This material is provided by and reproduced with permission of Dan A. Bailey. Mr. Bailey is a member of the Columbus, Ohio, law firm of Bailey Cavalieri LLP. Mr. Bailey specializes in D&O liability insurance, corporate and securities law. He is a frequent lecturer and has authored and coauthored several books dealing with D&O liability issues.

The material is not intended to provide legal advice as to any of the subjects mentioned but is presented for general information only. Readers should consult knowledgeable legal counsel as to any legal questions they may have.

As with other forms of liability insurance, excess policies are used as a means of increasing the overall limits of protection offered by the primary layer of coverage. It is not uncommon to have multiple layers of excess insurance depending on the total limits of protection required.

One of the most important issues with excess D&O insurance relates to the point in the claim process at which such excess liability actually attaches. In recent years, attachment disputes have arisen under excess D&O policies with increased frequency. In many large D&O cases, which settle for amounts involving multiple excess coverage layers, insurers often raise a variety of potential coverage defenses and then seek to negotiate with the insureds a discount off the limits of the implicated excess policies. In such situations, the primary insurer often will pay some, but usually not all, of its limit of liability in exchange for a release from the insureds, thereby creating a question as to whether liability attaches to the excess policies despite the primary insurer failing to pay its entire underlying limit.

In many large claims, all of the excess insurers potentially involved in a large settlement will collectively agree to contribute a portion of their limit towards the settlement, even though the underlying insurers have not paid their full limit. The rationale for such consensual agreement is that it often is in the mutual best interests of both the excess insurers and the insureds to resolve coverage issues pursuant to a compromise rather than engage in a protracted and expensive coverage litigation or arbitration proceeding pursuant to which the insurers could pay either nothing or their full limit of liability.

Such a mutually acceptable compromise between excess insurers and the insureds sometimes does not occur, yet the insureds nonetheless often will enter into an acceptable compromise with the primary insurer and perhaps other excess insurers. Such situations have led to a growing number of coverage lawsuits between insureds and non-settling excess insurers. In such coverage disputes, the excess insurers typically will argue that liability does not attach to the excess policy because the full amount of the underlying limit was not paid by the underlying insurers. On the other hand, insureds often argue that the full amount of the underlying limit has been incurred by the insureds, and therefore liability attaches to the excess policy (i.e., the excess policy is not dropping down, but is merely paying loss which is within the excess policy's layer of coverage).

Until recently, courts were somewhat divided on this issue. The courts which ruled that liability attaches to the excess policy even though the full amount of the underlying limit was not paid by the underlying insurers frequently relied upon a 1928 Second Circuit case applying New York law which held that an insured's settlement with a primary insurer constituted exhaustion of the primary limit of liability where the insured assumed the gap between the amount paid by the primary insurer and the full primary limit of liability. Zeig v. Massachusetts Bonding & Ins. Co.,23 F.2d 665 (2d Cir. 1928). However, other courts rejected or distinguished Zeig and held that Liability does not attach to an excess policy unless and until the underlying limit is paid in full by the underlying insurers.

As evidenced by four recent court decisions on this issue, there is now a clearer trend by courts to reject Zeig and to enforce the unambiguous attachment language in excess policies (which historically required payment by the underlying insurers of the full amount of the underlying limit as a condition to liability attaching to the excess policy). Those recent cases and their impact on future claims are summarized below.

Recent Cases

CitiGroup, Inc. v. Federal Ins. Co.

In CitiGroup, Inc. v. Federal Ins. Co., 649 F.3d 367 (5th Cir. 2011), the Fifth Circuit Federal Court of Appeals, applying Texas law, held that coverage under four excess policies did not attach where the insured settled with the primary insurer for less than the primary policy's limit of liability. In that case, the insured settled two lawsuits for $263 million and sought coverage for that settlement under the insureds' $200 million insurance program. The insurers initially denied coverage for the settlement, but the primary insurer eventually paid $15 million of its $50 million limit of liability to settle that coverage dispute. Four of the excess insurers then argued that liability did not attach because the primary insurer did not pay the full amount of the primary limit of liability. Each of the excess policies contained somewhat different attachment language. One excess policy stated that liability attached only after the underlying insurers “have paid in cash the full amount of their respective liabilities.” Another excess policy required that “the total amount of the Underlying Limit of Liability has been paid in legal currency by the insurers of the Underlying Insurance.” A third excess policy attached only after each underlying insurer “paid or have been held liable to pay the full amount of its respective limits.” The fourth excess policy stated that liability attached only after “exhaustion of all of the limits of liability of such Underlying Insurance solely as a result of payment of Loss thereunder.”

