Allstate Insurance Company and Granite Mutual Insurance Company v. Hamilton Beach/proctor Silex, Inc., Docket No. 04-6282-Cv

HALL, Circuit Judge. Allstate Insurance Company and Granite Mutual Insurance Company (collectively “Plaintiffs”) brought this subrogation action against Hamilton Beach/Proctor Si-lex, Inc. (“Hamilton Beach” or “Defendant”) seeking to recover approximately $97,000 paid to their respective insureds, Joseph Malboeuf and Michael and Gail Leggett, for covered losses sustained in a residential fire in St. Albans, Vermont. Plaintiffs alleged that a defective coffee *452 maker manufactured by Hamilton Beach caused the fire and asserted claims for products liability and breach of express and implied warranties. 1 Hamilton Beach moved for summary judgment arguing that Plaintiffs could not establish a defective condition in the coffee maker, an essential element of both claims. Plaintiffs argued that they had produced sufficient circumstantial evidence to show the coffee maker was defective to preclude summary judgment as to both claims. With respect to their strict products liability claim, Plaintiffs urged the District Court to adopt the Restatement (Third) of Torts: Products Liability § 3 (1998) (“Restatement § 3”), which under certain conditions permits recovery on a strict liability claim where there is only circumstantial evidence of a defect. With respect to their breach of warranty claim, Plaintiffs argued that Vermont law allows recovery where circumstantial evidence establishes that a defect in a product is the most likely cause of injury. Plaintiffs asserted, therefore, that summary judgment on their breach of warranty claim would be inappropriate regardless of whether the District Court adopted the malfunction theory. In his Report and Recommendation, Magistrate Judge Niedermeier determined Plaintiffs’ circumstantial evidence was not sufficient to show that a defect in the coffee maker was the more probable cause of the fire when compared to all other possible causes. Magistrate Judge Nied-ermeier declined, therefore, to consider whether the Supreme Court of Vermont would adopt the malfunction theory and recommended granting Hamilton Beach’s motion for summary judgment in its entirety. The United States District Court for the District of Vermont (Murtha, /.) adopted the Report and Recommendation without modification and dismissed the complaint. Plaintiffs appeal. BACKGROUND In May 2002, Malboeuf purchased from Ames Department store a coffee maker manufactured by Hamilton Beach, brought it home and placed it, still packaged, on his kitchen floor. The coffee maker remained there until the night of June 13, 2002, when Malboeuf removed it from its packaging and set it up. The following morning, Malboeuf used the coffee maker for the first time. Before leaving for work, he turned it off, but did not unplug it. Less than three hours later, a neighbor saw flames coming from Malboeufs home and called the St. Albans fire department. Although the fire department arrived just two minutes later and promptly brought the fire under control, it had caused substantial damage to both Malboeuf s property and that of his tenants, the Leggetts. Gary Palmer, St. Albans’s fire chief, conducted an initial investigation into the cause of the fire. Based on that investigation, he determined that the fire started to the right of the stove, where Malboeuf claims the coffee maker was located prior to the fire. Palmer ruled out the possibility that the fire was the result of arson or careless smoking. He did not, however, offer any theory of how or why the coffee maker started the fire. On June 17, 2002, David Eliassen, a cause and origin investigator retained by Allstate, visited the scene to undertake his own investigation. Eliassen noted that the coffee maker had been reduced to very small pieces and there was a very heavy char pattern on the splash board behind *453 the coffee maker. According to Eliassen, this char pattern — shaped like a “V” with the lowest point of heavy char directly behind the coffee maker — indicated that the fire originated in that area. While Eliassen testified in his deposition that there were three other potential electrical sources of ignition in the area — the electric range, the range hood, and the electric receptacle behind the coffee maker — he did not believe that the burn patterns were consistent with a fire in either the electric range or the range hood. Nonetheless, Eliassen recommended that Allstate retain an electrical engineer to examine each of these alternate sources of ignition so that they could be definitively ruled out. On June 28, 2002, Eliassen returned to the scene to continue his investigation. He was accompanied by Eric Chaine, the electrical engineer retained by Allstate on Eliassen’s recommendation, and Charles King, a fire investigator representing Hamilton Beach. King inspected the scene and spoke with Malboeuf. According to Eliassen, when he explained to King that one of the purposes for the visit was to examine closely the electric range and range hood, King stated that he could see that those items did not cause the fire and did not stay for that examination. Based on that interaction, Eliassen