RAPA | 2018 Annual Conferencereinsadmin.org/wp-content/uploads/2015/07/MASTER...the U.S. in the historical Louisiana Purchase in1803, and migrated by Americans, French, Creoles, Africans,

RAPA | 2018 Annual Conference

October 21st – 23rd, 2018

Agenda

8:30 – 9:00 Welcome and Introductions

9:00 – 10:00 An Introduction to Artificial Intelligence

10:00 – 10:30 Break

10:30 – 11:30 Claims: Reviewing the Current Ecosystem& Identifying

Priorities for Improvement.

11:30 – 12:30 Total Line/Jumbo – Is There a Problem to be Solved?

12:30 – 1:45 Lunch

1:45 – 5:30 Roundtable Session

Do Old Dogs Need to Lean New Tricks or Do New Dog Need to Learn

Old Tricks?

Design Thinking Sprint: The Puzzle Project

4:00 – 4:30 Break

Accelerated Underwriting

5:30 Adjourn Day 1

7:00 Cocktail Reception and Dinner

Antitrust StatementThe Association makes no warranties as to the accuracy of the information contained in discussion forums, meeting minutes or presenter materials. The posting of messages, meeting minutes or presentation materials does not constitute knowledge, endorsement or approval by the Association, nor do we accept any liability for the content of any posting. Individuals using these discussion forums do so at their own risk and shall also remain individually responsible for their actions and statements in using these discussion forums. Because the Association is committed to adhering strictly to the United States and Canada antitrust, copyright, trademark, securities and other federal statutes, as well as state or provincial common laws covering libel, slander, defamation, false advertising, invasion of privacy and violations of the rights of publicity, we strongly discourage users of our discussion forums or attendees of our meetings from verbally stating or writing anything that (1) sets or controls prices or terms of products or services and the manners in which products or services are sold; (2) violates the proprietary or personal rights of others; or (3) constitutes an advertisement. Your use of or participation in Association discussion forums or meetings is acknowledgement of your agreement with the above and your promise to use these forums in a professional and courteous manner.

Welcome to New Orleans (or-lins)

Originally inhabited by the Chitimacha tribe, New Orleans was founded in 1718 by the French as La

Nouvelle-Orléans, and named in honor of the Regent of France, Philip II, Duke of Orléans. It was sold to

the U.S. in the historical Louisiana Purchase in1803, and migrated by Americans, French, Creoles,

Africans, Irish, Germans, and Italians.

New Or-lins 10 interesting facts

10. Louisiana is broken into 5 subdivisions called parishes(not counties) - Orleans, Jefferson, Plaquemines, St. Bernard,

and St. Tammy.

9. The city was developed in a rectangle block known as the French Quarter, and well known for Cajun style culinary

dishes. Gumbo anyone?

8. New Orleans is NOT the home of the first Mardi Gras celebration in the United States. The first was in Mobile, Alabama.

Really??

7. Hurricane Katrina in 2005 – levee systems failed and 80% of the city flooded. More than 1,500 died and many were

permanently displaced.

6. The New Orleans Saints won their only Super Bowl appearance in 2010. Five years after the Hurricane Katrina

disaster. The chant "Who dat? Who dat? Who dat say dey gonna beat dem Saints?“ became a rally cry for fans and the city

of New Orleans post-Katrina.

5. The Superdome is the largest enclosed arena in the world. More than two football fields in diameter. That is Super!!

4. Alcohol is available in New Orleans at any moment of the day. Bars can stay open all night and liquor is sold in grocery

stores. Anyone have eggs and bourbon on their grocery list?

3. Lee Harvey Oswald, John F. Kennedy’s ‘assassin’, was born in New Orleans in 1939.

2. New Orleans is where the first opera was performed in the U.S. back in 1796.

1. New Orleans is the birthplace of Jazz and many famous musicians such as Louis Armstrong, Fats Domino, and Harry

Connick Jr.

Enjoy “The Big Easy”

Executive and Initiative Chairs

Eddie Martinez – Chair

Genevra Pflaum – Vice Chair

Greg LaRochell – Past Chair

Garfield McIntyre – Treasurer

Jill Dupuis – Secretary

Susan Whitehead – Membership

Kelly Priest – Planning

Karen Rotondi – Communications

Dalia Khoury – Education

Lynn Martone – Risk Management

Brittainy Jones – Post Level Term

Rhonda Nielsen-Jackson – Data Quality

Mitch O’Campo – Innovation

RAPA RECOGNITION

• Special RAPA Thank You to the spokes that make our

RAPA wheel turn

• Members of our Board

• Initiative Leads and Committee volunteers

• Transitions in-progress

• Treasurer role

• Communications

• Education

• Risk Management

Thanks to our Sponsors!

PLATINUM

BLACK

Thanks to our Sponsors!

SILVER

GOLD

BRONZE

CANADA

US

Membership Overview

• 110 Members

• New Membership Package established

to encourage non-member companies

• Equitable distribution between Direct

Writers and Reinsurers

Other Countries 3%

Canada 41%U.S.A. 56%

2018 Geographic Distribution

Australia, Barbados, Bermuda, Oman Canada U.S.A.

40

%

41

%

7%

12

%

INSURA NCE COMPA NIES REINSURERS RETROCESS IONA IRES V END ORS

INDUSTRY DISTRIBUTION

CommunicationsWebsite

Fully functional with on-going improvements

RAPA content continues to be updated – Whitepapers, SOA, CRC,

Newsletters, etc - Check back frequently

Links to social media

Paypal Integration (reduces cost for Association)

Treasurer’s Report

Healthy Balance remains stable year-over-year

Robust Due Diligence

▪ Budget

▪ Cash Management

▪ Financial Statements

Annual IRS 990 filing completed for the association

Artificial Intelligence

We’re smarter together

September 2018

The journey towards smarter automation has just begun

Ashley Broering, AVP Digital & Smart Analytics

13

What is AI?

