Armed and ready

Every day, the claims sector’s anti-fraud squad prepares to enter battle. The pressure to defeat fraudsters is mounting and the stakes have never been higher, as recent surveys show that the appetite for insurance scams is growing.

According to the National Fraud Authority (NFA), fraudsters are now costing the UK economy a staggering £38bn a year, with insurance fraud contributing £2.1bn of that. According to the ABI, fraud has added, on average, £44 to every household’s annual insurance bill.

That’s not all. Research by Swinton reveals a worrying change in policyholders’ attitudes. In a survey of 1,200 online customers, Swinton found that 22% had exaggerated a claim, while 37% of home insurance customers said they would be prepared to embellish their story for a successful claim.

Meanwhile, in Legal & General’s ‘FraudStoppers Report 2011’, more than half (57%) of brokers and advisers surveyed said a customer could get away with an exaggerated claim.

Insurers clearly need to arm themselves with better and more sophisticated weapons in the fight against fraud. But what weapons do they have at their disposal and, perhaps more importantly, how effective are they?

Share and share alike

Industry experts believe data sharing remains the best tactic for insurers to combat fraud. But while there is plenty of information available, mining or using that information remains problematic.

The main hurdle is insurers’ failure to achieve a more cohesive approach to data sharing.

Loss adjuster Crawford psychology director Nikki Grieve-Top says: “It frustrates me no end if we investigate a fraudster for a great length of time only to find out that another insurer knew about the fraudster already. If insurers shared fraudulent risk data more freely, then you could see the whole picture rather than just a part. Information really is the key.”

Loss adjuster GAB Robins head of counter fraud and investigations service Neal Daniel echoes this frustration. “Different companies have excellent databases, but they are not joined up,” he says, adding that today’s savvy fraudsters are fully aware of these gaps in insurers’ fraud detection and exploit them to the full.

Nonetheless, Daniel admits that there are barriers in the way of effective data sharing. “Clearly, you have insurers in competition with each other for products or customers, and data sharing can cause a problem there.” He adds that insurers are often concerned that they might give away more information than necessary when submitting data and that rivals can see how many claims they are receiving.

The sector is addressing this issue. For example, the ABI is launching an insurance fraud register to hold a record of fraudsters. AXA group fraud risk manager, Richard Davies, who is steering the project, says this will be a major step forward. “This will solve a big gap in our control structures. We know that opportunistic fraudsters will move from company to company, but until now we have been unable to share that information in a way that allows us to make better decisions about who we are going to cover.”

In addition to the register, the Insurance Fraud Bureau (IFB) is this year launching a pan-industry web portal, through which insurers can access a wider range of industry data.

It’s a start, but Grieve-Top explains that the perfect solution is a portal that grants access to a huge spectrum of cross-sector data, including data from public bodies. Davies says there is significant red tape attached to creating this central gateway. “Currently, insurers are in a better place to share data compared to public sector bodies, which may struggle to get agreement from the government,” he says.

Deception detection

Of course, databases are only one solution. With today’s technology, the insurance sector’s fraudbusters have several high-tech weapons at their disposal.

Loss adjusters such as Merlin and GAB Robins have started digitally recording interviews so the insurer can listen to them when they choose, rather then trawling through written reports. Loss adjusters have also started to use digipens to write statements from the claimant, which can be sent directly to the company’s database, helping the anti-fraud team quickly identify any ‘red flags’ or signs of suspect activity.

In addition, metadata software is increasingly being used to analyse information from digital photographs. The software can help determine when the photograph was taken, and this can trap fraudsters if the picture was taken after the alleged loss.

Furthermore, the internet has created a wealth of data that can help insurers. The rise of social networking in particular has created new opportunities for insurers to catch fraudsters.

But new technology can only go so far. Professor Laurence Allison, academic director at the University of Liverpool’s National Centre for the Study of Critical Incident Decision Making, warns that there is no magic bullet to tell if person is lying. He adds that the jury is still out on technologies such as voice stress analysis and brain mapping, which involves scanning the brain and voice tremors to detect deceit.

“Lying is a very subtle thing,” Allison explains. “There are all kinds of ways that people can lie so these tools are not sufficient to pick up deception,” he says.

He believes that better investigative methods, such as skilled interviewing, is a better strategy to root out fraud. Merlin head of special investigations Tim Richardson, meanwhile, believes the sector is moving towards specialist investigative strategies. He points out that of his team of seven, four are former police officers.

However, Allison believes that levels of fraud awareness and knowledge of fraud theory remains sporadic. “It seems very unsystematic to me,” he says. “Most insurance companies don’t have well-established mechanisms for training staff in deception detection.”

Train to gain

Grieve-Top points out that while many companies offer fraud-related courses and training, there is little in terms of set standards throughout the sector. The NFA announced in 2009 that it was considering standardising fraud-related qualifications and courses, but little progress has been made since then.

Allison says that more needs to be done when it comes to investing in research. “Given the economic impact of fraud, it is surprising the dearth of academic literature on it. I haven’t seen much academic literature on fraud over the last 10 years that has looked at the data and established empirically what type of people we are looking at and why they do it.”

Despite these gaps in insurers’ armoury, progress is being made. Last year, more than 2,000 dishonest insurance claims were detected every week. The value of these claims, at £840m, was 14% up on the previous year. The number of calls to the IFB’s cheat line soared by 70% over the past year, as policyholders become increasingly aware of how fraudsters are hitting their pockets by pushing up premiums.

The fight against fraud has become increasingly bloody. But the sector is becoming streetwise, and realises that only by fighting tough can it win the war. IT

Empower your claims team with the 21st century fraud fighter’s armoury

Fraud databases

The main industry database is Q or the Claims Underwriting Exchange. The ABI is working on an insurance fraud register that will provide a record of proven fraudsters across commercial, personal and life insurance lines, while the IFB is planning to launch a new pan-industry web portal later this year, which will provide access to a greater range of databases.

Insurers can also access external tools, the most popular of which include CIFAS, the association dedicated to helping prevent identity theft, and Equifax and Experian, which both provide data on policyholders’ credit history.

Meanwhile, the government has announced that insurers should have greater access to information from the Driver and Vehicle Licensing Agency within six months.

Live video streaming

The days of trawling through written reports are gone. Digital recorders can now record an interview and stream it directly to the loss adjusters’ headquarters or the insurer in another location.

Digipens

These allow the loss adjuster to attach the policyholder’s statement to the company’s claims database as it is being written.

Metadata analysis

This technology can scan for information within digital photographs, revealing when the picture of the object claimed for was taken.

Brain imaging

This is a powerful technique for looking at the brain when it is processing emotional and cognitive material. Some have suggested that this technology can potentially help insurers gauge whether people are lying.

Voice stress analysis

Supporters of this technology believe it can help claims handlers pinpoint variations in voice patterns or micro-tremors that indicate deception.

Text mining

This technology can find key words within documents and saves the user from having to manually trawl through documents looking for common words or names, or other fraud risk indicators.

Investigative techniques

Experts say that successful fraud detection is down to hard graft and good interviewing skills. While a number of companies now offer fraud-related courses and training, there is very little in terms of set standards throughout the sector. In general, fraud courses should be recognised by the Association of Certified Fraud Examiners.