INGENCO Holdings, LLC v. ACE American Insurance Company

Opinion

HONORABLE RICHARD A. JONES

UNITED STATES DISTRICT COURT AT SEATTLE INGENCO HOLDINGS, LLC and BIO ENERGY (WASHINGTON), LLC, Case No. 2:13-cv-00543-RAJ Plaintiffs,

ORDER ON CROSS MOTIONS v. FOR PARTIAL SUMMARY JUDGMENT (DKT. ## 241, 243) ACE AMERICAN INSURANCE Defendant. I. INTRODUCTION This matter comes before the Court on the parties’ cross motions for partial summary judgment. Dkt. ## 241, 243. Having considered the submissions of the parties, the relevant portions of the record, and the applicable law, the Court finds that oral argument is unnecessary. The motions are GRANTED in part and DENIED in part as set forth below. II. BACKGROUND Long ago, the parties moved for summary judgment. Dkt. # 142. The Court entered summary judgment in favor of Defendant ACE American Insurance Company (“ACE”) and against Plaintiffs Ingenco Holdings LLC and Bio Energy (Washington) LLC (collectively, “Ingenco”). Dkt. # 155. Ingenco later appealed the Court’s summary judgment order, and in the end, the Ninth Circuit affirmed in part, reversed in part, and remanded for trial. Dkt. ## 157, 166; see also Ingenco Holdings, LLC v. Ace Am. Ins. Co., 921 F.3d 803, 819 (9th Cir. 2019). On remand, the parties asked the Court for leave to file new motions for summary judgment. Dkt. # 205. Permitting them to do so, they said, would allow them to “address certain legal issues” “left unresolved by the Ninth Circuit.” Id. The Court granted their request, Dkt. # 210, and the parties submitted a joint statement setting forth the five legal issues that they wanted resolved, Dkt. # 219. This Order addresses those five issues. And given that the issues are legal, not factual, in nature, the Court need not explain the entire history of the case here. The Court has already done so in its previous summary judgment order. Dkt. # 142. Were that not enough, the Ninth Circuit’s opinion also provides a summary of the facts giving rise to this case. 921 F.3d at 806-08. For more background, the Court refers to those orders. Still, to resolve the legal issues, some facts are essential. The Court summarizes them here. The Court draws from its previous order and the Ninth Circuit’s opinion. The description here is meant to inform, not to supplement the record or to make new findings of undisputed fact. A. Gas Purification Plant, Diffuser Basket, and Adsorbent Beads Ingenco operates a gas purification plant at the Cedar Hills landfill located in King County, Washington. Dkt. # 142 at 2. At the plant, landfill gas is processed into natural gas. Id. The final step of that process involves the removal of excess nitrogen, which occurs in a nitrogen rejection unit, or “NRU.” Ingenco, 921 F.3d at 806; Dkt. # 142 at 2. The NRU is comprised of four pressure vessels. Dkt. # 25 ¶ 13. In the vessels are a “diffuser basket” and “adsorbent beads.” Id.1

1 This Court, the Ninth Circuit, and the parties use different names to refer to these components. These include diffuser “shield” and adsorbent “material.” These terms are interchangeable. In short, landfill gas flows into the top of each pressure vessel and is then filtered through adsorbent beads, to which the nitrogen adheres. Ingenco, 921 F.3d at 806-07. The diffuser basket reduces the force of gas flow on the beads. Id. Installed inside the pressure vessel where the gas first enters, the diffuser basket prevents the incoming gas from striking the beads directly. Id. It does so by allowing the gas to first hit the basket’s perforated bottom plate. Id. The gas hits the plate, is diffused, and then passes through the beads with reduced force. Id. On October 1, 2010, the diffuser basket in one of the four pressure vessels, pressure vessel number thirty-two (“V32”), broke. Id. at 807. Without a basket, the gas hit the adsorbent beads at full force and pulverized the beads into “dust.” Id. Days later, the plant shut down. Id. The failed diffuser basket was removed and replaced with a redesigned basket, and the plant resumed operations more than a week later. Dkt. # 142 at 2. Many months passed until, on March 29, 2011, the plant shut down again. Id. at 2. The dust generated months prior, on October 1, 2010, had infiltrated other parts of the system including other pressure vessels. Ingenco, 921 F.3d at 806-07. The dust degraded the beads in those vessels as well and caused a plant shutdown. Id. Ingenco later filed a property damage and business interruption insurance claim with its insurer, ACE. Id. The parties’ policy is an “all risks” policy, insuring the Cedar Hills facility among others. Id.; see also Dkt. # 244 ¶ 2 Ex. A; Dkt. # 1 ¶ 8 Ex. 1. ACE denied the claim, and Ingenco filed suit. Ingenco, 921 F.3d at 808. Through this action, Ingenco seeks coverage not for the failed diffuser basket itself but for the “ensuing losses” to the adsorbent beads and rest of the NRU. Dkt. # 25 ¶ 55. Summary judgment is appropriate if there is no genuine dispute as to any material fact and the moving party is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(a). The moving party bears the initial burden of demonstrating the absence of a genuine issue of material fact. Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986). Where the moving party will have the burden of proof at trial, it must affirmatively demonstrate that no reasonable trier of fact could find other than for the moving party. Soremekun v. Thrifty Payless, Inc., 509 F.3d 978, 984 (9th Cir. 2007). On an issue where the nonmoving party will bear the burden of proof at trial, the moving party can prevail merely by pointing out to the district court that there is an absence of evidence to support the non-moving party’s case. Celotex Corp., 477 U.S. at 325. If the moving party meets the initial burden, the opposing party must set forth specific facts showing that there is a genuine issue of fact for trial to defeat the motion. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250 (1986). The court must view the evidence in the light most favorable to the nonmoving party and draw all reasonable inferences in that party’s favor. Reeves v. Sanderson Plumbing Prods., 530 U.S. 133, 150-51 (2000). The scope of this Order is limited to the five issues raised in the parties’ joint statement. Dkt. # 219. 1. The parties seek to resolve the meaning of “Accident,” as set forth in the Boiler & Machinery Endorsement of the ACE policy and the effect of the exclusion which states that “Accident shall not include loss [f]rom depletion, deterioration, corrosion or erosion [and] wear and tear.” 2. Whether Plaintiffs bear the burden of proof for showing that the losses that are the subject of this lawsuit were fortuitous in nature or if ACE, as the insurer, bears the burden of proving the losses claimed by Plaintiffs were not fortuitous. 3. Whether the insuring language of the all-risks coverage provided by the ACE policy requires a separate showing that the losses claimed were the result of an external cause, or whether a showing that the losses were fortuitous is sufficient. 4. The construction and application of the exclusion in the ACE policy for “[g]radual deterioration, depletion, inherent vice, latent defect, . . . wear and tear . . . unless such loss is caused directly by physical damage not otherwise excluded in this Policy to the property covered.” 5. The parties also dispute the issues to be presented to the jury with respect to the exclusion for “loss, damage, or expenses caused by or resulting from . . . faulty or defective material, . . . [or] errors or omissions in the plan or specification design . . . unless loss by a peril not otherwise excluded ensues, and then only for the ensuing loss.” Id. The Court addresses each issue in turn. Apart from issue three, each issue receives its own section. Issue three, on the other hand, is divided across two sections. A. Issue 1 – Boiler and Machinery Endorsement Included in the parties’ all risks policy is a “boiler and machinery” endorsement. Dkt. # 244 at 16-20. Section 1 of that endorsement provides that ACE shall be liable for certain losses “resulting from” an “Accident” to an “Object.” Id. at 16. “Accident” is defined by § 1. Id. It is further limited by § 6, a section covering policy exclusions. Id. at 19. Under § 6(B)(1), referred to as the “wear and tear” exclusion, an “Accident” shall not include “loss . . . [f]rom depletion, deterioration, corrosion or erosion, [and] wear and tear.” Id. at 19. The parties seek guidance on the meaning of “Accident.” Dkt. # 219 ¶ 1. Specifically, they ask the Court to interpret the word “Accident” and to decide the validity of the wear and tear exclusion. Id. i. Meaning of “Accident” Section 1 defines “Accident”: “a sudden and accidental breakdown of an Object or a part thereof which manifests itself at the time of its occurrence by physical damage that necessitates repair or replacement of the Object or part thereof.” Dkt. # 244 at 16. Many terms within this definition are undefined, namely “sudden and accidental.” Ingenco asks this Court to interpret “sudden and accidental” to mean “unforeseen and unexpected.” Dkt. # 243 at 2-3. It also asks the Court to declare that the term “has no temporal element,” in that “it makes no difference whether the breakdown occurred abruptly or gradually.” Id. This reading of “sudden and accidental,” Ingenco says, is required under Washington law. It cites Anderson & Middleton Lumber Co. v. Lumbermen’s Mutual Casualty Co., 333 P.2d 938, 939 (Wash. 1959). In Anderson, the Washington Supreme Court interpreted an insurance policy requiring an insurer to pay for loss “occasioned by an accident” to a bandsaw wheel. 