Insurance Times together with our partner, law firm Keoghs, and associates, motor inspection specialist motorcheck, sponsor a public debate on insurance fraud at a time when the government is preparing anti-fraud legislation. Our panel of fraud experts answers questions from the floor
The panel
John Beadle, UK counter fraud manager, Royal & SunAlliance
Nick Starling, director of general insurance, ABI
John Freeman, director of special investigation, Capita Insurance Services
Damien Ward, partner, Keoghs
Nick Leeson, formerly of Barings Bank
Q Does the panel believe that the industry is now able and, most importantly, willing to deal with the problem of organised insurance crime?
John Beadle: Yes, we are able, and we are willing. We now have over 90% of the industry signed up to the new Insurance Fraud Bureau (IFB), which is going to tackle, coordinate and lead the industry strike against organised cross-industry fraud. It will be up and running by the end of the first quarter next year. We're currently going through a workshop to hone the processes and protocols that we're going to use and it will be a reality next year. The bureau will greatly enhance the industry's ability to tackle this 5% or 10% of the fraud spectrum which is involving organised criminality cross-border, cross-police area and certainly cross-insurer.
Nick Starling: The organised part of insurance fraud is a very defined and relatively small area. I distinguish it from other sorts of premeditated fraud, like burning down your business when it's having difficulties.
There's a huge opportunity for tackling that sort of fraud. The industry is capable of collaborating properly on that. But for it to be even more effective it needs to attract the attention of other stakeholders.
I'm sure that when you look at a house which has 200 insurance claims, there are other things going on there. The people in that house could be of great interest to the Department for Work and Pensions, for example. A recent police crackdown on uninsured drivers found an awful lot of people doing other illegal things, as well as uninsured driving.
Damien Ward: The industry is capable of collaborating effectively and organising the fight against organised crime. And it's doing it. Cooperation today is much much better than it's ever been and is continuing to improve.
The IFB is a real step forward. The industry is moving in the right direction. It's also important to understand that organised insurance fraud is only part of the organised crime picture. Almost inevitably, insurance fraud and fraud rings, whether they're involving large-scale staged accidents or any other kind of fraud, are often a front for other kinds of criminal activity. Securing the cooperation of the police and working with them and within the industry itself, in a joined up approach, is the key. And the IFB can only improve that process.
Nick Leeson: Again I don't think you have any choice. Any fraud can be combated, whether in banking or insurance. But you will have to collaborate to beat it. I've been in many different forums where my case was described as a devious and complex fraud. That's really not the case. I know that the products that I used to trade, or some of them, may have been described as complex years ago, but the frauds were anything but. It was just a case of very, very simple checks and controls not being done, and I'm sure that does translate in many respects into the insurance industry. There was certainly a degree of complacency during my time at the bank. That doesn't take any of the responsibility away from me, and I don't want to give that impression. But the checks that weren't done could have exposed my actions far earlier. But you have to collaborate to stamp out the possibility of fraud.
Beadle: There are complex issues around sharing the data and maximising the use of the industry databases that we currently have. We're probably considerably ahead of some of the other financial services sectors in this respect, in sharing and using state-of-the-art analytics on our data and using it in this way.
Q Given the extremely important role that insurance plays in the social and economic well-being of both the country and the lives of individuals, why is it that so many people believe it is acceptable to submit fraudulent claims? Can we change this mind set?
John Freeman: The root of this problem is that there is little or no recognition on the part of the public that insurance is such an integral part of the fabric of our society. It follows that the higher regard that people have for insurance, then the less likelihood of fraud being acceptable. The message we always bang out really is that insurance is a victimless crime. We probably have to accept that the public believes that insurers' true motive is perhaps to maximise profits and shareholders dividends, to the detriment of fair claims settlement.
On changing the mind set consider benefit fraud and TV licence evasion. There the focus has indeed shifted - it's no longer seen by the vast majority of the public that cheating is acceptable. They seem to have grasped the nettle that fraud or evasion has a direct impact on their own budget. We have to, therefore, I think, heighten the campaign and get the message across that fraud increases premiums for us all.
Starling: I don't quite know why there is this public acceptability. If you have any area where people think they can get away with things or that other people are getting away with it, then it becomes contagious.