Why now?

Why care?

Agenda

Wow, what should we do?

2

3

4

1?

14

What is AI?

2 3 41?

Artificial Intelligence

15

The concept of AI is often mistaken for robots overtaking humanity

16

Artificial intelligence(AI) is the intelligence of a machine that could successfully perform any intellectual task that a human being can.

AI is the computer science term for allowing computers to think like humans

perceiving

reasoning

learning

problem solving

17

High Level AI Spectrum

ARTIFICIAL INTELLIGENCEbeginning in the 50s

MACHINE LEARNINGfrom the 80s

DEEP LEARNINGthe last decade

AI represents a spectrum of capabilities whereby machines learn from data

18

Machine learning: algorithms that use data and statistics so that computers get better at performing a task

Reinforcement Learning

Unsupervised Learning

Supervised Learning

Mac

hin

e Le

arn

ing

Tech

niq

ues

New Data

These are birds

Input: Unlabelled Data

“Bird”

“Bird”“Bird”

Take Action

Observe / Get Reward

Check Mate

Input: Labelled Data

Beginning State End State

Got it!

It’s a Bird

I see a pattern

19

Deep learning: a type of machine learning, inspired by biology, that uses multiple software layers to form a “neural network”

INPUT LAYER

HIDDEN LAYERS OUTPUT LAYER

Artificial Neural Network

• Require larger datasets• Generally produce better

results

Key Features:

More Layers = “Deep” Network

20

AlphaGo: A Deep Learning Success Story

Solution: self-play & unsupervised training

Image from: Alphagomovie.com

Image from: https://blog.yellowant.com/before-google-duplex-5-ai-technologies-that-wowed-us-52aa2f942338

Lee Sedolwinner of 18 world titles

computer program

“AlphaGo” vs.

Problem: Once we pass human performance, there is no more good training data

21

There are many branches and applications of AI

Virtual Assistants

Speech to Speech

Translation

Natural Language Processing

Robotics

Recommendation Engines

Analyze textual data written in human language, from words, syntactic structure, to semantic meaning

Studies the computing that relates to, arises from, or deliberately influences emotion or other affective phenomena.

Process images and videos. Recognize objects, identifying depth of objects, direction of light, and the interpretation of images.

Process signals and transform waveforms into words and sentences.

Some examples you may have heard of…

Computer Vision

Affective Computing

Speech Recognition

22

AI models and techniques enable increasingly advanced analytics

Image from: https://becominghuman.ai/lets-talk-about-advanced-analytics-a-brief-look-at-artificial-intelligence-bf1c7a7d3f96

23

Why now?

2 3 41?

24

AI is not a new topicSony first introduced its Autonomous Entertainment Robot, AIBO, in 1999

AIBO ERS-210 from: https://www.sony.net/SonyInfo/News/Press_Archive/200010/00-050A/

AIBO ERS-110, released in 1999

• Displays happiness and anger • Detects distance from a certain object• Has a head sensor so owners can praise or discipline the dog

AIBO ERS-110 from: https://www.roboticstoday.com/robots/aibo-ers-110-description

AIBO ERS-210, released in 2000

• Recognizes its own name and simple words• Has additional sensors so it can show “an abundant array of emotions”• Reacts to external stimulus and makes its own judgements

25

An era of massive data & cheap storage makes AI more accessible

1970 1980 1990 2000 2010

Sto

red

Dig

ital

Info

rmat

ion 2.7 Zetabytes

of data exist today

48hoursof new video every

minute of every day

100terabytes of data uploaded

daily on

44x more data in 2020 than in 2009

The Digital Transformation

Data sizes from: https://analyticsweek.com/content/big-data-facts/

Storage costs from: https://www.computerworld.com/article/3182207/data-storage/cw50-data-storage-goes-from-1m-to-2-cents-per-gigabyte.html

1MB hard drive cost $1M in 1967, today

it’s about $0.02

26

AI is more relevant now because the data explosion has been coupled with an increase in processing power

“Recent advances in hardware have

exponentially increased the

amount of data that can be

processed through deep learning.

Graphical processing units (GPUs) leveraged by video games have increased the speed of

deep learning systems by orders of magnitude,

reducing running times from weeks to days. “

- Dr. Christian Klose, Dr. Nitin Nayak

Decoding the human genome

10 years 1 week

27

AI algorithms have been developing and improving over time

1950 19601940

ANOVA

Fisher's linear discriminant

Naive Bayes classifier

Artificial Neural Network

Perceptron

Logistic regression

Backpropagation

Quadratic classifiers

Hidden Markov models

Nearest Neighbor Algorithm

K-means algorithm

1980 19901970

Multinomial logistic regression

Support Vector Machines

Single-linkage clustering

Expectation-maximization algorithm

CART

CHAID

Self-organizing map

Word Embedding

Boltzmann machine

Bayesian network

Boosting

C4.5

Bootstrap (bagging)

Apriori algorithm

Random Forests

2000 2010

Conditional Random Field

Latent Dirichlet Allocation

MapReduce

Deep Learning

2018

Non-parametric Bayesian

source: Andrew Ng

28

Advances in algorithms, data, and computing power have propelled AI into the spotlight

https://www.mckinsey.com/business-functions/mckinsey-analytics/our-insights/an-executives-guide-to-ai

AlgorithmsMassive data & cheap storage

Processing power

29

Why care?

