anticipated experience assumption (a component of prudent estimate assumptions)

 Prudent estimate assumptions are company experience + margin

 Grade company experience with industry experience based on credibility of company

Step 1: Step 2: Step 3:

2 Company experience rates

3 Credibility and experience margins

4 Industry mortality table and margins

5 Blending company and industry

Expect mortality to vary by several factors, possibly including:

Example: Risk class and gender are key variables

 Determined from company’s own experience where available

 Company experience data:

Consistent with segment definitions, split experience by smoking status and

% A / E Based on 2015 VBT (SM / NS)

Smoker 100% 110% 97% 107%

Split FNS across risk classes

% A / E Based on 2015 VBT

2nd Best NS 83% 35% 100% 78%

Residual NS 111% 20% 145% 113%

Total NS 78% 100% 100% 78%

Note: Make sure weighted average of experience factor is at least as great as

 Credibility impacts margins and the industry grading period

Bühlmann Credibility Results

Smoker 33% 49%

Attained Credibility: Credibility: Credibility: Credibility: Credibility:

<45 20.4% … 13.7% 12.7% 11.6% … 2.3%

46 - 47 20.2% … 13.7% 12.7% 11.6% … 2.3%

48 - 49 20.0% … 13.5% 12.5% 11.4% … 2.3%

 Used 2015 VBT RR Tables

 2015 VBT Margins

Duration Number Deaths Duration Number Deaths

Credibility of Maximum # of Maximum # of Maximum # of years in which the

 Max Yrs to Begin Grading after last 7

 Max Yrs to Grade to 100% of Industry Min (17, 9 + 15 * 72% = 19.8) = 17

 First Duration at 100% of Industry 26

Experience Experience Experience Experience Industry Industry Industry Final

1-14 62% 62% * X** 12.70% 100% 100% 20.4% 0% 70% * X

15 62% 82% 12.70% 100% 100% 20.2% 0% 92%****

17 62% 82% 12.50% 90% 100% 20.0% 10% 95%

18 62% 81% 12.50% 80% 100% 20.0% 20% 97%

21.  (Aggregate) minimum reserve (MR) = (1) + (2) + (3)+(4)+(5)+(6)+(7):

•  VM S7.B.3: Requires AggregaLon Sub-Groups ~ Integrated Risk Management

à ImplementaLon Plan: will go the Board