475 N. Martingale Road, Suite 600
Schaumburg, IL 60173
P +1-847-706-3500
F +1-847-706-3599
SOA.ORG
2019 Student Research Case Study Challenge
Design an Autonomous Vehicle Insurance Policy
Client Overview
You and your team of actuarial consultants have been assigned to a new client project. Sofia J. Fibonacci,
FSA, chief actuary of the automobile insurance company Safelife, is preparing for her presentation to the
Board of Directors on how to design an automobile insurance product for autonomous vehicles, and she
needs your help.
Safelife has an 82-year history selling personal and commercial automobile insurance policies in the
country of Carbia, a developed nation that has similar laws and regulations to the majority of states in the
United States. At present, the top five automobile insurance companies in Carbia control about 70 percent
of the market. With a 34 percent market share, Safelife is in the lead. In 2018, Safelife paid claims of
approximately 4.5 billion Carbs (Ĉ 4.5 billion), the currency of Carbia.
However, Safelife anticipates that the current automobile insurance market may eventually decrease with
future public adoption of autonomous vehicles. Such a shift from traditional automobile products to
autonomous vehicle products is likely to usher in a new era of risk and liability considerations that will
demand big data capabilities and new actuarial models. At this time, no other company in Carbia has
announced plans to design an insurance policy for autonomous vehicle owners, leading Safelife to believe
there is potential for a first-mover advantage.
Accordingly, Sofia has been assigned to lead the design and development of a new insurance policy for
autonomous vehicles. Her current management goal, assuming the evolution of autonomous vehicles
grows as anticipated, is for the new policy to account for 20 to 25 percent of Safelife’s overall business by
the year 2030.
Project Objectives
Your objectives for this project are two-fold:
1. Recommend a design for a new autonomous vehicle insurance policy, and develop loss cost
estimates, also known as pure premiums, for the new policy. Recommend a launch date for the
policy and provide a ten-year forecast of pure premium from the point of launch.
2. Over this same time horizon, forecast what Safelife’s pure premium will be from their traditional
policies and the new autonomous vehicle policy combined, assuming they choose to develop your
proposed autonomous vehicle policy. Also include a baseline forecast of pure premium assuming
they do not develop an autonomous vehicle policy.
In anticipation of autonomous vehicles, the chief actuary and Board of Directors have kept a close eye on
federal efforts to create autonomous vehicle legislation. In completing these two project objectives, they
have asked you to consider how future legislation may impact insurance and liability regulations for
autonomous vehicles. The government is currently developing autonomous vehicle and insurance
regulations that mirror legislative efforts in the United States.
2019 Student Research Case Study Challenge 2 of 6
Deliverable Requirements
The expected deliverable is an executive report, informed by your research and analysis, summarizing your
team’s recommendations and addressing each of the following items in bold:
• Which risks should the new policy cover and for whom? For any party covered by the policy, please
describe their relationship to the vehicle and what the benefits of the policy will be.
Key considerations:
o Who is most likely to adopt autonomous vehicles in Carbia, how will they be used, and
what new risks will emerge as a result?
o What is the anticipated timeline for autonomous vehicle adoption?
o Which parties will face risk or liability exposure as a result of autonomous vehicle
adoption, and what is their relationship to the vehicle?
o To what extent should the new policy include commercial and/or personal coverage?
o How and to whom will liability be assigned in the event of a claim?
o How will current and anticipated regulations impact your policy design?
o To what extent should the policy cover traditional automobile insurance risks versus new
risks expected to emerge with autonomous vehicle usage?
• At what level(s) should Safelife expect the new autonomous vehicle policy’s pure
premiums?
Key considerations:
o What do you expect the frequency and size of claims to be?
o What do you expect the per-policy pure premium to be?
o How could the timing of market entrance impact pure premium?
• When should Safelife launch the policy, and what are the anticipated pure premium projections
based on anticipated sales? Please provide a 10-year forecast of pure premium, starting at the
launch date, to inform your recommendations.