concluded that Hamilton Beach had “no interest in preserving the range or the range hood.” Chaine, however, went forward with the examination on site and also took the remnants of the coffee maker back to his laboratory for further study. Eliassen and Chaine both submitted reports detailing the results of their respective investigations. Eliassen concluded that the “fire had a single point of origin at the ... coffeemaker,” but deferred to Chaine to identify a specific failure mode within the coffee maker. Eliassen based his conclusion on the lack of any evidence of arson or accidental ignition, as well as his professional opinion that “all ignition sources, except for the ... coffeemaker ha[d] been considered and ruled out.” Chaine’s report also ruled out the range, the range hood, and the receptacle as potential causes of the fire. The report noted that the plastic housing of the coffee maker had been consumed completely by the fire, leaving only “the bottom base plate, the heating element assembly (including the warmer plate), some remnants of the carafe, and pieces of stranded wires.” Despite this destruction, Chaine was able to identify the basic components of the coffee maker which were “still in a fair condition, with no signs of failure.” Chaine also observed that the wire strands probably came from the power cord and indicated multiple points of electrical arcing. This arcing, he concluded, was likely the result of a cord failure and the most probable cause of the fire. King died before filing a report or being deposed in connection with this action. To replace King, Hamilton Beach retained Scott Barnhill. Based on his investigation, Barnhill stated that he was “comfortable ruling out the range, the receptacle, [and] the hood....” In an apparent reference to Plaintiffs’ disposal of several component parts of these alternate sources following King’s site visit, however, Barnhill suggested that his conclusion might have been different had he had an opportunity to examine additional evidence. Barnhill also eliminated the coffee maker as a potential cause of the fire. Disputing Chaine’s conclusion, he asserted that any electrical arcing on the coffee maker’s power cord would not have been strong enough to ignite the coffee maker’s housing. Hamilton Beach moved to preclude Chaine’s testimony regarding the coffee maker’s alleged mode of failure. Plaintiffs agreed to limit Chaine’s testimony “to his *454 examination of the range, range hood and receptacle, and his elimination of these items as potential causes of the fire.” Thereafter, Hamilton Beach moved for summary judgment, arguing that Plaintiffs could not establish that the coffee maker was defective. Plaintiffs conceded that they could not identify a specific defect in the coffee maker. Relying on expert reports and testimony eliminating the other possible causes of the fire, however, they argued they had produced sufficient circumstantial evidence that the coffee maker was defective to withstand summary judgment on both claims. With respect to their strict liability claim, Plaintiffs’ argument assumed that the District Court would apply Restatement § 3, often referred to as the “malfunction theory.” The Vermont Supreme Court has not yet issued an opinion that adopts the malfunction theory, which allows a plaintiff to use circumstantial evidence to demonstrate that a product was defective where the incident that harmed the plaintiff: “(a) was of the kind that ordinarily occurs as a result of a product defect; and (b) was not, in the particular case, solely the result of causes other than product defect.” Restatement § 3. The malfunction theory relieves plaintiffs of the heavier burden of establishing a specific defect in a product. With respect to its breach of warranty claim, Plaintiffs asserted they had at least created a material issue of fact because existing Vermont law permits a plaintiff to use circumstantial evidence to prove that a defect in a product is the most likely cause of the injury. Thus, Plaintiffs argued that summary judgment was inappropriate even if the District Court declined to adopt the malfunction theory. In his Report and Recommendation, Magistrate Judge Niedermeier recommended granting Hamilton Beach’s motion for summary judgment. The recommendation was supported by two central determinations. First, Magistrate Judge Nied-ermeier refused to allow Plaintiffs to make use of their evidence that ruled out all sources of ignition aside from the coffee maker, reasoning that due to Plaintiffs’ failure to preserve several components of the potential alternate sources of ignition for examination by Hamilton Beach, it would be inequitable for Plaintiffs to make use of such circumstantial evidence or avoid the burden of proving a specific defect in the coffee maker. Second, he found that Plaintiffs had not presented any evidence that the coffee maker could not have been damaged in transit from the store after Malboeuf purchased it, or while it sat on Malboeufs floor before he used it. On the basis of those findings, Magistrate Judge Niedermeier concluded it was not necessary to consider whether the Vermont Supreme Court would adopt the malfunction theory, because