333 P.2d at 939. “Accident” was defined as “the sudden and accidental breaking” of the bandsaw wheel. Id. The term “sudden,” like here, was undefined. Id. Absent a definition, the insurer argued that “sudden” meant “instantaneously.” Id. at 941. The Anderson court disagreed. Id. at 940-41. It determined that “sudden” did not mean “instantaneous” but rather “unforeseen and unexpected.” Id. Other courts have upheld that interpretation. See, e.g., Time Oil Co. v. Cigna Prop. & Cas. Ins. Co., 743 F. Supp. 1400, 1408 (W.D. Wash. 1990) (adopting the Anderson definition and holding that “sudden” does not create a “temporal restriction” to coverage); Babai v. Allstate Ins. Co., No. 2:12-cv-01518-JCC, 2014 WL 12029279, at *2-3 (W.D. Wash. Oct. 8, 2014). Ingenco is correct to say, then, that “sudden,” when undefined in an insurance agreement, means “unforeseen and unexpected.” It is also correct to say that “sudden”— alone—does not impose a “temporal restriction” under Washington law. But here “sudden” does not stand alone, and the boiler and machinery endorsement does not rely on “sudden” for its temporal restriction. There are many components to the “Accident” definition. “Sudden” is just one of them. In all, an “Accident” requires a (1) “sudden and accidental” (2) “breakdown of an Object” (3) that “manifests itself at the time of its occurrence by physical damage” (4) that requires repair of the Object. Dkt. # 244 at 16. The boiler and machinery endorsement indeed contains a “temporal restriction.” Time Oil, 743 F. Supp. at 1408. But that restriction is not found in the word “sudden”; it is found in the phrase “at the time of its occurrence.” Thus, unlike the Anderson and Time Oil courts, this Court need not determine whether “sudden” standing alone implies a time element. Here, “sudden” does not stand alone, and another part of the “Accident” definition independently imposes a temporal restriction. To be sure, because the term “occurrence” has two parts, the temporal restriction also has two parts. “Occurrence” under the policy means either “an event” or “a continuous exposure to conditions which causes direct physical loss or damage or destruction to property insured.” Dkt. # 1 at 31. Hence, the “time of a breakdown’s occurrence” may be when a single “event” occurs, which would be instantaneous. Or it may be during a “continuous exposure to conditions” that causes physical loss. So, in the end, Ingenco is correct that a breakdown need not be instantaneous for it to be still considered an “Accident.” A breakdown may, all the same, occur during a “continuous exposure.” Altogether, the Court offers the following guidance on § 1 of the boiler and machinery endorsement: Ingenco has the initial burden “to show its loss falls within the scope of the losses insured under the policy.” Olivine Corp. v. United Capitol Ins. Co., 52 P.3d 494, 502 (Wash. 2002). To meet that burden, it must show that its losses were the result of an “Accident” to an “Object.” Dkt. # 244 at 16. An “Accident” requires Ingenco to show that an “Object” broke down; that the breakdown of the “Object” was sudden and accidental for it was unforeseen and unexpected; that the breakdown manifested itself by physical damage at the time of the event or the time of continuous exposure to conditions which caused the loss; and that the physical damage manifested by the breakdown required repair or replacement of the “Object.” ii. Validity of Wear and Tear Exclusion Just as § 1 sets forth what an “Accident” is, § 6(B) sets forth what it is not. Section 6 sets forth exclusions under the boiler and machinery endorsement. And the wear and tear exclusion under § 6(B)(1) states that an “Accident” does not include loss “[f]rom depletion, deterioration, corrosion or erosion, [and] wear and tear.” Dkt. # 244 at 19. Ingenco seeks an order from the Court that this exclusion does not apply as a matter of law. Dkt. # 243 at 2-3. According to Ingenco, the word “from” is legally distinct from the phrase “resulting from.” Id. at 8-10. As Ingenco tells it, what is excluded under the wear and tear exclusion is coverage for damage “from” wear and tear, which would be wear-and- tear damage to the insured property itself. Id. Damage “resulting from” wear and tear, it says, is a different matter and would not be excluded. Id. In less abstract terms, Ingenco believes that because the wear and tear exclusion uses the word “from,” not the phrase “resulting from,” it only excludes wear-and-tear damage to the diffuser basket itself, not damage caused by a worn-and-torn diffuser basket. Ingenco’s reasoning rests on its reading of Dickson v. U.S. Fidelity & Guaranty Co., 466 P.2d 515 (Wash. 1970). Before the Dickson court was an all risks insurance policy containing a latent defect exclusion. 466 P.2d at 517-18. The exclusion only said that the policy would not insure against latent defects. Id. The latent defect exclusion did not contain causation language, unlike other exclusions. Id. For example, the policy contained other exclusions for “[l]oss or damage caused by or resulting from the weight of a load exceeding the registered lifting capacity of any machine” and for “[l]oss or damage caused by or resulting from infidelity of [an] employee.” Id. at 518 (emphasis added). Given the “language . . . used in the other exclusionary clauses,” the court found the omission significant. See id. The policy “manifested” the insurer’s “obvious intent” that the latent defect exclusion was not to read “loss or damage caused by or resulting from . . . latent defects.” Id. “If that is what the insurer intended,” the court reasoned, “it should have said so” in the latent defect exclusion as it did in other exclusions. Id. Instead, the court read the latent defect exclusion to “apply only to the latent defect itself,” not losses stemming therefrom. Id. Ingenco argues that, like the latent defect exclusion in Dickson, the wear and tear exclusion here does not exclude losses “resulting from” wear and tear. Dkt. # 243 at 8- 10. Rather, it only excludes losses “from” wear and tear. Id. According to Ingenco, the boiler and machinery endorsement thus only excludes losses for wear and tear to the “Object” itself. Id. The “Object,” it says, is the diffuser basket. Id. And because Ingenco is not seeking coverage for the diffuser basket, the wear and tear exclusion does not apply as a matter of law. Id. This argument fails because the latent defect exclusion in Dickson is unlike the wear and tear exclusion here: the latent defect exclusion had no causation language; the wear and tear exclusion does. In Dickson, the latent defect exclusion did not contain causation language at all. The exclusion simply said, “This [p]olicy [d]oes [n]ot [i]nsure [a]gainst . . . latent defect.” 466 P.2d at 517-18. Because the Dickson court determined that the omission was intentional, it refused to inject such language. Here, there is no omission. The wear and tear exclusion contains causation language. Section 6(B)(1) plainly says that “Accidents” do not include losses “from” wear and tear. Dkt. # 244 at 19. Thus, the two exclusions are not analogous. It matters not, as Ingenco argues, that the exclusion here says “from” wear and tear rather than “resulting from” wear and tear. Just like the phrase “resulting from,” the word “from” alone indicates a “source, cause, agent, or basis.” From, Merriam-Webster, https://www.merriam-webster.com/dictionary/from (last visited Mar. 9, 2022) (emphasis added); Overton v. Consol. Ins. Co., 38 P.3d 322, 327 (Wash. 2002) (“Undefined terms in an insurance contract are given ‘plain, ordinary, and popular meaning’ as set forth in standard English language dictionaries.” (quoting Boeing Co. v. Aetna Cas. & Sur. Co., 784 P.2d 507 (Wash. 1990)). There is no distinction between the two under the case law. Dickson does not distinguish between “from” and “resulting from” let alone declare that the former is insufficient to denote causation. Thus, Ingenco’s argument that “from” and “resulting from” are “materially” different under Dickson fails. Dkt. # 243 at 8-10. In all, Dickson stands only for the proposition that if language is missing in one exclusion, but is present in other exclusions, a court should not inject that missing language into the exclusion. Here, the Court need not inject causation language to the wear and tear exclusion for it already contains such language. Hence, the wear and tear exclusion applies. B. Issue 2 – Burden to Prove Fortuity The Court was reversed on appeal because, among other reasons, it “failed to consider the role of fortuity in all risks insurance disputes.” 921 F.3d at 818. In the context of all risks policies, fortuity asks “whether a particular loss was certain to occur,” what the parties’ “perception of risk [were] at the time the policy issued,” and “whether the loss could reasonably have been foreseen.” Id. at 814-15. The Ninth Circuit explained the doctrine, and the Court need not do so again here. The Court must, instead, resolve a threshold issue. Dkt. # 219 ¶ 2. Under Washington law, who does the fortuity burden fall on? Must insureds prove that their losses were fortuitous? Or must insurers prove that the losses were not fortuitous? In a diversity action, federal courts must “approximate state law as closely as possible” to ensure that “the vindication of the state right is without discrimination because of the federal forum.” Ticknor v. Choice Hotels Int’l, Inc., 265 F.3d 931, 939 (9th Cir. 2001) (quoting Gee v. Tenneco, Inc., 615 F.2d 857, 861 (9th Cir. 1980)). Federal courts are “bound” by the “pronouncements of the state’s highest court on applicable state law.” Id. If the state’s highest court has not decided an issue, then “the task of the federal courts is to predict how the state high court would resolve it.” Dimidowich v. Bell & Howell, 803 F.2d 1473, 1482 (9th Cir. 1986). For “guidance,” courts may look to decisions by intermediate appellate courts of the state and by courts in other jurisdictions. Id. State intermediate appellate court decisions, specifically, are “data” “not to be disregarded by a federal court unless it is convinced by other persuasive data that the highest court of the state would decide otherwise.” Id. (quoting Estrella v. Brandt, 682 F.2d 814, 817 (9th Cir. 1982)). The Washington Supreme Court has not squarely decided the issue of who bears the fortuity burden. Hence, the Court must “predict” how the state high court would resolve the issue. The Court, for “guidance,” looks to relevant Washington Supreme Court decisions, Washington intermediate appellate court decisions, and federal court decisions applying Washington law. Unfortunately, the pertinent authority is split. On one hand, many courts have stated that the fortuity burden falls on the insured. Relevant here is a trio of cases. Frank Coluccio Const. Co. v. King Cty., 150 P.3d 1147, 1156 (Wash. 2007) (describing fortuity as a “condition precedent” and allocating the burden of proof to a, constructively, insured party); Churchill v. Factory Mut. Ins. Co., 234 F. Supp. 2d 1182, 1189 (W.D. Wash. 2002) (predicting that, under Washington law, “the insured bears the burden of showing that it suffered a loss and that the loss is fortuitous”); Underwriters Subscribing to Lloyd’s Ins. Cert. No. 80520 v. Magi, Inc., 790 F. Supp. 1043, 1048 (E.D. Wash. 1991) (predicting that, under Washington law, “[i]t is [an insured]’s responsibility to prove its loss is fortuitous”). And even this Court, relying on those decisions, has repeated the same. Canyon Ests. Condo. Ass’n v. Atain Specialty Ins. Co., No. 2:18-cv-01761-RAJ, 2021 WL 1208581, at *4 (W.D. Wash. Mar. 31, 2021). Only one decision, Frank Coluccio, belongs to a state intermediate appellate court. The rest were the decisions of fellow federal district courts. One Washington Supreme Court decision, however, casts doubt on that trio. In Aluminum Co. of Am. v. Aetna Casualty & Surety Co., 998 P.2d 856, 878 (Wash. 2000) [hereinafter Alcoa], the state high court explained that, in Washington, the fortuity principle is called the “known risk” principle or defense. It further described fortuity, under Washington law, as an “unnamed exclusion,” for it “has the effect of an exclusion” yet “never appears in insurance contracts.” Id. at 878-79. At issue in Alcoa, however, was fortuity under Pennsylvania state law. Id. at 879. The trial court applied Pennsylvania law to the parties’ insurance dispute. Id. at 862. On the issue of fortuity, the trial court was thus tasked with predicting “what rules of law governing fortuity a Pennsylvania court would apply.” Id. at 879. And the Washington Supreme Court, for its part, was tasked with reviewing the trial court’s prediction for error. Id. Ultimately, the state high court decided that Pennsylvania courts applying Pennsylvania law would place the fortuity burden of proof on insurers. Id. at 882. The high court’s reasoning: because under Pennsylvania law the burden of proving the applicability of an exclusion lies with insurers, and because “[f]ortuity is, in effect, an exclusion,” fortuity “logically should be the burden of the insurer to plead and prove the exclusion.” Id. (emphasis added). Relying on the above cases as guidance, the Court predicts that the Washington Supreme Court would follow its reasoning in Alcoa and hold that, under Washington law, the fortuity burden falls on the insurer. The Court makes this prediction for a simple reason. The three cases allocating the fortuity burden to insureds—Frank Coluccio, Churchill, and Magi—do not explain why they do so. Alcoa, on the other hand, explains why the burden logically falls on insurers. Thus, the Court finds that Alcoa is a better predictor of how the Washington Supreme Court would decide the fortuity burden issue.2 In Frank Coluccio, the Washington Court of Appeals, reviewing a trial court’s determination that a constructive insured3 met its burden to prove fortuity, found no error in the trial court’s conclusion. 150 P.3d at 1156. Implicitly, then, the court endorsed the trial court’s allocation of the fortuity burden to the insured. Id. The court also described fortuity as a “condition precedent” to coverage. Id. It gave no other explanation. Id.

2 To be sure, none of these four decisions are “binding” on the Court’s inquiry. The Court is only bound “by the pronouncements of the state’s highest court on applicable state law.” Ticknor, 265 F.3d at 939. Here, the Washington Supreme Court has made no such pronouncements. The Court treats all four decisions as guidance.

3 In Frank Coluccio, the defendant failed to buy an insurance policy, a requirement under the parties’ agreements. 150 P.3d at 1150, 1155-56. Because the defendant breached its obligation, it assumed all risk of loss, and the plaintiff could recover “the full amount that would have been covered by insurance, had the breaching party performed as specified.” Id. at 1155-56. Hence, the plaintiff in that case was not, strictly speaking, insured but constructively insured. Id. Similarly, citing non-Washington cases, the federal district court in Magi simply stated, without explanation, that it was an insured’s responsibility to prove that its loss was fortuitous. 790 F. Supp. at 1048. The court in Churchill did the same, citing Magi and non-Washington cases. 234 F. Supp. 2d at 1189. In all, no case in the trio explains why the fortuity burden must fall on insureds. That those courts gave short shrift to the fortuity burden is unsurprising. In those cases, the burden of proof was hardly an issue. The courts were describing the fortuity principle generally, as other jurisdictions have understood it. The precise issue of who bore the fortuity burden of proof was not squarely raised by the facts or litigants in those cases, unlike here. In any event, because those cases give little, if any, justification for placing the fortuity burden on insureds, the Court finds them only slightly persuasive. Cf. Davis v. Metro Prods., Inc., 885 F.2d 515, 524 (9th Cir. 1989) (relying on three Arizona Court of Appeals’ decisions to approximate Arizona law because those decisions contained “substantive analysis” that served as “forceful precedent”). Contrast Alcoa. There, the Washington Supreme Court explained why the fortuity burden logically falls on insurers. The state high court explained that Washington itself recognizes the fortuity principle and that Washington considers fortuity to be an “unnamed exclusion” given that it “has the effect of an exclusion. 998 P.2d at 878-79. It also said that fortuity, as an exclusion, “logically should be the burden of the insurer to plead and prove the exclusion.” Id. at 882. Of course, Alcoa must be read in context. Earlier in the opinion, the Alcoa court described the fortuity principle as it is known under Washington law. 998 P.2d at 878- 79. But later in the opinion, when discussing who bore the fortuity burden, the court was trying to predict how Pennsylvania law would allocate that burden. Id. The Alcoa court predicted that Pennsylvania law would place that burden on insurers. Id. at 881-82. The court did not say, however, that the outcome would have been the same under Washington law. Still, what the Court finds persuasive is not Alcoa’s holding, but its reasoning. The Court does not see why Alcoa’s “logic” would be good in Harrisburg but not Olympia. 998 P.2d at 882. Washington law, like Pennsylvania law, maintains that fortuity is an unnamed exclusion. Alcoa, 998 P.2d at 878. Washington law, like Pennsylvania law, places the burden of proving an exclusion on the insurer. Mut. of Enumclaw Ins. Co. v. T & G Const., Inc., 199 P.3d 376, 383 (Wash. 2008) (“The insured bears the burden of showing that coverage exists; the insurer that an exclusion applies.”). In Washington, that has been the insurer’s burden for at least sixty years. Labberton v. Gen. Cas. Co. of Am., 332 P.2d 250, 254 (Wash. 1958) (“If the question were doubtful, it must be remembered that the burden of proving that the loss is within the exclusionary clause is upon the insurance carrier.”). Alcoa’s syllogism should thus also apply under Washington law: The burden to prove an exclusion falls on the insurer. Fortuity is an exclusion. Thus, the fortuity burden falls on the insurer. The Court finds it unlikely that, given Alcoa, the Washington Supreme Court would endorse its logic for Pennsylvania but reject the same logic for its own state. In the end, the Court predicts that the state high court would allocate the fortuity burden to the insurer. C. Issue 3 – Sufficiency of Fortuity Ingenco’s policy insures its property against “all risks of direct physical loss or damage . . . from any external cause.” Dkt. # 244 at 5. Hence, to prove coverage, Ingenco must prove that damage to the adsorbent beads was the result of an “external cause.” Dkt. # 25 ¶ 53. On appeal, the Ninth Circuit pondered whether a finding of fortuity would “obviate” the need for an “externality inquiry” altogether. 921 F.3d at 816-17, 817 n.17. It noted that some courts have decided that a finding of fortuity, alone, would be sufficient. Id. at 816 (citing Second and Fifth Circuit opinions). The Ninth Circuit found that approach “persuasive.” Id. at 817 (explaining why it is “difficult” to identify an “external cause” and explaining that the difficulty “perhaps, more than anything, illustrate[s] the advantages of avoiding the ‘external cause’ question altogether”). Ultimately, however, the court left the question open. Id. It reasoned that the Washington Supreme Court had not spoken on the issue and that the resolution of the issue was not key to its holding. Id. Unsurprisingly, the Ninth Circuit raised a question, and the parties now want it answered. They ask the Court to determine whether the policy “requires a separate showing that the losses claimed were the result of an ‘external cause,’ or whether a showing that the losses were fortuitous is sufficient.” Dkt. # 219 ¶ 3. The Court declines their request because the fortuity inquiry would not, in fact, “obviate” the “external cause” inquiry as the Ninth Circuit surmised. When determining insurance coverage, fortuity comes after “external cause.” In Washington, an insured must first prove coverage and, only then, must the insurer prove that an exclusion applies. McDonald v. State Farm Fire & Cas. Co., 837 P.2d 1000, 1004 (Wash. 1992). “External cause” is part of Ingenco’s initial burden to prove coverage. Dkt. # 244 at 5. Fortuity, on the other hand, is an “unnamed exclusion” that ACE must prove. See supra Section IV.B. In order, then, Ingenco must first prove coverage by showing that damage to the adsorbent beads was the result of an “external cause,” and ACE in turn must show that Ingenco’s losses were not fortuitous. Thus, the “external cause” inquiry comes before the fortuity inquiry. The latter could not obviate the former. Because it comes after “external cause,” proving fortuity would not “obviate the need to examine whether that loss was caused by an external force,” and this Court is not free to “avoid[] the . . . question altogether.” Ingenco, 921 F.3d at 816. The Ninth Circuit’s suggestion that fortuity could supplant external cause was based on an unproven premise. The court assumed that the fortuity burden would fall on insureds. For example, it explained that “some courts . . . have held that an insured need only demonstrate that a fortuitous loss has occurred, notwithstanding ‘external cause’ policy language.” 