I'm a pedestrian in the City of London, and I have to deal with cyclists, and cyclists in the City of London have decided that the laws don't apply to them at all. They go through red lights, they cycle on the pavement, because they know they're not going to be stopped, and it's just become acceptable. So any cyclist who stops at a red light feels a bit idiotic and carries on through. And that's an example of when there's a perception that gets a hold, it's very difficult to counter. There's always the element of insurance, which is, that I've paid my premiums for all these years and I should get something back.
So there's a big issue about changing attitudes and values. Can you do it? Well I think you can. And if you think back about 20 years, drink driving was acceptable really. There was the idea of one for the road, and if you were caught you were unlucky. And now it's not acceptable.
Ward: There's a bit of a 'have a go' mentality out there. And of course if the public has a go and gets away with it, then they have a go again and it will be a bigger sum, and so it moves on. And of course if a particular fraudster has a go and doesn't get away with it, what happens? That has sunk into the public consciousness and exists there. I do agree that it could be changed, but also believe that it sits there very firmly in the public consciousness. It's just years of perception, and part of the perception is the lack of value that the public perhaps gets from insurers.
And there is the perception that it's a victimless crime, and perhaps as well that the industry is more remote than it used to be, from members of the public.
And finally of course, the lack of prosecutions. There are very few prosecutions in terms of small-scale insurance fraud, either whether they're successful or whether they've attempted fraud.
The irony is, of course, if the fraud is an attempt and it's defeated by the particular claims office of the insurance company, then it remains an attempt and the likelihood of a prosecution is extremely remote. And so the fraudster will come back and have another go.
Beadle: There was some ABI-sponsored research that showed that one of the perceptions of why people committed insurance fraud was that they associated insurance with part of what they called 'rip-off Britain', particularly in respect of motor insurance. With insurance premiums going up, the cost of petrol was also going up, road tax was going up, there was congestion charging and there were parking charges and clamping.
And they saw insurance as part of this 'rip-off Britain'. So, in making a fraudulent insurance claim they had the opportunity to get money back .
Q Does the panel believe that there is sufficient investment in identifying internal insurance fraud?
Leeson: From my own experiences I found that much of the focus was on external fraud and not on internal fraud. And the impact of the internal fraud can be far, far greater. Look at some of the financial scandals over the past 10 years - the ramifications were all very dramatic.
Sarbanes-Oxley and other corporate governance legislation has imposed a degree of personal responsibility that I'm sure nobody's taking lightly. I often use the example that if a Barings fraud were to happen again to the extent that it happened during my time there, the impact would be more widely felt within the bank's directors and the people who made the decisions.
I had a risk manager, I had a compliance officer, I had an accountant, all of whom had responsibilities within the organisation. But all of them failed to function, and it's very difficult to work out why.
I know the most difficult question I often get asked is why the management of the bank didn't ask more detailed questions at the time. The bank was only worth £250m. In December 1994 they sent me £500m in Singapore. The maths just didn't add up.
If you look at the level of margin that I had with the exchanges versus the volume of trading that I was supposed to be doing, it defies common sense. And these questions aren't being asked.
My summary would be that, certainly from a banking perspective, the money, the focus is more on things that can impact the bank externally, rather than internally. And they don't like spending the money on controls.
John Freeman: We have to accept, first, that no matter how tight your systems are, greed and dishonesty can never be got rid of - it's an unfortunate part of human nature. On the bright side, most of us in this room spend most of our time trying to stop the unstoppable and long may it continue.
But to really tackle this you have to have a complete strategy for the problem. And probably the biggest lesson we should learn from things like Enron and Barings is that an anti-fraud strategy should not be compartmentalised or confined to specific or different parts of a business or on an ad hoc basis.
A really effective fraud strategy should straddle the whole of the business and be embedded in its culture and philosophy, all of the parts should be linked to work as one.
Most of the efforts in combating fraud tend to focus on the claimant. That's not wrong, because we know that there's a big area of leakage. But to have an effective anti-fraud strategy you have to get your own house in order. That seems to be a blind spot for some organisations. The starting point is to change an organisation's culture. One way to do that is to develop and implement a formal fraud policy. It has to be highly visible, and across the business with accountability running from top to bottom, rather than what is often the case, the other way round.
