FACTS AND PROCEDURAL HISTORY

ITT’s Liability Insurance

ITT manufactured and supplied products to a variety of industries worldwide, including automotive and industrial components. During the relevant period, from 1959 to 1986, ITT purchased multiple layers of insurance coverage: (1) primary policies for different policy periods from various insurers, including Pacific Employers Insurance Company (PEIC); (2) umbrella policies from various insurers, including Affiliated FM Insurance Co. (Affiliated FM); and (3) excess policies from various insurers, including Munich and United States Fire Insurance Company (U.S. Fire). The layers of insurance coverage were structured to operate as if a single insurer had issued a single policy. If ITT’s primary insurance coverage were exhausted by a single claim or group of claims, ITT expected the first layer umbrella policy to take over as if it were the primary policy. If the umbrella policy were exhausted by the payment of claims up to the aggregate limit, the next layer of excess coverage would be activated, exhausting vertically thereafter until the entire amount of coverage was exhausted. ITT purchased “follow form” excess policies that incorporated the terms and conditions of the lead umbrella policy for substantially identical coverage terms. Thousands of product liability actions have been filed against ITT since the 1990s, seeking compensatory damages for progressive injury, disease, and death, alleged to have been

4 caused by asbestos exposure. Primary insurer PEIC initially assumed responsibility for product liability lawsuits against ITT arising from asbestos, paying the costs of defending and resolving the claims on ITT’s behalf.

Litigation Instituted and Phase III of Trial

PEIC began to dispute the extent of its obligations, leading ITT to file the instant action for declaratory relief against its primary, umbrella, and excess insurers in February 2003. Trial was conducted in six phases, but only phases relevant to the issues on appeal are described herein. ITT reached agreements with certain primary insurers requiring payment of ITT’s defense costs and most of the costs to resolve claims, but ITT’s primary coverage for the relevant time-period was exhausted by the end of 2013. Phase III of the trial was held in June 2017. The trial court issued a statement of decision on August 17, 2017. It is undisputed on appeal that New York law governs interpretation of the insurance policies.

A. Trigger of Coverage

The court found asbestos exposure causes a continuous injurious process. “Bodily injury,” within the meaning of liability insurance policies, takes place with the initial cellular injury that occurs upon exposure to asbestos fibers and progresses continuously as the damage cascades, over time, into disease, frequently ending in death.

5 The injury begins at the time of the claimant’s exposure to asbestos and continues thereafter, triggering each policy in effect from the time of the claimant’s first exposure to asbestos until the claimant’s diagnosis of an asbestos-related disease or death. If a claimant’s exposure to asbestos occurred prior to or during an insurer’s policy period, the policy’s coverage was triggered.

B. Allocation of Loss when Multiple Successive Policies are Triggered

A loss may start in one policy period, but extend beyond it. Allocation refers to distributing responsibility for a loss among multiple policies that have been triggered. The court considered two general approaches to allocation: all sums and pro rata. All sums allocation applies when each policy carries a separate, independent obligation to fully indemnify the insured up to the policy limits. Under the all sums approach, the insured can select a single policy year to respond to losses and collapse all of its costs into a single policy year. The insurers for that policy year must reimburse the entire loss up to the limits of the policies, subject to their contribution rights against the other triggered policies. Pro rata allocation apportions a loss among all the triggered policies. Each triggered policy provides coverage only for the portion of asbestos-related bodily injury occurring during the policy period, spreading the loss across all the triggered policy years.

6 C. Policies Containing Non-Cumulation Conditions

The policy language dictates the method of allocating loss among multiple successive policies. With the exception of one umbrella policy issued by Affiliated FM, all of the umbrella policies at issue contain a “non-cumulation condition” or “prior insurance and non-cumulation of liability condition.” The New York Court of Appeals has already determined that for policies with a non-cumulation condition, or a prior insurance and non- cumulation condition, all sums is the appropriate allocation method. (In re Viking Pump, Inc. (2016) 27 N.Y.3d 244, 260–261 (Viking Pump).) Conditions in insurance policies do not expand or restrict coverage; they are conditions on the promise to pay. A “prior insurance and non-cumulation of liability condition” in umbrella and excess policies prevents an insured from asserting a claim and obtaining indemnity for the same loss occurrence under horizontally related policies. The court found the allocation provisions of ITT’s policies expressly provided for indemnification of “all sums” and non- cumulation conditions, with the exception of the umbrella policy issued by Affiliated FM in effect between 1975 to 1977.

D. Affiliated FM Umbrella Policy

With respect to the Affiliated FM umbrella policy, the court concluded it was an “all sums” policy based on the plain language of the provisions. Affiliated FM agreed to pay the “ultimate net loss” for which ITT was found liable. The policy defined “ultimate

7 net loss” to include “all sums” which ITT became legally obligated to pay as damages and the defense expenses for such claims. Based on this language, the court found the policy agreed to pay “all sums.” The court noted that the Affiliated FM policy contained a pro rata liability provision, but the provision applied solely to ITT’s liability for advertising injuries. The Affiliated FM policy expressly covered personal injury liability for damages, including death at any time resulting therefrom, sustained by any person or persons. The policy defined an “occurrence” to include continuous or repeated exposure to conditions, which results in personal injury. In the event of exhaustion of the underlying limits though payment of losses, the Affiliated FM policy would continue in force as underlying insurance. The court found the Affiliated FM policy required indemnity for all sums that the insured became legally obligated to pay as damages, including death at any time resulting therefrom, as shown by the policy definition. The court concluded the express contract terms promised to protect ITT from liability for damages arising from injury that took place during, but continued beyond, the policy period. The promise to pay damages that continue beyond the policy period, such that multiple successive insurance policies could indemnify ITT for the same loss or occurrence, is the same as the inference from a non- cumulation condition to which the New York Court of Appeals found all sums allocation was appropriate. Although the Affiliated FM policy did not contain a “non-cumulation or prior insurance condition,” this single difference from the other

8 umbrella policies, although relevant, did not lead to a different interpretation with respect to the allocation method. The court concluded that all of ITT’s umbrella policies, either through a non-cumulation condition or through the language of other policy provisions, provided for the possibility that multiple successive insurance policies might indemnify the insured for the same loss. Where recoveries on continuing losses could overlap, the policy terms anticipated and addressed continuing coverage.

E. Excess Policies

The excess insurance policies either contained express terms identical to the underlying umbrella policy or “follow form” provisions adopting the warranties, terms, and conditions of the underlying umbrella insurance policy. The excess layers increased the limits of liability without adding to the underlying coverage. Munich argued that its excess policies contained their own insuring agreements and did not follow form to the underlying umbrella policy. The language of their policies did not support the argument. Therefore, all of the excess policies at issue promised to indemnify ITT for “all sums” that ITT became obligated to pay as damages due to liability from bodily injury or personal injury caused by an occurrence. With the exception of the excess policies following form to the two-year Affiliated FM umbrella, all of the excess policies included a non-cumulation and/or prior insurance condition. The excess policies following the Affiliated FM umbrella policy were reasonably construed to reflect the

9 parties’ mutual intent to provide coverage for the contractually bargained for policy limits for long-tail losses. The court concluded, in accordance with New York law, that non-cumulation conditions were inconsistent with pro rata allocation. ITT’s losses should be allocated on an all sums methodology, allowing the insured to collect its total liability under any policy in effect during the periods that damage occurred, up to the policy limits. “Once a policy has attached and is selected by the insured, each triggered policy has a separate and independent obligation to cover ITT in full for its liability arising from a given claim, up to the policy’s product liability limits.”

