S.A.F.E.S.T - Who is the Victim

WHO IS THE VICTIM?

Insurance fraud in America in the 21st century continues unabated. By this time, 30 years after the first Special Investigation Units were formed in the insurance industry, one would think that the industry would have made some real progress. So what is happening in the world of fraud and why haven't we seen more success in fighting fraud? Let us examine the road we have traveled and perhaps by doing so, we will be able to tell why progress has been lacking and perhaps we can tell just who the real victim is.

SIUs

Thirty years ago, a small nucleus of insurers saw that they were being cheated by claimants submitting phony claims. Most of the interest centered around bodily injury, arson and auto theft. The former Insurance Crime Prevention Institute, (ICPI) had been operational since about 1971. But, in the late 1970's, arson was sweeping across the nation and it looked like it was going to cost property carriers a lot of money. Several companies established specialized units to investigate questionable fire claims. Those units came to be known as Special Investigation Units. A number of other companies doing predominantly personal lines auto business added investigation units to their claims divisions and all of the companies thought they were going to make a real difference. Most of the units had hired x-cops, mostly experienced detectives from major metropolitan police or state agencies.

Over the next few years, the industry experienced a good deal of success. One large company proudly claimed that their investigation unit returned $15 for every dollar spent in support of their SIU. The average return on investment (ROI) was about $7.00 returned for every dollar spent in support of the unit. One would think that with a ROI of 7 to 1, the industry would invest a lot of money and energy hiring SIU's and really going after fraud. Why should they? What is the cost of fraud anyway? As a matter of fact, the cost of fraud is still more than 10%. According to the Insurance information Institute, "Fraud and "padding" added $4.3 billion to $5.8 billion to U.S. auto injury settlements in 2002, or 11 to 15 percent of all dollars paid for private passenger auto injury insurance claims. (From January 2005 report from the Insurance Research Council, http://www.iii.org/media/hottopics/insurance/fraud/ 8-18-2005)

A study, reported by Conning Insurance Research and Publications in 1996 said the payback for fraud fighting efforts from their fraud survey was 6.88 to 1 and if that figure holds true industry wide, the industry might be benefiting to the tune of $5 billion dollars per year based on an outlay of $675 million in support of fraud fighting which was the estimate by Conning for fraud fighting expenses by both the PC and Life/Health insurers. Insurance fraud, the quiet catastrophe 1996, Conning Insurance Research and Publications, Conning and Co, Hartford, CT, 1996 Page 81. The Insurance Information Institute reports, "A 2000 study by Conning Research & Consulting suggests that results vary widely. Using the ratio of "claims exposure reduction" to the expense of running SIUs, the study found ratios ranging from a low of 3 to 1 to a high of 27 to 1, depending on the year and line of insurance. http://www.iii.org/media/hottopics/insurance/fraud/ 8-18-2005 It would seem to be axiomatic that when an industry is hemorrhaging red ink, they would do all they could to stem the flow. So what has happened?

AUTOMATION

Over the past several decades, the industry has turned a lot of attention to automation, preferring to use databases to identify fraud and save the money that would otherwise go to pay salaries and logistical support for personnel. It was a typical industry trend; automation. And one would think that a combination of personnel and automation should work. Does it?

For one thing, the vast majority of automatic fraud identification systems depend on search engines scanning massive databases. The systems, by definition, require large volumes of data. That means that data has to be entered somewhere in the policy master file or in the claims file. And, the data has to be accurate. The criminals know that and they know that a change in one or more characters in a data field negates or at least reduces the computer's ability to match the data. Add to that the fact that much of the data is wrong and you have the beginnings of a conundrum. The data search engines are quite limited in their ability to find fraud.

One major insurance company told me that at least 1/3 of all the VIN's in their policy master data file were in error. Even if a VIN edit program catches the error, it can be and usually will be over-ridden and the erroneous VIN will be entered. It's hard to match data when the data is wrong. In addition, many companies don't even enter data into the major databases in the first place. Furthermore, the main use of the massive databases is to catch repeat offenders such as organized fraud rings. These rings are expensive to the industry, but the majority of fraud criminals are opportunists and the databases usually don't catch them.

The bare fact is that reliance just on data bases is misplaced judgment. Data bases by themselves have not worked, will not work and cannot work. They require good data input and they depend on companies designing their own systems to download that data systematically and constantly. Remember the adage, GIGO, Garbage In = Garbage Out. If the data is corrupt or never entered, you cannot expect the data bases to work. Dependency on automation by itself is poor judgment!

The Industry needs people on the street (in the form of SIU's) and Behavioral Flag Identification Systems. So why don't insurers invest some of their premium dollars in personnel and in Behavioral Flag Identification Systems? Keep note of that word, "premiums."

