¶24 (dissenting) — Despite acknowledging the proper interpretation of “ensuing loss” set forth in our opinion in the companion case of Vision One, LLC v. Philadelphia Indemnity Insurance Co., 174 Wn.2d 501, 276 P.3d 300 (2012), the majority concludes there is no coverage for Max and Krista Sprague’s collapsed deck. I disagree. This case is in all material respects the same as Vision One, and the distinctions the majority would draw rest upon a faulty application of the ensuing loss clause. I would affirm the Court of Appeals.
¶25 First, it must be remembered the Spragues had an all-risks policy with Safeco Insurance Company. As we *533explained in Vision One, such policies “ ‘provide coverage for all risks unless the specific risk is excluded.’ ” Vision One, 174 Wn.2d at 513 (quoting Steven Plitt, Daniel Maldonado & Joshua D. Rogers, Introductory Concepts of the Risk; Public Policy Insurability, and Causation, in 7 Couch on Insurance 3d § 101:7, at 101-17 (2006)); see also Findlay v. United Pac. Ins. Co., 129 Wn.2d 368, 378, 917 P.2d 116 (1996) (noting that in an all-risk policy, “any peril that is not specifically excluded in the policy is an insured peril”). During the relevant time frame, the Spragues’ policy contained no exclusion for the collapse of a structure.3
¶26 The policy listed several excluded perils under a section entitled “Building Losses We Do Not Cover.” Clerk’s Papers (CP) at 51. Relevant to this case, the policy excluded losses “caused directly or indirectly” by “deterioration,” “mold,” and “wet or dry rot.” CP at 51, 74-75. This exclusion was subject to a clause providing that “any ensuing loss not excluded is covered.” CP at 51. The policy also excluded losses “caused directly or indirectly” by “faulty, inadequate [,] or defective . . . workmanship [or] construction.” CP at 51-52, 74, 76. Just as did the exclusion for deterioration, mold, and rot, the faulty workmanship and construction exclusion provided that “any ensuing loss not excluded or excepted in this policy is covered.” CP at 52.
¶27 The Court of Appeals concluded that the ensuing loss clause covers the damage to the Spragues’ deck. The court explained that “the losses that are faulty construction and rot are not covered, but the ‘ensuing losses,’ those that result from such faulty construction or rot, are covered because such an ensuing loss is not excluded elsewhere in *534the policy.” Sprague v. Safeco Ins. Co. of Am., 158 Wn. App. 336, 341, 241 P.3d 1276 (2010). I agree. The ensuing loss clause in the Spragues’ policy provided that “any ensuing loss not excluded or excepted in this policy is covered.” CP at 52, 76. While the policy excluded losses associated with the defective construction of the deck, the ensuing loss clause preserved coverage for any losses not otherwise excluded under the policy. Because the policy did not exclude losses associated with collapse — collapse damages being the losses that ensued from the rot or defective construction— the policy covers the Spragues’ loss.
¶28 Safeco insists that even if its policy covers collapse, coverage should apply only when a structure actually falls down. The majority does not endorse this argument, and for good reason. Absent a policy definition, courts have generally rejected the fall-down notion of collapse in favor of the more liberal standard, “ ‘substantial impairment of structural integrity.’ ” Mercer Place Condo. Ass’n v. State Farm Fire & Cas. Co., 104 Wn. App. 597, 602 n.1, 17 P.3d 626 (2000). We implicitly adopted the more liberal definition of collapse in Panorama Village Condominium Owners Ass’n Board of Directors v. Allstate Insurance Co., 144 Wn.2d 130, 134-35, 144-45, 26 P.3d 910 (2001). And the federal district court for the Western District of Washington, applying Washington law, concluded that Washington would reject the fall-down definition of collapse in favor of the liberal standard. Allstate Ins. Co. v. Forest Lynn Homeowners Ass’n, 892 F. Supp. 1310, 1314 (W.D. Wash. 1995), opinion withdrawn from publication, Allstate Ins. Co. v. Forest Lynn Homeowners Ass’n, 914 F. Supp. 408 (W.D. Wash. 1996) (unpublished). This formulation is consistent with the dictionary definition of collapse, which defines “collapse” to include not only a falling down, but also “a breakdown of vital energy, strength, or stamina.” Webster’s Third New International Dictionary 443 (2002). Moreover, “[Requiring the insured to await an actual collapse would not only be economically wasteful, but would also conflict with the insured’s contractual and com*535mon law duty to mitigate damages.” Beach v. Middlesex Mut. Assurance Co., 205 Conn. 246, 253 n.2, 532 A.2d 1297 (1987) (citation omitted).4
¶29 While the majority does not endorse Safeco’s definition of collapse, it insists we need not decide if the deck collapsed because “[w]hether or not the deck had reached a state of collapse, its condition was the result of the excluded perils of defective workmanship and rot and did not constitute a separate loss apart from those perils.” Majority at 529. This analysis misses the mark. The fact that the deck collapsed is key to a proper application of the ensuing loss clause.
¶30 We explained in Vision One that the ensuing loss clause limits the scope of what is otherwise excluded under the policy. 174 Wn.2d at 514. This clause ensures “that if one of the specified uncovered events takes place, any ensuing loss which is otherwise covered by the policy will remain covered. The uncovered event itself, however, is never covered.” McDonald v. State Farm Fire & Cas. Co., 119 Wn.2d 724, 734, 837 P.2d 1000 (1992). In a sense, the ensuing loss clause creates an exclusion within an exclusion. Id.; see also Capelouto v. Valley Forge Ins. Co., 98 Wn. App. 7, 16, 990 P.2d 414 (1999).