F. Horizontal and Vertical Exhaustion

The court additionally found, relying on New York law, that vertical exhaustion of policy limits was consistent with the all sums allocation method, allowing an insured to seek coverage through the layers of insurance for a specific year. Therefore, exhaustion of the available underlying coverage for a particular policy period triggered the excess policies for the same policy period.

Phase IV of Trial

Phase IV of the trial was held in November 2018. The trial court issued a statement of decision on January 9, 2019. Resolute Management, Inc. administers claims for several insurer defendants (collectively referred to herein as Resolute). In Phase IV, the court determined, among other issues, the effect

10 of prior insurance and non-cumulation provisions. Resolute characterized ITT’s asbestos-related claims as a single loss and argued that the prior insurance and non-cumulation condition in the excess policies at issue reduced the aggregate limits available. The court found the testimony of Resolute’s expert Barrett Breitung unbelievable that a reimbursement payment for a single asbestos-related claim under a particular policy would, as a result of the non-cumulation condition, eliminate any other claim for asbestos-related liability loss, even claims resulting from a different occurrence, under later policies issued by different insurers at the same coverage level. The applicable non-cumulation condition stated: “It is agreed that if any loss covered hereunder is also covered in whole or in part under any other excess policy issued to the Assured prior to the inception date hereof, other than a policy that is specifically stated to be in excess of this Policy, the limit of liability hereon as stated in Item 2 of the Declarations shall be reduced by any amounts due to the Assured on account of such loss under prior insurance.” The court noted that language of the condition standing alone did not have a definite and precise meaning. The court found the testimony of defendants’ witnesses about how the non- cumulation condition operated was confusing and inconsistent. ITT’s interpretation that each asbestos claim arises out of a separate occurrence and therefore the term “loss” must be tethered to an “occurrence” was supported by ample case authority. Under New York law, each individual plaintiff’s continuous or repeated exposure to asbestos is a separate occurrence unless it can be shown that multiple plaintiffs’ exposure was at the same time and place. Based on these

11 findings, the trial court concluded the language of the non- cumulation condition was ambiguous, because ITT and the insurer defendants proposed reasonable, but inconsistent, interpretations. The court noted that an ambiguity must be construed against the drafter and in favor of coverage. The policy must also be interpreted to fulfill the reasonable expectations of the insured. The provision states that the liability limit shall be reduced by any amounts due to the insured, but does not explain the meaning of “amount due.” Because the policyholder can choose the policy year to cover its liability, the condition cannot mean that the same amounts are “due” under prior policies. The court found credible the testimony of expert witness Peter Wilson that the non-cumulation condition was developed as part of a new umbrella form in the 1960s to prevent a policy holder from obtaining a double recovery by pursuing coverage under two policies for the same liability. The condition was intended to address situations where an occurrence could be construed to overlap multiple policy periods and an insured could seek to collect “loss” arising out of the occurrence in more than one policy year. Wilson believed the non-cumulation condition limited the insured to a recovery of one year’s limits for the loss or occurrence. The insured could not “stack” a single claim or loss arising from a single occurrence under multiple policies. In the broadest sense, stacking means treating multiple policies that apply to a single loss cumulatively. Most often, stacking refers to the insured’s ability to obtain benefits from a second policy on the same claim. The drafters did not design the condition to apply to continuous triggers and all sums allocation.

12 Expert witness Brian Hibbert, who collaborated with Wilson to write policies for ITT, testified that the non-cumulation language was intended to address stacking when a single occurrence or loss could be claimed under more than one insurance policy. The underwriters did not intend that prior insurance could eliminate coverage under later-in-time policies for which the premium was being charged and paid. Only if a policyholder tried to “stack” policies after obtaining indemnification for a specific loss from an earlier policy in the relevant coverage layer would the later policy’s non-cumulation condition be implicated. The policy must be viewed as a whole. Viewing the policy terms as a whole, the court found the only reasonable interpretation of the loss referred to in the non-cumulation provision was as tethered to “occurrence.” Each separate claim alleging exposure to asbestos from an ITT product arises out of a separate occurrence. Only if ITT attempts to assert definite claims for reimbursement of the same “loss” under more than only policy at the same excess layer covering different years does the non-cumulation provision apply, and in those instances, the condition limits the “amounts due” on account of that same loss under the prior insurance. In short, the condition does not apply until the insured has obtained indemnification for that specific loss from another policy in the relevant coverage layer. For the vast majority of settlements, ITT kept the loss for each occurrence in one policy year. For 76 claims, ITT split the settlement amount equally between multiple policy years. For these claims, both policies contribute but neither policy pays out more than a reasonable proportion, as expected by the underwriters, brokers, and policyholder.

13 In March 2020, ITT and Resolute entered into a settlement agreement buying out certain ITT policies. ITT had billed more than $7 million to these policies, but settled before the amounts were paid. In exchange for a release of ITT’s claim, the Resolute settlement transferred $60 million to a qualified settlement fund (QSF).

Phase V of Trial

Phase V of the trial was held in November 2020. The trial court considered, among other issues, whether ITT was required to allocate proceeds from the Resolute settlement to losses attributable to specific asbestos-related claims, impairing the policies settled under that agreement, before accessing U.S. Fire Policy No. XS 2199 for the 1972-1973 policy year. The court entered a statement of decision for Phase V on January 25, 2021. The court noted the undisputed evidence showed ITT would not obtain a windfall as a result of the Resolute settlement. The evidence showed asbestos claims against ITT would extend to 2054, and many of ITT’s insurers faced insolvency. The settlement proceeds would be used to provide ITT the liquidity necessary to advance funds to meet ongoing defense and settlement obligations in light of the insurers’ chronic failure to timely pay, disputes over reimbursement standards, insolvencies, and other delays. ITT’s unreimbursed past settlements greatly exceeded the proceeds of the Resolute settlement. ITT had absorbed more than $64 million in unreimbursed settlements paid between 2005 and 2016, $44 million in unreimbursed defense costs since 2015, and $22 million in asbestos losses due to insolvencies of subscribers to the London policies. U.S. Fire had

14 failed to show that the settlement proceeds exceeded ITT’s obligations for past, present, and future asbestos-related losses, and therefore failed to show that ITT had been made whole and any further recovery would be a windfall or multiple recovery, such that the settlement proceeds were allocable to losses for which ITT was claiming a right of recovery against U.S. Fire. The court concluded that ITT’s claims were due and payable under the U.S. Fire Policy at issue. In December 2020, ITT entered into a more comprehensive settlement agreement with certain insurers (the CIP settlement). In Phase VI, the trial court considered Munich’s cross- claims for equitable indemnity and equitable contribution. The trial court explained that Phase V had determined whether ITT must allocate the Resolute settlement proceeds to losses attributable to specific asbestos-related claims before accessing U.S. Fire Policy XS 2199 in the 1972-1973 policy year. The court noted that Munich was given an opportunity to join with U.S. Fire in asserting this position, or to articulate its own “offset” issue with respect to the Resolute settlement, but had declined. Phase V proceeded without Munich’s participation. At the end of Phase V, the trial court rejected the contention that U.S. Fire was entitled to offset or further allocation of the Resolute settlement proceeds. The trial court considered Munich’s claims for equitable contribution or indemnification and concluded Munich had failed to show it had paid more than its fair share. Munich argued that every unexhausted policy issued a solvent insurers below the Munich layer should pay their fair share of payments Munich made to ITT under 1970 MRAm Policy No M-0083871. The court found the testimony of Munich’s expert was entitled to no weight.

