Citation Insurance Company v. Broan-NuTone LLC

Opinion

United States Court of Appeals For the First Circuit

No. 25-1426 CITATION INSURANCE COMPANY, as subrogee of Peter T. Damore, Plaintiff, Appellant,

v.

BROAN-NUTONE LLC; JAKEL MOTORS INCORPORATED, Defendants, Appellees.

APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS

[Hon. Margaret R. Guzman, U.S. District Judge]

Before

Barron, Chief Judge,

Lipez and Rikelman, Circuit Judges.

Mark E. Utke, with whom Cozen O'Connor, Patrick J.

Loftus, III, Katherine C. Loftus, and Loftus and Loftus, P.C. were on brief, for appellant.

Christopher A. Duggan, with whom Andrew D. Black and Smith Duggan Cornell & Gollub LLP were on brief, for appellees.

August 28, 2026

BARRON, Chief Judge. This appeal arises from a lawsuit that Citation Insurance Company ("Citation") has brought against Broan-NuTone LLC ("Broan") and Jakel Motors Incorporated ("Jakel") (together, the "defendants"). Citation is suing as the subrogee of its insured, a Massachusetts homeowner whose home was damaged in a fire. The suit alleges that the defendants, the manufacturers of a bathroom exhaust fan and fan motor in that home, are liable for that fire under Massachusetts law. The District Court granted summary judgment to the defendants. We affirm.

I.1

A.

On September 15, 2020, a fire broke out at Peter Damore's home in Burlington, Massachusetts. Damore was at his home that day, as was a Verizon technician who was working in the basement to upgrade the home's Wi-Fi system. Sometime after the Verizon technician arrived that morning and began working, he informed Damore that a fuse on the circuit breaker panel in the basement tripped when he tried to use it. Shortly thereafter, smoke detectors began to go off on one of the upper floors of the home, and Damore hurried upstairs to investigate. Upon opening

1 "We draw the facts from the summary judgment record that was before the [D]istrict [C]ourt, and we array [those facts] in the light most favorable to the nonmoving party." O'Horo v. Bos. Med. Ctr. Corp., 131 F.4th 1, 8 (1st Cir. 2025) (citation modified).

3

the stairway to the attic, Damore discovered heavy smoke and heard the crackling of a fire. He and the Verizon technician vacated the home, and Damore called 911.

The Burlington Fire Department responded and determined that the main fire was located in the ceiling of the master bathroom, above the bathroom ceiling fan. A lieutenant observed "a heavy glow of fire around the metal ceiling fan and darkened drywall on [the] ceiling next to that glow." After extinguishing the fire, the Fire Department prepared an incident report, which documented findings of "obvious charring in [the] area of [the] ceiling fan, roof joists and rafters immediately above the [bathroom ceiling] fan," as well as "a lot of charring of [the] plywood floor above [the] fan." The report recounted the determination of a trooper that the "fire started with [the] bathroom vent fan," was "accidental," and had "no connection to Verizon working on [the] scene."

Damore thereafter submitted an insurance claim pursuant to his policy with Citation. In response, Citation made approximately $726,000 in payments to Damore in connection with damage sustained as a result of the fire.

B.

Citation, as Damore's subrogee, filed suit against the defendants in federal district court in the District of Massachusetts based on diversity jurisdiction. Its complaint

4

alleged that the fire in the Damore residence "originated at or around a defective bathroom exhaust fan manufactured and placed into the stream of commerce by Broan." It further alleged that the fire was "caused by the catastrophic failure of [the fan's] electric motor," which was "designed, manufactured[,] and sold by Jakel."

As relevant here, the Jakel motor in the Broan fan contained a thermal cutoff device ("TCO") manufactured by Tamura.2 As the District Court explained, a TCO "is a thermal fuse that is supposed to melt open when [a] [m]otor reaches a certain unacceptable temperature," thereby causing power to the motor to be cut off so that it does not become too hot and ignite a fire. Citation Ins. Co. v. Broan-NuTone LLC, 793 F. Supp. 3d 299, 303-04 (D. Mass. 2025).

