Can you imagine finding a million dollars of surplus funds inyour ledger?

That's exactly what happened to us when we ran our reinsurancecontracts through a popular data-driven audit analyticsprogram.

Here's a little-known fact: Most reinsurance agreements haveerror correction features that have no statute of limitations(meaning you can dig back throughout the years to get your moneyback!) This makes it a very opportune area for a deep-dataaudit.

But seriously, who has the time, right?

Well, from our experience, time is money, and a lot ofmoney can be found with just a small investment of time — if it'sspent wisely and the right tools are used.

Our company is a Life and Property & Casualty (P&C)multi-line insurer with annual premiums totaling about $800million. Over the last 25 years, our internal audit department hasmade a conscious effort to audit reinsurance contracts both forpremiums ceded and losses recovered.

Due to our mid-size status, reinsurance contract administrationis not centralized, touches several departments and is a small partof many employees' jobs. In our risk assessment process we askourselves: "Who is getting graded on this?" Our risk ratingsinflate when we can't answer that question with just one or twonames.

At a large company there might be a dedicated reinsurancemanager or director who oversees the function, but even with thisstructure in place, it would still pay to audit reinsurancecontracts for potential profit.

Reinsurance is not just another contractaudit

How do you go about a reinsurance audit? Before you can evenbegin, you'll need to obtain, read and understand your contracts.You might need to get data from your policy and claimsadministration systems as well, because reinsurance is not justanother contract audit.

Our P&C reinsurance contracts are industry standard:catastrophe cover, property per risk, equipment breakdown andexcess of loss. However, our antiquated P&C systems werecode-driven. To apply reinsurance correctly, you had to know andenter a code, and be able to maintain it (fancy drop-down menusdidn't exist yet). In fact, you couldn't even extract a fulldescription of loss because the field only kept the first 15characters.

As our systems improved, we were able to get full lossdescriptions. However, the loss cause code-driven environment thatfed reinsurance reporting still applied. As a result, we found thatthe available codes often failed to specifically support asubsequent reinsurance agreement, or in many cases, what was keyedjust didn't make sense.

Continue reading…

(Photo: iStock)

Keyword matching was key

To rectify the potential disparities between the actual claimand the submitted code, we applied a simple keyword matching searchthrough the auditing software we use, ACL™GRC. Here are some examples of what we found:

• One loss cause code in our system was "Fire and Lightning." Akid playing with matches who accidentally started a fire didn'tqualify for catastrophe reinsurance. However, a lightning strikethat started a fire did! But we only had one code. It wasincredibly helpful to extract the description-of-loss field for theentire population of fire claims.
• Loss cause codes could be keyed as "theft," which by itself isnot reimbursable under the catastrophe agreement. A closer look atone loss cause code description revealed that a storm wasresponsible: "Wind knocked out power to alarm system and theftoccurred." This would be covered under the agreement.
• Similar problems were noted in our Equipment Breakdown (EBK)contract. Eligible claims have a specific EBK loss cause code.Claims without the EBK code but with certain keywords in the lossdescription field, like lightning, broken,failed or cracked, were run through ACL keywordmatching, yielding hundreds of thousands of dollars inrecoveries.

We found $1 million!

Overall, we found that under normal operations, the descriptionof loss was often not being matched to the code that drove theautomated portion of catastrophe loss recoveries. Upon review, weuncovered a small number of losses that had already been submittedto the reinsurer for reimbursement that were not eligible based onloss description.

Of course, we returned our reimbursement to the reinsurer forthese claims. However, we discovered a large number of claims thatwere eligible for reimbursement for which we had failed tobill the reinsurer. We subsequently did bill the reinsurer andreceived payment for the claims.

Using audit analytics programs like ACL's, small and mid-sizefirms like ours can complete a reinsurance audit rather easily,quickly and with little upfront cost — especially considering thereturn on investment.

By conducting a thorough, retroactive audit using modern dataanalytics tools, we were able to uncover more than $1 million ineligible claims that were previously overlooked. How many millionsof dollars are waiting for you?

Steve Hummer is the senior audit manager for the Farm BureauInsurance of Michigan. He can be reached by phone at 517-679-4722or by email at [email protected]. Opinionsexpressed are the author's own.