From faked deaths to real murder, cash-strapped policyholders, con artists and unscrupulous agents can go to extraordinary lengths in pursuit of a fraudulent life insurance payout. Increasingly, insurers are fighting back. RGA’s cross-functional Fraud Conference was designed to bring together professionals from multiple disciplines and industries to learn how to defeat these schemes. Still, while interest has been intense, actuaries seem less engaged, says Kathryn Cox, SVP, Business Development, RGA U.S. Mortality Markets. At a panel discussion with Brian Millman of MIB at the Society of Actuaries annual meeting, she called on her fellow actuaries to join underwriters, claim analysts and other industry professionals at the front lines. She recently sat down with RGA to explain why.
Fraud is an ever-present challenge for life insurers, and yet you seem to believe that life actuaries need to confront it with greater urgency. Why?
First, let me say that actuaries do take fraud seriously. We all acknowledge that our industry is facing a number of challenges: rising competition, growing digital disruption, escalating operational costs, the list goes on. We can’t afford to be complacent – fraud takes an enormous toll on individuals, the community, the carrier’s bottom line — and ultimately the consumer in the form of higher premiums. We all acknowledge this – and yet, our experience with the RGA Fraud Conference has been eye-opening.
Over the past six years, attendance at our conference has increased substantially every year. Fraud is an industrywide problem. We designed a conference to bring the industry together to help solve it: we welcome claims analysts, lawyers and risk managers, data scientists, behavioral economists, underwriters, members of law enforcement and private investigators. We welcome people from the life insurance industry and beyond. And, of course, we welcome actuaries. And yet in the last year, of the 100+ attendees, only three were actuaries. Fraud is rampant – the Coalition for Insurance Fraud pegs it at $80 billion dollars across all lines of insurance, equivalent to the GDP of some small countries.
Are actuaries just too busy?I'm not sure that's the issue. Perhaps we think that, while fraud is important, it should be a priority for another group? This isn’t an underwriting problem or a claims problem. It’s an industry problem and, I can say this as an actuary, we are a key part of this industry. Criminals collaborate – they don’t make distinctions between departments or units. And it will take all of us to be 100% committed to stay ahead of fraud – these schemes come in a full spectrum of shades. Fraudsters never stop changing colors, evolving techniques, adapting technologies … and, yes, innovating.
The analogy of the spectrum is challenging. If these schemes come in all of these different varieties, how can an actuary contribute – when so much of actuarial science relies on past experience?
Actuaries are exceptional at seeing patterns, identifying trends, taking problems to the root cause to help find proactive solutions. Think about the way we build our pricing models, and how that has evolved. In the past, we might have performed an analysis of our experience to determine the drivers of fraud and developed rules so that a claims analyst could identify and flag suspicious submissions based on prespecified criteria. At some carriers, these claims might be forwarded to a Special Investigation Unit for additional attention.
This still happens, but now we also employ some pretty bright data scientists and actuaries who are building predictive models that draw on a host of traditional and non-traditional data sources to help identify fraud. We can apply analytics in real-time or near-real-time. The capacity to work faster, with a more full-color perspective, is there – but the implementation is uneven and some of us are still operating in black and white. I see significant opportunity.
What you describe sounds like fairly high-tech investigative techniques?
It can be as simple as a quick online search. Some fraud schemes are not very clever, while others sound like something out of a bestselling crime novel. One of my favorites is the case of the drowning husband. It goes like this: a death claim was filed on a six-month-old policy. The cause of death was accidental drowning while swimming in a lake. The body was said to have been found and cremated. So far, so good right?
The wife was listed as the beneficiary, and she asked that the claim be expedited. Since the matter of death was accidental, the insurance company routinely conducts limited investigations, such as online research. The insurer searched online news sources and came up with absolutely no articles about the drowning. Then, there was the review of the online obituary, which showed the wife to have a different last name than the insured’s. Condolences posted with the obituary referenced a similar tragic death a few years prior. Reviews of Facebook showed the insured’s account was closed, posts indicated that the insured and beneficiary had recently been married and posts referenced the death of her previous husband, who had coincidentally died of accidental drowning.
The red flags generated by a simple online search led to a full contestable investigation. The investigation proved that the drowning was fabricated, the death certificate was fraudulent, and the funeral home was complicit.
You mention an online search, but yet we can’t go a month, it seems, without another hacking incident from some website. The Internet and cloud-based technologies make us more vulnerable to fraud, too.
Absolutely. The motives for fraud haven’t changed, but the methods have. The same high-tech tools are at the fraudster’s disposal. Documents can be easily falsified. Devices can enable untraceable communications. Online content opens up more opportunities to steal personally identifiable information, and even websites for commercial DNA testing provide new avenues for consumer anti-selection. As acceleration and automation also expedite the underwriting process and reduce human involvement, we could see more loopholes appear that criminals could exploit. And that’s not even addressing the more dangerous data for sale on the Dark Web.
If it is the case that the motives for fraud are the same, won’t fraud always be with us, no matter what technologies and techniques we employ?
Yes. But I still believe we can make a dent in it. While there are groups and individuals intentionally committing fraud, the greater threat may be more mundane: the little white lies and bending of truths from ordinary applicants. We all know these tend to add up. It’s one thing to add a Provision for Adverse Deviation (PAD) for volatility, but wouldn’t it be nice if we didn’t have to add a PAD for anti-selection? Here, we have new behavioral science tools that can provide a path forward from the product design to the application process. It’s not just about learning technologies; it’s also about looking for answers and strategies that work.
All this seems rather daunting.
In a perfect world, we wouldn’t need a fraud conference. Insurance applicants would be 100% truthful on their applications, all claims would be legitimate, and claims proceeds would go to the people they are intended to help. That’s not the world we’re living in today. But we can take comfort in fact that knowledge is power, and working together within and across industries we are immensely powerful.