As relevant to this appeal, Citation's complaint set forth claims against both defendants for negligence and breach of implied warranty of merchantability.3 As relief, Citation sought "damages sustained as a direct and proximate cause of the" fire, as well as "interests and costs."

2 Tamura is not a party in this case.

3 Citation also asserted a claim for breach of implied warranty in tort. The parties and the District Court treated Citation's implied warranty claims as the same. Neither party argues on appeal that the two claims should be considered differently, so we follow suit.

5

1.

Citation designated two experts in support of its claims: Stephen M. Harris, a fire analyst; and Michael Wald, an electrical engineer.

Harris submitted an expert report stating that the fire originated at "the mounting position for the master bathroom air exhaust fan." The report included Harris's opinion "that the fire originated within the exhaust fan[,] as there [were] no other competent ignition source(s)" at the location of the fire's origin. The report did not identify "the specific failure mode within the exhaust fan."

Wald also submitted an expert report. Because that report is central to the present dispute, we describe it in some detail.

In his report, Wald began by detailing his observations from an examination of the fan (completed jointly with Broan's representatives), which included the following:

▪ "The fire damage patterns . . .

demonstrate[d] a fire originating at the fan and then spreading up and out into the attic."

▪ "Examination of the fan motor coil revealed two distinct areas of localized damage to the coil windings." (Citation omitted.)

▪ "Examination of the [TCO] revealed that the internal element had melted open and one of the leads had separated from the

6

body of the device. This supports the conclusion that this device was exposed to temperatures well beyond its rating at some point during the event."

Wald then opined, in relevant part, as follows:

▪ "The damage on the fan motor coil is quite clearly the result of abnormal electrical activity (arcing)."

▪ "The presence of electrical activity on the motor coil, but nowhere else in the fan supply wiring, not only confirms that the fan was energized at the time of the fire but also precludes any damage inside the fan from having been caused by external fire attack. . . . Thus, the possibility of a fire originating at any other wiring [in the attic] can be eliminated."

▪ "[T]he short amount of time between when the breaker was being operated [by the Verizon technician] and the detection of the fire precludes the resetting of the breaker from having been associated with causing the fire."

▪ "The root cause of these failures is deficiencies in the design and manufacture of [Broan fans equipped with Jakel motors and Tamura TCOs] such that the motor coil can be subjected to temperatures which damage the insulation on the coil windings. Once this occurs there is an unintended flow of electricity from one winding to adjacent windings . . . . and eventually arcing failures develop."

▪ "There is no evidence that this fan was improperly installed, improperly maintained, or modified since it was purchased. It was being used in its originally intended manner. Thus, it is concluded that this fan was in a

7

deficient condition when it left the hands of the manufacturer."

Finally, Wald's expert report additionally stated:

▪ "The propensity of Broan fans equipped with Jakel motors and Tamura [TCOs] to start fires is well known and discussed in several papers on the subject . . . .

This investigator has examined dozens of fans that have caused fires which have th[e] same damage pattern [as the fan here]."

▪ "While the installation of a [TCO] can often detect an overheating event before it develops into winding failures, the Tamura [TCO] is known to be particularly unreliable. Testing performed by others has demonstrated the inability of this device to remove power from overheating windings. Moreover, as noted above, this investigator has seen numerous real-world examples of Tamura [TCOs]

failing to prevent coil winding failures."

The defendants deposed Wald. At that deposition, Wald described three specific deficiencies that have been documented to occur in TCOs, including: (1) degradation of the TCO as a result of exposure to the fan's normal operating temperature; (2) a failure that allows TCOs to continue conducting electricity even after they have melted open; and (3) improper bending of the TCO lead wire during installation. As to the fan in question, Wald testified that the TCO lead wire had "been bent in violation of the [TCO] instructions," but that he "could not tell after the fire" whether the bend had caused specific further damage likely

8

to lead the TCO to fail. He further stated, "I don't know why the TCO failed. I just know it did."

Wald also testified that numerous alternative designs would, in his view, improve fan reliability and/or reduce the risk of fire, including: encasing the fan motor in metal, using noncombustible or fire-resistant plastics, or installing a mechanical TCO or multiple TCOs. As to those alternatives, however, he stated that he did not know how much they would cost or what specific modifications to the subject fan's design would be necessary to accommodate them.