We have to ensure that adequate management controls are bolted into the front end of risky processes, and I'm not just talking about underwriting here or claims notification. I'm talking about reference qualifications - checking on people when they come through the industry.
Beadle: We haven't yet been brave enough to work together as an industry, either in the insurance or in the wider financial services sector, to share information about internal fraud.
This is probably because of the embarrassment factor, and nobody likes to necessarily wash their dirty linen in public, as it were. FSA regulation has certainly made us all examine our navels and sharpen our act up, in respect of internal fraud. Obviously Sarbanes-Oxley, for some companies, not for all, has some very stringent requirements.
And some work is now beginning to take place in terms of the industry owning up to the fact that we do really need to work together in respect of this. And we should, I think, be as a financial services sector, sharing data around individuals who are known fraudsters.
Certainly from my perspective I inherited somebody who had lost his job at an insurer through fraud, and then committed a fraud against us. When he came out of jail he worked for a broker and was sacked for doing the same thing there. Now that's got to be madness. Certainly we are much better now than we used to be, regarding things like screening of staff.
I think there is a piece of work that needs to be done and is in its early stages, around the insurance industry and the financial services sector looking at working together a bit more around this issue.
Q A recent ABI survey found that one in 10 motorists admitted lying on their insurance applications. Does the panel believe it is now time to switch the focus from detection in claims fraud, to prevention at the underwriting stage, so essentially physically validating and inspecting the vehicles before insuring them?
Beadle: It would be a wonderful thing if we could pre-inspect vehicles to ensure that they exist, that there was no pre-existing damage and a whole host of other factors. There would obviously be a cost to doing that, which would need to be factored into the premium.
The difficulty with the current marketplace is, unless all insurers did that, you might be actually selected against. These days people expect to have instantaneous insurance cover when they ring up, or go on-line. Which is a bit crazy as we're exposing ourselves to almost unlimited liability instantaneously.
If we started putting our premiums up to cover the cost of doing that on our own, people would just say: "I'll go to a cheaper insurer". It would be a great fraud prevention tool, but it needs to be taken up by more than one brave insurer.
Starling: The immediate answer to this question is that we've moved away from inspection, and we're going to continue to move away from it. I once took out buildings insurance with an insurer and slightly to my surprise one of their men said he was going to come round and see the house. I live in an area of London which is prone to subsidence. I said to him: "It's easy to find my house - it's the only one standing among all the other ones that have fallen down." There was this pause and he said: "Never, ever make a joke like that to an insurance salesman."
What insurers have done, of course, is improved their risk assessment methods. They now know the mapping of the area much better, they know characteristics and they have better knowledge. So it's moved away from physically inspecting. Now we have to be better at assessing the risk that relates to the customer and knowing the customer.
There may be some scope for more data sharing. If you find someone has made a fraudulent claim and you find that they're also cheating on welfare benefits or have made a false credit application, for example, then you might want to look at their insurance application more closely.
Leeson: Know your customer. We talk about that in the banking world, but we don't talk about that in insurance. Perhaps we should. There are two elements to this. One is that the public expects to get insurance immediately in real time, and the market has responded incredibly well to that. And the other point is that it would take the entire industry to move towards a different approach. I find it incredible that the public accept that it will take 24, 48 hours at least and a wealth of information to set up a bank account, and that bank account will have a nil balance.
I entirely accept that we are unlikely to move towards physical inspection of vehicles and property in the way that perhaps the industry used to do. But I certainly do believe that a more joined up approach from an underwriting perspective, which would have to be an industry driven thing, would be a good thing for anti-fraud.
Q What value do you see in adopting or taking up a nil or zero tolerance approach to managing fraud, whereby you pursue fraudsters through the criminal courts?
Ward: A zero tolerance approach is unlikely to have the effect that one might perceive it to have, for a number of reasons.