2 3 41?

30

AI is already all around us.

1.

31

Voice Recognition in Every Day LifeDeep Learning enables speech recognition systems

Deep Learning revolutionized the field of speech recognition.

Almost all current speech recognition systems are based on Deep Learning.

32

AI in HealthcareAI can assist doctors in cancer detection

Skin cancer is primarily diagnosed visually, beginning with an initial clinical screening and followed potentially by dermoscopic analysis.

Deep Learning achieves performance on par with tested experts, demonstrating a level of competence comparable to dermatologists.

33

Autonomous Intelligent VehiclesDeep Learning to solve self-driving car tasks

Street Mapping

•Where am I?

Scene Understanding

•Where is everyone else?

Movement Planning

•How do I get from A to B?

Driver State

•What’s the driver up to?

https://selfdrivingcars.mit.edu/

34


Street Mapping

•Where am I?

Scene Understanding


Movement Planning


Driver State



35


Street Mapping

•Where am I?

Scene Understanding


Movement Planning


Driver State



36

Autonomous Intelligent VehiclesDeep Learning to improve driver safety

Street Mapping

•Where am I?

Scene Understanding


Movement Planning


Driver State



37

Man’s best friend is back

Sony is re-launching its robot dog, Aibo, which now uses the same kind of AI that’s in self-driving cars

$2,899 bundle package:

Aibo will be in the US in time for the holidays!

Source: https://www.cnet.com/news/sony-new-aibo-robot-dog-is-finally-coming-to-the-us/

38

AI is changing what we insure…

… and how we insure it

2.

39

AI touches all areas in the Insurance Value Chain

Sales UnderwritingPolicy

AdministrationClaims

40

AI can increase sales and streamline the sales process

➢Reach customers through new distribution channels

➢Personalize communication and provide relevant advice

➢Understand existing and potential customers to

improve upselling & cross-selling

➢Prioritize leads to focus on closing the most promising cases

SalesUnderwritin

g

Policy Administrat

ionClaims

Insurance products can be embedded in Digital Ecosystems to provide relevant product options

AI-enabled Opportunities:

41

AI will fundamentally change the way we assess risk

Pressure from consumers, new data sources, and analytics capabilities have given rise to accelerated underwriting initiatives

➢Improve customer journeys and lower costs by

increasing straight-through processing

➢Increase price flexibility through a more personalized and accurate cost of claims prediction

➢Detect anomalies due to mistakes and nondisclosure

SalesUnderwritin

g

Policy Administrat

ionClaims


42

AI can help identify and manage risks in existing books of business

Smart home, smart mobility, and smart health devices engage consumers and provide new data streams for AI models

➢Engage policy holders to improve risk awareness and jointly prevent future risks

➢Understand if risk selection was as planned

➢Identify and manage factors that drive claims costs

➢Focus retention efforts to reduce lapse of good risks

SalesUnderwritin

g

Policy Administrat

ionClaims


43

AI can streamline claims processing

Satellite images can be used to expedite the estimation of damages and claims following a natural catastrophe

➢Improve fraud detection

➢Optimize current claims processing

➢Estimate expected costs from open claims

SalesUnderwritin

g

Policy Administrat

ionClaims


𝒑𝒊 𝒑𝒊

Satellite 𝑆𝑎𝑓𝑡𝑒𝑟 Satellite 𝑆𝑏𝑒𝑓𝑜𝑟𝑒

44

Source: http://captricity.com/technology/

SalesUnderwriti

ng

Policy Administrat

ionClaims

AI enables digitization and automation along the entire insurance value chainAI-enabled Opportunities:

➢Digitize data

➢Convert unstructured data into a structured format

➢Automate repetitive tasks

45

We can be part of the change

3.

46

Deep learning could account for

$3.5-$5.8 trillion in annual value, across industries …

Source: https://www.mckinsey.com/featured-insights/artificial-intelligence/notes-from-the-ai-frontier-applications-and-value-of-deep-learning

AI has the potential to make a big impact

…and $142-$312 billion in

insurance

47

There’s room for AI and humans to coexist

Source: https://hbr.org/2018/07/collaborative-intelligence-humans-and-ai-are-joining-forces

Humans Assisting Machines:

▪ Train machines to perform a task

▪ Explain the outcomes of machine tasks

▪ Sustain the responsible use of machines

Machines Assisting Humans:

▪ Amplify our cognitive strengths

▪ Embody human skills to extend our physical capabilities

▪ Interact with customers and employees, and free up our time

48

What can we do?

2 3 41?

Think about long-standing problems in new ways

Example and illustration from: https://hbr.org/2017/01/are-you-solving-the-right-problems

“Make the wait feel shorter.”Put up mirrorsPlay musicInstall a hand sanitizer

“Make the elevator faster.”Install a new liftUpgrade the motorImprove the algorithm

SOLUTION SPACE

SOLUTION FINDING“The elevator is too slow.”

PROBLEM FRAMING

Reframing the problem

“The wait is annoying.”

Imagine you own a building and your tenants are complaining about the elevator…

“The point of reframing is not to find the “real” problem but, rather, to see if there is a better one to solve.”

- Thomas Wedell-Wedellsborg, HBR, Jan-Feb 2017

50

Look for ways to test potential solutions quickly

Agile and Design Thinking methodologies help teams take a human-centered approach to innovation and develop solutions that people will actually use

• Fail fast

• Iterate on a prototype• Empathize with customers • Shorten the feedback loop

• Think “How might we?”

Key Concepts:

51

Data is the new gold

Deep learning algorithms have a

voracious appetite for data...

…but finding large datasets can be hard.