Plaintiffs’ evidence could not satisfy it. Therefore, he recommended dismissing Plaintiffs’ strict products liability claim. With respect to the breach of warranty claim, Magistrate Judge Niedermeier noted that Vermont law permits a plaintiff to establish the existence of a defect through circumstantial evidence by demonstrating that a defect is the more probable cause of the injury when compared to other possible causes. Referring to his earlier finding that Plaintiffs had not produced any evidence that the coffee maker was not damaged after it was purchased by Mal-boeuf, he concluded that no such showing could be made in this case and recommended the dismissal of the breach of warranty claim as well. Plaintiffs filed timely objections to the Report and Recommendation, and Hamilton Beach filed a timely response. On November 4, 2004, *455 the District Court (Murtha, J.) adopted the Report and Recommendation without modification. On appeal, Plaintiffs challenge the two determinations underlying the District Court’s judgment. Plaintiffs argue here that the District Court erred in finding that they did not provide Hamilton Beach an opportunity to examine the other potential ignition sources and that dismissing their strict products liability claim on that basis constituted an inappropriate spoliation sanction. In addition, Plaintiffs claim that the District Court erroneously drew a factual inference in favor of Hamilton Beach when it held that the coffee maker may have been damaged in transit or while it sat on Malboeuf s floor. According to Plaintiffs, the record contains circumstantial evidence sufficient to allow a jury reasonably to conclude that an unspecified defect in the coffee maker was the more probable cause of the fire and that the coffee maker was defective while in the possession and control of Hamilton Beach. Plaintiffs argue that the District Court erred by refusing to consider whether the Vermont Supreme Court will adopt some form of the malfunction theory, which would support a recovery on their strict products liability claim. In addition, Plaintiffs argue that Vermont law on breach of warranty requires that those claims be decided by a jury. We agree that both determinations underlying the District Court’s entry of summary judgment were in error. Construing the record in the light most favorable to Plaintiffs, as we must, we find there is circumstantial evidence sufficient to allow a jury reasonably to find: (1) that a defect in the coffee maker was the more probable cause of the fire; and (2) that the coffee maker was in substantially the same condition as it was when last in Defendant’s possession and control. We, therefore, hold the District Court erred in dismissing Plaintiffs’ breach of warranty claim. For the same reasons, and in light of Travelers Ins. Cos. v. Demarle, Inc., 178 Vt. 570, 574, 878 A.2d 267, 272 (Vt.2005), where the Vermont Supreme Court suggested that under Vermont law a plaintiff may rely on circumstantial evidence to establish causation in strict products liability actions in the same manner as he or she may in breach of warranty actions, 2 we hold that the District Court also erred in dismissing Plaintiffs’ strict products liability claim. DISCUSSION I. Standard of Review We review a district court’s grant of summary judgment de novo. Bouboulis v. Transp. Workers Union of Am., 442 F.3d 55, 59 (2d Cir.2006). “Summary judgment is only warranted upon a showing ‘that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law.’ ” Feingold v. New York, 366 F.3d 138, 148 (2d Cir.2004) (quoting Fed.R.Civ.P. 56(c)); see also Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). A genuine issue of material fact exists, where “there is sufficient evidence favoring the nonmoving party for a jury to *456 return a verdict for that party.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). “In assessing the record to determine whether there is such an issue, the court is required to resolve all ambiguities and draw all permissible factual inferences in favor of the party against whom summary judgment is sought.” Stern v. Trs. of Columbia Univ., 131 F.3d 305, 312 (2d Cir.1997). We review a district court’s imposition of spoliation sanctions under an abuse of discretion standard. West v. Goodyear Tire & Rubber Co., 167 F.3d 776, 779 (2d Cir.1999); Flury v. Daimler Chrysler Corp., 427 F.3d 939, 943 (11th Cir.2005). In so doing, we accept the district court’s factual findings in support of the sanctions unless they are clearly erroneous. West, 167 F.3d at 779. II. Breach of Warranty and Strict Products Liability under Vermont Law Under Vermont law, there are two elements necessary to establish causation with respect to both breach of warranty and strict products liability actions. The first is proof of a product defect. See Hershenson v. Lake Champlain Motors, Inc., 139 Vt. 219, 222, 424 A.2d 1075, 1077 (Vt.1981) (breach of warranty action); see also Webb v. Navistar Int’l Transp. Corp., 166 Vt. 119, 126, 692 A.2d 343, 346 (Vt. 1996) (stating that in strict liability actions the plaintiff must prove that harm or damages resulted from a “defective product”). The second is “proof that a defect existed in the