Driving for Profit: A view of the UK private and commercial motor insurance markets 2015

In this year’s Driving for Profit report, Milliman consultants Derek Newton, Vincent Robert, and Peter Moore present the results from their analysis of the United Kingdom’s private and commercial motor market.

The overall performance of the private motor market in 2015 has resulted in a pretax net insurance ratio of -0.4%. This ratio has historically been highly distorted by material releases or strengthening of prior years’ reserves held by the Direct Line Group. Excluding the Direct Line Group, the figures indicate that the rest of the market in aggregate is operating at a more pronounced loss with a pretax net insurance ratio of -4.5% in 2015.

The UK’s commercial motor market improved in 2015 to a pretax net insurance ratio of 0.7%, the first profit seen in seven years. The better performance in 2015 is largely down to releases from prior years’ reserves.

To download the entire report, click here.

Support documentation for predictive modeling

Insurance companies have many responsibilities involving the submission and approval of rate filings, including providing support for any predictive models used to develop rates. Determining the level of predictive modeling support required can be complicated because it varies widely by state.

In this Insight article, Milliman’s Eric Krafcheck discusses the types of predictive modeling support that companies should include in a rate filing to minimize the objections and amount of time needed to receive a regulator’s approval.

Here is an excerpt:

While there are a variety of predictive modeling support documents that a filer can choose to include in a rate filing, some are more standard than others. At the very least, in addition to the indicated rating factors, an actuarial memo should be included with the filing that describes the data used, the adjustments that were made to the data, and the overall modeling process, including a description of the model validation techniques. Additionally, the memo should include a discussion of the predictive modeling method used and give any necessary specifications of the models. For instance, if a generalized linear model (GLM) was used, include model specifications such as the target variable being modeled (e.g., pure premium, loss ratio, frequency, severity, etc.), the error distribution used (e.g., Poisson, Tweedie, etc.), the link function used, and the predictor variables included in the final model. As always, if you are an actuary, consult the Actuarial Standard of Practice No. 41, Actuarial Communications, for other items that you should consider adding to the memo.

Other types of support are useful to have on hand but might not be necessary to provide in every state. For instance, some states require the submission of various goodness-of-fit measures and other model validation statistics whereas other states may not. If you do not want to provide this information in every state unless necessary, you should at least have the information readily available in an exhibit format. This will save a lot of time down the road if goodness-of-fit information is requested in response to a filing.

Overview of Solvency II external audit considerations

With Solvency II’s public disclosure requirement, European countries are considering a variety of policies for the external audit of insurer information. A significant number of member states have not yet decided on making external audit an official requirement or the supervisor is not in the position to make external audit an official requirement. In this article, Milliman consultants Luca Inserra and Glennfor Hellement highlight which countries require an external audit, and provide more details on the scope of and reasons behind the external audit policies for Europe.

Can social media and data analytics expose insurance fraud?

Insurers are now perusing social media networks to detect insurance fraud and avoid paying large claims. Here’s some perspective from Michael Henk’s article “Insurance fraud and social media.”

With the immensely high rate of social media usage, some insurance companies are turning to this large source of public information to help combat insurance fraud. As of April 2016, there are 1.59 billion active Facebook users every month globally, while 400 million people use Instagram every month and 320 million users are active on a Twitter account monthly.5 Social media users account for a substantial portion of the population, and these users are generating a massive amount of data, an insurance company’s best friend.

Each day, 500 million tweets are posted—about 6,000 tweets a second.6 As of May 2013, 4.75 billion pieces of content were shared daily on Facebook. Currently, every 60 seconds on Facebook, 510 comments are posted, 293,000 statuses are updated, and 136,000 photos are uploaded.7 Unless social media users have adjusted the privacy settings on their accounts, all information that is posted is considered public. This means that insurance companies have free access to any non-private posts from their customers (or anyone else for that matter). The insurance industry thrives off of data for business purposes such as underwriting, and now, this social media “data” may help insurance companies succeed in combating insurance fraud.

Searching through these networks can become time consuming and expensive though. Implementing data analytics in addition can optimize insurers’ time and results by sifting through extensive data and expediting a claims investigation.

While analyzing social media can occasionally result in a big win in the fight against insurance fraud, the costs will not always outweigh the benefits. However, if paired with data analytics, the likelihood of uncovering fraudulent claims can greatly increase for a company. Data analytics gives a company the opportunity to process large amounts of data and quickly identify important outliers or other potential indicators of fraud.

