Insurance Fraud EU

June 2016, London, United Kingdom

Drive Forward Tactics to Win the War on Fraud

Fighting New Types of Crime in Insurance

 

The UK government estimates that insurance contributes “£25 billion to the national output, and believes that it is facing £1.3 billion of detected fraud, with a further £2.1 billion undetected” (Source: Gov.uk).

With the same document stating that it costs the industry around £200m a year to pursue fraud, a number of data sources have been created to help insurers identify and clamp down on offenders:

• Insurance Fraud Register (IFR) - records of current and former fraudsters. Proposals are under way to allow third-party access.

• Claims and Underwriting Exchange (CUE) – a central database of motor/ home/injury incidents reported to insurers and self-insured local authorities. Travel is also being considered.

• Motor Insurers Anti-Fraud and Theft Register (MIAFTR) – a register of vehicles stolen or beyond repair.

• Motor Insurance Database (MID) - can identify organised application fraud.

• MyLicence - joint initiative between insurers, Driver Vehicle and Licensing Authority (DVLA) and Department for Transport (DoT). Includes convictions and entitlements and is being enhanced to include a no claims discount. 

Source: Gov.uk

What these various resources are hoping to do, along with insurers internal data, is form a better view of known and potential fraud incidents. The aim is to help the industry to prevent, police and prosecute more effectively. This paper will examine the key priorities insurers have when it comes to fraud, what today’s fraudsters look like and how various agencies can pool their resources to deliver a more efficient fraud response.

Key areas of interest

Interviewees for this paper insist there is no one particular type of fraud that they are more interested in pursuing over another. “It’s about all fraud,” claims Adele Sumner, Head of Fraud Intelligence and Strategic Development in the Counter Fraud Unit at RSA. “Traditionally insurers focused in the past on motor insurance fraud but this is now a managed risk. The fraudsters are aware that we are targeting them.” She adds that motor is still a focus, along with liability and slip and trip claims.

Sumner does highlight the trend towards “The faceless, digital customer”, suggest¬ing that, because a large amount of transacting is now done online. RSA is looking at tackling those committing application fraud.

Ursula Jallow, Group Head of Financial Crime Intelligence, LV= agrees that there is no one particular type of fraud that the company is dedicated to, preferring instead to focus on tightening up the front end application procedures in the hope of weeding out potential trouble before it takes root.

However, Jallow notes that the claims opportunist is proving particularly hard to tackle. “Where they’ve added on some items to a genuine claim, you might have a strong suspicion but you can’t prove anything.” 

Jallow adds that the existence of claims management companies (CMCs) who practice so-called ‘claims farming’ is one of her key risk areas. “It’s a very fine line because we do have honest customers who have genuine injuries,” she says.

Interestingly, while Jallow notes that opportunistic claims fraud is one of the hardest to tackle, there may be encouraging signs that it is declining. Sumner says: “Socio-economic factors and a strong press presence from the insurance board and others means smaller, low-end opportunistic fraud is easing. People are starting to understand that there is a real consequence.”

Who is in the frame?

The Government’s insurance fraud task force identifies four loose categories of fraudster that warrant attention:

• Gangs - ‘crash for cash’

• Individuals - premeditated

• Opportunists - applications

• Opportunists - Claims inflation

A recent study - Insurance Fraudsters - commissioned by the Association of British Insurers highlighted research that showed the average household insurance fraud¬ster was almost as likely to be male as female, an average of 44 years old. It also noted that it was likely to be committed by ‘ordinary’ people because it required no skill. Unlike ‘gaming’ applications, copying car registrations or setting up bots to use multiple identities, all that was required was to add a couple of items to the list of a burglary for example.

These one-off, individual acts of fraud are still hard to trace. There is no precedent or trail of unusual activity to point them out to insurers and merit greater scrutiny. However, Jallow notes that this sort of behaviour can be seductive and the oppor¬tunist’s greedy nature has the potential to bring them eventually to the attention of the authorities.

“The propensity to claim is being tackled by the Insurance Fraud Board in its five year programme to bring as much data together as possible. Even if there is no proof of fraud, analysis of this data can pick up behaviours. When opportunists see how easy it was to do it once, they start doing it more often. But then the IFB data gets better,” Jallow says.

Organised fraudsters also play a part through crash for cash injury payouts and also exploiting online applications and the incentives mentioned above. However it would appear that it is largely a significant number of individuals committing intentioned or opportunistic fraud in volume that is the biggest target for insurers.

In other sectors, large scale, organised fraud is also a huge problem, but of equal concern is the volume of individuals trying to work the system in their favour - much the same as in insurance. Some steal identities to set up accounts to gain a handset, others to gain the use of the network until the debt collection process cuts them off. In this case, as with insurance, the best way to track these opportun¬ists is through data.

“Fraudsters are typically quite lazy and place several orders but if they’re asked to put in an email address they use the same one. We can build up lists where the same email has been used 20 times or more. Even if applications seem legitimate there are often factors that can become red flags,” states Vodafone’s Fraud Manager, Corporate Security, Andrew Spencer.

For Vodafone, that information comes from helpful location data that proves a fraudster claiming to have no knowledge of an account or living anywhere close to where the phone was registered - by kindly pointing out that he is calling from that location. Alternatively, their billing history will show that a number of texts have been sent to established contacts. Much can be done to identify fraudsters by the information that surrounds them. It’s just a question of being able to access that data.

