Tag Archives: auto insurance

Addressing actuarial and tax issues arising from premium rebates

An unexpected issue faced by property & casualty insurers during the COVID-19 pandemic has been premium refunds to policyholders – especially on personal auto policies.

The refunds and rebates are justified by substantial reductions in auto claims and losses because people have driven less as a result of being sheltered at home, and as stores and restaurants have remained closed or with restricted operations.

Beginning in late March, shortly after stay-at-home orders were widely imposed nationwide, many insurers began voluntarily paying partial premium refunds to personal auto policyholders. In addition, some state insurance commissioners have mandated auto premium refunds because of the reduction in miles driven. In addition, where permitted, some companies have begun offering discounts on renewal premiums to reflect the current better-than-expected loss experience instead of or in addition to refunds on current policies.

In this article, Milliman’s Susan Forray and Eric Krafcheck discuss actuarial considerations for premium refunds. They also provide perspective on the tax treatment of premium refunds from the standpoints of insurers and policyholders.

Foley & Lardner’s Richard Riley, Jr., also contributed to this article.

How could COVID-19 affect the auto insurance industry?

While the health and welfare of people around the world is the most important concern, the outbreak of the coronavirus will no doubt have far-reaching effects on all aspects of our lives, including our driving behaviors. As a result, auto insurers may likely see short- and long-term effects in their claims experience.

In the short term, the auto insurance industry will likely see a sudden and sharp decline in the volume of reported claims due to social distancing and stay-at-home mandates. Whether or not the effects of social distancing will be felt over a prolonged period will be dependent on how quickly the economy recovers once day-to-day life returns to normal. A sluggish economy could have lasting long-term effects on auto claims frequencies.

In this article, Milliman’s Paul Anderson, Eric Krafcheck, and Katherine Pipkorn discuss what auto insurers may be facing in the weeks and months ahead.

Can a bonus malus system evaluate motor liability insurance risk better?

Under a bonus malus system, motor liability insurers can adjust policyholders’ premiums based on individuals’ claims histories. For instance, a customer may receive a reduction, or “bonus,” on a premium if no claim is made during the previous year. Conversely, the customer may receive a premium increase, or “malus,” if a claim is made during the previous year.

In this article, Milliman actuary Diana Dodu provides an analysis of bonus malus systems in several European countries. She highlights the similarities and differences between the system designs in each country.

Here is an excerpt:

• Countries that do not have a specific system defined by law, such as Poland, where the system is fully liberalised and insurers have the liberty to provide own risk coefficients and load back the premium to obtain balance; Estonia, where insurers can design their own rules and where it may seem that the maximum bonus can be achieved within several years of claims free driving; and Lithuania, where according to the Law on Compulsory Motor Third Party Liability Insurance effective from 17 May 2007, premiums are fixed by the insurer and companies can take into account risk factors.

• Countries where the bonus malus system is defined in the legislature, requiring insurers to take into consideration loss history, but which grant freedom to insurance companies to design their own rules, which is present in countries such as the Czech Republic and Slovakia.

• Countries that are regulated by law, such as Italy and Romania… and Hungary, where according to law NGM 21/2011, the number of classes are predefined as well as movement between classes depending on the number of accidents, and insurers are obligated to issue accident and claim certificates, but are also allowed to use historical data for the purposes of classification to calculate additional correction factors, and Serbia, where the bonus malus system is defined in the law on compulsory traffic insurance but insurers can use correction coefficients if they do not contradict the ones mandated by law.

• Countries such as Croatia, where there seems to be a defined system, but companies offer extra benefits such as additional bonuses above the maximum and protection of the bonus (after several years of no claims, insureds can pay an additional premium to protect the bonus in case they have an accident in the subsequent year), and Slovenia, where you can also protect your bonus, while the future premium in cases of protection of the bonus would depend on the number of accidents in past periods.

• Latvia, where companies can set premiums based on prior histories with a conversion system in which it seems that it is more difficult to move towards bonus classes and where the system is evaluated annually (on September 15).





Personal lines coverage evolves with exposure data tracking

Increasingly, individuals are having their driving habits and living environments monitored electronically. A recent Insurance Journal article cited Milliman’s Sheri Scott discussing how exposure data tracking is shaping new underwriting practices for personal lines coverage like auto insurance.

Here’s an excerpt from the article:

Exposure tracking and the advent of autonomous vehicles are shifting personal auto insurance risk exposure from dependence on driver skills, estimated distances driven and garage location to the precise determination of vehicle locations, driving habits, driving distances and traffic conditions, all determined through the collection of trip data gathered in real time.

Yet even these underwriting considerations will soon be supplemented, if not supplanted, by the loss experience of automated vehicles and their manufacturers.

This transformation will not be without risks of its own, Scott said. In particular, she cited disruption of networked communications as a hazard, especially as vehicle occupants become dependent on automated control and less practiced at taking control of a vehicle.

“If some kind of communication goes down, there could be a very serious occurrence,” she said.





Pooling autonomous vehicle risk

The production of autonomous vehicles is shifting the responsibility of automobile-related insurance risk. Some manufacturers have already stated that they would assume liability for accidents caused by faulty technology in their cars. However, autonomous carmakers would likely spread that liability out among the suppliers of the cars’ software, systems, and devices.

According to Milliman’s Chris Kogut, a Supplier Product Liability Autonomous Share (SPLASh) insurance pool may help manufacturers and suppliers effectively manage the risk associated with autonomous car accidents.

Continue reading





Telematics and machine learning can help insurers employ usage-based insurance

Telematics can help auto insurers implement usage-based insurance (UBI) by obtaining valuable data related to individuals’ driving behaviors. However, producing actionable information through generalized linear model (GLM) methods has been difficult for auto insurers. Machine learning techniques provide insurers with a better way of analyzing big data.

Milliman consultants Marcus Looft, Scott Kurban, and Terry Wade provide perspective in their article “Usage-based insurance: Big data, machine learning, and putting telematics to work.”

For auto insurers, machine learning holds the promise of enabling carriers to explore hundreds, if not thousands, of factors involved in calculating the potential risk of individual customers. Moving beyond GLM and introducing machine learning techniques with telematics data may enable insurers to leverage key competitive advantages.

UBI pricing necessarily supersedes traditional GLM methods because the complex interactions of the factors in play require machine learning to uncover them within a reasonable timeframe in order to be cost-effective. Pricing differences often cannot be fitted by GLM distributions. And correlations between telematics and non-telematics effects will tend to disturb the clarity of results in a single GLM. Distribution over different frequency and severity models only confuses the analysis of differences within telematics policies.

GLM techniques will thus always show how a business differs in terms of its dependencies on a limited set of specific factors, such as age, or how much mileage goes on a car, or other very high-level factors. But if a whole model is taken and everything modeled together to try to understand the risk for all of the policies, in the end there tends to be only a mixed bag of different effects. An insurer still hasn’t been able to look very deep into its business.