For example, geospatial analysis can be utilized when looking for fraud in wind, hail, or flood damage claims. This statistical analysis helps identify what geographical locations were most affected by a storm and where the company would expect to receive claims. It could also aid in identifying the likely severity of a claim based on the proximity to a storm event. Receiving a claim from outside the affected area or a very large claim from a minimally affected area should trigger the need for further investigation.

Analysts can also examine historical claims to identify typical frequencies and severities of policyholder claims. When a current period’s claims are compared to historical data, one can identify cohorts of policyholders who are either making more frequent claims or claims with higher loss amounts. Investigators may want to look further into these types of policyholders to look for further signs of possible fraud.

Reading list: Florida’s private flood insurance market

Advances in catastrophe models and new state insurance regulations have opened the door for an affordable, risk-based private insurance market in Florida. This reading list highlights articles focusing on various issues and implications related to the market. The articles feature Milliman consultants Nancy Watkins and Matt Chamberlain, whose knowledge and experience is helping insurers to understand and price flood risk more precisely.

• Bradenton Herald: “Important for homeowners to compare flood insurance options
Florida homeowners must consider the issues related to the National Flood Insurance Program (NFIP) and private flood policies. Private insurers can use predictive modeling technology to determine a home’s distinct flood risk.

• Tampa Bay Times: “Remember the flood insurance scare of 2013? It’s creeping back into Tampa Bay and Florida
Real estate and insurance experts comment on the possible effects that high flood insurance rates may have on homeowners. Insurers express interest in the granular modeling of flood-prone territories.

• Tampa Bay Business Journal: “Why some Tampa Bay property insurers are offering flood coverage and others are not” (subscription required)
Insurers need to weight the risks and rewards associated with the underwriting of flood insurance. A few carriers have already decided to participate in Florida’s private flood insurance market.

The Data Science Game’s qualification round is complete

In July, teams of data science students from more than 50 universities around the globe competed in the qualification phase of the 2016 Data Science Game. Over 140 teams of four students were asked to develop an algorithm that could recognise the orientation of a roof from a satellite photograph by building on more than 10,000 photograph of roofs categorized through crowdsourcing.

Twenty-two teams have qualified for the final phase. The top three ranking teams were Jonquille (University Pierre and Marie Curie), PolytechNique (Ecole Polytechnique), and The Nerd Herd (University of Amsterdam). The final is being held in Paris on September 10-11, where the teams will compete in a big data analysis challenge.

Data Science Game finalists

For more information on the Data Science Game, click here.

Milliman is a sponsor of the 2016 Data Science Game.

Enterprise risk management: Evolution and transformation

Enterprise risk management (ERM) is evolving, and as insurance companies have moved through time, there has been greater ability to control information in a more sophisticated way. This video featuring Milliman’s Tony Dardis provides more perspective.

To learn more about Milliman’s ERM services, click here.

Milliman releases Arius 2.8, with new documentation tools and the ability to build more complex analysis projects

Milliman announced today that it has released version 2.8 of Arius®, its state-of-the-art loss-reserving system for property and casualty insurers. This is the 10th major release of new functionality since Arius’s introduction three years ago, and an 11th release of new features is planned for fall 2016. This latest update focuses on usability and project structure, allowing actuaries to build multilayered and interrelated analyses to more realistically reflect the complex nature of today’s insurers.

Milliman does reserve analysis very much like our clients, under similar time constraints and quality standards. We understand that in addition to actuarial or statistical functionality, sometimes the best way we can help our clients is by improving the processes and approaches to their analyses. Our latest release of Arius demonstrates Milliman’s continued commitment to providing the best solutions to enhance the efficiency of our clients’ work.

With this release, Arius adds a number of capabilities:

• Reserving projects can now include multiple reserving segments that are interrelated, including segments completely or partially derived from other segments in the project—for example, for summarizing regions, sublines, or coverages.
• New reserving segments can be easily created from existing segments in the same project file, while preserving the original’s selections and other items of actuarial judgment.
• Notes are now available throughout the system to allow users to document their work at the point that specific decisions are made.
• Additional flexibility is now available in setting up interim projects that require more advanced interpolation capabilities, to more realistically address insurance contracts based on accident, policy, report, and underwriting periods.
• The system’s stochastic models are updated with the latest yield curve tables for Swiss Solvency Test and European Insurance and Occupational Pensions Authority (EIOPA) risk-free rate term structures. Both sets of curves reflect multiple currencies.