product at the time that it left the possession and control of the defendant.” Hershenson, 139 Vt. at 222, 424 A.2d at 1077 (breach of warranty action); see also Webb, 166 Vt. at 126, 692 A.2d at 346 (stating that the doctrine of strict products liability renders manufacturers liable “for physical harm or property damages resulting from a defective product that reaches a user without undergoing substantial change”). In a breach of warranty action, Vermont law permits an injured party to establish causation by means of circumstantial evidence. Hershenson, 139 Vt. at 223, 424 A.2d at 1078. As the Vermont Supreme Court explained: “Circumstantial evidence may be resorted to ... if there can be drawn therefrom a rational inference that [a defect in the defendant’s product] was the source of the trouble. There must be created in the minds of the jurors something more, of course, than a possibility, suspicion or surmise, but the requirements of the law are satisfied if the existence of this fact is made the more probable hypothesis, when considered with reference to the possibility of other hypotheses.” Hershenson, 139 Vt. at 223-24, 424 A.2d at 1078 (quoting Patton v. Ballam, 115 Vt. 308, 314, 58 A.2d 817, 821 (Vt.1948)) (alteration in Hershenson). In Travelers Ins. Cos., as previously noted, the Vermont Supreme Court suggested that a plaintiff may rely on circumstantial evidence to establish causation in strict products liability actions in the same manner as he or she may in breach of warranty actions. See 178 Vt. at 574, 878 A.2d at 212. 3 *457 In sum, under Vermont law Plaintiffs prevail in their opposition to Defendant’s summary judgment motion if they can establish that: (1) their evidence would allow a jury reasonably to conclude that a defect in the coffee maker was the more probable cause of the fire; and (2) a jury could reasonably infer that the product was defective while still in the possession and control of Hamilton Beach and the defect was not due to any post-purchase mishandling or misuse. Because we hold the District Court should have taken into consideration Plaintiffs’ expert testimony that excluded all possible ignition sources aside from a defect in the coffee maker, and when we then view the record in the light most favorable to Plaintiffs, as we must, we find that the material facts in dispute preclude summary judgment in Defendant’s favor on Plaintiffs’ breach of warranty and strict products liability claims. III. The Exclusion of Plaintiffs’ Circumstantial Evidence On June 14, 2002, the day of the fire, Allstate sent Hamilton Beach a letter notifying it of the potential subrogation claim and offering to preserve the scene for Hamilton Beach’s inspection. Two weeks later, on June 28, 2002, Hamilton Beach sent King as its representative to inspect the fire scene. According to Elias-sen’s testimony, which Hamilton Beach does not challenge, the only items King indicated he wanted preserved were the coffee maker and the receptacle. King had no interest in even inspecting, much less preserving, the range or range hood. Thus, after Plaintiffs took detailed pictures of the range and the range hood and determined that those appliances had not caused the fire, those items were discarded. In evaluating Plaintiffs’ strict products liability claim, the District Court held that it would be inequitable to allow Plaintiffs to make use of their evidence that ruled out all possible sources of the fire other than the coffee maker. Plaintiffs assert that this holding, though not characterized as such by the District Court, amounts to a spoliation sanction. “Spoliation is the destruction or significant alteration of evidence, or failure to preserve property for another’s use as evidence in pending or reasonably foreseeable litigation.” West, 167 F.3d at 779. While acknowledging that a district court may impose such a sanction in its discretion, Plaintiffs argue that the District Court abused that discretion here because Hamilton Beach was provided a full and fair opportunity to examine the fire scene as *458 well as the alternate potential ignition sources. Hamilton Beach argues that Plaintiffs were obliged to preserve the range and range hood despite King’s indication that he was not interested in examining them. Citing our decision in Fujitsu Ltd. v. Fed. Express Corp., 247 F.3d 423 (2d Cir.2001), Hamilton Beach asserts that there is a broad duty to preserve evidence irrespective of the absence of a request by the opposing party. Fujitsu will not bear such a reading. While we noted that “[t]he obligation to preserve evidence arises when the party has notice that the evidence is relevant to litigation or when a party should have known that the evidence may be relevant to future litigation,” id. at 436, we did not hold that such an obligation continues indefinitely. Indeed, we affirmed the district court’s refusal to impose sanctions, in part, because the defendant had never asked to inspect the evidence at issue. Id. Here, not only did the defendant not request that Plaintiffs preserve the range and range hood, Hamilton Beach, through its representative, affirmatively disclaimed any interest in the evidence. Hamilton Beach did so, moreover, after