The more reliable the information that insurers can accumulate about risk, the more precisely they can manage it, whether from an underwriting, investment portfolio or operational risk perspective. That being the case, big data stands to have a profound effect on enterprise risk management (ERM). The challenge facing insurers is mastering their current data processing environments to be ready for the potential competitive benefits of big data.
To some extent, big data is the latest industry buzzword, and it’s understandably regarded with skepticism by the characteristically cautious insurance industry. Much of what’s billed as big data turns out to be about data and analytics-related offerings that fall short of the qualifying “three V’s” of big data: volume, velocity and variety.
Indeed, risk management is an area of insurance that has led the way with rapid processing of large amounts of data, but generally from restricted sources, most often data derived from internal systems. However, the third “V” — variety — will inevitably become important for ERM. Cisco Systems predicts that 1.4 zettabytes of data will circulate through the Internet during 2017 alone — a figure that exceeds the entire amount of data borne by the Internet from 1984 to 2012. Data on that scale will affect not only the velocity of analytics-powered business but the pace of change, asserted K.C. Wu, VP of IT at Cisco, at Pegasystems’ recent PegaWorld user conference.
It’s not reasonable to expect insurance to escape such a secular change, perhaps especially in the ERM function, which attempts to model future risks using all available information, says Chuck Johnston, research director in Celent’s insurance practice. “Big data has the potential to greatly increase the accuracy of risk models by increasing available data in depth and breadth,” he says. “For example, using big data-derived technologies, insurers can do policy-level stochastic modeling for financial risk, where in the past most insurers had to tranche and summarize data.”
Looking to external data, insurers can more closely quantify certain market risks by quantifying sentiment risk through social media scanning, hone operational risk models using environmental data and overall reduce the gap between known and unknown variables in the ERM equation, Johnston adds.
Insurance carrier risk managers are beginning to use the burgeoning unstructured data emanating from websites, email correspondence, contact center records, social media, blogs and even geographical information systems, such as those used by Google Maps, according to Stuart Rose, global insurance marketing manager at SAS. “Risk managers are using information from document repositories, spreadsheets, PDFs [and] XML documents to feed into risk engines to better understand the multiple different risk impacting insurers, including market risk, reputational risk, operational risk [and] technology risk,” Rose says.
Most insurers, however, are continuing to struggle with existing data environments, cautions Bill Spinard, executive director of Ernst & Young’s FSO Insurance Risk practice. “There are very big data challenges that limit many insurers’ ability to take advantage of data currently available within their enterprises, let alone the vast volumes of data predicted to circulate on the Internet,” he says.
The era of big data may be looming, but more often than not, carriers are improving their ERM capabilities in response to compliance requirements, Spinard suggests. “Many insurers are re-examining ERM because of the demands of regulators and rating agencies or preparation for ORSA [Own Risk and Solvency Assessment],” he says. However, some insurers are using the need for regulatory compliance defense to prepare for business offense. The insurance operation of Eika, an Oslo, Norway-based financial services company, selected SAS Risk Management for Insurance to meet Solvency II’s EU-harmonized requirements for calculating solvency and reporting. “Being a relatively small-sized company compared to our international competitors, it is decisive for us to improve our internal processes in response to Solvency II,” says Olav Høiby, risk manager in Eika’s insurance operations. “Our experience so far is that we can use insight from the projects [run from the SAS product] toward further development of the company and in tailoring new products in our different departments.”
Anthony O'Donnell has covered technology in the insurance industry since 2000, when he joined the editorial staff of Insurance & Technology. As an editor and reporter for I&T and the InformationWeek Financial Services of TechWeb he has written on all areas of information ... View Full Bio