921 F.3d at 816 (first emphasis added) (second emphasis omitted). If that were the case, then perhaps a fortuity finding would indeed render “external cause” unnecessary. Both would be Ingenco’s burden to prove, and perhaps proving one would eliminate the need to prove the other. But the Ninth Circuit’s assumption that the fortuity burden would fall on insureds was just that, an assumption. 921 F.3d at 816. The court did not conduct a diversity analysis to predict how the Washington Supreme Court would allocate that burden among insurers and insureds. Id. This Court did. See supra Section IV.B. And after it did, the Court predicted that the Washington Supreme Court would instead place the fortuity burden on insurers. Id. Given this Court’s prediction, the Ninth Circuit’s meditations on fortuity and “external cause” no longer apply. In this Court’s view, because it comes second, fortuity cannot obviate Ingenco’s need to prove “external cause.” What is more, there is yet another reason not to decide this issue. Washington courts are silent on the matter. No court has addressed whether a finding of fortuity could obviate the need for an “external cause” analysis. The Ninth Circuit said itself, “[T]he Washington Supreme Court has not spoken on this issue.” 921 F.3d at 816. Worse, the state’s intermediate appellate courts are equally silent. The parties, too, offer little help. ACE does not address the issue at all. And Ingenco, for its part, only says that because “external cause” is a “quagmire,” it is “best discarded in favor of fortuity.” Dkt. # 249 at 12. On unsteady ground, the Court will not decide state law issues of first impression needlessly. “Traditionally and properly federal courts are reluctant to decide questions of first impression in the interpretation and application of state substantive law in diversity jurisdiction cases.” See Black v. United States, 421 F.2d 255, 258 (10th Cir. 1970) (finding it “unnecessary and certainly undesirable to advance a foreguess” on whether the Utah Supreme Court would recognize the doctrine of strict liability). This is especially true given that the resolution of this issue does not advance the Court’s analysis and thus does not advance the resolution of the parties’ claims. See id. at 259 (“[T]o decide the controlling issues presented would serve no purpose other than to expound academic views into a field of state law that would serve as no controlling precedent.”). For these reasons, the Court will not determine, as a matter of law, whether the parties’ policy “requires a separate showing that the losses claimed were the result of an ‘external cause,’ or whether a showing that the losses were fortuitous is sufficient.” Dkt. # 219 ¶ 3.4 D. Issue 3 – “External Cause” Interpretation The parties’ all risks policy reads: This Policy covers the property insured hereunder against all risks of direct physical loss or damage occurring during the period of this Policy from an external cause, except as hereinafter excluded or limited. Dkt. # 244 at 5 (emphasis added). “External cause” is undefined. The parties now seek clarity: what does “external cause” mean? Dkt. # 241 at 10-18; Dkt. # 243 at 19-20. The Court previously, at the insistence of the parties, applied one interpretation of “external cause.” On remand, circumstances have changed. The parties no longer endorse their, and for that matter the Court’s, previous interpretation. They request a new one. The Court agrees that the previous interpretation will no longer do and arrives at a new interpretation. It does so by applying well-settled Washington contract interpretation 4 The Court understands that, in reaching its conclusion here, it effectively answers the parties’ question. Going forward, the Court will require Ingenco to prove that its losses were the result of an “external cause,” no matter how the fortuity inquiry falls. To that extent, fortuity and “external cause” require “separate showings.” For clarity, however, the Court’s conclusion is based on the facts of this case and the wording of the parties’ agreements. This conclusion is not the result of using “guidance” to predict how the Washington Supreme Court would rule on the issue. principles. The Court first explains its previous interpretation of “external cause,” the Ninth Circuit’s later treatment of that interpretation, and how the parties’ arguments have changed on remand. Then, the Court explains the problems with the previous interpretation. Finally, the Court provides a new definition. i. Previous Interpretation – Standard Structural Steel Moving for summary judgment long ago, the parties insisted that the term “external cause” be given the same meaning that courts in other jurisdictions have given it. Dkt. # 80 at 9; Dkt. # 87 at 9. Specifically, they asked the Court to apply the definition used by a Connecticut district court in Standard Structural Steel Co. v. Bethlehem Steel Corp., 597 F. Supp. 164, 193 (D. Conn. 1984). The Court agreed and applied that definition: “[a] cause is external if damage which arises from it does not result wholly from an inherent defect in the subject matter or from the inherent deficient qualities, nature and properties of the subject matter.” Dkt. # 142 at 20. In effect, this definition defines “external cause” by “what it is not.” Standard Structural Steel, 597 F. Supp. at 193. An “external cause” is one that is not “wholly” the result of an “inherent defect.” Or, as the Ninth Circuit later put it, under this definition “external cause” becomes “coterminous” with the lack of an “inherent defect.” 921 F.3d at 818. On appeal, the Ninth Circuit did not endorse or reject that definition. Instead, the court began by wondering why courts should look to “judicial interpretations of that phrase” in the first place, rather than the phrase’s “plain and ordinary meaning.” Id. at 813-14. It also criticized the “external cause” doctrine at large. Id. at 817. It explained that deciding what was an “internal cause” and an “external cause” was conceptually “difficult.” Id. Indeed, it said that the difficulty might justify abandoning the “external cause” question entirely. Id. (“These conflicts are difficult to resolve on the record before us, and perhaps, more than anything, illustrate the advantages of avoiding the ‘external cause’ question altogether.”). In the end, the Ninth Circuit refrained from concluding that the Washington Supreme Court would adopt the Standard Structural Steel definition. Id. at 816 n.15. Instead, the Ninth Circuit said that, even assuming the Standard Structural Steel definition was correct, there was a triable issue of fact. Id. at 816. Ingenco had offered evidence that the diffuser basket failure was not caused by an internal defect but was rather caused by unforeseen resonant vibrations. Id. at 816-18. Hence, whether Ingenco’s machinery suffered from an inherent defect was a question of fact. Id. at 818. So too, then, was whether Ingenco’s loss was “internal” or “external.” Id. Because this Court previously determined that Ingenco’s loss was internal as a matter of law, the Ninth Circuit reversed. On remand, the parties have changed course. No longer do they rely on Standard Structural Steel. ACE now asks the Court to use Dickson v United States Fidelity & Guaranty Company, 466 P. 2d 515 (Wash. 1970), to define “external cause.” Dkt. # 241 at 9-11. Ingenco, on the other hand, asks the Court to adopt the phrase’s plain and ordinary meaning. Dkt. # 255 at 9-11. ii. Problems with Previous Interpretation On remand, the Court reassesses its previous application of Standard Structural Steel. It finds two problems with its previous interpretation of “external cause.” First, the previous definition is circular and creates two burdens of proof on the same issue. While the parties’ policy provides coverage for losses from an “external cause,” it also excludes coverage for losses from an “inherent vice.” Dkt. # 255 at 7-8 (“This Policy does not insure against loss, damage or expense caused by or resulting from . . . [g]radual deterioration, depletion, inherent vice . . . .”). Were “external cause” the same as the lack of an “inherent defect,” as Standard Structural Steel suggests, then Ingenco would essentially have to disprove the inherent vice exclusion, only then for ACE to prove the inherent vice exclusion. Take the following illustration: Ingenco has the initial burden to prove coverage. To do so, it must show that its losses were the result of an “external cause.” Under Standard Structural Steel, “external cause” is defined as a cause not the result of an “inherent defect,” or, in this case, not the result of an “inherent vice.”5 Thus, under Standard Structural Steel, Ingenco must prove the lack of an “inherent defect.” Put differently, it must disprove the “inherent vice” exclusion. In turn, ACE would have to prove the “inherent vice” exclusion. This violates one of Washington’s most basic insurance maxims—that the burden to show an exclusion falls on the insurer, not the insured. McDonald v. State Farm Fire & Cas. Co., 837 P.2d 1000, 1003-04 (Wash. 1992). Second, no party explains why the Court may, let alone should, look to “judicial interpretations” of the phrase “external cause” in the first place. Ingenco, 921 F.3d at 813. Washington’s rules for interpreting insurance contracts are “well settled.” Quadrant Corp. v. Am. States Ins. Co., 110 P.3d 733, 737 (Wash. 2005). None of those rules invite the Court to look to other jurisdictions to interpret an undefined term in a contract. See id. And, as far as Washington courts go, the “external cause” doctrine has been given only glancing and peripheral attention. No court has given the term a consistent “judicial construction” that this Court can apply. Cf. Queen City Farms, Inc. v. Cent. Nat. Ins. Co. of Omaha, 882 P.2d 703, 725 (Wash. 1994) (“Our conclusion . . . is reinforced by the proposition that when a judicial construction is placed upon words or phrases prior to the issuance of a policy which uses those words and phrases, it is presumed that construction is intended by the parties.”). “External cause” does not have a widely recognized meaning under Washington insurance law, unlike, say, the word “accident,” which many courts have already construed. See, e.g., Grange Ins. Co. v. Brosseau, 776 P.2d 123, 125 (Wash. 1989) (“For insurance coverage, this court has said that ‘an accident’ is an unusual, unexpected, and unforeseen happening.”) (collecting