15 Several of the insurers that Munich sought contribution from were insolvent or not currently liable to ITT. Munich did not pay ITT timely under the terms of its own policy, and did not provide proper notice of its claims to other insurers of its intent to seek contribution. Allowing equitable contribution from settled insurers would defeat the purpose of settling and turn equity upside down. Munich had not paid more than its fair share and the relief that Munich sought was neither fair, nor equitable. After Phase VI of the trial, on August 18, 2021, the trial court entered judgment and a statement of decision resolving all of the claims and defenses asserted by ITT and its insurers. Munich filed a timely notice of appeal from the judgment.

DISCUSSION

Standard of Review

“Under New York law, insurance policies are interpreted according to general rules of contract interpretation.” (Olin Corporation v. OneBeacon America Insurance Company (2d Cir. 2017) 864 F.3d 130, 147–148 (Olin).) We review contract interpretation questions de novo. (MPEG LA, LLC v. Samsung Electronics Co., Ltd. (N.Y. App. Div. 2018) 166 A.D.3d 13, 17.) “In determining a dispute over insurance coverage, we first look to the language of the policy [citation].” (Consolidated Edison Co. of New York, Inc. v. Allstate Ins. Co. (2002) 98 N.Y.2d 208, 221–222 (Consolidated Edison).) “When construing insurance policies, the language of the ‘contracts must be interpreted according to common speech and consistent with the reasonable expectation of the average insured’ [citations].

16 Furthermore, ‘we must construe the policy in a way that affords a fair meaning to all of the language employed by the parties in the contract and leaves no provision without force and effect’ [citation]. Significantly, ‘surplusage [is] a result to be avoided’ [citation]. Moreover, while ‘ “ambiguities in an insurance policy are to be construed against the insurer” ’ [citations], a contract is not ambiguous ‘if the language it uses has a definite and precise meaning, unattended by danger of misconception in the purport of the [agreement] itself, and concerning which there is no reasonable basis for a difference of opinion’ [citation].” (Viking Pump, supra, 27 N.Y.3d at pp. 257–258.)

Allocation of Loss Among Multiple Successive Policies

Munich contends the plain language of the Affiliated FM umbrella policy and Munich’s following form excess policy required pro rata allocation of loss. We disagree.

A. Relevant Policy Provisions

1. Affiliated FM Umbrella Policy

The original policy period for the Affiliated FM umbrella policy was August 1975 to August 1978. Affiliated FM agreed to provide coverage on behalf of ITT for “ultimate net loss in excess of the retained limit hereinafter stated, which the insured may sustain by reason of liability imposed on the insured by law, or assumed by the insured under contract arising out of: [¶] (a) Personal Injury Liability [¶] For damages including damages for care and loss of services, because of personal injury, including

17 death at any time resulting therefrom, sustained by any person or persons; . . . .” The policy applied “only to occurrences during the Policy period anywhere; except that, with respect to Coverage I(c) [advertising injury], if any such occurrence began prior to the Policy period and continued during or after that period, the company’s liability shall be limited to that proportion of the total damages resulting therefrom which the use of injurious material or the number of injurious acts during that period bears to the aggregate use of injurious material or the total number of injurious acts.” An “occurrence” was “an accident, including continuous or repeated exposure to conditions, which results in personal injury or property damage neither expected nor intended from the standpoint of the insured.” (The provision did not limit coverage to personal injury during the policy period.) “There is no limit to the number of occurrences for which claims may be made hereunder, provided such occurrences happen during the Policy period, except as hereinafter provided.” Personal injury included “bodily injury, sickness, disease, disability, shock, mental anguish and mental injury[.]” The definition of personal injury was not limited to injuries occurring during the policy period. “Ultimate net loss” was expressly defined as “the total of the following sums with respect to such occurrence: [¶] (1) all sums which the insured . . . became legally obligated to pay as damages, whether by reason of adjudication or settlement, because of personal injury . . . to which this Policy applies, and [¶] (2) all expenses incurred by the insured in the investigation, negotiation, settlement and defense of any claim or suit seeking

18 such damages, . . . provided ‘ultimate net loss’ shall not include any damages or expense because of liability excluded by this Policy.” The policy stated that it did not apply to defense, investigation, settlement or legal expenses covered by underlying insurance.

2. 1975-1978 AmRe Excess Policy

American Re-Insurance Company Policy No. M-1027596, to which Munich succeeded, was a following form excess policy in effect from December 31, 1975, to December 31, 1978 (the 1975-1978 AmRe excess policy). The declarations page of the 1975-1978 AmRe excess policy states that the underlying insurance is the first layer umbrella provided by Affiliated FM. The policy also stated, “Ultimate net loss, as used herein, shall be understood to mean the sums paid in settlement of losses for which the Insured [is] liable after making deductions for all recoveries, salvages and other Insurances (other than recoveries under the underlying insurance policies of co-insurance, or policies specifically in excess hereof), whether recoverable or not, and shall exclude all ‘Costs.’ ” The policy further provided, “This Certificate applies only to accidents or occurrences happening between the effective and expiration dates shown in Item 2 of the Declarations, unless otherwise cancelled.”

3. Substitution of Underlying Insurance

Effective August 5, 1977, the underlying umbrella policy for the 1975-1978 AmRe excess policy was replaced: “In

19 consideration of payment of the premium charged, it is understood and agreed that Item [Number] 4 – Underlying Insurance, of this Certificate, is amended in part to read: [¶] First Layer Umbrella provided by London (Policy No. to be advised). [¶] And an additional First Layer Umbrella provided by National Union Insurance Company. Policy Number 1128365.” The newly substituted umbrella policies contained non- cumulation provisions. No additional premium amount was charged.

B. Law Applicable to Allocation of Loss

1. General Principles

Under the “all sums” method of allocating losses, the insured may recover its total liability under any policy that was in effect during the period that the damage occurred, up to the limits of that policy. (Viking Pump, supra, 27 N.Y.3d at p. 255.) “The burden is then on the insurer against whom the insured recovers to seek contribution from the insurers that issued the other triggered policies [citation].” (Id. at pp. 255–256.) Under the pro rata method of allocation, “an insurer’s liability is limited to sums incurred by the insured during the policy period; in other words, each insurance policy is allocated a ‘pro rata’ share of the total loss representing the portion of the loss that occurred during the policy period [citation].” (Viking Pump, supra, 27 N.Y.3d at p. 256.) Dividing liability when a claim alleges gradual and continuing harm reflects that it is difficult to determine what transpired during any particular policy period. (Ibid.)