being provided a full opportunity to inspect the items. See Thiele v. Oddy’s Auto and Marine, Inc., 906 F.Supp. 158, 162-63 (W.D.N.Y.1995) (denying a spoliation sanction requested by the primary defendant who had inspected the evidence before destroying it, while granting a spoliation sanction in favor of a third-party defendant who was not afforded a similar opportunity to inspect the evidence), cited with approval in Fujitsu, 247 F.3d at 436; Howell v. Maytag, 168 F.R.D. 502, 505-08 (M.D.Pa.1996) (concluding that sanctions more severe than an adverse inference jury instruction were unnecessary where plaintiff destroyed the scene of the fire, including other possible sources of ignition, before it was inspected by the defendant, but preserved the presumptive cause of the fire); see also Baliotis v. McNeil, 870 F.Supp. 1285, 1290 (M.D.Pa.1994) (noting that “[t]he scope of the duty to preserve evidence is not boundless,” but requires, at a minimum, that the defendant be provided an opportunity for inspection (internal quotation marks omitted)). The District Court, therefore, abused its discretion in precluding consideration of Plaintiffs’ evidence offered for purposes of eliminating the possible alternate ignition sources. If allowed to consider such evidence, a jury could reasonably conclude that a defect in the coffee maker was the more probable cause of the fire in comparison to all other possible causes. See Hershenson, 139 Vt. at 223-24, 424 A.2d at 1078. Hamilton Beach argues the evidence is nonetheless insufficient to allow the jury to infer that a defect in the coffee maker caused the fire because Plaintiffs have failed positively to establish an actual defect in the coffee maker. The two Vermont cases on which Hamilton Beach relies, however, did not involve breach of warranty or products liability claims, were decided well before Vermont adopted strict products liability, and establish only that expert opinions may properly be excluded when based on conjecture rather than evidence in the record. See State v. Teitle, 117 Vt. 190, 206, 90 A.2d 562, 573 (Vt.1952) (upholding exclusion of expert testimony that a malfunction in the motor or wiring of equipment located near the point of origin of a fire possibly caused that fire where the record was devoid of evidence that the wiring or motor was in fact defective, or that such equipment may inherently malfunction even when properly maintained); Bliss v. Moore, 112 Vt. 185, 190, 22 A.2d 315, 317 (Vt.1941) (excluding expert testimony that an electrical overload *459 could have been caused by the motors to an electrical burner, a circulator and an electrical refrigerator coming on simultaneously where there was no evidence in the record on which that opinion was based). Here, in contrast, Plaintiffs’ expert testimony that the coffee maker started the fire was based on evidence — the burn pattern, the condition of the coffee maker, and the condition and location of the other possible ignition sources — all of which should have been considered in connection with the summary judgment motion. While the Vermont Supreme Court has stated that Plaintiffs may rely on circumstantial evidence to prove causation in products liability and breach of warranty actions, it has yet to opine on the kind of circumstantial evidence required to create a jury question regarding the cause of a fire. Cases from other jurisdictions that allow plaintiffs to rely on circumstantial evidence to prove causation lead us to conclude, however, that the evidence Plaintiffs could present, if their case were to go to trial, is sufficient. See, e.g., Gen. Accident Fire & Life Assurance Corp. v. N. Am. Sys., 658 N.Y.S.2d 757, 759-60, 240 A.D.2d 920, 921-23 (3d Dep’t.1997) (upholding New York Supreme Court’s determination that a coffee maker was defective and the defect was a substantial cause of a fire based on expert testimony that failed to point to a specific defect in the coffee maker but eliminated all possible sources of ignition other than the coffee maker); Klein v. Gen. Electric Co., 714 S.W.2d 896, 899-900 (Mo.Ct.App.1986) (finding expert testimony that failed to point to any specific defect in a coffee maker, but eliminated all other possible sources of ignition was substantial evidence from which the jury could reasonably conclude that a defect in the coffee maker caused the fire); Cassisi v. Maytag Co., 396 So.2d 1140, 1143, 1152-53 (1981) (reversing grant of defendant’s motion for summary judgment based on plaintiffs expert’s testimony that a clothes dryer was the “ ‘source from which the fire pattern and heat source eminate[d]’ ” even though expert failed to point to a specific defect in the clothes dryer and failed to eliminate all other possible ignition sources). The Vermont Supreme Court’s decision in Webb, 166 Vt. 119, 692 A.2d 343, suggests that the court would not adopt any stricter approach to the evaluation of circumstantial evidence. There, the court stated that the purpose of the doctrine of strict products liability is to “lessen the burden of proof for plaintiffs injured by defective products” in order to motivate “manufacturers to produce safe products” and because “manufacturers are in the best position to spread the cost of injury resulting from defective products by passing it on to consumers as a cost of doing business.” 