...Some Tips on Detection, Prevention and Dealing with Insurance Fraud

The insurance sector is one silent arena where fraud has time and again taken place with perpetrators getting away with it most of the time. Fraud has negatively affected the insurance sector and has often gone un-reported, under-reported or un-noticed at all. Time has come for this situation to be reversed and protect premiums contributed by policy holders and the insurer’s bottom-line. Two broad classes of insurance fraud may exist:

·        Hard fraud. This involves people obtaining funds illegally from insurance companies by faking claims for risks that never took place. It also occurs where employees issue fake receipts for premiums paid and pocketing the funds.

·        Soft fraud. This arises out of deceit to the insurer with the aim of concealing certain information for financial gain. This out rightly contravenes the principle of utmost good faith but the insurer may not notice this at the first instance.

In most insurance fraud, the key areas affected include:

·        Claims processing and payment: Claims are the worst hit by insurance fraud and constitute over 50% of the fraud in the insurance sector. Claims fraud can occur in multi-varied ways. One instance would be where the risk being claimed never existed at all (e.g. a fake death certificate or a police report!). The other instance is where the risk claimed was stage managed. The other form is where claims are submitted for policy holders who are non-existent. The worst form would be where there was collusion between the policy holder and the insider employees (or agents) to fake claims (insider fraud). In some cases, the claims are inflated or accompanied by bogus documentation.

·        Underwriting and policy issuance: in this case the insurer’s employee, policyholder and/or other parties collude to draft a policy that includes certain clauses that will be used to the benefit of all parties involved. For instance, a clause allowing the sharing of profits at a certain percentage. How the proceeds of the “profit” are shared, it is the parties who know best.

·        Remittance of premiums: This can occur out rightly or indirectly. It may arise where remitted premiums never hit the insurer’s accounts or where the premiums are paid as per the company’s systems and the bank but the amount paid is not the correct amount payable under a certain policy. In other cases, employees may issue fake receipts for premiums paid.

Assessing fraud risk in insurance can be a thorny issue. This is because, fraud in insurance affects not only the insurers but it also affects the policy holders. It is well known that fraud in insurance increases the cost of insurance and this may lead to lost business to competition and policy holders paying exorbitant premiums.

Detecting insurance fraud

To detect insurance fraud, the first stage is to examine the nature of data on policies, premiums and claims. This is meant to ensure consistency and accuracy. Of great importance is capturing the POLICY NUMBER. Yes, I mean the POLICY NUMBER. This is one of the most important codes in any insurance business. It is so important such that failure to capture the policy number means you cannot track that policy, you can track the premiums paid on that policy and worst, you cannot confirm whether the claim submitted for that policy holder is authentic. Many may argue that you can use the name of a policy holder to track policies, premiums and claims. But alas, NO. What if the name you are using is shared by more than one policy holder? What if that policy holder has taken multiple policies with the same insurer?  

The next step is to proactively identify suspicious claims from either a GAP analysis or using the Benford approach. Depending on the aim, a GAP analysis may be useful in instance where you want to examine a transaction after another while the Benford’s approach helps where you want to examine a group of transactions at a go.

The other way of detecting insurance fraud is through fraud reporting mechanisms (e.g. whistle blowing hotlines, anonymous emails, etc)

Once detected, the next step involves investigating the fraud by carrying out further analysis, which we shall discuss at a later forum.

Dealing with insurance fraud

Below are some suggestions of dealing with insurance fraud:

·        Regular review of policies. This helps in weeding or cleaning up out policies with vague clauses and so forth.

·        The use of integrated information systems helps in consistency especially in capturing unique codes referring to a certain policy holder, policy details, premium details and ultimately claim details.

·        Using a combination of profiling rules to filter fraudulent transactions with advanced analytics software. The IDEA case ware tool comes in handy when carrying out the data analytics. It can carry out among other analytic tasks, gap analysis, and duplicate detection and perform a Benford test to check on the pattern of claims under a certain cover extension.

·        Denial of the claims of policy holders if they fail to submit their papers after their discharge from hospitals within the stipulated timeframe.

·        Screening the policy holders. This is likened to Know Your Customer (KYC) reviews carried out mainly by commercial banks. Screening can also be carried out on employees prior to employing them. Insurance agents and other third parties should also be subjected to screening as part of due diligence processes prior to contracting them. Screening should be continual and not a once-off exercise and if flaws are established, action should be taken immediately.

·        Physical visits to location where the risk is purported to have occurred.

·        Having a dedicated anti-fraud department.

...a bit more on Forensic Investigations

When organizations discover fraud, sometimes they tend to panic. They will call the first forensic accounting firm on the telephone directory and give a brief that they require a forensic audit done in a week’s time or two weeks or in just a few days. The organization will expect the investigator to come up with names and potential loss within that time. Few investigators will turn down a job just because the turnaround time expected is short – and there lies the beginning of a not soo good experience/relationship between the client (the organization) and the forensic investigator. I will explain.

The nature of forensic investigations is to bring forth facts relating to a certain allegation(s). Research has shown that most fraud occurrences are not one off events. Fraudsters will start with one small amount (for whatever reason) and rationalize that they will pay it back or convince themselves that they deserve the money (equity). Off course, this never happens and at some point they will slip and be caught. The point of the foregoing is that in most cases fraud will have been going on for some time before it is discovered. People committing the fraud will have had a good grasp of the business process and know where they can exploit. The expectation that an investigator will be able to put together a case/report that can stand in court in a week’s time is unreasonable. In my experience such harried investigations end up destroying the case.

The most important aspect of any investigation is planning. You get this wrong and you end up with another internal audit report. Yes, time is of essence in any investigation to ensure that evidence is not destroyed – however securing of documents and the ‘crime scene’ should be well thought out. You do not want to alert your target before the time is right.

A good investigator should therefore be able to calm his client and ensure that they understand the implications of a hurried and unplanned investigation. Given that forensic investigations are expensive undertakings a good plan ensures that appropriate time is assigned to each activity and time is not spent going back to issues that should have been dealt with earlier……