20 “Some jurisdictions have expressed a preference for the all sums method, usually relying on language in policies obligating an insurer to pay ‘all sums’ for which an insured becomes liable [citations]. Others have, instead, utilized the pro rata method, emphasizing language in the insurance policies that may be interpreted as limiting the ‘all sums’ owed to those resulting from an occurrence ‘during the policy period,’ or public policy reasons supporting pro rata allocation, or a combination of the two [citations].” (Viking Pump, supra, 27 N.Y.3d at pp. 256–257.) In determining whether all sums or pro rata allocation applies to a particular policy, New York courts have focused on whether the plain language of the policy expressly extends the protection of the policy beyond the policy period for continuing injuries, potentially overlapping with coverage under successive policies, or whether the policy limits liability to only the portion of the loss within the policy period, such that no two insurance policies in separate policy periods would indemnify the same loss or occurrence.

2. Consolidated Edison

The New York Court of Appeals first considered the appropriate method for allocating loss when a continuous harm triggers successive insurance policies in Consolidated Edison, supra, 98 N.Y.2d at p. 221. Consolidated Edison involved property damage due to decades of environmental contamination, which triggered coverage under multiple successive policies. (Id. at p. 215.) The court’s analysis focused on the policy terms “all sums” and “during the policy period.” (Id. at p. 222.) The relevant

21 policy provided coverage for all sums that the insured was liable to pay for damages on account of “ ‘property damage, caused by or arising out of each occurrence.’ ” (Id. at p. 222.) But the policy applied only to occurrences “ ‘happening during the policy period.’ ” (Id. at p. 222.) And an occurrence was defined as “ ‘an event, or continuous or repeated exposure to conditions, which causes injury, damage or destruction during the policy period. All such exposure to or events resulting from substantially the same general conditions during the policy period shall be deemed one occurrence’ (emphasis supplied).” (Id. at p. 222.) In other words, the policy’s definition of an occurrence limited coverage to liability for injury or damage during the policy period. The insured asserted that because the policies at issue covered liability for “all sums,” it could select any one of the policies in effect during the 50 years that property damage occurred and collect its total liability from that insurer, up to the policy limit. (Consolidated Edison, supra, 98 N.Y.2d at p. 222.) The insurers argued, however, that the phrase “during the policy period” limited coverage to all sums incurred during the policy period. (Consolidated Edison, supra, 98 N.Y.2d at pp. 222–223.) The Consolidated Edison court concluded that all sums allocation, which would permit the insured to select a particular policy to collect the total liability, was not consistent with the policy language providing indemnification for all sums of liability resulting from an accident or occurrence “during the policy period.” (Id. at p. 224.) “Pro rata allocation under these facts, while not explicitly mandated by the policies, is consistent with the language of the policies. Most fundamentally, the policies provide indemnification for liability incurred as a result of an

22 accident or occurrence during the policy period, not outside that period [citation].” (Ibid.) Focusing solely on the term “all sums” would read this qualification out of the policies. (Ibid.) The Consolidated Edison court did not conclude, however, that pro rata allocation was the appropriate allocation method among successive insurance policies in every case. (Viking Pump, supra, 27 N.Y.3d at p. 257.) The court relied on general principles of contract interpretation and found the language of the particular policy at issue in a case controls the allocation method. (Ibid.)

3. Viking Pump

In Viking Pump, the umbrella policies at issue agreed to indemnify the insured for “ ‘all sums in excess of the retained limit which the insured shall become legally obligated to pay, or with the consent of the [insurer], agrees to pay, as damages, direct or consequential, because of . . . [¶] personal injury . . . [¶] with respect to which this policy applies and caused by an occurrence’ (emphasis added).” (In re Viking Pump, supra, 27 N.Y.3d at pp. 251–252.) The definition of “occurrence” included “ ‘injurious exposure to conditions, which results in personal injury,’ ” which, in turn, is defined as “ ‘personal injury or bodily injury which occurs during the policy period’ (emphasis added).” (Viking Pump, 27 N.Y.3d at p. 252.) To determine the limits of the insured’s liability, the policies provided that all personal injury “ ‘arising out of continuous or repeated exposure to substantially the same general conditions . . . shall be considered as the result of one and

23 the same occurrence.’ ” (Viking Pump, supra, 27 N.Y.3d at p. 252.) The excess policies at issue in Viking Pump either followed form to these umbrella policy provisions or provided for substantively identical coverage. (Ibid.) The umbrella policy contained a non-cumulation provision stating: “ ‘[i]f the same occurrence gives rise to personal injury, property damage or advertising injury or damage which occurs partly before and partly within any annual period of this policy, the each occurrence limit and the applicable aggregate limit or limits of this policy shall be reduced by the amount of each payment made by [the umbrella insurer] with respect to such occurrence, either under a previous policy or policies of which this is a replacement, or under this policy with respect to previous annual periods thereof.’ ” (Viking Pump, supra, 27 N.Y.3d at p. 252.) Excess policies that did not follow form to the umbrella policy’s non-cumulation provision contained a similar prior insurance and non-cumulation provision. (Ibid.) The New York Court of Appeals concluded that all sums allocation was the appropriate method for a policy containing a non-cumulation provision. (Viking Pump, supra, 27 N.Y.3d at pp. 260–261.) The non-cumulation provision, by acknowledging and addressing situations where a loss or occurrence may also be covered under another policy, plainly contemplates that multiple successive policies could cover the same loss or occurrence. (Id. at p. 261.) Pro rata allocation would be inconsistent with a non- cumulation provision and render the non-cumulation provision surplusage. (Ibid.) “[T]he very essence of pro rata allocation is that the insurance policy language limits indemnification to losses and occurrences during the policy period—meaning that no two

24 insurance policies, unless containing overlapping or concurrent policy periods, would indemnify the same loss or occurrence. Pro rata allocation is a legal fiction designed to treat continuous and indivisible injuries as distinct in each policy period as a result of the ‘during the policy period’ limitation, despite the fact that the injuries may not actually be capable of being confined to specific time periods. The non-cumulation clause negates that premise by presupposing that two policies may be called upon to indemnify the insured for the same loss or occurrence.” (Viking Pump, supra, 27 N.Y.3d at p. 261.) Several of the excess policies in Viking Pump also contained “continuing coverage clauses within the non- cumulation and prior insurance provisions, reinforcing our conclusion that all sums—not pro rata—allocation was intended in such policies. The continuing coverage clause expressly extends a policy’s protections beyond the policy period for continuing injuries. Yet, under a pro rata allocation, no policy covers a loss that began during a particular policy period and continued after termination of that period because that subsequent loss would be apportioned to the next policy period as its pro rata share. Using the pro rata allocation would, therefore, render the continuing coverage clause irrelevant. Thus, presence of that clause in the respective policies further compels an interpretation in favor of all sums allocation [citations].” (Viking Pump, supra, 27 N.Y.3d at p. 262.) Based on the language of the policy at issue and persuasive arguments that pro rata allocation is inconsistent with non- cumulation and non-cumulation/prior insurance provisions, the Viking Pump court held that all sums allocation is appropriate in

25 policies containing non-cumulation provisions such as the ones at issue in that case. (Viking Pump, supra, 27 N.Y.3d at p. 264.)