2.

The defendants filed a motion for summary judgment. The motion argued that Citation had not introduced sufficient expert evidence to prevail on its implied warranty claims under either (1) a manufacturing defect theory or (2) a design defect theory.

As to Citation's implied warranty claims insofar as they were based on a manufacturing defect theory, the defendants pointed to Wald's deposition testimony stating that he could not opine as to the specific TCO defect that caused the fire. They argued that the absence of expert evidence on that score was fatal to a manufacturing-defect-based implied warranty claim. See Fernandes v. Union Bookbinding Co., 507 N.E.2d 728, 735 (Mass. 1987) ("In order to invoke the implied warranty of merchantability under [Mass. Gen. Laws Ann. ch. 106,] § 2-314, a plaintiff must

9

demonstrate that the damages complained of were proximately caused by a defect or breach which existed at the time of the sale.").

As to Citation's implied warranty claims insofar as they were based on a design defect theory, the defendants argued that Citation also could not prevail. Here, they pointed to the fact that Wald had testified that he was not aware of the costs or ramifications of adopting the alternative designs he had identified. See Evans v. Lorillard Tobacco Co., 990 N.E.2d 997, 1024 (Mass. 2013) ("In claims alleging . . . a design defect in breach of the implied warranty of merchantability, the plaintiff must show an available design modification which would reduce the risk without undue cost or interference with the performance of the product . . . ." (citation modified)).4 Citation opposed the defendants' summary judgment motion. First, as to its evidentiary burden under a manufacturing-defect-based theory of its claims, Citation argued that Massachusetts courts do not require plaintiffs to identify "the exact" cause of an accident so long as "they can rule out other potential causes." (Quoting Collins v. Sears, Roebuck &

4 The defendants argued that the negligence claims similarly failed because Citation failed to show (1) that a defect caused the fire or (2) that the defendants were negligent in failing to adopt an alternative design. On appeal, Citation does not dispute that it cannot prevail on its negligence claims unless it makes one of those showings, and so we proceed to treat its negligence and breach of implied warranty claims together.

10

Co., 583 N.E.2d 873, 874 (Mass. App. Ct. 1992).) Citation contended that, "[i]n Massachusetts, negating all other sources of injury is not treated as a distinct legal theory but is instead considered another factor in determining whether the plaintiff can show a greater probability than not that the relevant accident or injury resulted from the defendant's negligence." (Citing Walker v. Gen. Elec. Co., 968 F.2d 116, 119-20 (1st Cir. 1992) (citation modified).)

Citation argued that summary judgment was not warranted under that standard on a manufacturing defect theory of its claims because it had shown "'a greater probability than not' that the fire was caused by the [f]an's faulty TCO." It first pointed to the fact that Harris, the Burlington Fire Chief, and Wald had all concluded that "the fire originated in the area of the bathroom exhaust fan." It then noted, citing Wald's deposition, that Wald had "consistently reiterated his expert finding that the fire was caused by a manufacturing defect in the TCO." Citation acknowledged that Wald "also identified three well-documented deficiencies with the TCO at issue," but it argued that it was "unnecessary" for him to specify which of those deficiencies caused the fire, as "such specificity is not necessary" under Massachusetts law "where, as here, the fire destroyed important pieces of evidence . . . that would help to prove how the coil in the TCO failed."

11

Citation then argued as to a design-defect-based theory of its claims that all it needed to do to survive summary judgment was "offer evidence that th[e alternative] design is currently being used in similar products in the market or, at least, that such a design could have reasonably been produced." (Citing Evans, 990 N.E.2d at 1015.) It contended that it had satisfied that standard because Wald identified several design alternatives in his deposition testimony that "are already being used or are available in the market."