166 Vt. at 126, 692 A.2d at 346. Those interests would be undermined by an approach to circumstantial evidence that would prevent a plaintiff from recovering where, because a product is damaged so severely by a manufacturer’s defect, the plaintiff is unable to identify the specific defect that caused the fire. We find, therefore, that Plaintiffs’ expert testimony both eliminating all possible sources of ignition other than the coffee maker and opining that the coffee maker was the ignition source, by way of the burn pattern, constitutes circumstantial evidence sufficient to allow a jury reasonably to conclude that a defect in the coffee maker was the more probable cause of the fire. See Hershenson, 139 Vt. at 223-24, 424 A.2d at 1078. At oral argument, Hamilton Beach argued that Plaintiffs’ failure to offer some credible theory of how the coffee maker could have started the fire was fatal to their circumstantial case. Its argument *460 relied on Truck Ins. Exch. v. MagneTek, Inc., 360 F.3d 1206 (10th Cir.2004), in which the Tenth Circuit, applying Colorado law, held that in order to establish through circumstantial evidence that an allegedly defective appliance “caused a fire, not only must one find that the appliance was in the area of origin,” id. at 1216, but also one must “ ‘rule in’ ” the appliance “ ‘as a scientifically plausible cause,’” id. at 1215 (quoting Hollander v. Sandoz Pharm. Co., 289 F.3d 1193, 1211 (10th Cir.2002)). While Truck Insurance Exchange suggests that the evidence Plaintiffs adduced here would, under Colorado law, be insufficient to create a jury question, Webb suggests that Vermont would not adopt so strict an approach. Such an approach would conflict, moreover, with what the Vermont Supreme Court has suggested is required of a plaintiff to prove causation in breach of warranty, see Her-shenson, 139 Vt. at 223-24, 424 A.2d at 1078, and strict products liability cases, see Travelers Ins. Cos., 178 Vt. at 574, 878 A.2d at 272, because it would prevent a plaintiff from surviving a defendant’s motion for summary judgment where the plaintiff adduces enough circumstantial evidence from which a jury reasonably could conclude that a defect in a product was the more probable cause of a fire, but fails to adduce evidence ruling in the product as a scientifically plausible cause. In addition, Truck Insurance Exchange is distinguishable from the case at hand. There, the plaintiffs evidence failed to rule out all other possible sources of ignition, and the defendant presented scientific evidence that the fire could not have been caused by the plaintiffs claimed ignition source. 360 F.3d at 1215. Here, in contrast, Plaintiffs’ expert ruled out all possible sources of ignition other than the coffee maker and testified, based on the burn pattern, not only that the fire originated in the area where the coffee maker was located, but also that the coffee maker itself was the source of ignition. We, therefore, find more persuasive cases from jurisdictions that have evaluated the sufficiency of circumstantial evidence in situations closer factually to the one presented here, and under a standard more in accord with both the principles that motivated the Vermont Supreme Court to adopt strict products liability and the causation standards already adopted by the Vermont Supreme Court. IV. Whether a Jury could Reasonably Infer an Absence of Post-Purchase Misuse or Mishandling As an alternate ground for dismissing Plaintiffs’ strict products liability claim, the District Court held that Plaintiffs had failed to produce any evidence that the defect in the coffee maker, if any, was not caused by misuse or mishandling after it was purchased. That rationale was the sole basis for the District Court’s dismissal of Plaintiffs’ breach of warranty claim. Plaintiffs argue that, in reaching that conclusion, the District Court not only ignored Malboeuf s testimony, but improperly resolved on summary judgment a disputed factual issue. Malboeuf testified that he purchased the coffee maker in mid-to-late May, took it home, and placed it, still in its box, on his kitchen floor. According to Malboeuf, the coffee maker remained on the floor while he finished moving into the house and undertook various renovation projects. On June 13, the day before the fire, he removed the coffee maker from the box, rinsed the glass carafe, and filled the coffee maker with ground coffee. The following morning, Malboeuf plugged in the coffee maker for the first time, brewed a pot of coffee, filled his travel mug, turned off the coffee maker, and left for work. The fire started shortly thereafter. Malboeuf *461 also asserted that he did not drop the coffee maker before or after setting it up, and he denied dunking the coffee maker in water. The District Court implied, and Hamilton Beach argues, that the coffee maker could have been damaged while it sat on the floor in the midst of Malboeufs renovation projects. To be sure, that is an inference that might reasonably be drawn from the evidence. The District Court erred, however, in drawing that inference