4. American Precision

A third relevant case currently wending its way through federal court, American Precision Industries, Inc. v. Federal Insurance Company (W.D.N.Y. 2022) 648 F.Supp.3d 442 (American Precision), also concerns asbestos-related personal injury lawsuits. The insurer defendants issued primary comprehensive general liability policies to the plaintiff. (Id. at p. 443.) The policy language limited coverage to bodily injury or property damage that occurred “ ‘during the policy period.’ ” (Id. at p. 445.) The policies at issue expressly defined bodily injury as “ ‘bodily injury, sickness or disease sustained by any person which occurs during the policy period, including death at any time resulting therefrom,’ ” or they contained functionally equivalent language. (Id. at p. 446.) The parties filed cross-motions for summary judgment concerning the allocation of defense and indemnity costs. (American Precision, supra, 648 F.Supp.3d at p. 443–444.) The insured argued that, similar to the non-cumulation provision relied on in Viking Pump, providing coverage for “death at any time” extends the policy protection beyond the policy period, which is inconsistent with the theoretical basis of pro rata allocation of long-tail claims. (Id. at p. 446.) The magistrate judge agreed, finding all sums allocation was the appropriate allocation method for both defense and indemnification costs. (Id. at p. 445.)

26 The district court held, however, that the policies required pro rata allocation of indemnity costs and all sums allocation of defense costs. (American Precision, supra, 648 F.Supp.3d at p. 445.) The court concluded the phrase “death at any time” in the definition of bodily injury did not compel all sums allocation. (American Precision, supra, 648 F.Supp.3d at pp. 446–447.) The district court noted that the language at issue was significantly different from the language of the non-cumulation provision relied on by the Viking Pump court, which clearly articulated that multiple successive insurance policies could indemnify the same loss or occurrence. (American Precision, supra, 648 F.Supp.3d at p. 447.) “First, the relevant language in this case does not explicitly refer to any other or additional policies. Second, even if this Court were to presume that the parties contemplated potential interaction with other policies during their negotiations of the definition of bodily injury in each of the Primary Policies, which it does not, the language at issue does not explain how any reductions of liability should occur, if at all, in the event of overlapping coverage by multiple policies.” (Ibid, fn. omitted.) The district court did not believe pro rata allocation of indemnity would render the “death at any time” language surplusage, although the court acknowledged that pro rata allocation could complicate the calculations. (American Precision, supra, 648 F.Supp.3d at pp. 447–448.) The court suggested a secondary level of pro rata allocation could be used to apportion liability for damages associated specifically with a death to each insurer on the risk during the pendency of the injury. (Id. at p. 448.)

27 The district court discussed the reasoning of a Massachusetts appellate court that “ ‘[t]he bodily injury definition simply sets forth the unremarkable proposition . . . that in the typical case where the time of injury is easily determined, the policy in place when the injury occurs will cover all consequential damages, even those taking place after the policy period.’ ” (American Precision, supra, 648 F.Supp.3d at p. 448.) As a result of the district court’s holding that pro rata allocation applied to indemnity, the court found the insured was liable for periods that it went without insurance coverage. (Ibid.) The district court found the policy language providing for the insurers to pay all costs and expenses arising from defending any claim seeking damages on account of a covered occurrence required an all sums allocation of defense costs. (American Precision, supra, 648 F.Supp.3d at p. 449.) The duty to defend is broader than the duty to pay. (Ibid.) When a claim triggers multiple policies, the court may decline to order pro rata allocation of defense costs, and the insurer may seek contribution from other applicable policies. (Ibid.) The insured was entitled to a full defense under the policy terms, while contribution flowed from equitable principles. (Ibid.) As a result, the district court granted the insurers’ motion for summary judgment in part and granted the insured’s cross- motion for summary judgment in part. (American Precision, supra, 648 F.Supp.3d at pp. 450–451.) On November 20, 2023, the district court amended its order to certify several questions for immediate interlocutory review by the Second Circuit, including (1) whether Viking Pump precludes pro rata allocation of indemnification where the insurance policy definition of bodily injury includes “death at any time” language, and (2) whether

28 insurers can be required to pay defense costs for long-tail claims on an all sums basis where the policy language limits indemnification to harms occurring during the policy period. (American Precision Industries, Inc. v. Federal Insurance Company (W.D.N.Y., Nov. 20, 2023, No. 14-CV-1050) 2023 WL 8014382, at *8.)

C. Analysis

The question before us in this appeal is whether the language of the policies at issue limits indemnification to losses and occurrences during the policy period—meaning that no two insurance policies, unless containing overlapping or concurrent policy periods, would indemnify the same loss or occurrence – or whether the policy agrees to provide coverage for losses extending beyond the policy period. (See Viking Pump, supra, 27 N.Y.3d at pp. 261–262.) We agree with the trial court that the Affiliated FM umbrella policy promised to indemnify losses extending beyond the policy period, and the 1975-1978 AmRe excess policy followed form. Affiliated FM agreed to pay the ultimate net loss for ITT’s liability for personal injury damages, “including damages for care and loss of services, because of personal injury, including death at any time resulting therefrom, sustained by any person or persons.” “Ultimate net loss” with respect to an occurrence was expressly defined to include “all sums which the insured . . . became legally obligated to pay as damages, whether by reason of adjudication or settlement, because of personal injury . . . to which this Policy applies[.]”

29 The insuring agreement expressly promises to pay “all sums” and extends protection beyond the policy period to include damages for future care and death at any time. Therefore, in the context of a continuous harm triggering multiple successive policies, it is clear that more than one policy may indemnify the same loss or occurrence. The insuring agreement does not contain language limiting coverage to damages during the policy period. Although coverage is limited to occurrences happening during the policy period, an occurrence is defined as “an accident, including continuous or repeated exposure to conditions, which results in personal injury or property damage neither expected nor intended from the standpoint of the insured.” Personal injury was defined to include “bodily injury, sickness, disease, disability, shock, mental anguish and mental injury[.]” There is no temporal limitation in the definition of an occurrence or personal injury which restricts coverage to damages during the policy period. In comparison, in Consolidated Edison, the policy at issue limited coverage to occurrences “ ‘happening during the policy period,’ ” but an occurrence was defined as “ ‘an event, or continuous or repeated exposure to conditions, which causes injury, damage or destruction during the policy period.” (Consolidated Edison, supra, 98 N.Y.2d at p. 222.) This policy language limiting coverage to damages during the policy period led the Consolidated Edison court to conclude that the parties did not intend to extend protection beyond the policy period. In this case, the provision limiting advertising injuries supports our conclusion that the Affiliated FM policy is an “all sums” policy providing coverage for liability extending beyond the policy period. When an occurrence begins prior to the policy

30 period and continues during or after that period, the policy expressly limits coverage for advertising injury to a proportion of the total damages resulting from the occurrence. The policy contains no similar limitation, however, on coverage for personal injury liability. The terms of the policy, read in context, make clear that the parties intended to extend coverage beyond the policy period and not limit it to damages occurring during the policy period. The 1975-1978 AmRe excess policy follows the insuring agreement and conditions of the underlying Affiliated FM policy and the successor umbrella policy. Nothing in the 1975-1978 AmRe policy conflicts with the language of the Affiliated FM policy. Therefore, the trial court properly found that the 1975- 1978 AmRe policy required the insurer to indemnify ITT for personal injury continuing after the termination of the policy, and therefore, was also subject to all sums allocation. Having concluded that the 1975-1978 AmRe excess policy required all sums allocation based on the language of the policy provisions, including those of the Affiliated FM umbrella policy, we need not address the effect of substituting a different umbrella policy with a non-cumulation provision that was also subject to all sums allocation.