The District Court granted summary judgment to the defendants on all claims. Citation Ins. Co., 793 F. Supp. 3d at 303. In doing so, the District Court began by analyzing the "[t]hreshold [q]uestion" of whether Citation had violated Federal Rule of Civil Procedure 26(a)(2)(B). Id. at 309. That Rule requires parties, in disclosing expert witnesses, to provide "a written report" containing, among other things, "a complete statement of all opinions the witness will express and the basis and reasons for them." Fed. R. Civ. P. 26(a)(2)(B)(i). The District Court determined that Citation had violated that Rule with respect to Wald because Wald had opined on three matters that had not been clearly identified in his expert report: (1) an alleged specific manufacturing defect; (2) how any such defect caused the fire at issue; and (3) potential design alternatives and their feasibility. Citation Ins. Co., 793 F. Supp. 3d at

12

309-10. The District Court concluded that the "most appropriate" sanction for those violations was exclusion of those portions of Wald's deposition testimony. Id. at 311.

The District Court reasoned that those exclusions provided a sufficient basis for granting summary judgment to the defendants. Id. That was so, the District Court explained, because "the first two" pieces of excluded testimony "relate to essential elements of [Citation]'s manufacturing defect claim" -- namely, the identification of a specific defect that caused the fire. Id.; see also id. at 309-10. The District Court further explained that the third piece of excluded evidence "relates to an essential element of [Citation]'s design defect claim" -- namely, available design modifications and their potential costs and ramifications. Id. at 311; see also id. at 310.

The District Court went on to consider, "in the interest of completeness," the merits of Citation's claims had it not excluded the testimony in question. Id. at 311. It began by explaining that, "given the complexity of the product at issue," expert testimony was necessary to prevail on either a manufacturing or design defect theory. Id. at 313. It then found that the defendants were entitled to summary judgment under a manufacturing defect theory for "two independently sufficient reasons": (1) the record did not establish that the particular fan at issue "deviated

13

from its intended design" and (2) expert testimony was necessary on the issue of causation, yet Wald had "conceded at [his] deposition that he did not know whether the alleged manufacturing defect that he identified caused the fire." Id. at 313-14. The District Court then concluded that the defendants were also entitled to summary judgment on a design defect theory because of the absence of expert testimony as to whether an alternative design "would be unduly costly and/or would affect performance of the machinery." Id. at 314.

Citation timely appealed.

II.

We review the District Court's grant of summary judgment to the defendants de novo. See Miceli v. JetBlue Airways Corp., 914 F.3d 73, 80 (1st Cir. 2019). "Summary judgment is appropriate where the record, viewed in the light most favorable to the nonmoving party" -- here, Citation -- "discloses no genuine dispute as to any material fact and demonstrates that the movant[s]" -- here, the defendants -- are "entitled to judgment as a matter of law." McCoy v. Town of Pittsfield, 59 F.4th 497, 504 (1st Cir. 2023) (citation modified); accord Fed. R. Civ. P. 56(a). In conducting our review, "[w]e are not bound by the [D]istrict [C]ourt's reasoning" but "may affirm the entry of summary judgment on any ground apparent from the record." McGunigle v. City of Quincy, 835 F.3d 192, 202 (1st Cir. 2016) (citation modified).

14

"A federal court sitting in diversity applies state substantive law." Levin v. Dalva Bros., 459 F.3d 68, 73 (1st Cir. 2006). Under Massachusetts law, a warranty of merchantability is implied in contracts for the sale of goods when sold by merchants of such goods. See Mass. Gen. Laws Ann. ch. 106, § 2-314 (West 2026).

To prevail on a breach of warranty of merchantability claim under Massachusetts law, a plaintiff must establish the following elements:

(1) that the defendant manufactured or sold the product; (2) that a defect or unreasonably dangerous condition existed at the time the product left the defendant's hands so that it was not reasonably suitable for the ordinary uses for which goods of that kind were sold;

(3) that at the time of his injury, the plaintiff was using the product in a manner that the defendant intended or that could reasonably have been foreseen; and (4) that the defect or unreasonably defective condition was a legal cause of the plaintiff's injury.

Lally v. Volkswagen Aktiengesellschaft, 698 N.E.2d 28, 43 (Mass. App. Ct. 1998) (citation modified).