The example that was given about cyclists in central London is an excellent one. The prevalence of speed cameras is such that people have a fear of getting a speeding ticket, but they don't particularly have a fear of speeding in the same way. You know where the speed cameras are on the motorway roadworks, and you're doing 80mph between them and 40mph where it says 40mph. And I don't think that the mere fact that there is a penalty, albeit a financial one in that case, or some points on your licence, necessarily changes the attitude of the driver to the criminal offence, which is what it is.
And the other point is that if we talk about the figure the ABI research has thrown up, which is that 47% of the public out there would be quite happy to commit an insurance fraud, and if you extrapolate that into claims and you even take a very small slice of that, 10% of people in the general insurance market are actually submitting an element of the claim which is false. If you prosecute in every case where the claim seems to be false, you would be locking up enormous numbers of people and putting enormous numbers of people through the courts. I really don't believe that is the answer, although it's part of the answer.
The greatest fear that the public has is not being able to get insurance in the future or being blacklisted, or the cost of the insurance becoming very, very substantial indeed.
I accept entirely that the perception of the victimless crime has to change, I don't believe, however, that you can prosecute everybody.
People can still be prosecuted for insurance fraud. The codified law will make it easier to prosecute people and the law will be easier to understand.
And the irony is, of course, that the better the insurance industry becomes at detecting fraud, means that more people will be prosecuted for attempted fraud, rather than perpetrated frauds. And what we would be doing is prosecuting people, putting people through two-week trials in the criminal courts for an attempted fraud.
I'm not saying it shouldn't happen - far from it, but the political will is unlikely to be there. And zero tolerance in the criminal sector probably isn't the way to go about it.
Beadle: The issue here is that insurance fraud - or fraud in general - is not in the national policing plan. Police forces are not resourced for it, it's not in their objectives, and therefore it's little wonder that we have difficulty in getting the police to take an interest in it. As for whether we should be zero tolerant, in respect of the cases that are typically going to be taken on by the insurance fraud bureau, which are organised crime and involve real criminality, we should be absolutely zero tolerant. And we will be, both in terms of pursuing criminal prosecutions, seeking innovative civil solutions and disrupting their networks. We will be absolutely zero tolerant on that.
But I'm sure we can all remember cases where we might have a degree of sympathy towards the person who has committed it, and a zero tolerant policy may not necessarily create a particularly good image for the insurance industry.
Starling: It's a difficult one. Insurance fraud will stop when people realise they're going to be caught. How do you reach that sort of state? My understanding of zero tolerance is that first little line of graffiti. You slam on it. So in practical terms you cannot have that sort of zero tolerance. A few high profile cases will help, I have to say, but that's not quite the same as zero tolerance.
Q What are the biggest fraud risks in the insurance industry over the next three years?
Beadle: Injury fraud is currently the biggest single area of fraud that we're exposed to. In seeing where we might go in the next three years, we probably couldn't go far wrong if we look on the opposite side of the Atlantic. They have big issues, and have done over recent years, with medi-fraud, which is a big area for the US. We certainly copied the US propensity for staged motor accident fraud. I'd say we'd probably need to look at the potential for medi-fraud.
Starling: The cynical answer is that if we're successful over the next three years, the people who are committing organised fraud will find someone else to commit fraud on. And the people who are organised will always find new things to do.
Ward: The only answer I could give would rather depend on the economic stability of the UK economy. When there's been a downturn in the economy, certain types of fraud become more prevalent, and property fraud, arson claims, that kind of claim might see a resurgence. But I don't see any major change, apart from what John Beadle has identified, which is medical and dental fraud. They will, by definition of the size of the market, the speed at which it is growing and the cost of those claims, probably be a growth area. But I think motor fraud will continue in the same way that it does now.
Chair: That's the end of the formal question time, so now I'm now going to throw it open to the floor.
Jack Brownhill: I was interested by a few things that came out following the comments by Bernd of motorcheck. Most of the focus on the floor tonight, quite rightly, has been on claimant fraud. But getting a policy for the wrong premium for the insured is fraud in itself. That's one thing that comes out of the way that we manage certain lines of business at the point of sale. The point was also made that you would have to pay for the vehicle checking service which might competitively disadvantage you. That's obviously true, but again it may benefit you, because if you are eradicating fraud, you could set the premiums back down again and they may even be lower than they were before.