With appropriate diligence around data use, security, and privacy, it is possible to utilize and share valuable information

• Public data sources• Datasets that other departments may

maintain and use

• External partnerships to share and jointly analyze data

Consider:

52

What AI really means

Why we’re hearing about AI now

Why we should care about AI

We now understand…

What we can do to be part of the change

2

3

4

1?

53

Thank you!

General Public Release

Legal notice

55

©2018 Swiss Re. All rights reserved. You are not permitted to create any modifications or derivative works of this presentation or to use it for commercial or other public purposes without the prior written permission of Swiss Re.

The information and opinions contained in the presentation are provided as at the date of the presentation and are subject to change without notice. Although the information used was taken from reliable sources, Swiss Re does not accept any responsibility for the accuracy or comprehensiveness of the details given. All liability for the accuracy and completeness thereof or for any damage or loss resulting from the use of the information contained in this presentation is expressly excluded. Under no circumstances shall Swiss Re or its Group companies be liable for any financial or consequential loss relating to this presentation.

Sponsored by

Cocktails & Dinner

RAPA | 2018 Annual Conference

October 23, 2018

Agenda

8:45 – 9:00 Welcome and Agenda Review

9:00 – 10:00 Transformation & Innovation

10:00 – 10:30 Break & Conference Evaluation Survey

10:30 - 12:00 Initiative Updates

- Education

- Risk Management

- Data Initiative

- Post Level Term

- Innovation

12:00 – 12:30 Project and Initiative Breakout Meetings

12:30 Adjourn

2019 Conference

Mark your calendar for the 2019 Annual Conference!!

October 20th – 22nd, 2019

Savannah, GA Marriott - Mansion on Forsyth Park

Sponsorship Opportunities

$5,000

Platinum

$3,000

Gold

$2,500

Silver

$1,500

Bronze

$1,000

2018 Conference Evaluation Survey

Sponsored by

Break

Initiative Strategy

Purpose: To improve the effectiveness and efficiency of RAPA member's reinsurance

understanding and processes. The initiatives work streams provides the

opportunity for members to work together on education, training, and to benefit

from robust collaboration with industry experts within the working groups.

Want to join an Initiative? Just contact one of

the leads

Education: Dalia Khoury

[email protected]

Data: Rhonda Nielsen-Jackson

[email protected]

Risk: Lynn Martone

[email protected]

Post Level: Brittainy Pratt

[email protected]

Innovation: Mitch O’Campo

[email protected]

Thinking of a new initiative?

Submit your suggestion to Eddie Martinez

[email protected]

RAPA

DATA

EDUCATION

RISK

PLT

INNOVATION

mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]

EDUCATION INITIATIVE

Education Initiative

Purpose

To develop training material for different

processing functions and ultimately have case

studies.

The training material includes the following topics:

▪ Treaty Fundamentals and Provisions

▪ Glossary of Reinsurance Terms

▪ Workflow of New Business

▪ Workflow of Maintenance


Our 2018 mandate is to develop a webinar to inform RAPA

members and non-members of the educational material

available, how to access it and the costs (if any).

Future plans under discussion include a webinar about

Best Practices and another about Post Level Term.


Deliverables

4Q2018 – RAPA/Education Initiative Webinar

1Q 2019- Post Level Term

If you have ideas for webinars or podcast, let us know


WHAT IS RAPA?

WHY SHOULD YOU BECOME A RAPA MEMBER?

WHO WILL BENEFIT?

WHY SHOULD YOU ATTEND THE ANNUAL CONFERENCE?

http://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwjil7r9pOjdAhWd2YMKHe6_A54QjRx6BAgBEAU&url=http://www.jatengpos.com/2012/10/dprd-jogja-pertanyakan-efektivitas-forpi-337365&psig=AOvVaw1D7_geVLB7zTcBvpiGh2Yw&ust=1538587767466664http://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwj-wY7avujdAhVF34MKHTAGCFQQjRx6BAgBEAU&url=http://tradezyone.com/&psig=AOvVaw2QcI5VY0z7FftdAiiNHmQo&ust=1538594872042119https://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwjVwo3Rw-jdAhVk5YMKHQJgDhsQjRx6BAgBEAU&url=https://laurelplace.ca/delta-chamber-networking-evening/&psig=AOvVaw3d_BLF9nGr4owmHoIhU9dl&ust=1538596174306731


New Members are welcome

http://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwja35bQxejdAhVJ9YMKHYF0DQcQjRx6BAgBEAU&url=http://www.inlandempireafp.com/index.php?page%3D82&psig=AOvVaw3aHSsbuG8IFdKPfDsyUVpz&ust=1538596706415237

RISK INITIATIVE

Risk Initiative

Lynn Martone, Swiss Re

Amalia Figueiras, Prudential Financial

Risk Initiative

The RAPA Risk Management initiative team works to develop tools and techniques related to risk identification,

assessment and mitigation, as well as assessing audit and compliance reviews of the reinsurance administration

function. The key goal is to provide common reinsurance guidelines and methodology approaches for direct

writers, reinsurers and retrocessionaires.

This initiative is divided into two sub-teams with focus on:

Reinsurance Administration Risk Assessment:

The Risk Assessment sub-team has developed a risk assessment matrix (spreadsheet) detailing and defining

strategic, operational and compliance risks. We have concentrated our efforts on assessing the key sub risks to the

function and defining how it leads to risk mitigation. During 2018, our team completed the following:

• Confirmed all risk classifications in the matrix have been reviewed and populated appropriately.

• Assigned control names and risk categories to relevant key controls

• Took a deep dive into the Operational Sub Risks including; Quality Assurance, Collusion, Fraud, STOLI,

and Outsourcing

This updated document will be added to the RAPA website for your review.