5 See infra note 8. cases); Allstate Ins. Co. v. Raynor, 143 Wash. 2d 469, 481, 21 P.3d 707, 714 (2001) (“While the policy did not define ‘accidental loss,’ ‘accident’ has a very specific meaning in Washington’s insurance law.”) (Talmadge, J., concurring). Persuasive as other courts’ interpretations may be, the Court must principally follow Washington’s contract interpretation framework. These two problems compel the Court to reassess and adopt a new definition, one that returns to Washington’s well-settled rules of insurance contract interpretation. Askins v. U.S. Dep’t of Homeland Sec., 899 F.3d 1035, 1042 (9th Cir. 2018) (“The law of the case doctrine does not preclude a court from reassessing its own legal rulings in the same case.”); United States v. Cuddy, 147 F.3d 1111, 1114 (9th Cir. 1998) (permitting a court to “depart from the law of the case” if “changed circumstances” exist). On remand, circumstances have changed. The parties no longer advocate for the definition of “external cause” found in Standard Structural Steel. What is more, the conceptual problems above and the guidance from the Ninth Circuit, militate in favor of a different interpretation. iii. New Interpretation In Washington, interpretation of an insurance contract is a question of law. Quadrant, 110 P.3d at 737. Courts must consider a policy “as a whole,” giving it a “fair, reasonable, and sensible construction as would be given to the contract by the average person purchasing insurance.” Id. (quoting Weyerhaeuser Co. v. Commercial Union Ins. Co., 15 P.3d 115 (Wash. 2000)). Courts must “give effect to each provision.” Moeller v. Farmers Ins. Co. of Washington, 267 P.3d 998, 1001 (Wash. 2011) (emphasis in original) (quoting Allstate Ins. Co. v. Peasley, 932 P.2d 1244 (Wash. 1997)). Undefined terms in an insurance policy are given their “plain, ordinary and popular meaning.” Queen Anne Park Homeowners Ass’n v. State Farm Fire & Cas. Co., 352 P.3d 790, 793 (Wash. 2015) (quoting Queen City Farms, 882 P.2d at 718). If an insurance policy is ambiguous, then the interpretation “most favorable to the insured must be applied, even though the insurer may have intended another meaning.” Id. (quoting Dairyland Ins. Co. v. Ward, 517 P.2d 966 (Wash. 1974)). A policy is ambiguous when “it is fairly susceptible to two different interpretations, both of which are reasonable.” Am. Nat. Fire Ins. Co. v. B & L Trucking & Const. Co., 951 P.2d 250, 256 (Wash. 1998). Because the parties’ all risks policy does not define the term “external cause,” the Court gives the term its plain and ordinary meaning. For that, the Court turns to the dictionary. Overton v. Consol. Ins. Co., 38 P.3d 322, 327 (Wash. 2002). A “cause” is “something that brings about an effect or a result.” Cause, Merriam-Webster, https://www.merriam-webster.com/dictionary/cause (last visited Mar. 9, 2022). And “external” is an adjective used to denote the spatial relation of one thing to another: “of, relating to, or connected with the outside or an outer part” or “arising or acting from outside.” External, Merriam-Webster, https://www.merriam- webster.com/dictionary/external (last visited Mar. 9, 2022). Together, an “external cause” is a cause that arises “from outside.” That is hardly illuminating. The obvious question is, for purposes of coverage, what is “inside” and what is “outside”? Or what must a cause be “external” to, exactly?6 6 To define “external cause,” ACE points the Court to Dickson v. U.S. Fidelity & Guaranty Co., 466 P.2d 515, 520 (Wash. 1970). Dkt. # 241 at 9-11. There, an insured’s crane was damaged when the crane’s boom collapsed while pulling beams out of the ground. The crane contained a defective weld, which, in part, caused the boom to collapse. There was also evidence, however, that an “external force” also caused the collapse, namely the earth falling onto the beams as the beams were being lifted out of the ground. The Dickson court held that this falling earth was the “direct,” “violent,” “efficient,” and “responsible” cause of the loss, even though the latent defect also caused the loss. At best, Dickson helps determine, given two or more causes of loss, which is legally significant. It provides almost no guidance on what is “internal” and “external,” the question before the Court. Finding that the cause of Ingenco’s harm here was “direct” or “efficient” would not help to answer that question. And the Court struggles to see how the facts of Dickson—cranes, beams, defective welds, falling earth—are sufficiently analogous to this case so that the Court could meaningfully apply Dickson’s “external force” analysis here. The policy sheds no light. The policy covers property damage to the entire Cedar Hills property, including its “building” and “contents,” which include “all machinery, equipment, tanks, inventory, fixtures and furniture, office equipment, and improvements and betterments.” Dkt. # 244 at 9-10; see also Dkt. # 25 ¶¶ 57, 59. It does not say, for purposes of coverage, where the boundary between an “internal cause” and “external cause” lies. Drawing that line is difficult. In theory, because the entire Cedar Hills property— down to its constituent elements—is insured, the line could be fixed in any number of places between the edge of the property line and the edge of each adsorbent bead in pressure vessel V32. The Ninth Circuit explained this framing problem: On the one hand, landfill gas was certainly essential to the operation of the Cedar Hills facility. At the same time, however, landfill gas was not an essential component of the facility or of the machinery itself, which existed before the first inflow of gas was ever piped in. Furthermore, although the landfill gas indisputably “arose” from outside the insured facility, whether it “acted” internally or externally depends on whether the borders of the insured subject matter begin at the property line or at the edge of the pressure vessel. These conflicts are difficult to resolve on the record before us, and perhaps, more than anything, illustrate the advantages of avoiding the “external cause” question altogether. See Morrison Grain, 632 F.2d at 430. 921 F.3d at 817 (first emphasis in original) (second emphasis added) (footnotes omitted). Perhaps, as the Ninth Circuit suggested, this problem is endemic to the “external cause” doctrine. Endemic or not, the problem falls on this Court’s shoulders. Contract interpretation is a matter of law, only for courts to decide. Neither party argues that the interpretation of “external cause” is a factual question, only capable of being resolved through extrinsic evidence. Ingenco, 921 F.3d at 814 n.8. Here, the all risks policy, undoubtedly, applies only when a loss is the result of an “external cause.” Dkt. # 244 at 5. The Court is not free to abandon the word “external” merely because it poses difficult questions. The Court must view the insurance contract in its entirety and must give effect to each provision. Moeller, 267 P.3d at 1001 (Wash. 2011).7 To solve this difficult problem, the Court starts with first principles. First, the Court deems the term “external cause” ambiguous. The undefined term, “on its face,” is “fairly susceptible to two different interpretations, both of which are reasonable.” Am. Nat. Fire Ins., 951 P.2d at 256. Indeed, the policy is not susceptible to just two reasonable definitions. It is potentially susceptible to an infinite number of reasonable definitions. The Ninth Circuit said that the “externality inquiry” presents two extremes. Both of which would be improper. On appeal, Ingenco argued that the “internal” and “external” line should be drawn based on its insurance claim. 921 F.3d at 817 n.17. Because it only seeks coverage for damage done to the adsorbent beads, Ingenco said that is where the externality inquiry should begin. Id. The Ninth Circuit rejected that argument. That view, the court held, would be “so fine-grained as to render virtually any damage to any particular component” “external.” Id. Likewise, the court suggested that ACE’s comprehensive characterization was “no less extreme,” rendering any damage “internal.” Id. Hence, the “externality inquiry” has two unreasonable extremes. Id. On one end, the line cannot be drawn at the edge of the adsorbent material contained in pressure vessel V32. That would, in theory, render any damage to the beads “external.” On the other end, the line cannot be drawn at the edge of the Cedar Hills property line. That would similarly render any damage to the beads “internal.” Between those ends are an infinite number of places to draw a line. Some would be reasonable; some would not. In any event, the Court can conclude that the term “external cause” is ambiguous given that many reasonable definitions are possible.