¶31 The majority wrongly requires that there be a “separate” cause of an ensuing loss, distinct from the excluded peril. Majority at 530. But, such a requirement would render the ensuing loss clause inoperative because the clause is implicated only when a covered loss results or ensues from an excluded peril. This presupposes causation. *536What the majority appears to be searching for is a physical line of demarcation. Thus, it seizes on the fact that in Vision One, the shoring installation was physically separate from the floor slab and related concrete work damaged in the collapse. Majority at 530.
¶32 I acknowledge that the nature of the ensuing loss is easier to see in Vision One, but the same principles are at play here. The engineers who examined the Spragues’ deck concluded that the “decayed wood framing in the deck piers” was caused by a combination of (1) “[inadequate flashing between the deck beams and the deck piers,” (2) “[p]ossible inadequate flashing between the deck guardrails and the deck piers,” and (3) “[inadequate ventilation of the deck piers.” CP at 109. They determined that the deck had been in a state of imminent collapse since before 2003. CP at 106. The engineering report described the likely process that undermined the structural integrity of the piers:
Water drained through the gaps between the spaced decking boards and onto the deck beams below. The water then seeped through the cracks along the sides of the deck beams and under the small metal flashings over the deck beams into the pier assembly. Once inside the pier assembly, the water came in contact with the wood framing. The absence of ventilation in the deck piers prevented the framing from drying. The moist conditions were conducive to the growth of fungi in the wood that causes decay and, over time, resulted in the gradual deterioration of the wood framing.
CP at 227.
¶33 While the majority reduces the circumstances here to the “natural process of decay,” majority at 531, in fact it is possible to identify the progression of events from faulty workmanship and wood rot to the imminent collapse. Just as in Vision One, the covered loss ensued from excluded perils. To the extent the ensuing loss is harder to visualize in this case, perhaps that is because it developed inside the deck fin walls rather than out in the open, as in Vision One.
*537¶34 Finally, the majority insists that the ensuing loss clause applies only to personal injury or property damage that results from a collapse, noting that here, “[t]he only loss was to the deck system itself.” Majority at 531. We rejected a similar, separate property argument in Vision One. 174 Wn.2d at 517-18. While the majority agrees with our observation there that an event can often be characterized as either a loss or a peril, it fails to see how the collapse at issue here is a loss. Majority at 529. Why? The ensuing loss of collapse is strikingly similar to the textbook example of ensuing loss, i.e., the fire that results from faulty wiring. In either case, the only property loss may be to the very structure affected by the excluded peril of poor workmanship. For example, a fire caused by faulty wiring may burn only the wall in which the wiring is contained. On the other hand, the fire may cause additional damage to the contents of the structure, to a neighboring structure, or to a person. Either way, the covered ensuing loss is the fire. It is the same with collapse. The ensuing loss is properly understood as the collapse, regardless of whether only the deck is damaged or whether it falls on someone or something. Significantly, the coverage expert in Vision One acknowledged at trial that there is no meaningful difference between a fire loss and a collapse loss under an ensuing loss clause. 174 Wn.2d at 518. And, Safeco’s adjuster noted that “if collapse occurred as an ensuing loss to the faulty construction exclusion, coverage would have been triggered.” CP at 168.5
¶35 In sum, the Spragues’ all-risk policy contemplated losses caused by collapse — including substantial impairment of structural integrity. The ensuing loss clause provides coverage because the deck collapse ensued from the *538defective construction of the deck, and the policy did not exclude coverage for collapse. I would affirm the Court of Appeals conclusion that there is coverage. Because the Spragues have demonstrated coverage under the policy, I would also hold they are entitled to fees under Olympic Steamship Co. v. Centennial Insurance Co., 117 Wn.2d 37, 51-54, 811 P.2d 673 (1991).
C. Johnson, Chambers, and Owens, JJ., concur with Stephens, J.Reconsideration denied July 9, 2012.
Beginning in 2003, Safeco added an endorsement to the Spragues’ homeowners policy excluding “collapse” as a peril. Clerk’s Papers (CP) at 98. Under the new endorsement, the policy did not cover any losses “caused directly or indirectly” by “collapse.” Id. The endorsement also contained a definition of the word “collapse,” stating that only “an abrupt falling down or caving in of a building or any part of a building” qualified as a collapse. CP at 99. Some limited coverage for collapse was provided under the policy’s section for “Additional Property Coverages.” CP at 98-99.
Safeco asserts that its pre-2003 policies did in fact define collapse as requiring the structure to fall down because the personal property section clarifies that “[collapse] does not include settling, cracking, shrinking, bulging or expansion.” CP at 54, 79. Safeco reads this provision to exclude any form of collapse short of the actual falling down of the structure. A majority of courts have rejected this argument, even where the policy defines collapse to not include settling and cracking. Am. Concept Ins. Co. v. Jones, 935 F. Supp. 1220, 1227 (D. Utah 1996) (collecting cases). As the federal district court explained, a policy of this type may exclude “mere settling or cracking,” but it does not exclude “settling or cracking that results in substantial impairment of a home’s structural integrity.” Id.
The amount of coverage presents a different question. The majority seems bothered that recognizing coverage for collapse would seem to provide coverage for the very peril Safeco’s policy excludes, namely wood rot and defective construction. But, there is a difference between recognizing coverage and calculating the loss. The extent to which damages from excluded perils may be subtracted from ensuing loss coverage is not before us.