concurring in part and dissenting in part.
I agree with the majority that the $40 million first excess policies should not be stacked, but not because of some purported trend in the Illinois case law. There is ample support on other, more compelling, grounds for declining to stack two policies that appear separate in form, but are one and the same in substance. As the majority points out, the premium for the $40 million policy in force before the sale of Great Lakes by Itel to Blackstone did not change after the sale had taken place. If it were, the intention of the parties that coverage be increased to $80 million as a result of the transaction, there would have been no way to escape the additional premium that would go with the additional coverage. In my view, these circumstances provide good and sufficient support for an anti-stacking result on the specific facts of the duplicate policies. But by attempting to support this proposition with Illinois case law purportedly rejecting stacking in general, the majority heads down the wrong path, and calls into question the Illinois Supreme Court’s authoritative pronouncement on the subject.
There is agreement here that Illinois law governs the issues that are before us. And on the question of stacking, Illinois law has been authoritatively announced in Zurich Insurance Co. v. Raymark Industries, Inc., 118 Ill.2d 23, 112 Ill.Dec. 684, 514 N.E.2d 150 (1987). The majority inappropriately seeks to escape the implications of that opinion. Essentially, the majority tries to limit Zurich by seeing it as tied to the particular circumstances and policy language of that case, rather than as standing for a general principle. I do not agree with that interpretation. Zurich expressly rejected the premise underlying the pro rata (“time on the risk”) approach outlined in Insurance Co. of N. Am. v. Forty-Eight Insulations, Inc., 633 F.2d 1212 (6th Cir.1980), aff'd. on rehearing, 657 F.2d 814 (1981), cert. denied, 454 U.S. 1109, 102 S.Ct. 686, 70 L.Ed.2d 650 (1981). There, the Sixth Circuit held, in an asbestos case, that the insurance policies were triggered only by claimants’ exposure to *799asbestos and that thus there was a reasonable means of allocating liability among the triggered policies — based on the number of years of exposure. Rejecting this approach, the court in Zurich noted that the trial court had “found nothing in the policy language that permits proration.” 118 Ill.2d at 57, 112 Ill.Dec. 684, 514 N.E.2d at 165. Rather than indicating a specific reliance on the policy language to bind it to its conclusion that “stacking” was appropriate, the Zurich court seems to indicate that — absent policy language to the contrary — joint and several liability is the rule in Illinois.
The majority looks to other cases (from intermediate Illinois appellate courts) that do not require stacking and concludes that the Supreme Court of Illinois would probably adopt anti-stacking as a general rule if now presented with that question. This is particularly likely, the majority opinion states, because other jurisdictions have done so in well-reasoned opinions (New York, Colorado, Minnesota, New Jersey and Utah). The majority adds that even if we were to apply maritime law, the reasoning of Olin Corp. v. Insurance Co. of N. Am., 221 F.3d 307 (2d Cir.2000) and Sybron Transition Corp. v. Security Ins., 2001 WL 788624 (7th Cir. July 12, 2001), support the anti-stacking view. But we are not applying maritime law or the law of any other state; we are applying Illinois law. And the relevant intermediate appellate courts in Illinois have distinguished Zurich on the grounds that the cases before these lower courts, unlike Zurich, involved what have been characterized as single continuous occurrences that implicated successive policy periods and that in such a situation a pro rata, time-on-the-risk allocation is appropriate. See Missouri Pacific R.R. Co. v. Int’l Ins. Co., 288 Ill.App.3d 69, 79-80, 223 Ill.Dec. 350, 679 N.E.2d 801, 808 (1997); Outboard Marine Corp. v. Liberty Mut. Ins. Co., 283 Ill. App.3d 630, 642-45, 219 Ill.Dec. 62, 670 N.E.2d 740, 748-50. I do not believe that these intermediate appellate courts have gone so far as to reject joint and several liability (stacking) when the policy language and the specific facts do not preclude it.
As the district court noted, Outboard Marine and Missouri Pacific were “single continuous occurrence” cases. Thus:
In these cases, the facts, as viewed by the appellate court, involved damage continuously caused and continuously sustained. The cause of the damage and the damage caused were essentially contemporaneous, with the causative agent and the resulting damage occurring repeatedly and continuously. In such cases, because both the damage-causing agency and the damages the agency caused occurred continuously in each policy period, a rule which required the policy in effect when the first damage occurred to cover damages caused in that and successive policy periods would make no sense. The sensible rule, as the appellate court held, is to attempt to make each policy respond to the damage that occurred during its policy period.