So it could be an advantage. We also have to be very careful about issues such as pricing people out and blacklists etc, particularly in terms of motor insurance, because obviously it's compulsory liability and we have to make sure everybody on the road is actually properly insured and we're doing lots to make that happen.
Thanks to our colleagues across the water, we now have a 14-day cooling off period, which is a bit one way, because the customer can do it, but the insurers don't think they can do it either. But there are some things that perhaps insurers can be doing in those 14 days, or after the sale, such as checking.
One big gap on motoring insurance is that we don't get near to the customer at all - we don't even see handwritten proposal forms any more. So we can't get any feel from the documentation as to the type of customer he may be. One big area is that we don't check licence details either. That's a massive gap and I'm surprised we're not pushing for a liaison with the DVLA to allow us to do that as part of the contract closure process. Even if it's done post-closure,then at least then you can tell the customer that the story that he gave you isn't correct.
Beadle: There is probably a silver bullet, but we're some way away from it. And that is real-time access to data sharing on a wide scale - DVLA, all insurance data and so forth - at the point of sale. That would answer a lot of the things that we've spoken about this evening. It's a huge issue and frankly we are some way away from it. But if the IFB is a success, that might be an indication of a way forward. One consequence of sharing data is that you are going to identify people who the insurance company is going to find it very difficult to insure. If you have a customer who is a known fraudster and you discover this through data sharing, there is a question of what are we going to do. Will this create an underclass of people who can't get insurance? It's a real potential political hot potato, but that's a consequence which I think needs to be grasped, if you are going to have data sharing.
And finally, there is a lot more the industry can do in terms of prevention as well as cure. The obvious place to start, if you're going to tackle insurance fraud, is at the claims end, because you see immediate indemnity cost saving from your pound of spend.
But there is an optimum level at which you can do that before it doesn't become cost-effective. And you end up spending more than you're saving, which of course is passed on to the customer in any event. So you do need to look at your risk appetite and the moral risk that you're exposed to at the underwriting cycle.
Freeman: Of course there is no silver bullet. When Nick arranges his insurance he'll probably pass all those validation checks. But the problem I have is what if he decides to exaggerate his claim?
Ward: Can I just come back to this idea of blacklisting. This is knowing those people who are going to attempt fraud against you, and the idea of that is of course knowing your risk even more - it's not pushing those people out of compulsory insurance.
Beadle: The difficulty is that in the US and other areas it's routine if you want a car insured. You drive it to an inspection centre and you can't get it insured unless you do so. We do not have that culture of inspection. As Nick says people want cover now, they want it online, they want it on the phone, and that's where we've moved to. It's a brave company that's going to be the first, I guess.
Starling: The system is squeezing people all the time. If you take our car insurance, you go on the web and put in your details and the registration number and it will tell you what car you have, for example. The law is about to be changed from the offence of driving an uninsured vehicle, to being the keeper of an uninsured vehicle.
At every single point there is more and more squeeze on people's ability to defraud at the underwriting stage. There are still the issues around staged accidents and the abuse of the 'driving other cars' extension. But I think there is a series of incremental steps, mostly around data, and the better ways of manipulating that data that are helping.
Chair: We have another question from the floor. Nick Leeson, did you lie on your CV when you went into banking? This links to doing background checks on employees.
Leeson: I didn't lie about anything on my CV and I was headhunted anyway by Barings.
The speculation came from when I was working in Indonesia. Someone was living in the house that I have in the UK - something wasn't paid and there was a county court judgment for about £450. That was imposed when I was in Indonesia and already in the employ of Barings. So as far as I'm aware there was nothing that was incorrect on that CV. I've spoken at a couple of events where people have brought that up as an example and I've corrected them.
But there are some people who have fairly elaborate identity fraud and it shocks me how bad that is. I was even shocked when John mentioned about the employee who has gone from insurance company to insurance company, via prison, on each occasion. That strikes me as an example where clearly you should be sharing information. That guy is potentially very risky.
Beadle: I wonder who had the professional indemnity cover on the headhunters who selected you.
Leeson: On another point, you were talking about creating an underclass of people who wouldn't be able to get insurance. And my question to the insurance industry as a whole would be, don't you already do that with people who have a life-threatening illness? I can't get life cover for instance, so you're already creating that. What would your problem be with creating it for something else?