Impact of Non-Cumulation Provision on Aggregate Limit

Munich joined in the contention of another insurer, who has since settled and dismissed its appeal, that the trial court did not properly apply the policies’ prior insurance and non- cumulation conditions. Specifically, Munich contends that under the non-cumulation conditions contained in American

31 Reinsurance Company Policy No. M-0083871, effective from July 1, 1967 to July 1, 1970 (the 1967-1970 AmRe excess policy), and American Excess Insurance Company Policy No. EUL 5001441, effective from December 31, 1978 to December 31, 1979 (the American Excess policy), the annual aggregate policy limit is reduced when a loss covered under Munich’s policy is also covered under a prior policy. Munich contends the term “loss” is used broadly, referring to the insured’s total aggregate losses in a policy period, not simply the loss from a single occurrence. Munich also contends the condition applies to any loss for which coverage arguably existed under a prior policy, regardless of whether the insured received payment. Munich similarly contends that condition applies to any loss that is arguably covered under the subsequent policy, even if the insured has not sought reimbursement for that loss under Munich’s policy. Even assuming that Munich did not waive this issue and has identified the correct policy language,1 we conclude the policy language is not reasonably susceptible of Munich’s interpretation.

1 Munich contends the trial court’s finding that the 1967- 1970 AmRe excess policy follows form to Home Insurance Co. policy number HEC 9555421 is incorrect, because the policy language of the 1967-1970 AmRe excess policy expressly follows form to the immediately underlying excess policy, which is Lamorak Policy No. E16-8324-001. The parties agree, however, that both underlying policies contain substantially similar non- cumulation conditions and the analysis is the same under either policy.

32 A. Relevant Policy Language

1. 1967-1970 AmRe Excess Policy

The 1967-1970 AmRe excess policy states that there are two underlying insurance policies: Home Insurance Company Policy No. HEC 9555421 (the Home policy) and a policy issued by Employer’s Liability Assurance Corporation, which was succeeded to by Lamorak Insurance Company (the Lamorak policy). The declarations of the 1967-1970 AmRe excess policy provide that the coverage limit is $5 million. The insuring agreement of the 1967-1970 AmRe policy states that the insurer will indemnify the insured against “ultimate net loss in excess of and arising out of the hazards covered” as defined in the underlying insurance. In addition, it states that the coverage provided shall follow the insuring agreements, conditions, and exclusions of the underlying insurance immediately preceding the layer of coverage.2

2 The policy states in relevant part: “1. The Company hereby indemnifies the Insured against ultimate net loss in excess of and arising out of the hazards covered and as defined and in excess of the underlying Insurance as shown in Item 4 of the Declarations (hereinafter referred to as ‘underlying insurance’) but only up to an amount not exceeding the limit(s) shown in Item 5 of the Declarations. [¶] 2. Except as may be inconsistent with this Certificate, the coverage provided by this Certificate shall follow the insuring agreements, conditions and exclusions of the underlying insurance (whether primary or excess) immediately preceding the layer of coverage provided by this Certificate . . . .”

33 Under the heading “loss payable,” the policy provides in pertinent part: “The Company’s obligation to pay any ultimate net loss and costs with respect to any accident or occurrence falling within the terms of this Certificate shall not attach until the amount of the applicable underlying limit has been paid by or on behalf of the Insured on account of such accident or occurrence.”3

2. The Lamorak Policy

Munich contends the Lamorak policy is the immediately preceding layer of coverage to which the 1967-1970 AmRe excess policy follows form. The insuring agreement of the Lamorak policy promises to indemnify the insured for “all sums which the Insured shall be obligated to pay . . . [¶] . . . [¶] caused by or arising out of each occurrence . . . .” The Lamorak policy contains a prior insurance and non- cumulation of liability condition: “It is agreed that if any loss covered hereunder is also covered in whole or in part under any other excess policy issued to the Insured prior to the inception date hereof the limits of liability hereon as stated in Items 5 and 6 of the Declarations shall be reduced by any amounts due to the Insured on account of such loss under such prior insurance.”

3 Ultimate net loss is defined as “the sums paid in settlement of losses for which the Insured is liable after making deductions for all recoveries, salvages and other insurances (other than recoveries under the underlying insurance policies of co-insurance, or policies specifically in excess hereof), whether recoverable or not, and shall exclude all ’Costs[.]’ ”

34 “Subject to the foregoing paragraph and to all the other terms and conditions of this policy in the event that personal injury or property damage arising out of an occurrence covered hereunder is continuing at the time of termination of this policy the Company will continue to protect the Insured for liability in respect of such personal injury or property damage without payment of additional premium.” The policy additionally stated that ultimate net loss meant “the sums paid in settlement of losses for which the Insured is liable after making deductions for all recoveries, salvages and other insurance (other than recoveries under the Underlying Insurance or policies specifically in excess hereof), and shall exclude all ‘Costs’[.]” The policy stated that it was subject to the same warranties, terms and conditions, except as otherwise provided, as were contained in the underlying insurance.

3. American Excess policy

The American Excess policy provides coverage for the insured’s “ultimate net loss” in excess of the total limits of all underlying insurance and arising out of the hazards covered in the underlying insurance up to the policy limits.4 Ultimate net loss is defined as “the sums paid in settlement of losses for which

4 The relevant provision provides coverage for “ultimate net loss which is in excess of the total limits of all underlying insurance . . . and which arises out of the hazards covered and defined in the underlying insurance as shown in Item 5 of the Declarations, but only up to an amount not exceeding the limit(s) shown in Item 6 of the Declarations.”

35 the Insured is liable after making deductions for all recoveries, salvages and other insurances (other than recoveries under the underlying insurance, policies of co-insurance, or policies specifically in excess hereof), whether recoverable or not, and shall exclude all ‘Costs’.” The American Excess policy contains its own prior insurance and non-cumulation of liability condition, which states: “It is agreed that if any loss covered hereunder is also covered in whole or in part under any other excess policy or certificate issued to the Insured prior to the inception date hereof, the limit of liability hereon as stated in Item 6 of the Declarations of this Certificate shall be reduced by any amounts due the Insured on account of such loss under such prior insurance.”

B. Law Applicable to Non-Cumulation Conditions

1. General Principles

“Generally, non-cumulation clauses prevent stacking, the situation in which ‘an insured who has suffered a long term or continuous loss which has triggered coverage across more than one policy period . . . wishes to add together the maximum limits of all consecutive policies that have been in place during the period of the loss’ (12 Couch on Insurance 3d § 169:5; see 1 Barry R. Ostrager & Thomas R. Newman, Handbook on Insurance Coverage Disputes § 11.02 [e] [16th ed 2013]). Such clauses originated during the shift from ‘accident-based’ to ‘occurrence- based’ liability policies in the 1960s and 1970s, and were purportedly designed to prevent any attempt by policyholders to recover under a subsequent policy—based on the broader

36 definition of occurrence—for a loss that had already been covered by the prior ‘accident-based’ policy (see Jan M. Michaels et al., The “ ‘Non-Cumulation’ ” Clause: Policyholders Cannot Have Their Cake and Eat It Too, 61 U Kan L Rev 701, 717 [2013]; Christopher C. French, The “ ‘Non-Cumulation Clause’ ”: An “ ‘Other Insurance’ ” Clause by Another Name, 60 U Kan L Rev 375, 386 [2011]).” (Viking Pump, supra, 27 N.Y.3d at p. 259.)