Findings of Fact, Conclusions of Law and Order at 39. The language of these cases distinguishes them from the “triple trigger” approach of Zurich, which found that the policies were triggered at the time of the original exposure to asbestos, again at the time asbestosis appeared and also at times in between when the claimant manifested illness. See Missouri Pacific, 288 Ill.App.3d at 79, 223 Ill.Dec. 350, 679 N.E.2d at 807-08 (citing Zurich, 118 Ill.2d at 44, 112 Ill.Dec. 684, 514 N.E.2d at 165); Outboard Marine, 283 Ill.App.3d at 641, 219 Ill.Dec. 62, 670 N.E.2d at 748. The case before us involves similar progressive damage, with at least one trigger (the 1991 tunnel damage) that, like exposure to as*800bestos, results in some damage immediately to the tunnel and later damage to the flooded premises. On the other hand, Missouri Pacific and Outboard Marine do not concern a single trigger to which the later damage may be attributed. And, since the initial damage here may cause progressively increasing injury over time, “it makes sense to hold the policy in effect at the time of the initial injury responsible for all the claimant’s damages, even though the progression of the damage could over time trigger additional policies.” Findings of Fact, Conclusions of Law and Order at 39.
On the question whether the $60 million first-period second excess policy is available to cover the flood damage, the proper interpretation of the policy language, as well as the impact of Zurich, leads me to a different conclusion than that reached by the majority. The majority believes that the $60 million first-period policy was not triggered because an insurance policy is not triggered unless loss to the claimant occurred while the policy was in force. This is contrary to the most reasonable reading of the policy language. The majority opinion tells us that the policy defines “property damage” as an “accident or happening or event [that] ... results in a ... physical injury to ... tangible property ... during the policy period” (emphasis added). That is not exactly the case. The policy actually defines “property damage” as:
(a) physical injury to or destruction of tangible property, including the loss of use thereof at any time resulting therefrom; and/or
(b) loss of use of tangible property which has not been physically injured or destroyed; and/or
(c) evacuation losses arising from actual or threatened physical injury to or destruction of tangible property or bodily injury.
The temporal limitation appears, not in the definition of “property damage,” but in the definition of “occurrence,” which is “an accident or happening or event or a continuous or repeated exposure to conditions which unintentionally results in ... Property Damage ... during the policy period. All such exposure to substantially the same general conditions existing at or emanating from one premises or location shall be deemed one Occurrence.” (Emphasis added.) This temporal limitation appears only under the definition of “occurrence.” But the policy provides coverage for “damages on account of ... Property damage ... caused by or arising out of each occurrence happening anywhere in the world.” (Emphasis added.) Thus, any damage “caused by or arising out of’ any occurrence will be covered, regardless whether the damage occurred during the policy period. The policy language clearly supports the position of Great Lakes.
The majority then moves on from its creative parsing of policy language to rummage for Illinois case law to support its position. Again, the opinion cites cases from lower Illinois courts for a proposition that is at odds with the Illinois Supreme Court’s holding in Zurich. For example, the majority relies on Pekin Insurance Co. v. Janes & Addems Chevrolet, Inc., 263 Ill.App.3d 399, 200 Ill.Dec. 843, 636 N.E.2d 34 (4th Dist.1994), in which the Illinois appellate court clearly distinguished Zurich:
Plaintiffs argue under Zurich coverage under a general liability policy is triggered, not when the wrongful conduct takes place, but when the complained-of damage occurs. This is contrasted with the protection provided by an “occurrence” or “acts and omissions” policy, which provides coverage for the negligent acts or omissions which occur dur*801ing the policy period, regardless of when the injury occurs or the claim is made. In this case, the insured was covered under a general liability policy and coverage is triggered when the injury occurs.
263 Ill.App.3d at 404, 200 Ill.Dec. 843, 636 N.E.2d at 38 (citations omitted). The district court noted that this language indicated a possible misunderstanding by the Pekin court about the nature of occurrence policies. Whether the Pekin court erred or not, the district court concluded — and I agree — that the policies interpreted in Pe-kin are significantly different in language from the policies before us.
The majority has no good reason to depart from the Illinois Supreme Court’s pronouncement in Zurich, which indicated that events characterized by a long delay between the wrongful act and the manifestation of harm may be covered both by policies in force at the time of the wrongful acts and those in force when the harm becomes apparent. The fact that Travelers Insurance Co. v. Eljer Manufacturing, Inc., 2000 WL 1763322 (Ill. Dec. 1, 2000) is being reheard certainly does not change Zurich’s standing as authoritative Illinois law. In Eljer, the court found that installation of a potentially defective product into a home constituted injury to tangible property within the meaning of the insurance policy. See id. at *2 (interpreting New York law). This conclusion in Eljer is not at odds with Zurich; it does not even address Illinois law on a subject relevant here.1
According to the majority, we need not wait for Illinois to resolve the question whether a policy in force at the time of the wrongful act covers liability for injuries that are later manifested. For the majority argues that Zurich and Eljer are distinguishable on the ground that they involve damage to a single property (or person) over multiple periods. Here, by contrast, the City suffered injury in the first period, and the businesses in the Loop (as well as the City) were injured in the second period. Thus, the majority argues that there was no trigger for the property damage from the flood pertaining to the first-period policies, because no “occurrence” relating to flood damage happened with respect to any policies until the second period. To distinguish the controlling Illinois authority on the ground that it involved damage to the same property (or to the same person) over multiple periods is not persuasive. First, this distinction is not one indicated or relied on in the language of Illinois cases, nor is it otherwise apparent that the distinction has significance. Second, to view the 1991 tunnel damage and the 1992 flood damage as unrelated denies reality: there was a progressive series of consequences flowing from the 1991 damage, culminating in the flood of 1992. How quickly or sluggishly the consequences were revealed should not be material. As in Zurich, this case involves the manifestation of damage (asbestosis or a flood) resulting from a single occurrence (inhalation of asbestos fibers or cracking of a tunnel wall).