on summary judgment because it favors the moving party. See Stem, 131 F.3d at 312. The error was compounded by the court’s disregard of Malbeouf s testimony which, when considered in a light most favorable to Plaintiffs, is sufficient to create a triable issue of fact as to whether his post-purchase handling and use of the coffee maker caused the alleged defect. V. Whether a Jury could Reasonably Infer that the Coffee Maker was Defective When in Defendant’s Possession and Control Hamilton Beach also argues that while Plaintiffs may have produced evidence that the defect was not caused by post-purchase misuse or mishandling, they have nonetheless failed to trace the defect to Hamilton Beach because they have failed to produce any evidence eliminating the possibility that the coffee maker was damaged while in transit from Hamilton Beach to Ames or while in Ames’s possession. Hamilton Beach argues that this deficiency precludes recovery under either of Plaintiffs’ claims and, with regard to strict liability, that we may affirm the grant of summary judgment without determining whether Vermont would adopt the malfunction theory. As stated above, under Vermont law “proof that a defect existed in the product at the time that it left the possession and control of the defendant” is a necessary element of both breach of warranty and strict products liability actions. Hershen-son, 139 Vt. at 222, 424 A.2d at 1077. The Vermont Supreme Court, however, has not had occasion to address what evidence an injured party must produce to trace a defective condition to a defendant where the product, as is often the case, was not transferred directly from the defendant to the injured party. The courts in other jurisdictions that have addressed the issue have reached varied conclusions. Several courts have required the injured party to produce evidence showing that there was “no reasonable opportunity for the [product] to have been tampered with” and that it “was carefully handled by all those who obtained possession or control over it after it left the hands of the defendant.” Barbean v. Roddy Mfg. Co., 431 F.2d 989, 991, 994 (6th Cir.1970) (applying Tennessee law); see also Kerr v. Coming Glass Works, 284 Minn. 115, 169 N.W.2d 587, 589 (1969). Most other courts, however, have set a lower bar. For example in Mondido v. Cory Corp., 483 F.Supp. 26 (E.D.N.Y. 1979), the plaintiff was injured when the glass carafe from her coffee pot broke. The carafe itself was manufactured by Corning, but it was converted into a coffee decanter and sold by Cory, a separate company. Id. at 27. After both companies were found liable, Corning moved for judgment notwithstanding the verdict arguing that the plaintiff failed to produce evidence sufficient to support the jury’s determination that the carafe was defective when it left Coming’s plant. Id. The court rejected the motion, noting the plaintiffs evidence that Cory converted the Corning carafes into coffee decanters through a regular and orderly process and that this decanter remained in the Cory packaging until shortly before the acci *462 dent. Id.; see also V. Mueller & Co. v. Corley, 570 S.W.2d 140, 143 (1978) (finding that plaintiff successfully traced the defect to the manufacturer through evidence that the product was stored in the manufacturer’s sealed container and appeared to be in good condition when it was removed); McKisson v. Sales Affiliates, Inc., 416 S.W.2d 787, 792 (Tex.1967) (“When it is shown that the product involved comes in a sealed container, it is inferable that the product reached the consumer without substantial change in the condition in which it was sold.”); Shoshone Coca-Cola Bottling Co. v. Dolinsky 82 Nev. 439, 420 P.2d 855, 858 (1966) (holding that a plaintiff need not prove that there was no reasonable opportunity for tampering, but only introduce evidence suggesting that the defect existed while in the defendant’s control which would be sufficient for a jury to find an absence of tampering); Kroger Co. v. Bowman, 411 S.W.2d 339, 342 (Ky. 1967) (holding that plaintiff had successfully traced the defective condition to the manufacturer where there was no evidence of tampering by the retailer); Escola v. Coca Cola Bottling Co. of Fresno, 24 Cal.2d 453, 150 P.2d 436, 439 (1944) (“It is not necessary, of course, that plaintiff eliminate every remote possibility of injury to the [product] after defendant lost control, and the requirement is satisfied if there is evidence permitting a reasonable inference that it was not accessible to extraneous harmful forces and that it was carefully handled by plaintiff or any third person who may have moved or touched it.”). In addition, several courts have held that an unspecified defect may be traced to a manufacturer even after the product has been removed from its original packaging and put to use for a short period of time. For example, in Cassisi, 396 So.2d 1140, the plaintiff brought an action based on strict liability, negligence, and breach of implied warranty regarding a fire allegedly caused by a malfunctioning clothes dryer. Id. at 1142^43. There, the plaintiff/owner’s testimony that the dryer had never been serviced or repaired and had been normally