Doing the math on cyber riskReprints
Risk managers are exploring quantification techniques to better understand their cyber risk exposures and how much cyber insurance to buy, as well as other risk mitigation techniques.
However, experts say quantifying cyber risk is in the very early stages and lags the quantification used in more traditional lines of business.
“We're just starting to scratch the surface in terms of quantifying cyber risk,” said Brent Rieth, San Francisco-based senior vice president at Aon P.L.C.'s financial services group. “We've made a tremendous amount of progress as an industry in developing stronger risk models for cyber, but as data continues to aggregate” this process will improve only over time.
Experts say while there has been particular progress in measuring the effect of data breaches on personal health care information, measuring its effects more broadly in other areas such as on intellectual property or how ransomware has affected business has been slower to develop.
Meanwhile, cooperation between risk managers and those directly involved in information technology is critical to companies' success in the process.
Ways to quantify cyber risk vary, including proprietary approaches developed by brokers, those developed by independent consultants for their clients and the development of open-source software.
The quantification is needed so companies can efficiently determine their budgets, observers say.
Julian Waits Sr., CEO of Baltimore-based PivotPoint Risk Analytics, which has developed a program for its clients, said, “The whole reason we started the business was to help people with the whole focus of cyber security budgets being out of control” and executives facing the prospect of being personally liable for breaches.
One challenge in the process is how well risk managers and their IT counterparts cooperate, with company size being a factor.
Mark Greisiger, president of Gladwyne, Pennsylvania-based cyber risk management and information security provider NetDiligence, said, in meetings, there are cases where the IT person and risk manager are “exchanging business cards because they don't know each other.”
Risk managers “really don't quantify their risk for cyber,” but they do conduct research within their own organizations on the issue, said Meredith Schnur, Madison, New Jersey-based senior vice president and professional risk national practice leader at Wells Fargo Insurance Services USA Inc. “That can be anybody” from the company's information security officials to financial executives.
“We try to get (risk managers) as involved as possible, because at the end of the day, they must make a decision on how much risk transfer to purchase,” she said. “Risk managers have really stepped it up over the past two years” by increasing their involvement in this area, she said.
“Risk managers have had access to a variety of analytics around privacy-related related issues for several years now,” said Robert Parisi, managing director and national cyber product leader at Marsh L.L.C. in New York.
What has developed more slowly are the financial analytics involved, which “we take for granted” in property/casualty, directors and officers liability “and the more mature lines of coverage,” although those are beginning to develop as well, he said.
This includes, for example, better understanding of threats that disrupt a manufacturing process and translating them into their financial impact, he said.
The other piece of this is “looking at an inventory and being able to put together what are the crown jewels of the company, what would be the impact of the loss or disruption” of intellectual property or data, Mr. Parisi said.
New and evolving risks also complicate the process, observers say.
“There's very little actuarial data around to help understand the size of the risk, and the risk is very dynamic,” said Ben Beeson, cyber risk practice leader at Lockton Cos. L.L.C. in Washington.
“It doesn't sort of stand still, and the threats keep changing” and losses keep evolving, he said. It is no longer just a third-party liability risk, but also has become a first-party risk, he said.
“There are very few data points around to tell you what the size of your risk might be from a ransomware attack for example, because you don't know ultimately right now the probability of that happening,” which Mr. Beeson said is a challenging issue.
“That's not to say the industry isn't trying to solve this, and risk managers clearly are looking for answers, in conjunction, really, with the (chief information officer) in particular, and there are some solutions starting to emerge,” he said.
“There are nontechnology-related approaches that are very sort of fundamental,” such as business impact analyses, which involves developing different scenarios and suggesting their financial impact, he said.
Another approach, which has been focused over the past several years on personally identifiable information and personal health information, has been to try to use some of the broader surveys, such as that of Traverse City, Michigan-based Ponemon Institute L.L.C., to calculate response costs and potential liability, Mr. Beeson said.
In addition, “You've got a combination of certain technological tools emerging ... which are not perfect yet, but these approaches are starting to get a lot of attention,” he said.