Beadle: I think the point on the motor was that it's compulsory. I'm not saying that it would happen, I'm just posing the question, that if we're going to share data to that extent, it might be a consequence of it.
Chair: We've talked about staff fraud and external fraud. Does the panel believe that we are exposed to, with our staff, collusion and coercion, as has been seen in the credit industry? And if so, are there any ways in which that can be tackled?
Beadle: We would be naive to think that we wouldn't be affected, I know it's a big issue currently in the banking industry, where Trojan Horse employees are put in there by organised crime so they can work out a way round the bank's systems and controls.
The risk and reward in insurance terms is likely to be less than you would get in banking. If you looked at it on a pure risk basis, it's much more likely to happen where the rewards are higher - potentially in a banking environment.
Freeman: I'm sure there is some exposure there, but I agree with John - I don't think we know how big that exposure is, so the check is to make sure you have the balances and the processes in place to detect them.
Chair: What role, if any, does the panel see suppliers having in fraud prevention and detection?
Freeman: I'm not convinced that validation and fraud detection can't sit together. What I am concerned about, as many are, is that fraud clearly does exist within the supply chain itself, and I'm not sure we've tackled it enough. And I question whether we're sharing good practice with the supplier partners. Suppliers are also in the group where they need to embrace a cultural fraud awareness, just as insurers have done, and so should be encouraged and perhaps even incentivised more to look at training, process, audit and validated management information on their results.
Essentially we need to see more work done to ensure that fraud detection is seen as part of the suppliers' balance scorecard. There are some good things in the industry that we can share. If we take the Chartered Institute of Loss Adjusters as a good example, it operates under a charter and therefore the consumer has high level of protection. One would debate for hours I suppose, about impartiality.
But I'm not sure whether all the suppliers are actually working under an agreed code of practice or code of conduct. Fraud detection and validation are really two elements, but I worry about the customer experience if you separate them. Let's not lose sight of the fact, because you have a conference or a session like this and everybody walks out the door thinking everyone's a fraudster. The vast majority of people we deal with are honest policyholders. So we have to make sure that we apply the 80/20 process rule - that 80% of the customers at least have a very pain-free experience as they go through our process.
So I reckon they can sit together. Let's just make sure that we have the appropriate controls, checks and balances again, and develop a process whereby if you find some strong fraud indicators, you can at least get out of the validation and use a different skill set to investigate it.
Beadle: I agree with John Freeman absolutely - the supply industry in insurance terms is huge, so the potential exposure is massive, and you do need systems and controls in respect of your supply industry as a whole. In terms of them assisting in the identification of fraud, this is a partnership between us and our supply industry, and of course it has the technical skills that we don't have.
So in a lot of the supply industry we wouldn't identify the fraud, but use its technical know-how. And there will be some, hopefully, that would be identified at that stage, so I can imagine some will be caught there. It's only fair that should be the case.
It's a partnership, but the vast majority of customers will get a very pleasant, efficient and effective customer experience.
Chair: The next question is about public image and the perceptions of consumers and policyholders. Eliot Spitzer fined insurers $1bn for misrepresenting the values of policies, and there's widespread use by insurers of unauthorised individuals to limit values of claims. Correctly documented claims can also be intimated to be overvalued. Does the industry not need to look at itself about its moral fraud, if you like, towards legitimate claims of policyholders? After all, Mr Spitzer didn't impose billion dollar fines for nothing at all.
Insurers have a long long way to go before they understand what policyholders understand about insurers. At the moment we have engineers who are under-valuing vehicles, because the instructions from the insurers tell them to limit the value of the vehicle to a percentage of the market value.
Engineers are not FSA authorised and carry out regulated activities for insurers who ship a lot of their work out. You have data, which we know is one of the key ingredients for successful fraud, going outside the EU from insurers.
Beadle: There is an image problem with the insurance industry. I can only talk from my company's perspective, in that we certainly don't instruct motor engineers to value cars at a percentage lower than market value. We do of course, along with a lot of insurers, have a performance-related pay and reward scheme.