operated during its 19-month use, along with expert testimony that an unspecified defect in the dryer was the cause of the fire, constituted evidence of the dryer’s defective condition at both the time of injury and the time of sale sufficient to defeat defendant’s motion for summary judgment. Id. at 1152-53; see also Bailey v. Montgomery Ward, & Co., 6 Ariz. App. 213, 219, 431 P.2d 108, 114 (Ariz.Ct. App.1967) (fact that father took pogo stick out of original packaging to look at it before placing it back in packaging and wrapping it as a Christmas gift did not prevent submission of the case to the jury where product malfunctioned during initial use). Based on the evidence presented in this case, a jury could reasonably find the following facts: the coffee maker was packed in a box with Styrofoam when Malboeuf purchased it; the coffee maker remained in that box until the night before the fire; the coffee maker appeared to be in good condition when he removed it from the box; the coffee maker in question was packaged and sealed by Hamilton Beach in Mexico; and retailers do not usually open the boxes or alter the products in any way. The packaging itself could be considered evidence that the coffee maker was not tampered with, and its general appearance bolsters that conclusion. See Mondido, 483 F.Supp. at 27; McKisson, 416 S.W.2d at 792. These facts would allow a jury reasonably to infer that the product defect that caused the fire existed while the coffee maker was still in Defendant’s possession and control. Again, for the reasons already stated, we believe that under Vermont law the evidence of record on this *463 point creates at least a triable issue of fact so as to preclude summary judgment. VI. Conclusion In sum, Plaintiffs’ circumstantial evidence ruling out all possible ignition sources other than a defect in the coffee maker was erroneously excluded. Because that evidence should be available to be considered, it would permit a jury reasonably to infer that an unspecified defect in the coffee maker was the more probable cause of the fire, thus satisfying the causation requirement for strict products liability and breach of warranty actions under Vermont law. When viewing those facts in a light most favorable to Plaintiffs, moreover, a jury could reasonably infer that the defect in the coffee maker was not due to any post-purchase misuse or mishandling and existed when the coffee maker was still in Defendant’s possession and control. For the foregoing reasons, we vacate the District Court’s entry of summary judgment and remand Plaintiffs’ breach of warranty and strict liability claims for further consideration in light of this opinion. Footnotes [1] Plaintiffs voluntarily withdrew their negligence claim after Defendant moved for summary judgment. [2] In Travelers Ins. Co., the Vermont Supreme Court affirmed the dismissal of, inter alia, breach of warranty and products liability claims with respect to baking mats that the plaintiffs claimed were defective and had contaminated plaintiff Greyston Bakery’s food products. In holding that the plaintiffs had failed to adduce evidence from which a jury could reasonably determine that the baking mats caused the contamination that led to the plaintiffs’ losses, the Court stated in clear language "we agree with plaintiffs that causation in a products liability or warranty case can be proved through circumstantial evidence.” 178 Vt. at 574, 878 A.2d at 272. [3] Plaintiffs argue that the Vermont Supreme Court would adopt the malfunction theory and urge us to so predict and then apply that theory to Plaintiffs’ strict products liability claim. As embodied in Restatement § 3, the malfunction theory provides: It may be inferred that the harm sustained by the plaintiff was caused by a product defect existing at the time of sale or distribution, without proof of a specific defect, when the incident that harmed the plaintiff: *457 (a) was of a kind that ordinarily occurs as a result of product defect; and (b) was not, in the particular case, solely the result of causes other than product defect existing at the time of sale or distribution. It seems likely that the Vermont Supreme Court would adopt the malfunction theory in light of the fact that; (1) circumstantial evidence is treated similarly under both that theory and Vermont’s breach of warranty law; (2) the Vermont Supreme Court has suggested that a plaintiff may rely on circumstantial evidence to establish causation in products liability actions in the same way he or she may in a breach of warranty action, see Travelers Ins. Cos., 178 Vt. at 574, 878 A.2d at 272; and (3) the malfunction theory is consistent with the policy considerations that motivated Vermont to adopt strict products liability. See Webb, 166 Vt. at 129, 692 A.2d at 348. We need not and do not decide, however, whether the Vermont Supreme Court would in fact adopt the malfunction theory, because we find under the causation standards already adopted by the Vermont Supreme Court that Plaintiffs have submitted evidence in opposition to Defendant's motion for summary judgment sufficient to defeat that motion with respect to Plaintiffs’ breach of warranty and products liability claims.