With more clients going live with a point-of-quote integration and retrospective value analysis (retros) continuing for LexisNexis® Vehicle Build, and the ADAS risk classification coming to completion with more insurers in the UK, we wanted to give an update on the progress and the people involved.
When we conduct a retro, we play back actual data from a client’s book of insurance business (a data append) from the recent past and we’re able to join that with the Vehicle Build classification to show real-world benefits for an insurer. An insurer will provide a file that is representative of their book of business, where they have had exposure to claims (normally for 12 months).
Let’s start with few observations about how we got to the current point: why is it that ADAS data holds such promise for the future of auto pricing and underwriting? What are the types of technology combinations and clusters that are really predictive of insurance loss? What type of path is it creating for motor insurance rating in the future?
The Vehicle Build ADAS classification system standardises vehicle safety features for motor insurance providers. It is no industry secret that there are multiple sets of proprietary ADAS features and their diversity is further complicated by car manufacturer naming conventions, both as individual items and when offering them as a ‘pack’.
The logical sequencing and classification of hundreds of these variations into a common taxonomy enables insurers to more easily ascertain how these features affect a vehicle’s risk profile. From these many variations, the classification derives approximately 65 individual features, grouped into 31 feature groupings. Out of this total, 12 core feature groupings have been found to be most material in insurance outcomes and impact on claims, and all of which are delivered through Vehicle Build.
Why is ADAS important?
At LexisNexis Risk Solutions we have normalized a huge amount of car manufacturer data so that we can actually return this to the insurance provider as actionable insights in form of attributes. This identifies whether the ADAS features were fitted as standard, or if they are optional extras, because Vehicle Build is a VIN-level solution and confirms fitment on that precise vehicle.
But we don’t just identify if the feature is present or not. We are also able to tell the insurance provider how the feature behaves. Is it there to warn the driver? Is it there to assist the driver? Or, in more advanced systems, we identity when a feature is there to actually take over operational control of the vehicle (such as automatic emergency braking).
With Vehicle Build we have developed the infrastructure that will be required to connect insurance to the growing roster of vehicle tech features and vehicle autonomy in the future. We wanted to be able to tell the insurance provider, so, how does a specific feature perform, what’s its purpose? Is that feature there to avoid a collision or to reduce the severity? Or in some circumstances, is it there to maintain safe driving conditions, on the road, or even to create a better outcome in a post-collision circumstance?
For example, where likely impact is detected, the ADAS features can trigger hazard warning lights and pulse the rear light clusters to draw attention to the vehicle. Ensuring that windows are closed and doors are unlocked can take place, and also applying the brakes to prevent the vehicles being forced into any knock-on collisions. Occupant protection can be achieved by pre-tensioning of seat belts and bracing to help reduce further injury.
Within Vehicle Build we’ve created the 12 core feature groupings mentioned above, and we studied features individually and how they impact on insurance claims when interacting together. The classification of each core feature group was created from research on how ADAS moved the needle in claims frequency, plus an understanding of the automotive and insurance markets.
Forward Collision Mitigation: ADAS feature group example
Our research* has shown a highly complex picture in terms of the impact on insurance loss cost, seen when looking across different feature groupings and types of coverage (for example third party liability or TPL, windscreen cover and other coverage types).
Looking at TPL cover, cars with at least one core feature grouping showed between 5 and 30 percentage points lower claims frequency compared to the average of the market book. These vehicles also showed between 5 and 15 percentage points reduction in loss ratio.
The really interesting thing that we found, that while there was a decrease in claim frequency, the impact on claim severity was different and variable when looking at the correlation to each ADAS feature grouping, the correlation to different types of coverage, and claims costs overall.
There’s also a lot of variability between typical vehicles in the UK and the other markets where we work, in the EU and US. There are many pockets of opportunity for improved accuracy in underwriting, risk selection in pricing.
Insurance providers also need to understand how vehicles compare, not just how each ADAS feature affects driver behaviour. This can be achieved through the Vehicle Build ADAS Rating Indicator, which gives a value from 0-5 in terms of reducing claims frequency based on the type and combination of features on a specific vehicle. The higher the number, the better the performance in terms of reducing claims frequency. This single variable allows for rapid integration for an insurer and is especially valid where the likely application is to apply a percentage premium discount where important ADAS features are present.
This type of data enrichment will be business critical in the short term as ADAS becomes more prevalent in the driving behaviour of newly-manufactured vehicles. The insights shown below highlight the impact and claim relativity when a vehicle is equipped with at least one core ADAS feature grouping.
