Various groups within the NAIC are beginning to study the way in which insurers are using big data. On the property and casualty front, the Market Regulation (D) (Market Reg) Committee is reviewing the use of big data in claims settlements and the Consumer Liaison Committee has been looking into the use of big data in price optimization tools. On the life side, the Life Actuarial Task Force (LATF) is going to examine the emerging trend of accelerated underwriting using big data.
At the Summer National Meeting, the Market Reg Committee held a hearing on the use of consumer data in settling property and casualty claims. The hearing was held at the request of NAIC consumer representatives who are concerned that consumer data is improperly being used to settle claims. An industry representative explained to the Committee how big data helps insurers gather facts to more accurately and quickly determine which claims to pay and which claims to investigate and identify potential fraudulent claims. A consumer representative acknowledged that big data holds great promise for consumers and industry, but warned that because the data collected and used includes information from social media on buying habits, hobbies, and interests, the databases may not be protected by the Fair Credit Reporting Act. Because the data was not free from defect, and consumers likely do not know about the type of data being used and how it is being used in settling claims, protection is critical. The representative also cautioned that the algorithms being used to process the data may not be accurate and do not eliminate bias.
Big data may reveal a person’s buying habits. As a California regulator asked during the Consumer Liaison Committee’s discussion of "Non Traditional Rating Factors" during the Summer National Meeting, if a person buys a lot of hot dogs, does it reflect that the person is unhealthy or, is buying them for a child’s baseball team? Will this transaction be recorded and later impact this person’s insurance rates?
Regulators, amazed at the level of data being accessed, raised questions about transparency to consumers. They sought information on how the use of big data in settling claims helps consumers and what happens if, based on the data, the insurer believes a claim is fraudulent. According to the regulators, the conversation will continue.
At a LATF Summer NAIC meeting, a representative from the Society of Actuaries (SOA) informed LATF on the growing use of accelerated underwriting and simplified issue by insurers. The representative explained that a variety of data is being used in complex predictive modeling algorithms as part of insurers’ accelerated underwriting process. SOA raised concerns about the need for LATF to understand predictive analytics’ ability to predict mortality and noted that mortality experience will not emerge for several years. SOA asked LATF to consider a charge to revisit VM-51 to collect data from insurers in order to understand how their accelerated underwriting processes work, including collecting information on the algorithms and data being used. The Experience Reporting Subgroup will further review accelerated underwriting.