2. Hiraldo

In Hiraldo v. Allstate Ins. Co. (2005) 5 N.Y.3d 508, 511 (Hiraldo), the New York Court of Appeals found the non- cumulation condition limited the insurer’s liability for a single loss occurring over multiple policy periods by preventing the policyholder from obtaining the per occurrence limit from successive policies. The insurer in Hiraldo had issued three successive liability policies to owners of a building, each providing $300,000 of coverage. A tenant who was continuously exposed to lead paint during all three policy periods obtained a judgment against the landlords for $700,000. (Id. at pp. 511–512) The insurer paid $300,000, and the landlords brought an action for the remainder. The policies at issue stated that they applied only to losses that occurred during the policy period, and each contained a non- cumulation condition stating: “ ‘Regardless of the number of insured persons, injured persons, claims, claimants or policies involved, our total liability under Business Liability Protection coverage for damages resulting from one loss will not exceed the limit of liability for Coverage X shown on the declarations page. All bodily injury, personal injury and property damage resulting

37 from one accident or from continuous or repeated exposure to the same general conditions is considered the result of one loss.’ (Emphasis added.)” (Hiraldo, supra, 5 N.Y.3d at pp. 511–512.) The Hiraldo court noted that the plaintiff’s injuries from continuous exposure to the same general conditions was considered “one loss” within the meaning of each policy. (Hiraldo, supra, 5 N.Y.3d at p. 512.) The court concluded that under the plain language of the non-cumulation condition, regardless of the number of policies involved, the insurer’s total liability resulting from one loss could not exceed the $300,000 liability limit. (Id. at p. 513.)

3. Nesmith

In Nesmith v. Allstate Ins. Co. (2014) 24 N.Y.3d 520, 525 (Nesmith), the New York Court of Appeals found a non- cumulation provision prevented the policyholder from recovering the per occurrence limit from multiple successive policies, even though multiple claims were involved, because the claims resulted from the same general conditions and were therefore considered one loss under the terms of the policy. The liability policy at issue in Nesmith was issued to the landlord of a house painted with lead paint. (Nesmith, supra, 24 N.Y.3d at p. 523.) The policy provided a $500,000 limit for each occurrence and contained a non-cumulation condition stating: “ ‘Regardless of the number of insured persons, injured persons, claims, claimants or policies involved, our total liability under the Family Liability Protection coverage for damages resulting from one accidental loss will not exceed the limit shown on the declarations page. All bodily injury and property damage

38 resulting from one accidental loss or from continuous or repeated exposure to the same general conditions is considered the result of one accidental loss’ (emphasis omitted).” (Id. at pp. 523–524.) The Young family lived in an apartment in the house from November 1992 to September 1993, and the Patterson family lived in the apartment beginning September 1993. Two separate actions were brought on behalf of the children for personal injuries caused by lead paint exposure. (Nesmith, supra, 24 N.Y.3d at p. 524.) The Young action was settled in 2006 for $350,000, which the insurer paid. (Ibid.) The insurer paid $150,000 on behalf of the Patterson children as the remaining coverage on the policy. (Ibid.) Nesmith, who brought the action on behalf of the Patterson children, brought a declaratory relief action asserting that a separate $500,000 limit applied to each family’s claim, such that the Patterson children could recover an additional $350,000. (Ibid.) The Nesmith court concluded that the renewal of the policy did not make an additional limit available, and under the plain terms of the non-cumulation clause, the number of claims and claimants did not make an additional limit available. (Nesmith, supra, 24 N.Y.3d at p. 525.) Because all the children were injured by exposure to the same general condition, their injuries were considered one accidental loss under the policy definition, and as a result, only one policy limit was available to the two families. (Ibid.)

4. Olin

In Olin Corp. v. OneBeacon Am. Ins. Co. (2d Cir. 2017) 864 F.3d 130, 148 (Olin), the Second Circuit found the non-

39 cumulation condition prevented the policyholder from avoiding the per occurrence limit by making claims under multiple successive policies for a loss from the same occurrence, even when the policies were issued by different insurers. In that case, the plaintiff Olin brought a coverage action seeking indemnification for environmental contamination at multiple manufacturing locations. The defendant insurer OneBeacon had issued three excess policies to plaintiff for the same policy period. (Id. at p. 137.) The policies provided coverage for “ ‘all sums which the Insured shall be obligated to pay by reason of the liability . . . imposed upon the Insured by law . . . for damages, direct or consequential and expenses . . . on account of . . . Property Damage . . . caused by or arising out of each occurrence. . . .’ ” (Olin, supra, 864 F.3d at p. 137.) An occurrence was defined as “ ‘an accident or a happening or event or a continuous or repeated exposure to conditions which unexpectedly and unintentionally result[s] in . . . property damage . . . during the policy period.’ ” (Olin, supra, 864 F.3d at p. 137.) Each excess policy also contained a “prior insurance” provision and a “continuing coverage” clause in “Condition C” of the policy, stating: “It is agreed that if any loss covered hereunder is also covered in whole or in part under any other excess policy issued to the Insured prior to the inception date hereof, the limit of liability hereon . . . shall be reduced by any amounts due to the Insured on account of such loss under such prior insurance. [¶] Subject to the foregoing paragraph and to all other terms and conditions of this Policy in the event that personal injury or property damage arising out of an occurrence covered hereunder is continuing at the time of termination of this

40 Policy, [the insurer] will continue to protect the Insured for liability in respect of such personal injury or property damage without payment of additional premium.” (Olin, supra, 864 F.3d at pp. 137–138.) “The first paragraph constitutes the prior insurance provision and the second is the continuing coverage clause.” (Id. at p. 138.) Each excess policy was preceded in time by prior insurance in the same layer of coverage (the prior excess policies) providing substantially the same coverage for the same manufacturing sites for all sums the insured became legally obligated to pay for property damage during the policy period caused by an occurrence. (Olin, supra, 864 F.3d at p. 138.) The Olin court concluded the language of the “prior insurance” provision applied to any other excess policy within the same layer of coverage and was not limited to prior policies issued by the same insurer. (Olin, supra, 864 F.3d at p. 148.) “Rather, the general language of the prior insurance provision suggests that the clause is designed to apply whenever both earlier and later polices cover the same loss, just as the focus of noncumulation clauses is whether more than one policy provides coverage for identical loss within the same layer, unaffected by the identity of the insurer.” (Id. at p. 148.) “[T]he critical factor is whether the loss covered by a policy dictating all sums is also covered by another policy in the same coverage layer, which itself has already provided indemnification to the insured for the loss.” (Id. at p. 149.) “As Olin argues, application of the all sums method means that it may attribute the full amount of its loss to a single policy year and demand coverage from a single insurer up to the insurer’s policy limits (OneBeacon in this case). Yet, this same