Key considerations:
o How many autonomous vehicle policies do you expect Safelife to issue over the 10-year
time horizon from the point of introduction?
o How many traditional vehicle policies do you expect Safelife to issue over the 10-year
time horizon from the point of introduction?
• Over the same time horizon, provide a baseline of Safelife’s pure premium projections, assuming
they choose not to develop the new policy.
Key consideration: how will the introduction of autonomous vehicles affect the frequency and size
of claims over time for Safelife’s traditional personal and commercial auto?
2019 Student Research Case Study Challenge 3 of 6
• Please include a sensitivity analysis on management’s goal that the new policy account for 20 to 25
percent of overall business by the year 2030. Please make recommendations based on your analysis.
What are the range of potential risks and rewards associated with following each of your
recommendations?
Key consideration: what is the impact on pure premium for insurance policies for both
autonomous and traditional vehicles.
• Identify the key assumptions underlying your policy design and pricing that Safelife will need to test
in order to bring your proposed policy to market successfully.
Sofia requests that you include discussion of data limitations where relevant throughout your executive
report. For example, how complete is the data underlying your assumptions and conclusions? What
limitations were involved in completing your analysis?
Automobile Insurance Policies
Carbia is primarily an urban and suburban country, with very little rural area. Carbian law currently requires
that all automobile owners purchase automobile insurance. Because the law is strictly enforced, all
automobiles registered in Carbia are insured. Refer to Case Study Appendix 1: Automobile Insurance in
CarbiaCase Study Appendix 1: Automobile Insurance in Carbia for more information about insurance
policies in Carbia. Safelife policies reflect these requirements.
Available Data
Sofia has provided you with ten years of Safelife claims data history for existing personal and commercial
automobile policies. Please refer to Case Study Appendix 2: Historical Claims Data for more information
about the data. Assume that the data is complete.
In developing predictions relating to the new autonomous vehicle insurance policy, you may use the
current Safelife data, build your assumptions from scratch based on external research, or use some
combination of these two approaches.
Report Specifications and Evaluation Criteria
Sofia has explained that Safelife evaluates its consulting teams on their methodological soundness.
Therefore, Safelife reviews deliverables for the following criteria:
• Organization, form, clarity and cohesiveness of the report
• Clear responses to all questions,
• Thorough analysis,
• Creative and strategic recommendations,
• Consideration of secondary research—including, as necessary, an appendix showing a reference
list of all sources cited throughout the report,
• Consideration of the provided data, including documentation of data limitations, and
• Documentation of assumptions, including currency assumptions.
Your report and any supporting calculations and/or code should be submitted in PDF and/or Excel format.
To ensure every request is met, please carefully review the submission requirements on the following page.
2019 Student Research Case Study Challenge 4 of 6
Submission Requirements
Your team’s submission should contain a written report and supporting calculations that meet the criteria
outlined below. Please refer to the 2019 Student Research Case Study Challenge Official Rules for further
information.
Written Report
Submit a written executive report that meets the following criteria.
1. Includes the team name, names of each team member and an email address for each team
member
2. Addresses each item identified under “Deliverable Requirements”
3. Written in English
4. In PDF format (without any passwords)
5. Includes the team name and brief description in the filename of submitted files for quick
identification
6. In 12-point, double-spaced font with standard margins
7. A maximum of 2,500 words, excluding the following items, which do not count toward the 2,500-
word limit:
a. Figures of graphs/charts with captions summarizing key findings
b. Citations of external research
c. Reference list of cited sources at the end of your report
d. Supporting calculations and/or code
8. The report and supporting calculations must be original work of the submitting team. Previously
published work should not be repurposed or submitted.
Supporting Calculations
Submit an Excel and/or PDF file containing all calculations, models and/or code completed during your
analysis. Include the team name and brief description in the filename of submitted files for quick
identification.
When and Where to Submit
Final submissions must be sent to resear[email protected] by 11:59 pm Central Daylight Time, Friday, March 22,
2019.
