Alex Manov looks at the risks involved with named drivers and the questions insurers should be asking them
The motor insurance sector has become increasingly sophisticated in its use of data to determine risk, but there remains one major gap in our knowledge - the issue of named drivers. This knowledge gap leaves the sector prone to fraud, and customers at risk of their policy being invalidated.
Around half of all motor insurance policies have at least one additional driver who could be driving the vehicle for a significant amount of time, so a finer differentiation of risk for the named driver could be hugely beneficial. To determine whether additional information on named drivers would truly help to differentiate the risk of a policy, LexisNexis Risk Solutions analysed more than 200,000 policies with named drivers, using information from proprietary data sources covering policy history, public information and quote history.
Analysis of the LexisNexis Risk Solutions motor policy database indicates that up to half of named drivers have prior policy history as a policyholder. Furthermore, historical policy information linked to these named drivers is correlated with the loss ratio on the 'current' policy. For example, we found that the loss ratio of a portfolio was reduced by 25% if the named driver on a policy had, as a previous policyholder, been associated with their vehicle for more than three years.
Additionally, negative public records such as county court judgments (CCJs) or bankruptcy are usually associated with a higher propensity to claim. Our analysis indicated that for policies where a named driver had a CCJ and the policyholder didn't, the portfolio had a 17% higher loss ratio.
The prime reason for this analysis was to determine whether knowing more about named drivers on a policy could help insurance providers to predict the associated risks of making claims more accurately - and therefore price more accurately. The next stage was to look at how we could help insurance providers identify possible cases of 'fronting'.
Fraud and the named driver
Fronting is where somebody insures a car as a main policyholder and lists the de facto main driver of the car as a named driver in a bid to reduce the premium. This illegal practice is often only discovered when there is a claim, leading to the policy being cancelled and punitive actions taken against the people involved. It may result in a large burden of debt for the policyholder and named driver, as well as huge administration costs to the insurance provider.
Those who are 17-20 years old are nearly twice as likely as other age groups to manipulate the details they provide for an insurance quote. However, recent research on consumer attitudes to data manipulation emphasises the scale of the challenge: seven out of 10 motorists think it is acceptable to manipulate the information they provide when obtaining a quote for motor insurance from a comparison site, which have evolved to make it easier for consumers to shop around for cover. In addition, 41% of parents with children who drive said they would consider fronting to save their child money, according to research from Go Compare.
By connecting and comparing thousands of motor insurance quotes from across the market, it is now possible to offer insurance providers a clear indicator of possible fronting during the quoting process. This includes looking at cases where the named driver on a quote also appeared as the policyholder on other quotes for the same vehicle within a certain period, as well as other forms of data manipulation relating to the policyholder or named driver, such as changes to birthdates and changes to declared claims information.
Quote manipulation is an indicator of a higher risk of claim; if insurance providers can see that information provided has been changed at the point of quote and to what degree, they can make more informed decisions about the proposed risk. This helps to reduce their exposure to fraud, and enables actuaries to model pricing based on a more holistic view of risk.
Alex Manov is risk solutions statistical modeller on the EU analytics and statistical modelling team at LexisNexis