Reinsurance Audit & Compliance Reviews:

The Audit & Compliance sub-team has identified what criteria should be used when selecting audit candidates.

Our areas of concentration focused on taking a deep dive into the administration and audit of term conversions.

During 2018, we completed the following:

• Reached out to committee members, as well as industry contacts, to discuss the world of term conversions.

• Researched and documented industry facts and insights regarding term conversion administration, audit

practices, pain points, etc.

• Drafted the Administration view/practices to complete best industry approaches (and insights) when auditing

term conversions.

• Drafting the Audit view/practices (and insights) of white paper to be shared with the industry.

Risk Initiative

Next Steps - 2019

Risk Assessment:

• Gather feedback from larger population

regarding: what are your pain

points/biggest risk?

• Treaty negotiations, administration

and set up?

• Ongoing adherence?

• Financial reporting?

• KPI’s?

• Based on feedback, this subcommittee will

continue to take a deep dive into the next

relevant sub risk.

Audit & Compliance:

• Continue writing of Audit practices and

combine with Administration practices to

complete white paper for final review.

• Publish Term Conversion white paper on

RAPA site.

• Continue discussions and research on next

key audit topic identified by team in 2018 /

PLT.

Thank you to all who took time from

their day job to support this initiative!

Lynn Martone, Swiss Re

Amalia Figueiras, Prudential

Laura D’Andrea, Prudential

Garfield McIntyre, Munich Re

Anthea Cote, Munich Re

Ricky Peterson, Munich Re

Carolyn MacPherson, Swiss Re

Carol Arada, Swiss Re

Janice Lawrence, Pac Life

Alan Woodlard, RGA

Barbara Beicker, USAA

Kelly Priest, LOGIQ3

Karl Martone, Martone & Martone

Antonio Layne, RBC

Thomas Trent, Transamerica

Jennifer Atlee, Hannover Re

Kim Langstaff, Hannover Re

Glenn Beuschel, Munich Re

Chantal Lessard, Optimum Re

John Whitaker, RGA

Bethany Stivenson, Axa

Troy Leach, Canada Life

Zayda Marie, Associated Consulting

DATA QUALITY INITIATIVE

Data Initiative

Initiative Participants

➢ Initiative Chair: Rhonda Nielsen-Jackson, Hannover Re

Dean Santos, Munich Re Toni Ollerton, Aurigen Re

Sumitra Kumhare, Pacific Life Re Diana Aversa, Pacific Life Re

Genevra Pflaum, Hannover Re Grace Sirianni, Munich Re

Eileen Ah-Fat, Canada Life Kate Coyle, Ohio National

Duane Pfaff, Voya Lisa Clarke, Logiq3

Mindy Epstein-Hinshaw, Logiq3 Saline Smith, Swiss Re

Melinda Bynoe, BMO Reinsurance Dzan Dinh, Munich Re

Karen Lipka, RGA Allison Forde, RBC

Maureen Headley, Met Life Tom Kuenzel, Ameriprise Financial

Paul Winters, Q-Perior Tom Hartlett, Logiq3

Purpose

• Create a document for Guidelines for Reinsurance Reporting, in essence a document of

best practices for specific areas of reinsurance administration and data quality.

• The document includes information, samples and realistic scenarios. Topics:

▪ Reporting Issues

▪ Communication/Notifications

▪ Data Quality

▪ Conversions

▪ Taking a Treaty from paper to system implementation

▪ Samples of typical reinsurance reporting, i.e. Policy Exhibit, Transaction files, etc.

Data Initiative

Data Initiative

Deliverables & Milestones

1) Reinsurance-Reporting-Guidelines-and Best-Practices living document versions

1-6 have been rolled out between October 2014 and November 2017

❖ Currently available on the RAPA website @

http://www.reinsadmin.org under Initiatives and Data Management

2) Reporting Guidelines and Best Practices status:

a) Complete

▪ Reporting Issues

▪ Communication/Notifications

❖ Communicating New System, Data or Administration to Business

Partners is available on RAPA website and not required to be a

RAPA Member to use.

▪ Conversions

▪ Taking a Treaty from paper to system implementation

▪ Samples of typical reinsurance reporting, i.e. Policy Exhibit,

Transaction files, etc

b) Same topic – new twist

▪ Data Quality

http://www.reinsadmin.org/

Initiate

RAPA: Data Quality Initiative

Support

Create a Team of Advisors

SMART Goal Set

Develop a Plan

Execute

Complete Objective

RAPA Data Quality Initiative: Project Cycle

Big Data

Tre

aty

Com

pli

an

ce

The alignment of accurate data with its intended

use

Data Quality

Initiate

RAPA: Data Quality Initiative

Focus:Data Quality: Treaty Compliance

• How can we better state the administration requirements written in the treaty?

Dean Santos Director, Reinsurance Administration Individual Reinsurance OperationsMunich Re Canada (Life)

Sumitra KumbhareAVP OperationsPacific Life Re | Retro

Toni OllertonSenior Business Analyst, North America LifePartnerRe

Diana Aversa, ACSSenior Treaty AnalystPacific Life Re | Retro

Support

Create a Team of Advisors

SpecificClear and concise goals.

•Create a Mission Statement

•Identify purpose and objective

MeasurableAbility to track progress.

•Develop plan withmilestones for completing the documentation with the Data Quality focus team

AchievableSet challenging, yet achievable goals.

•Plan allows for review and feedback by the committee

•Distribute to participants of the RAPA - Data Quality Initiative

RelevantGoals that are relevant to our overall plan.

