Wednesday, March 31, 2010

The Fraud Detection Threshold

In today's über competitive insurance marketplace, property and casualty insurers must sometimes sacrifice accuracy at the altar of customer service.   Claims handling becomes a process to be streamlined, automated and maximized, especially for large companies.  Such companies make attractive targets for petit fraud, however, because their expedited underwriting and claims processes can allow fraudulent claims of the smaller variety to go undetected.  Smaller particles through a larger sieve, so to speak.  

Perhaps that's what Sidney Edwards of Buffalo, New York, was counting on.   He reportedly:
  • took out a homeowners policy on a Kenmore, New York home that he didn't own,
  • provided Allstate with a fictitious bank account number for payment of the policy premiums,
  • submitted a claim for damage to the home's undamaged garage door,
  • received a check -- most likely not to the insured location's address -- from Allstate for $1,600,
  • called Allstate to complain that he had not received Allstate's claim payment check,
  • received a second check for $1,600 from Allstate, again probably not sent to the insured address, and
  • cashed both checks at a local store.
Investigation by the New York State Insurance Department's Insurance Frauds Bureau revealed that the home's actual owner did not know Edwards and was unaware that he had used the address to obtain a homeowners policy.  But consider the opportunity for fraud that presented itself to Edwards:  an underwriting process that did not initially verify ownership; an accounting system that perhaps did not quickly enough flag the policy and submitted claims; and a claims process that did not include an actual inspection of the loss and permitted the issuance of not one but two claim payment checks, presumably sent to other than the insured address, that could be cashed at other than a bank. 

Maybe it was Edwards' greed in requesting a second check that caused Allstate to take a closer look at the claim.  In insurance fraud work, the maxim "pigs get fat, hogs get slaughtered" is often apropos.  Something Edwards did tripped the fraud detection threshold, leading to his eventual arrest.  What I wonder, though, is whether Edwards came up with this idea all on his own, or had he learned of a similar hit and run from someone else? 

Fast, cheap, good.  Can have two but not all three.  In project work, and in insurance fraud detection.


Roy A. Mura said...

Over at LinkedIn's SIU Professionals Group, John Bull, Special Investigator at North Carolina Farm Bureau Insurance, offered these insightful comments on this claim:

Good article. Something for all of us to remember and use in training when meeting with Agents, Underwriters and Claims Management. It's a team problem and fight from the time ink is put on the application for Insurance. I'm fortunate that our Company does, by and large, a pretty good job confirming ownership etc. prior to writing a Homeowners Policy. However, we, like many other carriers, can become complacent and let our guard down. When we do, we set ourselves up to be a victim.

I've found one of the hardest things to do is convince our own people the importance of paying attention to detail and just how prevalent fraud is. Insurance, by "nature of the beast" is still a trusting business. We rely on people to give us correct and truthful information in order to properly rate a policy. It doesn't take a genius to figure out that if I lie about "A,B,C" on an application for Insurance, it will reduce my premiums. It also doesn't take a genius to catch some of this stuff. The process starts with the Agent. They, are the proverbial gate keeper. Poop in = Poop out.

Companies have to ensure we make every effort to prevent and fight fraud. We owe our policyholders that much. It's part of being a good steward to that which we are entrusted. I feel much of this goes back to a Soap Box that I've stood on many times before. It's called Proactive Investigation verses Reactive Investigation. In many cases that I investigate, I will go back and request a copy of the Underwriting File. I will review the Application for Insurance etc. Seldom, do I NOT find an issue or concern from the Application. More times than not, it's not some small issue, it's something pretty significant that if the Agent or someone else in the process should've caught. I'm not vain enough to feel like I'm just that smart and/or observant. It's simply doing the right thing the right way.

I'm not sure about all States, but in ours, it's much easier to rescind and/or reform policies within the first few days of inception. The wind can be taken from the sails before it ever turns into a hurricane. It's also the old adage of "quantity verses quality". There are many ways to skin the cat, but it's a lot easier with a sharp knife.

Roy A. Mura said...

And Lin Werkheiser, Director of Southwest Risk Advisors, Inc., had this to say:


This is an excellent article. Companies today live for metrics and customer service ratings with no concern over whether or not the claim is legitimate. As SIU's become more heavily populated with claims metrics managers, the SIU folks fall into the same "make the customer happy at any cost" mentality.

I am sure that somewhere along the line an actuary has factored this "cost of doing business" into the rate making scheme. What the public isn't told is that are paying dearly for the quick claim payment. That is because they are paying for these "little frauds" in their premiums.

In answer to the question, I do not think that the sieve can be closed without slowing down today's calls for faster and faster payment OR faster and faster underwriting with less and less underwriters.

I have also found that non standard carriers are paying more attention to this type of loss in the auto lines than most standard carriers, at least right now.