Most Insurers See Predictive Analytics as Key to Fraud-Fighting: FICO/PCI Survey

With 77% of insurers estimating that fraud costs represent 5% to 20% of their claims volume, we can expect to see more implement predictive analytics within their organizations, according to a FICO/PCI survey of 143 insurance companies.

Forty-five percent of those surveyed identified predictive analytics as the most effective fraud-fighting initiative. Other spotlighted technologies included link analysis for detecting fraud (31%), business rules for stopping known fraud types (26%), and external databases (29%).

"Early detection is the key to mitigating fraud losses for insurers," says Russ Schreiber, who leads FICO's insurance practice.

[Despite the value of the technology, only 17% of P&C insurers have predictive analytics systems implemented in their claims organizations, according to Towers Watson.]

More than half of insurers surveyed expect to see an increase in the cost of fraud this year on personal insurance lines, attributable to sustained economic hardship by policyholders. Six in 10 see increases in personal property fraud, workers' compensation fraud and/or personal auto fraud. Three-quarters believe there is increased risk of fraud in no-fault states compared to states with tort systems.

"It is clear insurers understand the scope of the insurance fraud problem, and are taking steps to reduce it," said Robert Passmore, senior director of personal lines policy at PCI. "However, we also want that the public and policymakers to recognize that consumers are paying what amounts to a "fraud tax" that is far too expensive for hard-working citizens."

[Why cooperation between insurers is a fraud-fighting must.]

Nathan Golia is senior editor of Insurance & Technology. He joined the publication in 2010 as associate editor and covers all aspects of the nexus between insurance and information technology, including mobility, distribution, core systems, customer interaction, and risk ... View Full Bio