LEGISLATION

Fraud was receiving a lot of attention from the public at large in the 1980's and 1990's. Many states have since passed laws requiring insurers to do something about insurance fraud. Some of the states have gotten quite detailed about what the companies have to do to fight fraud. Other states are somewhat lax and though they have passed laws requiring companies to fight fraud, they have failed to fund enforcement mechanisms. The laws should have been successful. However, a good number of companies found that all they needed to do was take a few people from their Claims Division, call them Special Investigations, and presto, they would have an SIU.

However, that is not typical of the industry. By and large, claims operations are dedicated, sincere and serious in their desire to fight fraud. They are competent, but they are severely limited in the resources that they can bring to bear in terms of SIU staff, claims processing time and claims representatives. The sheer volume of claims that come across their desk in any given time causes many claims divisions to suffer a 30% turnover per year in personnel. It is tough to keep good people when they are pressed so hard to handle so many claims. The Claims Divisions find themselves having to be very circumspect in their selection of the claims that are assigned for Special Investigations. Therefore, the failure of the fight against fraud within the industry does not appear to rest with the Claims Divisions. It doesn't appear to rest with legislation, though legislation hasn't solved the problem either.

PERSONNEL SELECTION

What about SIU personnel? Could they be the limiting factor? With a payoff of between 3 and 27 to 1, if a unit is functional at all, it would appear to be a good investment of premium to hire more SIU staff. After all, the majority of SIU staff has come from police department detective staffs. They are not retired security guards. They are not retirees looking for a place to rest. Nearly 2/3 of the companies have hired former detectives, experienced, knowledgeable, capable of developing an interview, skilled in tracking leads and experienced in evidence and in court procedures. In some other companies, the SIU staffs are populated by some of the youngest and brightest of the Claims Division employees who have shown a propensity and have demonstrated a willingness to fight fraud. One would think a company would do well to spend more premium to save some claim dollars. Keep an eye on the word "premium".

COMMITMENT

With all the statistics that show insurance fraud as costing at least 10% of premium, it would be logical for an insurer to invest a fair amount of premium to save some claims dollars. Put yourself in the position of General Motors. If GM had a problem that was costing them 10% of their income, would they not expend every option to reduce that hemorrhage before Toyota could acquire their customers? The products are of similar quality so the customers would naturally migrate to the vehicle with less cost. Why don't insurers find that it works the same way with them?

For one thing, the American Insurance market is a captive audience. The United States is the largest single insurance market in the entire world. Approximately 1/3 of all the insurance that is sold worldwide is sold in the United States. The companies in the US are massive and there is no foreign competition. Aside from a few states mandating strict regulations on how to fight fraud, there is very little in the line of state regulations and there are no federal regulations requiring insurers to do a detailed accounting on what they are doing to fight fraud. There is no competition among insurers in fighting fraud. In fact, as far as the industry is concerned, if fraud were absolutely rampant, their premium dollars would only go up and profits would track accordingly.

Put in terms of supply and demand economics, the American Public is a captive audience. You cannot drive a car without insurance. You cannot own a house without insurance. Try getting sick without insurance. In a supply/demand graph, when consumption is fixed by edict, the only thing that controls price is the supply side and in the U.S., the supply side is controlled by the industry. When a peril such as fraud is allowed to run free or even encouraged by a lack of resistance, the expense of the covered peril is bound to go up. When the expense of the peril goes up, the supply side shifts as suppliers leave the market place or raise their prices to cover costs. At that point, premium rises and as premium goes up, so does investment and so does profit.

INDUSTRY EXPENDITURES TO FIGHT FRAUD

Let us take a look at how much insurance companies invest in anti-fraud activities. The expenditure by the entire industry to support the National Insurance Crime Bureau as of 2004 was on the order of 31 million dollars per year. Additionally, there are around 600 special investigation units populated by about 4,600 investigators industry wide, predominantly in the P/C insurance industry. The average salary per investigator is somewhere around $65,000 per year. Add benefits and perks and the cost of SIU amounts to approximately $373,750,000. Add support in terms of clerical staff, computers, office administration, cars and general travel and the total cost of SIU operations for the industry should be in the neighborhood of $616,687,500 per year, plus or minus. Add the cost of NICB and you are right in the neighborhood of $647,687,500. That would have been the total expenditure of the entire P/C insurance industry to fight fraud in 2004

The property/casualty industry in 2003 enjoyed a premium income of about $574.6 billion dollars according to the Insurance Information Institute. There were 3,330 property/casualty companies in the U.S. in 2002, though many of those companies were subsidiaries of larger companies. In 2003, according to IAA, the industry's rate of return on a statutory basis was 10.2%. After tax, the net gain in 2003 was 30.4 billion dollars. Based on that, the total industry expenditure to fight fraud is approximately 1/10 of 1% of premium income. Isn't it curious that an industry so heavily populated with accountants, actuaries, attorneys, statisticians and industry leaders would not see that an investment in any activity that shows a 7 to 1 return on investment would be a great deal?? Does it make sense to do less?