7 Ingenco asks the Court to scrap the “external cause” inquiry “altogether” because it results in a “doctrinal morass.” Dkt. # 255 at 11. The Court rejects that argument. The Court must give effect to all the policy’s terms. The second rule of interpretation that the Court relies on is, when a policy is ambiguous, the Court must resolve the ambiguity in favor of the policyholder. Among different reasonable interpretations, the Court is not required to treat each equally. In fact, the opposite is true. Under Washington law, “where the clause in the policy is ambiguous, a meaning and construction most favorable to the insured must be applied, even though the insurer may have intended another meaning.” Morgan v. Prudential Ins. Co. of Am., 545 P.2d 1193, 1195 (Wash. 1976); Sprague v. Safeco Ins. Co. of Am., 276 P.3d 1270, 1272 (Wash. 2012) (“As with any contract, ambiguous policies are construed against the drafter.”). Thus, the Court must choose the line that favors Ingenco, even if ACE, as the drafter, intended for that line to be drawn somewhere else. These principles in mind, the Court resolves the difficult question before it as follows: For purposes of coverage, the Court draws the “internal” and “external” line at the edge of pressure vessel V32. At trial, the parties may present evidence that Ingenco’s loss was caused by forces external to or internal to pressure vessel V32. For example, Ingenco may argue that the gas came into pressure vessel V32 too fast or at coincidentally the right speed to match the diffuser basket’s natural vibration frequency or resonant frequency, causing the basket to break. It may argue that this cause is external to pressure vessel V32 and resulted in Ingenco’s loss. If Ingenco is successful, then the burden shifts to ACE to prove that an exclusion applies, for example by showing that Ingenco’s loss was the result of an inherent vice. It might argue, perhaps, that the diffuser basket, located inside pressure vessel V32, broke because of a defect. These are factual issues not for this Court to decide on summary judgment. The Court only provides a fulcrum, on either side of which the parties may pile their “external cause” arguments. This interpretation follows all the rules set forth above. It treats the interpretation question as one of law and answers it. It reads the policy as a whole, giving effect to each clause. It gives “external cause,” an undefined term, its plain and ordinary meaning. It concludes, relying on the Ninth Circuit’s observations, that more than one reasonable interpretation of “external cause” could exist and deems the term ambiguous. It then adopts the interpretation most favorable to Ingenco, the insured. Perhaps most importantly, it adopts a fair, reasonable, and sensible construction based on the perspective of the average person purchasing insurance. A person buying an all risks policy does so to “allocate risks to the insurer.” Vision One, LLC v. Philadelphia Indem. Ins. Co., 276 P.3d 300, 306 (Wash. 2012). It is difficult to believe that an insurer would expect “all risks” coverage to apply only (a) in instances of obvious external risks, such as “lightning bolts,” “planes fall[ing] out of the sky,” or “meteorites” or (b) if the insured could, philosophically, characterize his loss as “external” to an insured component of the system. Oral Argument at 5:09, Ingenco Holdings, LLC v. Ace Am. Ins. Co., 921 F.3d 803, 806 (9th Cir. 2019) (No. 16-35792), https://cdn.ca9.uscourts.gov/datastore/media/2018/06/12/16-35792.mp3. The Court finds that the average person buying insurance would not view his bargain that way. Or, as the Ninth Circuit put it: [I]t would appear that all risks insurance arose for the very purpose of protecting the insured in those cases where difficulties of logical explanation or some mystery surround the (loss of or damage to) property. It would seem to be inconsistent with the broad protective purposes of “all risks” insurance to impose on the insured . . . the burden of proving the precise cause of the loss or damage. 921 F.3d at 816 (quoting Morrison Grain Co. v. Utica Mut. Ins. Co., 632 F.2d 424, 430 (5th Cir. 1980)). By drawing the line as it does, the Court crafts an interpretation that takes into account the purpose of all risks policies and the expectation of the average insurance purchaser. In sum, after applying a plain and ordinary meaning analysis to the term “external cause,” the Court concludes that the externality inquiry begins at the edge of pressure vessel V32. E. Issue 4 – Exception to Inherent Vice Exclusion The issue of “external cause” is one of coverage. This section marks a shift in the Court’s analysis. For the remainder of the Order, the Court interprets policy exclusions. The parties’ all risks policy contains an exclusion for loss resulting from an “inherent vice.” Dkt. # 244 at 7-8. This Policy does not insure against loss, damage or expense caused by or resulting from: . . . inherent vice . . . , unless such loss is caused directly by physical damage not otherwise excluded in this Policy to the property covered[.] Id. This exclusion, referred to as the inherent vice exclusion, contains an exception. Id. Under that exception, the exclusion does not apply if “such loss” is caused by “damage not otherwise excluded” under the policy. Id. The parties ask the Court to interpret both the exclusion and the exception. Dkt. # 219 ¶ 4. i. Burden to Prove Exception to Exclusion The Court first addresses the threshold issue of who bears the burden to prove the exception. The Court determines that ACE, the insurer, has the burden to prove the exclusion, while Ingenco, the insured, has the burden to prove the exception. Though the Washington Supreme Court has clearly placed the burden on the insurer to prove an exclusion, it has not determined who bears the burden to prove an exception to an exclusion. Other courts in this district have predicted that the state high court would place that burden on the insured. Berkshire Hathaway Homestate Ins. Co. v. SQI, Inc., No. 2:14-cv-00868-JLR, 2015 WL 1481728, at *7 (W.D. Wash. Mar. 30, 2015) (“If the insurer shows that an exclusion applies, the burden shifts back to the insured to demonstrate that the loss fits within an exception to the exclusion.”) (collecting cases). The majority of jurisdictions hold the same. 17A Couch on Insurance §25 4:13 (3d ed. 2021) (“The trend clearly appears, however, to place the burden on insureds to prove that an exception to an exclusion applies to restore coverage[.]”) (collecting cases). The Court predicts that the Washington Supreme Court would join the majority view and place the burden of proving an exception on insureds. ii. Meaning of “Such Loss” Caused by “Physical Damage Not Otherwise Excluded” Next, the Court turns to the meaning of the inherent vice exclusion and its exception. Again, the exclusion provides that if Ingenco’s loss was caused by an inherent vice, then its loss is excluded from coverage. Dkt. # 244 at 7-8. On the other hand, the exception provides that if “such loss” was caused by “physical damage not otherwise excluded,” then the exclusion does not apply. Id. The parties do not dispute what it means for a loss to be “caused by” or “result[] from” an inherent vice. Nor do they dispute the meaning of “inherent vice.” Thus, the meaning of the exclusion is plain. A loss caused by an inherent vice is excluded under the policy. But the parties seriously disagree about the meaning of the exception to the exclusion. What does it mean for the exclusion not to apply if “such loss” was caused by “physical damage not otherwise excluded”? The parties offer two competing interpretations. Before explaining those interpretations, the Court provides a brief clarification. The phrase “physical damage not otherwise excluded” is synonymous with “covered peril.” The parties’ policy is an all risks policy. Such policies “cover[] any peril that is not specifically excluded in the policy.” Nw. Bedding Co. v. Nat’l Fire Ins. Co. of Hartford, 225 P.3d 484, 487 (Wash. 2010); Findlay v. United Pac. Ins. Co., 917 P.2d 116, 121 (Wash. 1996) (“In an all-risk homeowners insurance policy, any peril that is not specifically excluded in the policy is an insured peril.”). Hence, under the parties’ policy, perils specifically excluded are “excluded perils,” while perils not specifically excluded are “covered perils.” Here, a loss caused by an inherent vice is excluded, so inherent-vice loss is an excluded peril. Dkt. # 244 at 7-8. The exception, however, mentions “physical damage not otherwise excluded.” Id. Under an all risks policy, physical damage not otherwise excluded is covered, making such damage a covered peril. Thus, the Court treats the phrase “physical damage not otherwise excluded” as equivalent to “covered peril.” Returning to the parties’ arguments, according to Ingenco, the exception requires ACE to show that an inherent vice was the “sole cause” of Ingenco’s loss. Dkt. # 243 at 21-23; Dkt. # 249 at 15-19. The exclusion does not apply, Ingenco argues, when another “non-excluded event” “contributes to the loss.” Dkt. # 249 at 17. A “non-excluded event,” as just explained, would be a covered peril under the parties’ all risks policy. Hence, in Ingenco’s view, if a covered peril contributed to its loss, then the exclusion would not apply. As a result, to prove the exclusion, ACE would have to prove that an inherent vice, and no covered peril, caused Ingenco’s loss. ACE has a different view of the inherent vice exclusion. ACE argues that if Ingenco suffered a loss that was caused by an inherent vice (an “inherent-vice loss”), then that loss would not be covered under the exclusion. But if the inherent-vice-loss was, itself, caused by a covered peril, then the inherent-vice loss would indeed be covered, thanks to the exception. Dkt. # 251 at 10-13. The Court agrees with ACE’s interpretation. That interpretation prevails because, unlike Ingenco’s, it respects the parties’ burdens of proof and considers the policy as a whole. What is more, ACE’s interpretation is confirmed by Ingenco’s own cited authority. iii. Ingenco’s “Sole Cause” Interpretation To start, Ingenco’s interpretation fails to account for the parties’ respective burdens of proof. Ingenco acknowledges that the exception is indeed an exception. It also acknowledges that, under the general rule, the burden “falls to insureds to prove exceptions.” Dkt. # 249 at 15-19. Yet its interpretation assumes that it is up to ACE, the insurer, to disprove the exception. Ingenco does not explain why. It only says that this rule “does not apply to the subject exclusion” because of how the “language operate[s].” Id. How the language operates, however, does not change the rule that exceptions are on insureds to prove. Nor does it change the fact that this exception is, without dispute, an exception. For this independent reason, Ingenco’s interpretation fails. Burdens of proof aside, Ingenco’s interpretation is not reasonable because it does not consider the policy as a whole. Surely, Ingenco’s interpretation has intuitive appeal. Under the inherent vice exclusion, the “[p]olicy does not insure against loss, damage or expense caused by or resulting from: . . . inherent vice.” Dkt. # 244 at 7-8. But under the exception, the exclusion does not apply if “such loss is caused directly by physical damage not otherwise excluded in this Policy to the property covered.” Id. On first read, that might result in the following: If Ingenco’s loss was caused by an inherent vice, then Ingenco’s loss would be excluded. But if Ingenco’s loss was also caused “by physical damage not otherwise excluded” (hence, a covered peril), then the inherent vice exclusion would be negated. In theory, then, Ingenco could defeat the inherent vice exclusion merely by showing that a covered peril, at least in part, caused its loss, even if the loss was also caused by