February 2014 may ultimately be seen as the month when the cyber insurance coverage market really began. Although certain insurance companies are writing cyber coverage, and some insureds have acquired that coverage, neither group is currently comprised of significant numbers of entities. That may be about to change.
Two court rulings, one from Washington state and another from New York, have held that liability exposures resulting from the hacking of personally identifiable information (“PII”) are not covered under traditional (i.e., noncyber) liability insurance coverage forms. These rulings alone are likely to drive greater inquiry into cyber coverage. On top of that, the recent media reporting that the malware involved in the Target data breach could have been introduced through one of Target’s vendors has put a focus on a risk corridor that many entities may never have considered, but which they may now actively seek to have insured.
While large data breaches involving tens or hundreds of millions of a company’s records are generally recognized to result in significant liabilities, even the hacking of a relatively small number of records is expensive — potentially cripplingly so for a small or medium-sized business. The average out-of-pocket cost to a company experiencing a small breach, involving under 100,000 records, is $2.4 million, with an additional $3 million being the average lost revenue. See, Ponemon Institute, "2013 Cost of Data Breach Study: Global Analysis," pages 1-2 and page 16 (May 2013).
Indeed, the costs incurred to notify even just 24,000 individuals, a very small relative number, that their PII was exposed averages just above $500,000. Business owners and executives are often shocked to discover their exposure from such relatively small breaches and are left wondering whether their company can survive a seven-figure hit to the bottom line as a result of an uninsured cyber loss.
Many companies have only their traditional liability insurance programs to fall back on in the event of a cyber loss, particularly when the breach involves the hacking of PII. One of the most commonly looked-to coverages following a breach and exposure of PII, is the traditional liability coverage of “personal and advertising injury coverage,” which often covers the oral or written publication of material that violates a person’s right of privacy. That is the very policy provision under which several Sony Corp. entities — SonyCorporation of America (“SCA”) and Sony Computer Entertainment America (“SCEA”) — were seeking coverage from Zurich American Insurance Co. as a result of the hacking of the Sony PlayStation network and resulting theft of PII.
In a Feb. 21, 2014, ruling, the Supreme Court of New York held there was no coverage available to the Sony entities under the Zurich policy since the acts constituting the insurable offense were not conducted by the policyholder, Sony, but were instead undertaken by unknown third parties. As the trial court noted:
[F]or coverage to actually get triggered it would have to be, the acts have to be conducted or perpetrated by the policyholder. ... In this case I have a situation where we have a hacking, an illegal intrusion into the defendant Sony’s secured sites where they had all of the information (Court Tr. at 76).
The court held that coverage could apply only in the circumstance:
[W]here the defendants, Sony, SCA or SCEA, commits or perpetrates the act of publishing the information. In this case they didn’t do that (Court Tr. at 80).
Accordingly, the court ruled that Zurich had no coverage obligations to the Sony entities. The court noted that its decision is the first to consider the coverage implications of a major hacking incident involving coverage being sought for the alleged public exposure of PII (Court Tr. at 29-30). Risk managers and company executives do not want to risk the future financial health of their respective company’s on such an unsettled area of the law.
A second decision, issued a week after the Sony order, set up another impediment to coverage under traditional liability policy forms for an exposure of PII. On Feb. 28, 2014, the U.S. District Court for Seattle ruled in the case of National Union Fire Insurance Company of Pittsburgh v. Coinstar Inc., holding that the common liability policy exclusion for a “violation of statute in connection with sending, transmitting or communicating any material or information” served to preclude any coverage for the hacking of PII from Coinstar’s video rental kiosks — known to the movie renting public as “Redbox” (Order at 5).
Given the breadth and depth of federal and state statutes pertaining to disclosure of PII, it may be the rare exposure of PII that will take place in a fashion that does not somehow violate any of the existing statutes. Accordingly, as with the Sony decision, this case sets up a significant road block to coverage under traditional liability policies for a hacking incident involving the theft of PII.
In addition to these recent legal impediments to coverage under traditional liability forms for cyber losses, the possibility that malware may be carried into a company’s systems through a trusted vendor raises additional concerns. The terms and conditions of vendor contracts are undoubtedly getting fresh language pertaining to mandatory data-security standards to be met by vendors, likely also including a requirement for indemnification by those vendors. However, since currently available cyber coverage often indemnifies for the transmission of malicious software, one avenue to help mitigate the financial impact of this risk is for a company to require its vendors to carry appropriate cyber coverage with specified minimum limits.
While both of the recent court rulings will in all probability be appealed, the ultimate result of those appeals will not be binding on any state other than New York and Washington, respectively, and, since the Coinstar case is in federal court, even that precedent is not necessarily binding on Washington state courts.
These recent decisions may well provide the impetus for companies to look more closely at the emerging cyber insurance market. For their part, the insurance companies will need to be sure they are providing a valuable product to the market. Given the risk, it is not especially compelling to a prospective insured to be offered, for example, $5,000 in notification expense coverage when the average cost for consumer notification for even a small breach exceeds $500,000. As the insurance companies and the prospective insureds contemporaneously realize both the need and the market for such insurance — and that it has a very real value worthy of the payment of premium dollars — it may well be that February 2014 will be the tipping point for this type of coverage.