•Complete the documentation on the focus for RAPA - Data Quality Initiative

•Opportunity of standardization by creating the basis of expected data

•Communicate and promote approach through RAPA

TimelyTarget finish time.

•Recruit participants for the committee

•Assign accountabilities to committee members

•Set up touch points with participants of the RAPA - Data Quality Initiative

•Update RAPA on the focus and approach for the Data Quality Initiative

•Promote initiative and next steps

SMART Goal Set

Data Quality: Treaty Compliance

Our mission is to ensure that the highest quality of data is

delivered through the coordination of our efforts for the purpose of lowering the

risk of systemic issues.

Data Quality Mission Statement

SMART Goal Set

Purpose: Ensure that data produced adheres to the treaty and satisfies both business and the technological requirements.

✓ Treaty, Plan fields identification

✓ Data requirements and validation

Client Data Files

Data Transformation

Internal Stakeholders

✓ File types (Inforce, Transaction, Billing, Claims)

✓Meeting the needs of our stakeholders

SMART Goal Set

CESSION ADMINISTRATION

1. REPORTING, BILLING AND PAYMENT OF ACCOUNTS

Reporting

(1) The reporting period will not be less frequent than quarterly.

(2) The Cedant will provide the Reinsurer with the minimum data requirements within 30 days of the end of the applicable reporting period, reflecting all reinsurance transactions taking place during the reporting period, with respect to Policies reinsured under this Agreement as set forth in the Reinsurer's Minimum Data Reporting Requirements, as amended from time

to time, in an electronic data interchange (EDI) format acceptable to the Reinsurer. the Reinsurer's current version of its Minimum Data Reporting Requirements as at the Effective Date

is attached as an Appendix to the Agreement. If the information reasonably required by the Reinsurer for the maintenance of proper records and appropriate actuarial reserves changes,

the original format agreed upon by the Cedant and the Reinsurer for the details reported by the Cedant will be updated to the extent possible. Any changes to the data field layout last

agreed to with the Reinsurer must be communicated in writing to the Reinsurer at least 30 days before the end of the applicable reporting period.

Sample: Current

Objective: Reestablish the requirements in the treaty to clearly state the needs of the Reinsurer to better utilize technology and to satisfy the needs of our stakeholders. Develop treaty language to be used as a source of reference with a minimum data requirements template that is promoted by RAPA.

SMART Goal Set

Jul 2018•Create deck and plan on the Data Quality focus

Aug 2018•Discuss advisory committee review

• Incorporate feedback

Sep 2018• Share the deck with

RAPA Data Quality group

Oct 2018• Discuss at the RAPA

Conference

Feb 2019• Prepare a draft

• Advisory committee review

Mar 2019• Complete proposal

Apr 2019• Discuss at the RAPA

Spring Conference

• Determine next steps

Develop Plan

Treaty: Cession Administration

Minimum data requirements

Frequency of the reporting

Consistent file name and data

format

File type provided (Inforce,

Transactional, Billing, Claims)

Machine readable format

Plan allocation

Communication of changes to:

Data transmission

File format

Actuarial reserves

Execute

Minimum data

requirements

Frequency of the

reporting


Data transmission

File format

Actuarial reserves

Elements for the ProposalCurrent

Factors for completeness:

1. Utilizing the Treaty as the agreement to support our administrative requirements.

2. Developing a sample of Treaty wording for Cession Administration that can be used as reference.

3. Promoting a base of reference in communicating the needs of a reinsurer through RAPA.

Complete Objective

Minimum data requirements

Frequency of the reporting

Consistent file name and data

format

File type provided (Inforce,

Transactional, Billing, Claims)

Machine readable format

Plan allocation


Data transmission

File format

Actuarial reserves

Administration Requirements

POST-LEVEL TERM INITIATIVE

POST LEVEL TERM

STARTING AT THE END

RAPA PLT Committee

2018

Data Inconsisten

cies

Shock Lapse

Timing Issues

Premium Inaccuracies

Data Analysis & Reporting

System Administrati

on

Standardization

Product Innovation

Addressing Financial Volatility

For many years, level term policies entered into administration systems have been entered as quickly and efficiently as possible. Often times the need to enter plans

with descriptive titles and actual term periods has been overlooked and due to time constraints there has been a

lack of forethought to what happens at the end of the level period. Conventional thinking has been that the policy would terminate prior to the post level period.

What happens if it does not?

So that is where we start…at the end.

Why do we need to discuss PLT?

Experience shows that policies are approaching and entering the post level term period more frequently and that is leading to some industry wide issues:

• Data Inconsistencies

• Premium Inaccuracies

• Administration Inconsistencies

• Shock Lapses

• Difficulty Modeling

• Financial Volatility

• Communication Gaps in Reinsurance Partnerships

• New Product Implementation Administration Challenges

• Resourcing Gaps to Handle Administration Challenges

Where do we begin?

As we reviewed issues and challenges presenting themselves as a result of Post Level Term Administration we arrived at four distinct categories that covered the issues presented.

Administration & Processing

Communication

FinanceEmerging

Trends

Data Integrity

• Conduct system data integrity reviews to identify inaccuracies

• Create log of identified issues and develop corrective action plans

Corrections

• Develop timelines for all corrections and communicate plans and timeline to Internal Customers, Reinsurers and Retros

• Understand implications/downstream effects of corrections within the reinsurance system

Implementation

• Develop governance and guidelines to ensure products are implemented into the system accurately

• Create a repeatable process that includes appropriate testing of administration

Controls

• Develop appropriate controls to ensure previous issues are mitigated in the current environment

• As corrections and implementations occur ensure proper signoff from downstream consumers of data

Addressing Data Issues – Past and PresentMany current day data issues are the result of historical system migrations, upgrades and poor administrative practices. As we move toward leveraging our data throughout our organizations, the accuracy and integrity of the data becomes increasingly more important.