an inherent vice, an excluded peril. Under this reading, for ACE to prevail, it would have to show that an inherent vice was the sole cause of Ingenco’s loss. But that reading would render the entire exclusion meaningless. To see why, take the following: Coverage is a two-step analysis. Step one, Ingenco must show coverage. To pass step one, Ingenco must show that its loss was a covered peril. Dkt. # 244 at 5. Under an all risks policy, it may do so by proving that its loss was the result of a risk not explicitly excluded under the policy, that is, a covered risk. Id. Only if Ingenco can prove that its loss was the result of a covered risk, and thus a covered peril, would the Court proceed to step two. In step two, the burden would then shift to ACE to show that the inherent vice exclusion applied, thereby avoiding coverage. Afterwards, Ingenco could reinstate coverage by proving the exception to the inherent vice exclusion. In its view, Ingenco could prove the exception by showing that, even if an inherent vice in fact caused its loss, a covered risk also caused its loss. So long as a covered risk “contributed to the loss,” it says, the exclusion would not apply. Dkt. # 243 at 21-22. That view, however, nullifies the exclusion. Say the parties met their respective burdens under steps one and two, and all that was left was to determine whether the exception applied. By that time, Ingenco would have already proven that its loss was caused by a covered risk. It would have done so in step one. Indeed, the Court could not have proceeded to step two had Ingenco not. In Ingenco’s view, it could prove the exception by proving that a covered risk, at least partially, also caused its loss. But that would merely be returning to step one. By the time Ingenco, ACE, and the Court reach the exception, Ingenco would have already proven that its loss was the result of a covered risk. Were Ingenco’s proof in step one automatically sufficient to defeat ACE’s proof in step two, as Ingenco seems to think, then there would be no need to go to step two at all. Under Ingenco’s interpretation, the exclusion would become a nullity. Because this interpretation does not consider the policy as a whole and would render express language meaningless, it is unreasonable. iv. ACE’s Interpretation ACE’s interpretation, on the other hand, is reasonable. According to ACE, the inherent vice exclusion and exception operate as follows: The policy contains an exclusion for loss caused by an inherent vice. Thus, if Ingenco’s loss was caused by an inherent vice, then ACE would not owe Ingenco coverage. But if the inherent-vice loss was—itself—caused by “physical damage not otherwise excluded” (a covered peril), then the exception would negate the exclusion. Put simply, if an excluded peril was itself caused by a covered peril, then the excluded peril would be covered. To support this interpretation, the Court need look no further than Ingenco’s own cited authority. Ingenco cites several cases to support its proposition that “[t]he [inherent vice] exclusion does not apply when any other, non-excluded event contributes to the loss.” Dkt. # 249 at 17. But these cases favor ACE’s interpretation, not Ingenco’s. One illustrative case is Allianz Insurance Co. v. RJR Nabisco Holdings Corp., 96 F. Supp. 2d 253 (S.D.N.Y. 2000). There, a food products manufacturer, Nabisco, obtained an all risks insurance policy. Id. at 254. The policy contained an exclusion for loss caused by contamination. Id. The contamination exclusion, like here, contained an exception for “such loss” that “result[ed] from a peril not otherwise excluded.” Id. Nabisco hired a company to store its food products in the company’s warehouse. Id. Later, customers complained that their Nabisco products had a chemical odor and flavor. Id. After investigating, Nabisco discovered that the warehouse contained chemicals that had contaminated the food products. Id. Nabisco destroyed the contaminated products and sought coverage from its insurer. Id. at 254-55. Raising the contamination exclusion, the insurer denied coverage, and Nabisco sued. Id. at 255. The parties agreed that Nabisco’s losses were caused by contamination, an excluded risk. Id. But they disagreed about whether the exception applied. The court explained that the exception was susceptible to only one reasonable reading: “contamination losses resulting from perils otherwise covered by the policy are likewise covered and contamination losses resulting from perils expressly excluded by the policy are likewise excluded.” Id. In other words, if the losses caused by contamination, themselves, resulted from covered perils, they too would be covered under the exception. Conversely, if the losses caused by contamination resulted from excluded perils, they too would be excluded. That is precisely the conclusion that this Court reaches here. If Ingenco’s inherent-vice losses were caused by covered perils, they too would be covered; if its inherent-vice losses were caused by excluded perils, they too would be excluded. Yet another one of Ingenco’s cited cases, Adams-Arapahoe Joint Sch. Dist. No. 28-J v. Cont’l Ins. Co., 891 F.2d 772 (10th Cir. 1989), supports this interpretation. There, an all risks policy contained an exclusion for loss caused by corrosion. Id. at 773-74. That corrosion exclusion, like here, contained an exception for “such loss” that “result[ed] from a peril not excluded in th[e] policy.” Id. Later, the insured’s roof collapsed because of corrosion damage. Id. at 734. The insured sought coverage, the insurer denied the claim under the corrosion exclusion, and the insured sued. Id. The Tenth Circuit later held that the exception applied. Id. at 777. It applied because the corrosion damage, though an excluded peril, was itself the result of a “covered risk,” namely defective design and construction. Id. The exception, the court reasoned, meant that if an excluded cause was itself caused by a “non-excluded cause,” then the corrosion exclusion would be negated. Id. So, too, here. If an inherent vice (excluded cause) was caused by a covered peril (non-excluded cause), then the inherent vice exclusion would be negated. These two cases illustrate the exact interpretation that ACE advances: “where the excluded cause of loss results from a covered cause of loss, the loss is covered.” Dkt. # 251 at 12. Put another way, if an “excluded loss resulted from a covered risk,” then the excluded loss would also be covered under the exception. Id. at 11-12. These two cases do not, as Ingenco purports, stand for the proposition that a covered peril must be the sole cause of Ingenco’s loss. Given the facts here, if Ingenco’s loss was caused by an inherent vice, then it would be an excluded loss under the inherent vice exclusion. But Ingenco may negate that exclusion if it can prove that the inherent-vice loss was, itself, caused by a covered peril. Because the inherent vice exception has only one reasonable interpretation, it is not ambiguous. And the Court interprets the exception as set forth above. F. Issue 5 – Defective Material Exclusion and Ensuing Loss The parties’ all risks policy contains an exclusion for “defective material” and “faulty workmanship.” Dkt. # 244 at 7. This exclusion, referred to as the defective material exclusion, also contains an “ensuing loss” provision. This Policy does not insure against loss, damage or expense caused by or resulting from: . . . [f]aulty or defective material, faulty workmanship, faulty methods of construction, errors or omissions in plan or specification design or errors in processing, unless loss by a peril not otherwise excluded ensues, and then only for the ensuing loss. Id. The parties ask the Court to interpret this exclusion. Dkt. # 219 ¶ 5. Specifically, both parties offer different coverage and exclusion theories that they want the Court to endorse. Dkt. # 241 at 13-16; Dkt. # 243 at 23-25. To be sure, the Court does not interpret the defective material exclusion in isolation. The Ninth Circuit, on appeal, has already interpreted the exclusion. 921 F.3d at 818-20. The Court begins by summarizing the Ninth Circuit’s holding. Then, it addresses each of the parties’ coverage and exclusion theories in turn. i. Ensuing Loss Exception on Appeal The Ninth Circuit began its analysis by explaining how ensuing loss exclusions work under Washington law. 921 F.3d at 818-20. “[E]nsuing loss clauses ensure that, where an uncovered event takes place, any ensuing loss which is otherwise covered by the policy remains covered, even though the uncovered event itself is never covered.” Id. (citing Vision One, LLC v. Philadelphia Indem. Ins. Co., 276 P.3d 300 (Wash. 2012)). The “dispositive question” is whether a loss that ensues from an excluded event is, itself, covered or excluded. Id. If the ensuing loss is itself an excluded peril, then there is no coverage for the loss. Id. Conversely, if the ensuing loss is not excluded, then there would be coverage. Id. Turning to the parties’ arguments, the Ninth Circuit identified two losses: Ingenco’s “bead-related loss” and “loss of the V32 diffuser shield.” Id. Ingenco conceded that loss of the diffuser shield was not covered. Id. It argued, however, that coverage would still obtain. Id. The loss of the diffuser shield occurred before the bead- related loss, making the bead-related loss an “ensuing loss,” Ingenco said. Id. Hence, even if the loss of the diffuser shield was excluded, that would not hinder coverage for the beads. Id. Damages to the beads would be covered under the ensuing loss provision, even though failure of the diffuser shield would be excluded. Id. The Ninth Circuit agreed with Ingenco. It said that, even if the loss to the adsorbent beads indeed ensued from some excluded event—for example, because the beads were damaged by a defectively designed diffuser basket—the loss to the beads would still be covered under the ensuing loss provision. Id. The court’s reasoning was simple: Assuming that loss to the basket was excluded, loss to the beads would be an ensuing loss. Id. And, because the parties’ all risks policy does not contain a “specific exclusion regarding the loss of the beads,” bead-related loss would be a covered peril under the ensuing loss exception. Id. “Thus, even if it were conclusively established that the diffuser shield suffered from some inherent defect,8 the subsequent destruction of the adsorbent beads would be covered under the policy’s ensuing loss exception.” Id. Still, the court did not completely resolve the issue. In theory, ACE could argue that “the beads themselves suffered from an inherent defect,” and thus loss to the beads would, in fact, be an excluded peril. Id. at 819 n.19. To the extent ACE pursued that

8 The parties’ all risks policy contains several exclusions, including an exclusion for “faulty or defective material” (¶ b) and a separate exclusion for “inherent vice” (¶ f). Dkt. # 244 at 7-8. This Order analyzes both. See supra Sections IV.D, IV.E, & IV.F. In this section, however, the Court analyzes the defective material exclusion. Specifically, this section analyzes the possibility for a defective diffuser basket or defective adsorbent beads to be subject to the defective material exclusion. When it analyzed the defective material exclusion, the Ninth Circuit made several references to “inherent defects.” 921 F.3d at 819 & n.19. This should not be confused with this Court’s earlier analysis of “inherent vices.” See supra Section IV.D.ii. There, the Court suggested that “inherent vices” and “inherent defects” were interchangeable. To be clear, however, the Court’s analysis in that section arose under the inherent vice exclusion, an entirely separate exclusion from the defective material exclusion analyzed here. This distinction is significant. The defective material exclusion has an ensuing loss provision; the inherent vice exclusion does not.