The Fifth Circuit held that all of those attachment provisions were unambiguous and required actual payment by the primary insurer of the full amount of the primary limit of liability. Because the primary insurer paid less than its full limit in exchange for a release from the insureds, the court held that none of the four excess insurers were liable for any portion of the $263 million settlement.

JP Morgan Chase & Co. v. Indian Harbor Ins. Co.

JP Morgan Chase & Co. v. Indian Harbor Ins. Co., 2011 N.Y. Misc. LEXIS 2767 (N.Y. Sup. Ct. May 26, 2011) involved a somewhat more complicated version of the Citigroup case. Instead of a single tower of insurance, JP Morgan Chase involved two separate insurance towers: (i) a $250 million D&O tower; and (ii) a $175 million bankers' professional liability (BPL) tower. The insured settled several lawsuits for a total of $718 million and submitted the loss to both its D&O and BPL carriers. Zurich issued a $15 million excess policy in the BPL tower, and Steadfast (a Zurich affiliate) had a $15 million quota share participation in two $100 million excess policies in the D&O tower. Zurich settled with the insured for a total payment of $17 million. In the settlement agreement with Zurich, the insured added a self-serving statement that Zurich's payment “exhausted” the Zurich limits under both towers even though Zurich paid less than its limits of liability under both towers.

Twin City, the carrier that issued the policy directly excess of Zurich's layer in the BPL tower, contended that liability did not attach to the Twin City policy by reason of the Zurich settlement. The Twin City policy stated liability attaches only if the underlying insurers “shall have duly admitted liability and shall have paid the full amount of their respective liability.” The insured argued that Zurich's $17 million payment more than exhausted its $15 million excess policy in the BPL tower. After reviewing law from jurisdictions across the country (including Zeig and its progeny), the court determined that Twin City's policy unambiguously required the underlying carriers to have admitted liability and paid the full amount of their respective policy limits before coverage under the Twin City policy would attach. Ultimately finding in Twin City's favor, the court noted that Zurich and the insured did not allocate the $17 million settlement between the BPL and D&O policies, making it impossible for the insured to prove the amount paid by Zurich with respect to its excess BPL policy.

Importantly, the court ruled that Zeig did not apply because the Twin City attachment language was unambiguous and that public policy arguments do not overrule clear policy language. This was the first decision by a New York court which rejected the holding in Zeig.

Goodyear Tire & Rubber Co. v. National Union Fire Ins. Co.

In Goodyear Tire & Rubber Co. v. National Union Fire Ins. Co., 211 U.S. Dist. LEXIS 121866 (N.D. Oh, Sept. 19, 2011), the court also held that liability does not attach to an excess D&O policy where the underlying insurer pays less than its full limit of liability. In that case, the insured tendered to the D&O insurers approximately $28 million in fees and expenses incurred in responding to an SEC investigation and an internal investigation. The D&O insurers denied coverage for the investigation costs, and the insured company filed a coverage lawsuit against the insurers. Two years into that coverage litigation, the insureds and the primary insurer reached a settlement pursuant to which the primary insurer paid $10 million of its $15 million limit of liability. The excess insurer then argued that it was not liable because the excess policy stated that coverage “shall attach only after the insurers of the Underlying Insurance shall have paid in legal currency the full amount of the Underlying Limit.”

The court granted the excess insurer's motion for summary judgment, rejecting the insureds' argument that Ohio public policy favors settlements and that the insureds should be allowed to “fill the gap” between the amount paid by the primary insurer and the primary policy's full limit of liability. The court ruled that the attachment language in the excess policy was unambiguous and that public policy does not trump that unambiguous language.

The court also rejected the insureds' argument that the excess insurer must prove prejudice from the failure to exhaust the underlying limit. The court held that the excess carrier did not need to prove prejudice, but nonetheless noted that the excess insurer would suffer prejudice if it could not rely on its attachment language since the excess insurer “based the premium it charged… on that expectation, not some lesser amount.” The court also noted that the excess insurer suffered prejudice because it had been forced to litigate the coverage issues for over two years based on the insureds' insistence that the excess limit was implicated. Finally, the court noted that as a sophisticated insured with the capacity to hire counsel to advise it in placing and bargaining for coverage, the insured company could have sought excess coverage with different attachment language.