Important Note
Please keep in mind the audiences who will read your report. While Safelife’s chief actuary and her
actuarial team will be interested in your in-depth calculations and analyses, corporate executives and the
Board of Directors will only have the time and expertise to read your written narrative. Therefore, please
be sure to include any conclusions you wish to highlight in the body of your written report.
2019 Student Research Case Study Challenge 5 of 6
Case Study Appendix 1: Automobile Insurance in Carbia
Carbian law requires that all automobile owners purchase automobile insurance. Before automobile
ownership may be transferred legally, the new owner must provide proof of insurance that includes each
of the following coverages. Safelife policies reflect these requirements.
• Bodily Injury Liability: Safelife pays for reasonable expenses for necessary medical and funeral
services for others who are injured or die in an automobile accident when the insured is driving,
regardless of fault. Safelife also pays for injury to others, including death, when an insured is
legally liable for damages. If an insured is sued, Safelife provides legal representation. The
maximum amount payable to a single person is Ĉ150,000 and the maximum amount payable per
accident is Ĉ300,000.
• Personal Injury: Safelife pays the insured's necessary and reasonable medical expenses that are
not covered by medical insurance and have been incurred as a result of an automobile accident,
regardless of who caused the accident. The maximum amount payable to a single claimant is
Ĉ30,000.
• Property Damage: Safelife pays for physical damage to, or destruction of, tangible property of
another, including loss of use, for which an insured is legally liable. If an insured is sued, Safelife
provides legal representation. The maximum amount payable to a single claimant is Ĉ150,000.
• Collision: Safelife pays for repair or cash value of the insured's vehicle if it collides with another
vehicle, flips over, or crashes into an object (except animals), regardless of who caused the
accident. The policyholder is responsible for the first Ĉ3,000 of cost, and the maximum amount
that Safelife will pay in a single incident is the cash value of the vehicle less Ĉ3,000.
• Comprehensive: Safelife pays for repair or cash value of a vehicle from incidents other than a
collision including: fire, theft, windstorm, vandalism, flood, damage from falling objects or hitting
an animal. The policyholder is responsible for the first Ĉ3,000 of cost, and the maximum amount
that Safelife will pay in a single incident is the cash value of the vehicle less Ĉ3,000.
2019 Student Research Case Study Challenge 6 of 6
Case Study Appendix 2: Historical Claims Data
Safelife has provided you with ten years of historical claims data from their personal and commercial
automobile policies. Assume that the data is complete. This data can be used to forecast what the future
might look like based on how the business looks today. To make predictions about how the future will look
after the launch of a new autonomous vehicle insurance policy, you may adapt the current-state data or,
alternatively, build your assumptions from scratch based on your research.
The data show the following items for each calendar quarter from 2009 through 2018:
• Calendar year and quarter
• Risk Class: Safelife classifies risks according to three categories, each with three classifications, for
a total of 27 risk classes. The classes are coded by the first letter of the classifications within each
category, in the following order:
Driver Age
Driver Risk
Young
Low
Middle
Average
Senior
High
For example, a policy that covers a large car with a primary driver who is middle-aged and has a
driving record that indicates a higher likelihood of future claims-causing incidents would show a
risk class code of LMH.
• Policy type
- Personal: policies owned by individuals for their automobiles that are primarily for non-
commercial use.
- Commercial: policies owned by businesses for automobiles that are used primarily for
business purposes.
• Number of car years of exposure: the annual equivalent of the number of insured vehicles in the
quarter. A single vehicle covered for a full quarter is ¼ of a car year of exposure.
• Number of Claims Incurred: the number of claims that occurred in the quarter.
• Amount of Claims Incurred: the total amount paid for claims that occurred in the quarter,
including payments that were made in subsequent quarters. Once a claim is received, Safelife pays
it in full and closes the case; no further payments will be made.
1
Amounts are in Carbs (Ĉ) and are
net of deductibles and copayments. Safelife uses neither reinsurance nor coinsurance.
1
Closing claims cases upon payment is a simplified approach adopted for the Student Research Case Study Challenge.
In reality, cases may be reopened for supplemental claims many months or years later.