41 principle also limits Olin’s ability to tap multiple insurers for the same loss. Just as the all sums allocation method allows Olin to seek recovery from any insurer of its choosing up to the limits of the relevant policy, by the same token, it also requires reducing the limits of liability on the OneBeacon policy at issue by amounts paid under any prior insurance policy at the same level of coverage that did, in fact, provide coverage to Olin for the same loss. See Stonewall Ins. Co. v. E.I. du Pont de Nemours & Co., 996 A.2d 1254, 1260 (Del. 2010) (examining substantially similar language in light of the all sums approach and concluding that ‘interpreting the noncumulation clause to limit how much [an insured] may seek from the selected tower of insurance by subtracting any amounts received by or payable to [the insured] from prior excess insurers [ ] is the only proper interpretation’). To conclude otherwise would be to strip the prior insurance provision of its bargained-for effect, as evinced by its plain language, and permit Olin to recover multiple times for a single loss by pursuing multiple insurers within the same layer of coverage. [Citation.]” (Olin, supra, 864 F.3d at pp. 149–150.) The Olin court rejected the insurer’s contention, however, that because the plaintiff “could recover from prior insurers whose policies provide coverage for loss at these sites and who sit in the same layer of coverage as OneBeacon, Olin may not recover under the OneBeacon policies.” (Olin, supra, 864 F.3d at p. 150.) The court concluded OneBeacon was misreading Condition C. “As explained earlier, Condition C permits an insured to pursue indemnification from any insurer whose policy was triggered (under the framework described above) as a result of continuing property damage. The prior insurance provision works in conjunction with the overarching approach dictated by

42 Condition C to prevent the insured from stacking policies once it has already obtained indemnification for that specific loss from another policy in the relevant coverage layer.” (Id. at p. 150.) “Thus, under Condition C, the insurer offering coverage for the selected policy year is generally required to demonstrate the existence of valid claims against other available policies and to pursue claims under them. [¶] The prior insurance provision, however, also offers some contractual protection for the insurer. This provision allows the insurer to offset its indemnification obligations by amounts already paid to cover the loss by another insurer in the same coverage tier.” (Olin, supra, 864 F.3d at p. 151.)

5. Hopeman

Munich relies on Hopeman Brothers, Inc. v. Continental Casualty Company (E.D. Va. 2018) 307 F.Supp.3d 433, 441 (Hopeman), a federal district court case applying New York law. The plaintiff Hopeman purchased multi-layered insurance coverage from 1971 to 1977. (Id. at p. 439.) Hopeman was subject to multiple personal injury claims due to asbestos exposure from its products, and received payments or otherwise resolved coverage as to all but the defendant insurers. (Id. at pp. 438–439.) The Hopeman court concluded that each asbestos-related claim was a separate “occurrence.” (Hopeman, supra, 307 F.Supp.3d at p. 449.) Based on the decisions in Viking Pump and Olin, the court found the non-cumulation conditions applied only to actual payments under prior policies in the same coverage tier. (Id. at p. 452.) The non-cumulation conditions did not apply

43 simply because the policyholder could theoretically recover a loss under a prior insurance policy in the same coverage tier, if the policyholder had not yet received actual payment for that specific loss. (Id. at pp. 452–453.) In addition, the Hopeman court concluded the insured was not required to group any claimants together in that case. (Hopeman, supra, 307 F.Supp.3d at p. 454.) Although the policy provided that all personal injury arising out of exposure to the same general conditions was the same occurrence,5 the court concluded that “because the injuries in the instant case involved multiple claimants at multiple locations over multiple years, each featuring varied and unique exposure patterns, that these claims constitute separate occurrences under the ‘unfortunate event’ test, and therefore should not be aggregated.” (Ibid.) The Hopeman court found the non-cumulation condition in the policy at issue was limited to recoveries involving the same occurrence, and each individual who alleged bodily injury from exposure to asbestos was a “separate occurrence” for purposes of the non- cumulation condition. (Ibid.) The court also found that the non-cumulation condition did not require the plaintiff to apply losses to the earliest triggered policy year or to exhaust policies in chronological order. (Hopeman, supra, 307 F.Supp.3d at p. 454.) Under Viking Pump and Olin, the policyholder was allowed to control the order in

5 The relevant policy language stated: “ ‘[f]or the purpose of determining the limits of the company’s liability: (1) all personal injury . . . arising out of continuous or repeated exposure to substantially the same general conditions . . . shall be considered as the result of one and the same occurrence.’ ” (Hopeman, supra, 307 F.Supp.3d at p. 454, emphasis omitted.)

44 which to seek payment from triggered policies under the all sums method. (Id. at p. 456.) The parties disputed whether the term “loss” in the non- cumulation condition of a different policy should be construed narrowly or broadly. (Hopeman, supra, 307 F.Supp.3d at pp.457– 458.) The Hopeman court concluded that the plain meaning of the term “loss” should be construed broadly enough “to mean the gross amount that a policyholder is seeking in its claim under a policy.” (Id. at p. 458.) Therefore, the court found the term “loss” in the non-cumulation conditions unambiguously referred to “the gross amount Hopeman is seeking under each policy.” (Id. at p. 459.) We note that the Hopeman court’s interpretation of the term “loss” in the non-cumulation condition at issue is not binding on this court and was rejected by the New York appellate court in Carrier Corporation v. Allstate Insurance Company (N.Y. App. Div. 2020) 187 A.D.3d 1616, 1622 [“loss” is defined narrowly in subject prior insurance and non-cumulation provisions].) In addition, we note that the Hopeman court was not asked and did not find that the term “loss” is not related to “an occurrence.”

C. Analysis

It is clear from the policy language and the case law that the prior insurance and non-cumulation condition applies when an insured seeks reimbursement for a loss resulting from the same occurrence under multiple successive policies. The condition applies when “any loss covered hereunder is also covered in whole or in part under any other excess policy issued to the Assured prior to the inception date hereof.” To be covered,

45 a loss must result from an occurrence, and for the loss to be covered under multiple successive policies, thereby implicating the non-cumulation condition, it must result from the same occurrence. In each case discussed above, the court considered whether a loss paid under a prior policy reduced the policy limits of a subsequent policy when the policyholder sought payment for loss from the same occurrence. As stated above, courts have found that before the non- cumulation provision applies, the policyholder must have actually recovered amounts due under the prior policy, not merely theoretically or arguably covered under the prior policy. Similarly, the non-cumulation condition does not apply until the policyholder seeks reimbursement under the subsequent policy for a loss from the same occurrence that was paid under the prior policy. In this case, it is undisputed that ITT has not sought reimbursement from Munich for any loss resulting from an occurrence that was paid under a prior policy.

Settlement Proceeds

Munich also purported to join in a contention that ITT should have allocated the Resolute settlement proceeds to specific asbestos claims before seeking reimbursement from particular policies. However, Munich declined to join in this issue in the trial court or assert its own “offset” argument, and therefore, may not raise the issue for the first time on appeal. (In re Aaron B. (1996) 46 Cal.App.4th 843, 846 [party is precluded from raising on appeal any point not raised in the trial court].) Munich has not challenged the trial court’s ruling on Munich’s equitable claims.

46

DISPOSITION

The judgment is affirmed. Respondent ITT LLC is awarded its costs on appeal.

NOT TO BE PUBLISHED.

MOOR, Acting P. J.

We concur:

KIM, J.

LEE, J.*

* Judge of the Superior Court of San Bernardino County, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution.

47