Federal Ins. Co. v. Estate Irving Gould

Federal Ins. Co. v. Estate Irving Gould, 2011 U.S. Dist. LEXIS 114000 (S.D.N.Y. Sept. 28, 2011) addressed whether excess insurers must drop down when an underlying insurer does not pay covered loss due to its insolvency. In that case, the first and fourth excess layers in the company's $51 million D&O insurance tower were issued by Reliance Insurance Company and the third and sixth layers were provided by The Home Insurance Company. Both Reliance and Home were insolvent at the time the insureds sought coverage for losses under the insurance program. The court ruled that Federal Insurance Company, which issued the second excess layer (directly above the Reliance first excess layer), was not obligated to drop down and pay losses within the insolvent Reliance layer based on both the “maintenance of underlying insurance” clause and the attachment language in the Federal policy.

The Federal policy's “maintenance of underlying insurance” clause stated that “the Underlying Policies shall be maintained during the Policy Period …. Failure to comply with the foregoing will not invalidate this policy but [Federal] shall not be liable to a greater extent than if this condition had been complied with.” Judge Sullivan (who, interestingly, was general counsel at Marsh, Inc. prior to becoming a federal judge) ruled that this language was enforceable and is consistent with New York and Pennsylvania law, which generally does not require an excess insurer to fill gaps in coverage created by the insolvency of an underlying insurer.

The court likewise ruled that Federal was not obligated to drop down into the underlying Reliance layer based upon the attachment language in the Federal policy, which stated that coverage under the Federal policy was triggered “only in the event of exhaustion of all of the limit(s) of liability of such Underlying Insurance solely as a result of payment of losses thereunder.” The court found that language to be unambiguous and to clearly require payment of the underlying losses before liability attaches to the Federal excess policy. Unlike the other cases summarized above, the insured company here was in bankruptcy, and therefore the insured directors and officers were arguing that liability attached to the Federal excess policy even though no insured paid the full amount of the underlying limit. In rejecting that argument, the court recognized that the excess insurers demonstrated “impressive foresight” by including the exhaustion by payment requirement, and further noted that if the directors and officers “were able to trigger the Excess Policies simply by virtue of their aggregated losses, they might be tempted to structure inflated settlements with their adversaries…that would have the same effect as requiring the Excess Insurers to drop down and assume coverage in place of the insolvent insurers.”

Practical Implication

This clear trend by courts to enforce the attachment language in excess policies as written regardless of the consequence to insureds appears on the surface to be a significant victory for excess insurers. Unquestionably, excess insurers can be much more confident today that their unambiguous attachment language will be enforced as written, and the risk of an excess insurer dropping down into an unpaid underlying layer of coverage is very unlikely. However, in reality, these new cases will only provide marginal benefit to excess insurers in the future for two reasons.

First, in response to the increased frequency of attachment disputes in recent years and as a result of the current D&O insurance market, brokers and insureds in the last few years have forced virtually all excess D&O insurers to amend their attachment language to recognize erosion of the underlying limit not only by payments of covered loss by the underlying insurers, but also by insureds, Side-A excess DIC insurers and, in some instances, any other source. A common attachment provision in an excess D&O policy today states that liability attaches to the excess policy “only after the insurers of the Underlying Insurance, the Insureds, an excess Side A DIC insurer or any other source pays in the applicable legal currency the full amount of the Underlying Limit.” This now common language would have eliminated the attachment dispute in the CitiGroup, the JP Morgan Chase and the Goodyear cases and would have required the excess insurer to pay the covered loss within the excess insurer's layer of coverage even though the insured rather than the underlying insurer paid the losses within the underlying limit. This new attachment language, though, would not require the excess insurers to drop down into an unpaid underlying layer if no one has paid the full amount of the underlying limit. As a result, if an underlying insurer becomes insolvent (like in the Estate of Irving Gould case summarized above), liability would not attach to the solvent excess insurers unless and until the insureds or some other source pays the loss which is within the insolvent underlying insurer's layer of coverage.

Second, to the extent liability does not attach to an excess D&O policy by reason of these recent cases, there is a greater risk of coverage being triggered under an excess Side-A DIC policy. Those DIC policies typically require the Side-A insurer to drop down and fill gaps created by an underlying insurer failing or refusing to pay a covered Side-A loss. In other words, these recent cases increase that DIC exposure, but the broader attachment language which is now common in excess policies should offset that increased DIC exposure. In any event, though, this DIC exposure, of course, applies only to Side-A D&O losses, which remain relatively infrequent.