Shock Lapse

Inconsistency in the administration of policies entering the post level term period create difficulties for Reinsurers and Retrocessionaires. For Reinsurers and Retrocessionaires tracking direct writers method of administration can be helpful. Implementing predictive analysis to address each direct writers preferred method can provide the ability to produce more accurate estimates for accruals and reduce the amount of adjustments required.

Policies are kept active but remit premiums based on the last year of the term period

PLT A

ctive Policies

Policies are suppressed for 2-3 months to off-set the “Shock Lapse” activities

PLT

Po

licy

Sup

pre

ssio

n

PLT

Pre

miu

m P

ayin

g

Post level term premiums are paid and lapses are applied as they occur

Premium Inaccuracies

Correct Premiums

Ensure correct premium rates at renewal and verify rates within reinsurance system are accurate and complete.

Conduct trending and predictive analysis to identify large premium increases on level term blocks to signify the entrance of policies into the post level term period.

Premium Rate Implementation

Implement premium increases utilizing appropriate methodology and governance to ensure appropriate timing and communication to internal stakeholders, reinsurers and retrocessionaires.

Ensure testing addresses all use cases to mitigate the need for recoding or corrections.

Repricings

Initiate communication with Reinsurers or Retrocessionaires early in the repricing process to ensure engagement and ease negotiations.

Ensure appropriate treaty setup in preparation for repricings by leveraging reinsurance system test environments to test rate application and premium calculations.

Data Analysis & Reporting

Timely Updates

Prepare and communicate timely updates of both assumed company and reinsurance administration systems to ensure accurate premiums, reserves, financial modeling and reporting.

Resource Allocation

Create a resource allocation and training plan to ensure the appropriate resource skill mixture is available to work on administration projects related to Post Level Term products.

Risk Mitigation &

Analysis

Implement risk reviews and quality assurance checks and analysis to obtain the appropriate comfort with data being received and reported.

Reporting Consistency

Implement a practice of providing descriptive and consistent naming conventions for plan/treaty codes in the reinsurance administration system. Eases administration for both direct writers, reinsurers and retrocessionaires. Catalog or keep a library of treaty mapping to plan and treaty codes to aid responses to reinsurance partners.

System Administration

Term Campaigns− Often times campaigns are unique and difficult to administer.

− Provide/Obtain detailed information during negotiations to ensure assuming administration team understand the product(s) and the challenges administration may present.

Addressing Direct Writers Flexible Products

‒ Direct Writers like to offer flexible products to their clients at competitive prices and this can create challenges for administration as they strive to retain business.

‒ Reinsuring products may require complex treaty setup to be administered correctly in the reinsurance system. Administration representatives should receive notifications throughout negotiations and should start writing business requirements early to determine if the complexity can be handled with base reinsurance system functionality or if custom modifications are required.

Administration System Setup

‒ Leverage reinsurance system test environments to ensure proper setup of treaties, pointers and rates.

‒ Leverage project implementation plans to create a repeatable process with appropriate controls and signoffs to mitigate the risk of incorrect system setup.

Better Data, Better Analysis

Data

Multiple options have been added to handle an

organization’s preferred method of administering PLT policies on the administration

system

Delayed Renewal Processing & Automatic Recalculation in 1st

year of Post Level Term are a few of those methods

System

Leverage System Updates to improve administration

processing, data quality and analytics

For example reinsurance systems have been updated to include switches that indicate

the level term period on a treaty code to aid in

communication to reinsurers and retrocessionaires

Analytics

Utilize proactive monitoring of level term business in

conjunction with historic data/ lapse and mortality studies to facilitate the adjustment and

scaling of estimates

Leverage new data fields in predictive analytics to estimate

financials and approximate accruals as they achieve post

level term period

Create industry standards around…

PLT Administration & Premium Reporting

Graded Periods

Scaled Premium Rate

Implementation

Lapse Practices

Create guidelines to standardize reporting files to reinsurers, so the need for translation/interpretation is minimal

Create guidelines to standardize graded periods used in pricing new products and the method of implementation in reinsurance systems

Ensure that scaled premium rate implementations adhere to specific guidelines to allow for consistency in modeling

Develop products and practices to mitigate shock lapses with insureds

Financial Volatility

As treaties and amendments are negotiated ensure representatives from all areas are present – Administration, Finance, Actuarial

Fluctuations in cash flow caused by paying the first year PLT premium and having the policy lapse a month or two later.

Direct writers decrease premiums charged to insureds but do not notify the reinsurer of the change in rate structure.

Incorrect premium rates are coded in the reinsurance system and PLT policies that truly remain inforce are paying the incorrect premium. When this is discovered there is often a fairly significant financial impact.

Run audit checks of premium calculations against entire blocks of term business by sample selection or by treaty code to effectively review large blocks. Ensure rates are populated in the reinsurance system beyond the term periods final duration.

Adjust parms and invest in system modifications to communicate specific arrangements/rate structures/unique product terms via reporting to reinsurers and retrocessionaires.

Today we see… Tomorrow we should…

CommunicationIn order to manage PLT business efficiently and effectively special attention is required to how you implement and evaluate change with in the organization. Imploring a consistent methodology can increase reinsurance partner communication, while minimizing the time and effort required to do so.

Initial ResearchDetermining

Course of ActionImplementation

Initial Research: Identify Business✓ What is the level term period being

analyzed? Where is the administration performed?

✓ Are there key dates when a large count of policies will enter PLT?