Industry collaboration

So it’s clear that by far the industry’s greatest weapon against fraud, organised or opportunist, is data. Insurers have a wealth of historical, transactional data plus modeling, allowing insurers to follow trends including abnormalities in purchasing behaviour or economic conditions impacting the desire to commit fraud. But to be able to make a complete picture that will help them identify risks, insurers need to work together as well as with outsiders.

Efforts towards solving the data sharing issue appear to have been moving slowly, if steadily. In 2013 a working group from the Chartered Insurance Institute (CII) met to address concerns about data sharing.

“Data is the next phase of the insurance industry. We have to be careful from a data protection perspective. It’s important that we’re doing it for the right reasons and that we’re compliant,” Jallow insists. The guidelines from the Insurance Fraud Bureau, part of which are reproduced in the table below, are in response to the group’s findings that there were “unman¬ageably high volumes” of requests; “poor quality” and “lack of industry consistency”. Unsurprisingly, this will have contributed to “poor response rates to requests” (source: OutLaw).

“The insurance industry is incredibly collaborative and has been for a number of years. Helped by the IFB and others we do share information and work to share patterns and trends. Where we need to improve in this is with the government agencies and the task force. It’s a great opportunity to start improving the detec¬tion of fraud with government backing,” Sumner says.

Making sure you have the systems in place to intelligently interrogate data and act on it swiftly is vital. As Vodafone’s Spencer illustrates: “I really believe that if you have an inhouse system you can respond to a trend. When you have a big company they can be slow to make changes. You need to respond quickly and the only way forward is with an inhouse team where you don’t have to get budget approvals and sign off. You can just do it.”

LV=’s Jallow agrees with the sharing aspect but notes that in a way, this helpful¬ness is in some ways also a challenge in itself. “The difficulty is that we have all these pots of data that you can draw an ultimate understanding of the potential fraudster from. But it’s understanding how much value all those pots really add. Three sources of data might be just as useful as 10.”The more unified and possibly centralised these shared data sources can be, she suggests, the better for insurers to act swiftly and decisively.

Prevention is better than cure

If it is difficult to comprehensively identify individual, opportunistic fraudsters after the event, insurers are having more success in creating protocols and models that can prevent them from becoming customers in the first place.

“Our front end controls are fairly strong,” insists the RSA’s Sumner. “We have seen a reduction in our application fraud. We’ve seen a 100% uplift in detection rates since we implemented a fraud solution covering social, predictive and risk rules.” But, she adds, the front end is not enough and detection measures must be put in place across the whole claims lifecycle.

“The core is to minimise the chances of fraud before you get to it. Historically, the industry’s approach to combating fraud has been claims focused. This goes back to the basic principles of how you operate as a business,” LV=’s Jallow adds.

In this case, insurers need to use all the data tools at their disposal - registering customer journeys, seeking third party data such as that from the DVLA regarding convictions or previous claims histories and propensity modelling - to fight from the front end. Recognising these high risk customers before they ‘purchase’ allows insurers to price accordingly or, as is the case for many, refuse to quote.

The view of insurance as a ‘victimless crime’ and the culturally accepted sense that it’s alright to cream a little off the top or exaggerate a loss or physical injury, is changing. “Habitually it’s been seen as OK to add the extra TV. The customer thinks ‘I’ve paid my premiums for 10 years, I’m owed.’That’s the root cause of this type of fraud and the one way we’re going to tackle it is through consumer communica¬tions,” Jallow says.

This is not appealing to the customer’s better nature, it is instead focusing on their sense of self preservation. Recent research commissioned by theABI, which inter¬viewed insurance fraudsters, noted that alongside their sense of insurance fraud being victimless, the majority believed there was a very low risk of getting caught, therefore they were willing to commit fraud for relatively low values. However, as Jallow points out, being able to publicise the fact that arrests are made for intent to commit fraud is critical. “In the interviews that took place, they admitted that had they known there was a chance of getting caught and going through what they did, from a personal vulnerability perspective, they would never have done it.”

Insurers realise that their best weapon against fraud is data. Keeping data clean, integrating different data sources and seeking complementary data from other insurers and third parties all allows for the sophisticated modelling and tracking that can put these organisations on the scent of the fraudster.

Equally, data has allowed insurers to become proactive in deterring the low-level fraudster, the opportunist or the individual who imagines that a little leeway here or there is undetectable. By understanding analytics around general consumer behaviour and motivation it has allowed insurers to see that the message must centre around two key facts: With their extensive data capabilities and analysts, insurers can pinpoint who individual fraudsters are and secondly, that they will be pursued.

There are ways to reduce the £3bn-plus fraud bill and increasing numbers of organ¬isations are being established to help. The industry does not want for data; it wants for a centralised, logical way to manage and disseminate it. This will be the commu¬nity’s next task and once accomplished, will give fraudsters fewer places to hide.

Disclaimer 

Views expressed by our experts represent their sole thoughts on the topic of Insurance analytics. They do not necessarily represent the views of their current organizations and should not be seen as an endorsement of any group, product or strategy. 

The information and opinions in this document were prepared by Insurance Nexus and its partners. Insurance Nexus has no obligation to tell you when opinions or information in this document change. Insurance Nexus makes every effort to use reliable, comprehensive information, but we make no representation that it is accurate or complete. In no event shall Insurance Nexus and its partners be liable for any damages, losses, expenses, loss of data, and loss of opportunity or profit caused by the use of the material or contents of this document.

No part of this document may be distributed, resold, copied or adapted without Insurance Nexus prior written permission.

 

 

comments powered by Disqus