The majority cites Illinois cases for the proposition that an occurrence policy is not triggered unless loss to the claimant occurred while the policy was in force, but those cases are inapposite. See Pekin, 263 Ill.App.3d at 404, 200 Ill.Dec. at 847, 636 N.E.2d at 38 (and discussion of the case, *802supra); Seegers Grain Co. v. Kansas City Millwright Co., 230 Ill.App.3d 565, 566-67, 172 Ill.Dec. 50, 595 N.E.2d 113, 114 (1992) (“The property damage covered under completed operations coverage was expressly defined in the policy to include only property damage which ‘occurs during the policy period.’ ”); Great American Ins. Co. v. Tinley Park Recreation Comm’n, 124 Ill.App.2d 19, 21-23, 259 N.E.2d 867, 868-69 (1st Dist.1970). Great American turns on the definition of the term “accident,” which constituted the only relevant occurrence, and the court concluded that an “accident” had not occurred until the injury was manifest. 124 Ill.App.2d at 21-23, 259 N.E.2d at 868-69. Further, Seegers and Great American are pre-Zurich, policy language-specific, lower Illinois court cases, and thus cannot defeat Zurich’s, clear mandate.
In the case before us, there is no mention of a “to the claimant” limitation on coverage in the policies and no Illinois law to support such a requirement. To read the policies as not requiring damage to the ultimate claimant during the policy period would not only be reasonable, but it would also be the most plausible reading of the policy language. Coverage for property damage “caused by or arising out of’ an occurrence supports an unrestricted reading. A similar “to the claimant” argument was made in Travelers Insurance Co. v. Penda Corp., 974 F.2d 823 (7th Cir.1992), in which the insurer argued that coverage was not available because the underlying claimant did not actually own the damaged property — even though the policy provided coverage for liability incurred “because of property damage.” 974 F.2d at 830. This court observed that “Illinois cases do not consider who owned the property in question when determining if a claim is within policy coverage.” Id.
These points demonstrate the irrelevance of the fact that different property belonging to different claimants was damaged in 1992 than was the case in 1991. At the very least, these arguments demonstrating irrelevance show that such a reading of the policies at hand is reasonable. And where there is reasonable disagreement, we construe the policy against the insurer. “If the language of a policy is ambiguous or otherwise susceptible to more than one reasonable interpretation, it will be construed in favor of the insured.” Int’l Minerals & Chem. Corp. v. Liberty Mut. Ins. Co., 168 Ill.App.3d 361, 370, 119 Ill.Dec. 96, 522 N.E.2d 758, 764 (1988); see also Allen v. Transamerica Ins. Co., 115 F.3d 1305, 1309 (7th Cir.1997); Travelers, 974 F.2d at 828.
There is good reason for this policy— especially here, where cause and effect are clear. Without this approach, insurers could completely escape liability for damage caused by an event — an occurrence— that happened “on their watch.” This would be an indefensible result, given that the flood damage resulted directly from an occurrence that happened while the questioned policy was in force. It is mere fortuity that the tunnel wall took a few months to collapse. No one would question London Insurers’ liability if the tunnel had promptly given way to the blow of the pile driver.
The majority goes on to bolster its conclusion by pointing out that here the flood loss was preventable (unlike-asbestosis) because inspection and repair of the tunnel could arguably have prevented the flood. Thus the majority notes that in February 1992 (when the second-period policies were in effect) inspectors saw cracks in the tunnel but failed to prevent the roofs collapse. I do not understand the relevance of this circumstance. Failure to repair has never been argued as an independent intervening *803cause of the collapse. The cause continues to be an occurrence resulting in damage to property that took place during the first period. That some inspectors saw a crack is certainly not an adequate reason to absolve the insurers of all liability.
Therefore, on the issue of the availability of the $60 million first-period first excess policy to respond to the flood damage, I must reject the outcome reached by the majority. The policy language and the controlling Illinois law point to a different result. I respectfully dissent.
. The court did address Illinois law in Eljer, but not in any way relevant to this case. It held that coverage for potentially defective plumbing systems was not triggered for systems that did not leak, i.e., systems that did not manifest any defect. See 2000 WL 1763322, at *2. The court did not address the question whether, for plumbing systems that proved faulty, installation triggered coverage under Illinois law.