Perhaps we need to see fraud in a slightly different light. Perhaps insurance fraud is not a problem to the big players. If insurance fraud actually costs at least 57 billion dollars per year (10% of premium) then if it isn't hurting the industry, who does it really hurt? Does it make sense at all not to fully fund anti-fraud activities? The answer of course is yes. It makes perfect sense not to fund anti-fraud activities. And in some companies, though not all, that is exactly what is happening.

There are many companies that prefer to fight fraud even if it is only on an ethical basis. They hate to be cheated and they don't like paying out their customer's premium dollars to criminals. However, if the entire industry does not take a stand, then to the companies that do fight fraud, it is like being stung by a million mosquitoes. The criminals keep coming back and coming back and sooner or later the hemorrhage will hit the books. Those who fight fraud will eventually drive away their own customers by taking a stand against bogus and even questionable claims. The fraud fighters' income levels drop as premiums drop. The companies who take a strong stand against fraud will find their investment cash diminishing and those companies who do not fight fraud and wish only to bring in high premium volume will have that much more premium volume to invest. Is there a solution?

At one time, a program was developed that would potentially eliminate fraudulent total auto theft claims. It was a behavior based program and it took 2 to 3 minutes per claim to analyze the claim. It had the potential of returning nearly 11 to 1 on investment. The company refused to install it. In another project to reduce auto theft, a system was developed that virtually eliminated the theft of a certain number of high risk vehicles. The system was not accepted for use. I had the opportunity to interview a vice-president from a large company and I asked him why such systems would not be employed. He said, "You don't understand. Auto theft is what we insure against." In other words, if we eliminate the covered peril, we eliminate the cause to have insurance.

Expand on that thought to include other lines of business and it makes perfect sense NOT to seriously attack fraud. If you eliminate it, a part of each covered peril would be eliminated and at least 10% and in some cases as much as 20% of premium would not be available for investment. 20% of 574 billion dollars is a lot of money. It is $114.8 billion every year that the industry would not have to invest.

And that is the key. Insurers make no money on the combined ratio. There is almost never an underwriting profit. Insurers regularly pay out well over 100% of what they bring in in premium dollars. It is the investment of the premium dollars that brings profit. Julian Levy, writing in "The US Non-Life Insurance Market" wrote that between 1984 and 1994, the average combined ratio was 108.38%. That means that the companies lost $108.38 for each $100 of premium that they brought in. The only way to have made a profit during that period of time was on the investment of premium for the generation of return on investment.

So, fraud is not really a cost to the industry. It is a money generator. So then, who pays for fraud? You do. You pay for it in increased rates that result from a failure to competently address insurance fraud. This is not a conspiracy among the companies. Some companies really try. But it does not appear that the majority of the industry is making a serious effort.

FUTURE OF FRAUD

The future does not bode well if the industry does not make a commitment to fight insurance fraud. If fraud is not stopped by all the players, it will still exist and the premiums to invest will continue to rise along with the fraud. The end result will be that many American's will need to "run bare". They won't be able to afford the premiums. The property casualty insurance industry gained 41.24% in net premiums written between 2000 and 2004. The alternative is regulation.

Can the insurance industry put up a serious fight against insurance fraud? Of course it can; with substantial commitment to anti-fraud activities including a combination of personnel and automation and a substantially increased commitment to such organizations as NICB.

It will also take a commitment to public education. By and large, the public, including the various law enforcement agencies that are tasked with fighting insurance fraud, all believe that insurance fraud is an industry problem and it only steals the industry's money. According to the Insurance Research Counsel in July, 2003, one in three American's say it is acceptable to inflate insurance claims. The industry and the government need to do more to educate the public that it is not the insurance company's money that is being stolen. It is their own.

Additionally, the industry needs to invest more time, energy and money in the correct application of computer processes. It can not depend on database search engines alone. They need to revert to the old tried and true behavioral indicator analysis that has been used by NICB for 30 years. Data driven search engines alone have not and never will be the final answer. To catch the majority of fraud, analysis of behaviors is the key. And those behavioral indicators can be reduced to computer driven software that assists the claims representative to target their limited resources on the most productive claims.

The bottom line is, if the industry does not want regulation, which is invariably non-profit to the industry, it will need to invest some of it's premium in the fight against fraud through hiring dedicated anti-fraud personnel, support of anti-fraud organizations, public outreach and investment in both data driven and behavioral analysis software.