To learn more about Arius, click here.

Walking through a minefield in $500 heels: Rent the Runway and the risks of renting haute couture

Ryan-DanaWedding season is in full swing, and recently I looked into renting a designer dress for a wedding I was attending. I didn’t want to wear a dress everyone had already seen me in, but I also didn’t want to spend a lot of money on a dress I would likely only wear once or twice. Renting a dress seemed like a perfect fit.

Fashion rental companies such as Rent the Runway, Bag Borrow or Steal, and ArmGem are on the rise because of the high expense of designer clothes and the desire to wear the latest fashion. According to a Business Insider article, Rent the Runway has grown since it was founded in 2009 to a company with 5 million members and $1 billion in inventory. With these rental companies growing is there a need for designer rental insurance?

As someone who works in the insurance industry, I started to ponder what would happen if I damaged or lost a rental, some of which can be worth thousands of dollars. If I’m being honest, I am very klutzy. I tend to spill things when I’m eating and have been known to catch a piece of jewelry on my dress, leaving a pull or tear. Immediately my mind tends to think of the worst-case scenarios. What would happen if I accidentally leaned into a candle at the dinner table and left a burn mark on the dress? If I rented earrings, what would happen if one of them fell out on the dance floor and got lost? What if makeup spilled in the designer bag I rented and destroyed the inside of the bag? Suppose I stumbled in my new heels and scuffed them up or broke a heel. The possibilities seemed endless.

Concerned about the risk of damaging or losing a rental I started to question if I could purchase insurance through the rental company. According to Rent the Runway’s website, an insurance policy can be purchased with a rental for a small fee. This insurance policy covers only minor stains or damages and does not cover theft, lost items, or damage beyond minimal wear and tear. Other companies did not offer any insurance at all.

Continue reading

Pokémon GO and the non-augmented reality of risk

Pikachu and his friends have caused quite a frenzy recently. While people are enthralled with Nintendo’s Pokémon Go, the GPS-based augmented reality (AR) game presents several risks to its developer and its gamers. In his article “Pokémon Go and augmented reality: Not all fun and games,” Milliman consultant Michael Henk discusses some of these AR technology-related risks.

Concerning personal injury risks:

Firstly, AR products like “Go” provides yet another “distraction.” We’re all aware of the dangers of being “distracted.” Texting while driving is illegal in a number of cities and states throughout the country. However, drivers aren’t the only ones being distracted. Distracted walking is a growing problem, one that has arisen naturally with the increasing dependence on mobile electronic devices and one that “Go” is already contributing to. There are anecdotes all over social media about players so engrossed in catching virtual monsters that they’re running into walls and walking in traffic. …

…“Go” may lead to an increase in distraction-caused injuries and pedestrian-vehicle injuries, which is currently the fifth-leading cause of death for children ages 5 to 19. It’s not inconceivable to imagine an incident in which both the driver and the pedestrian are distracted, maybe by the same “rare” Pokémon.

What about cyber risks?

Aside from “IRL” (in real life) dangers, there’s a data security concern with some early installs. Some iOS installs of the software require the user to provide the app with full access to their google accounts, which allows access to their Gmail (theoretically being able to send e-mail from your account), files stored on Google Drive and Google Photos, among other content. The developer has responded and said this was done erroneously, and that permissions will be corrected soon, but it’s important to make sure that users know exactly what programs on their devices have access to. There are other concerns about downloading the program from non-official app stores as well, but that stands for all programs and is definitely not a “Go”-specific concern.

Legal risks?

…There’s a significant risk for trespass with AR games that utilize real-world locations. It remains to be seen whether an AR developer placing cyber-content on your property constitutes trespassing or if AR users are “engaged on a cyber plane on which you have no exclusive property claim.” There’s another legal concern with “attractive nuisance,” which states that property owners are responsible for eliminating dangerous conditions on their property which may attract children. “An individual who fails to rectify an attractive nuisance on their property is civilly-liable to injury a child sustains on it, even if the child was trespassing.” Sounds like something that may happen in the pursuit of a rare Pokémon.