✓ Who are the reinsurers?✓ Are we repricing business? ✓ Has my direct writer contacted me about

repricing business?

Determining Course of Action✓ Communicate Intent

• Reprice or grade direct premiums?

• Allow policies to enter PLT ungraded?

✓ Provide Details • Assumptions for potential

grading decision• Describe administrative changes

✓ Identify Risk/Concerns• Will reinsurers want to opt out

Implementation✓ Execute Treaty Amendments✓ System Implementation

• Share specifics• New TAI Treaty codes• New Rates

• Settle-up vs Programming Changes

✓ Audit

Conclusion

As term business approaches the post level term period, there are administrative implications and business decisions that both the direct insurer and participating reinsurers must be aware of. Predictive and smooth earnings with minimal volatility of premiums paid, while maintaining sound administrative practices that adhere to the governing treaties is always the goal. Insurers, Reinsurers and Retrocessionaires can use effective communications, standardized administrative processing guidelines and innovation to ensure administration runs smoothly, accurate reporting is produced and business profitability is achieved.

INNOVATION INITIATIVE

169

Co

ok

ho

use

Lab

LEVERAGING COLLABORATION ACROSS THE RAPA COMMUNITY

169

Co

ok

ho

use

Lab

170

Co

ok

ho

use

Lab

• A treaty management hub to help automate the treaty management process

• Industry standard language to simplify the complexity of treaty language

• A contributory database to give the industry a baseline for benchmarking

• A premium validation tool to address the risk of future financial impact

INDUSTRY IDEAS GENERATED BY RAPA COMMITTEE

171

Co

ok

ho

use

Lab

WE SURVEYED THE RAPA BOARD & BROADER MEMBER COMMUNITY TO DISCOVER WHICH IDEA WOULD BR ING THE MOST VALUE

172

Co

ok

ho

use

LabOUT OF MORE THAN 100 SURVEY RESPONDENTS…

A premium validation tool to address the risk of future financial impact37%

173

Co

ok

ho

use

Lab

WORKSTREAMS IDENTIFIED

5

Co

ok

ho

use

Lab

PMO / LOGISTICS

174

Co

ok

ho

use

Lab

6

Co

ok

ho

use

Lab


PMO / LOGISTICS BUSINESS PROCESS

175

Co

ok

ho

use

Lab

7

Co

ok

ho

use

Lab


PMO / LOGISTICS BUSINESS PROCESS TECHNOLOGY

176

Co

ok

ho

use

Lab

8

Co

ok

ho

use

Lab


PMO / LOGISTICS BUSINESS PROCESS TECHNOLOGY MARKET / DATA

INDUSTRY

177

Co

ok

ho

use

Lab

9

Co

ok

ho

use

Lab


PMO / LOGISTICS BUSINESS PROCESS TECHNOLOGY MARKET / DATA

INDUSTRYCOMMUNICATION

178

Co

ok

ho

use

Lab

SLIDE 6

10

Co

ok

ho

use

Lab

WHAT WOULD THIS TOOL NEED?

179

Co

ok

ho

use

Lab

GENDERSMOKER

CLASSAGE BASIS

UNDER-WRITING

PLT

BENEFITS RATINGS DURATIONSELECT/

ULTIMATE

1791

Co

ok

ho

use

Lab

WHAT WOULD THIS TOOL NEED?

180

Co

ok

ho

use

Lab

1802

Co

ok

ho

use

Lab

WHAT DID WE CONSIDER NEXT?

181

Co

ok

ho

use

Lab

SCALABILITY

1813

Co

ok

ho

use

Lab


182

Co

ok

ho

use

Lab

SCALABILITY TRANSPARENCY

1824

Co

ok

ho

use

Lab


183

Co

ok

ho

use

Lab

TRANSPARENCYSCALABILITY USABILITY

1835

Co

ok

ho

use

Lab


184

Co

ok

ho

use

Lab

TRANSPARENCYSCALABILITY USABILITY AFFORDABLE

1846

Co

ok

ho

use

Lab


185

Co

ok

ho

use

Lab

TRANSPARENCYSCALABILITY USABILITY AFFORDABLE

1857

Co

ok

ho

use

Lab


186

Co

ok

ho

use

Lab

CARRIER

REINSURER 1

REINSURER 2

REINSURER 3

Ceded Premiums

Ceded Premiums

Ceded Premiums

1868

Co

ok

ho

use

Lab

PROPOSED SOLUTION

187

Co

ok

ho

use

Lab

1879

Co

ok

ho

use

Lab

W E E N V I S I O N T H I S T O B E A N I N D U S T R Y - W I D E P L A T F O R M T H A T C A N A L S O B E U S E D T O G E N E R A T E B E N C H M A R K I N G D A T A

188

Co

ok

ho

use

Lab

20

Co

ok

ho

use

Lab

MORE DETAILS AVAILABLE IN THE WHITEPAPER

THE ROAD TO BETTER DATA

AN INDUSTRY WHITEPAPER FOR L IFE REINSURANCE

BREAKOUT SESSION

Breakout Sessions

1. Executive - Eddie Martinez

2. Planning - Kelly Priest

3. Data Quality - Dean Santos

4. Risk Initiative - Amalia Figueiras

5. Post Level Term - Brittainy Jones

6. Education - Dalia Khoury

7. Innovation - Mitch O’Campo

8. Audit & Compliance - Lynn Martone

Documents

RAPA | 2018 Annual Conferencereinsadmin.org/wp-content/uploads/2015/07/MASTER...the U.S. in the historical Louisiana Purchase in1803, and migrated by Americans, French, Creoles, Africans,