Based on our analysis of more than three million vehicles across Europe, on average in 2021 there are eight ADAS safety features per vehicle in the UK compared to Spain which has six and Italy which has four.
Examples: ADAS feature combinations and impact on claims frequency
Overall industry claim relativities when one or more core ADAS features is present: bodily injury -27% claim frequency, property damage -19% claim frequency*.
We have also carried out a multivariate analysis to see the impact of combinations of ADAS features on claim frequency. We used a decision tree machine learning algorithm, and this considers interactions between variables.
We found that when vehicles are equipped with Forward Collision Mitigation, Adaptive Cruise Control, Blind Spot Mitigation and Lane Departure Mitigation we see a reduction of 25% in property damage claim frequency relativity. When vehicles are equipped with Blind Spot Mitigation, Lane Departure Mitigation and Adaptive Cruise Control we see a 13% reduction in property damage claim frequency relativity.
However, some of these features are optional extras, so just having model type level information isn’t sufficient. It’s crucial that we have VIN level information so that the insurance provider can know exactly what’s on the vehicle. Therefore, the Vehicle Build solution is at VIN level.
When we look at different types of losses, it’s necessary to keep in mind that insurance claims can involve an ADAS-equipped vehicle, two or more ADAS-equipped vehicles, and older lesser-equipped vehicles. Our research* provides the necessary technical analysis on where the underwriting ‘lift’ effect can be found.
However in general we’ve been urging insurance providers to use caution about applying generalities to any specific insurance book of business. We can work with any insurer on their own R&D and applying the data according to their own strategy and risk profile.
TPL claims at fault for example have a significantly different behaviour than not at fault claims. While ADAS features aim at warning the driver from their actions and potentially taking control of the vehicle itself in order to prevent accidents, these new technologies have a limited effect at preventing accidents derived from other drivers’ actions.
For this reason safety features have a significant effect on reducing at fault TPL claims in particular. ADAS technologies may have an effect of reducing the overall impact of accidents pushing claims costs to reduce. For example, it may not be possible to avoid a collision in all cases but the impact speed will be significantly reduced and a side impact could be reduced to a glancing blow.
In the event of an accident, the presence of ADAS features will result in the need for system calibration once the repair is completed. ADAS sensors or a radar unit may require replacement if severely damaged and certain car manufacturer repair methods can differ for ADAS-equipped vehicles. However, more body-shops and repair centres are obtaining the equipment to perform the calibration work in house and as car manufacturers remain committed to make their vehicle components and repair processes competitive, as with any new technology, prices can reduce over time.
Therefore it is going to be important for insurance providers to use the available data – and to use their market power in the overall claims and repair elements of the value chain – to keep as much control and visibility of these costs as possible. The same goes for using the power of this knowledge in contracts related to subrogation agreements and negotiations with other insurers for specific types of claims.
ADAS risk insights from UK, EU and US vehicles
What we found from our analysis was that the core ADAS feature groupings have the biggest impact on claim frequency with lesser impact on severity. In the US for example, which is more evolved in this space with a younger average vehicle age, the impact on claims severity was stronger compared to the UK and correlated more to underwriting ‘lift’.
As we know, in the UK eight out of ten new cars are sold in the UK with some form of ADAS. As technology develops, we would expect this percentage to increase. The crucial thing though, when we analysed a sample of European insurance policies, was a 14% reduction in pure premium loss ratio when the vehicle had an ADAS Rating Indicator of one or more. So in summary, our US analysis shows the sharpest reduction in claim frequency, and our European analysis shows that our ADAS features offer additional segmentation beyond the current predictions of insurance providers.
Ultimately Vehicle Build is seeking to deliver insight into the insurance market, on these safety features that are fitted to any individual vehicle that insurers are offering cover for, today. Taking into account the current traditional demographic score model, we’re able to say with certainty that Vehicle Build contributes to delivering a more accurate insurance quote, based on the specific claim relativities for any individual vehicle (bodily injury, property damage and collision costs) that we’ve identified in the market today.
It is an ongoing exploration with data that is continuing to deliver better and better underwriting results as we continue to grow our data sample. It is already starting to really help insurers with their pricing and underwriting decisions.
*Based on LexisNexis® pan-European research of more than three million vehicles and actual insurance losses in real world driving.
Follow the link to the LexisNexis Risk Solutions website to find out more about how we support insurance providers.