That said, the Court does not rule out the possibility that a defective diffuser basket and defective adsorbent beads could be subject to both exclusions. The Court only means to reconcile the terminology used by the Ninth Circuit and this Court. theory, the court said, “there is a triable issue of fact regarding the integrity of the beads.” Id. The Ninth Circuit’s ruling in hand, the parties offer legal conclusions that they want this Court to accept. They both advance coverage and exclusion theories that they believe can be decided as a matter of law. ii. Ingenco’s Defective Beads Theory According to Ingenco, the ensuing loss provision all but guarantees coverage for the shutdown of the Cedar Hills facility. Even if ACE can prove that the beads were inherently defective, Ingenco says that it would still be entitled to coverage. The ensuing loss provision, it says, comes to its aid yet again. Ingenco’s theory goes: If ACE can prove that the adsorbent beads were defective, then loss to the beads would be excluded. Dkt. # 243 at 24. But that would only relieve ACE of the duty to replace the beads. Id. The Court would then have to conduct yet another ensuing loss analysis. Ingenco says that, if the beads were defective, then ACE would still be liable for “dust clogging” because such clogging would be an ensuing loss. Id. Recall that, in Ingenco’s view, after the diffuser basket failed, the incoming gas pulverized the adsorbent material into dust, which over time clogged the whole system, causing the shutdown of the Cedar Hills facility. Dkt. # 25 ¶¶ 21, 54-55. This dust clogging, Ingenco argues, would be distinct from the inherently defective beads. Dkt. # 243 at 24. And because the policy does not specifically exclude dust clogging, clogging would be a distinct peril insured under the ensuing loss provision. Id. Thus, Ingenco argues that even if ACE can prove that the beads were defective, ACE would still be liable for the shutdown of the Cedar Hills facility. Id. This theory fails on summary judgment because it raises questions of fact. The Ninth Circuit treated “bead-related loss” and “loss of the V32 diffuser shield” as distinct perils. But it is not clear that the same would be true for “defective adsorbent beads” and “dust clogging.” Remember, the Court here is addressing a new hypothetical theory, one in which ACE proves that the “excluded event” is the failure of the defective beads. The Ninth Circuit addressed an entirely different theory, one in which the “excluded event” is the failure of the defective diffuser basket. 921 F.3d at 818-20. On appeal, the defective basket theory assumed that the “excluded event” would be the failure of the defective basket and the “ensuing loss” would be the damage to the adsorbent beads. Id. A defective beads theory, however, presents new variables. Under a defective beads theory, the “excluded event” would be the failure of the defective beads. What the “ensuing loss” would be is an open question. Ingenco says that the “ensuing loss” would be “dust clogging.” Dkt. # 243 at 24. It assumes that defective beads and dust clogging would be distinct perils, one ensuing from the other. But it does not explain why. And this elision is significant. If defective beads and dust clogging are not distinct perils, then the ensuing loss provision might not apply at all. Under a defective beads theory, if ACE indeed proves that the adsorbent beads were defective, then the dust clogging that resulted might not be an “ensuing loss,” but the very excluded loss itself. At least one Washington Supreme Court case suggests that defective beads and dust clogging would not be distinct perils. In Sprague v. Safeco Insurance Co. of America, 276 P.3d 1270 (Wash. 2012), a couple bought a home and insured it. They later installed a deck system. Id. at 1271. The deck system was held up by supports, known as “fin walls.” Id. More than a decade later, they discovered that the fin walls “were in an advanced state of decay,” caused by “inadequate flashing and other construction defects that caused the supports to rot.” Id. The homeowners submitted a claim to their insurer, characterizing their damage as “collapse.” Id. The insurer, in turn, denied the claim because the policy contained exclusions for “construction defects” and “rot damage.” Id. The state high court sided with the insurer: “collapse” was not an ensuing loss from “rot” because the two were not separate perils. Id. at 1272. The court explained that “rot typically results in the complete deterioration of the rotting material,” and in the case of a wooden structure, that “natural process” results in collapse. Id. at 1272-73. Advanced deterioration does not transmute the rotting process in some sort of alchemical fashion to a new and separate state of “collapse.” A “collapse,” whether consisting of a loss of structural integrity or a plunge to the earth, is the end result of the deterioration that constitutes “rot.” It is not a new and different peril. Id. at 1273. Here, the Court cannot say as a matter of law that “defective beads” and the resultant “dust clogging” would be distinct perils. The Court simply does not know the “natural process” that flows from defective beads. Nor does it know whether defective beads “typically result” in dust clogging. Perhaps, like “rot” and “collapse,” “dust clogging” is simply the “end result” of defective beads. Resolving this issue requires facts, making the matter unfit for summary judgment. In sum, the Court rejects Ingenco’s request. The Court cannot conclude as a matter of law that, were ACE able to prove that the adsorbent beads were defective and loss to the beads excluded, “dust clogging” would be a distinct peril covered as an “ensuing loss.” iii. ACE’s Defective Replacement Basket Theory ACE offers a new theory of its own. Dkt. # 241 at 13-16. To explain that theory, the Court offers some context. Recall that, on October 1, 2010, the bottom metal piece of the diffuser basket in V- 32 broke off. Dkt. # 87 at 7. Within days, Ingenco shut the plant down and discovered that adsorbent beads had been pulverized to dust. Dkt. # 25 ¶ 18. Ingenco removed the failed basket and replaced it with a redesigned basket. Dkt. # 87 at 8. About a week later, Ingenco resumed operations. Dkt. # 25 ¶ 25. Many months later, Ingenco shut the plant down again because the dust generated from the October 1 basket failure infiltrated other parts of the gas purification system. Id. ¶ 54. Initially, ACE argued that the original diffuser basket was defective. See Dkt. # 87 at 13-14. But, the Ninth Circuit said, even if that were so, that would not relieve ACE from providing coverage for damage to the adsorbent beads, which would be covered under the ensuing loss exception. ACE now tries a new theory. It says that not only was the original diffuser basket defective, so was the replacement basket. Dkt. # 241 at 13-16. To be sure, ACE does not dispute that it would be liable for damage to the adsorbent beads caused by the failure of the first diffuser basket. Such damage is, indisputably, a covered ensuing loss. But, its theory goes, if ACE can prove that the replacement basket was also defective, it would not be liable for any damage to the beads that occurred after the replacement. In that case, it says, any damage to the beads that resulted from the defective replacement basket would not be an “ensuing loss.” Rather, post-replacement damage to the beads would be the excluded loss itself. Dkt. # 254 at 9-10. It would appear obvious, at first, for this new replacement basket theory to go the same way as ACE’s original basket theory. But the replacement basket theory presents a new wrinkle. Unlike the original basket, the replacement basket did not break off. Dkt. # 254 at 10. Indeed, the replacement basket, however defective, suffered no damage. This distinguishes the second theory from the first. In the first theory, there were two losses: “bead-related loss” and “loss of the V32 diffuser shield.” The second theory, on the other hand, presents only one loss: bead-related loss resulting from the faulty replacement basket. ACE suggests that, because there is only one loss in the second theory, there is no “ensuing loss” to the beads. The only loss would be the loss to the beads caused by the defective replacement basket. That loss, ACE argues, would be excluded under the defective material exclusion. The Court rejects this theory because, even accepting that there was no damage to the replacement basket, the damage to the post-replacement beads would still be an “ensuing loss.” In short, the Ninth Circuit’s reasoning applies as much to ACE’s original theory as it does to its new theory. Both defective basket theories share the same material facts. Under both theories, the baskets (original and replacement) were defective, and the adsorbent beads were damaged as a result. Under both theories, the failure of the defective baskets would be “excluded events” and therefore “never covered.” 921 F.3d at 818-19. Under both theories, the damage to the adsorbent beads (whether caused by the original basket failure or the replacement basket failure) would be “ensuing losses” because “there was no specific exclusion regarding the loss of the beads.” Id. at 819. That the replacement basket did not break off or suffer any damage is immaterial. Were there, in fact, damage to the defective replacement basket, then ACE would not be liable for that damage. But it does not follow that, because there was no damage to the defective replacement basket, the Court must make use of the defective material exclusion somehow and search for another loss to exclude. ACE provides no authority for that approach. And that approach would only inject confusion into the Ninth Circuit’s ensuing loss ruling. For the reasons stated above, the Court GRANTS in part and DENIES in part the parties’ cross motions for partial summary judgment. Dkt. ## 241, 243.

DATED this 10th day of March, 2022. A The Honorable Richard A. Jones United States District Judge