As relevant here, a product may be considered defective under the second element "because of [(1)] a manufacturing defect, [(2)] a design defect, or [(3)] a warning defect." Evans, 990 N.E.2d at 1010. A manufacturing defect occurs when a product deviates from its intended design due to an error in manufacturing. See Back v. Wickes Corp., 378 N.E.2d 964, 970 (Mass. 1978) (stating

15

that, in "a case involving a manufacturing defect, the jury might simply compare the propensities of the product as sold with those which the product's designer intended it to have"). A design defect, in contrast, occurs when a product is manufactured as intended, but the product's design poses "foreseeable risks of harm" that "could have been reduced or avoided by the adoption of a reasonable alternative design." Evans, 990 N.E.2d at 1011 (internal quotation marks omitted) (quoting Restatement (Third) of Torts: Products Liability § 2(b) (A.L.I. 1998)).

III.

On appeal, Citation argues that the District Court erred in granting summary judgment to the defendants on the breach of implied warranty claims by premising its ruling on Citation's failure to provide admissible expert evidence as to a specific manufacturing or design defect that could support such a claim. In so contending, Citation faults the District Court for failing to apply the "malfunction theory." In support of this challenge, Citation contends that Massachusetts courts have recognized the malfunction theory and that the theory permits a plaintiff to show circumstantial evidence of a defect without showing that defect to have been caused by the product's design or how it was manufactured.

The defendants dispute whether Massachusetts law recognizes the malfunction theory at all. They further dispute

16

whether Massachusetts law does so even when, as here, no expert opined that destruction of the assertedly malfunctioning product made it impossible to identify the defect in that product that caused the malfunction. We need not resolve either dispute, however, to reject Citation's challenge on appeal. As we will explain, even if we were to resolve each of these disputes in Citation's favor, we still would reject its challenge to the grant of summary judgment.

A.

We can easily dispense with the portion of Citation's malfunction-theory-based challenge to the grant of summary judgment that takes aim at the District Court for faulting Citation for failing to provide admissible expert evidence sufficient to show a design defect. See Citation Ins. Co., 793 F. Supp. 3d at 314. The reason that we may do so has to do with the arguments that Citation made to the District Court in opposing the defendants' motion for summary judgment.

In doing so, Citation did not rely on the malfunction theory that it now invokes on appeal. It instead argued that it had introduced sufficient evidence of alternative designs to create a genuine dispute of material fact as to a design defect theory of its claims. And, in support of that contention, Citation relied solely on Wald's deposition testimony.

17

Citation does not challenge on appeal, however, the District Court's exclusion of Wald's deposition testimony. It instead contends that it introduced "sufficient circumstantial evidence and evidence excluding secondary causes of the fire" to survive summary judgment under the malfunction theory. Nor does Citation develop an argument that other evidence in the record suffices to show a genuine dispute of material fact as to reasonable alternative designs.

Citation did not argue below that, under the malfunction theory, it did not need to introduce evidence of alternative designs to prevail on a design defect theory of its claims. As a result, it cannot now challenge the District Court's ruling on that ground. See Morales Feliciano v. Rullan, 378 F.3d 42, 49 (1st Cir. 2004) ("It is a bedrock rule that when a party has not presented an argument to the district court, he may not unveil it in the court of appeals." (citation modified)). We therefore see no basis for concluding that the District Court erred in granting the defendants' summary judgment motion as to Citation's design-defect-based theory of its implied warranty claims.

B.

There remains to address Citation's challenge to the District Court's grant of summary judgment to the defendants as to the manufacturing-defect-based theory of those same claims. Here, the District Court faulted Citation for failing to provide

18

admissible expert evidence supportably showing a manufacturing defect. See Citation Ins. Co., 793 F. Supp. 3d at 313-14. Citation contends that the District Court erred in so ruling. Citation argues that the malfunction theory applies to those claims and that, under that theory, it did not need to provide such expert evidence.

Citation is in a stronger position in advancing this challenge than it was in advancing its design-defect-related challenge. After all, it did advance the malfunction theory below as to its manufacturing-defect-based theory. Nonetheless, we still conclude that the challenge is without merit, given the District Court's reasons for granting summary judgment and the way that Citation deployed the malfunction theory in the proceedings below.

Specifically, in granting summary judgment with respect to Citation's manufacturing-defect-based argument, the District Court reasoned in part as follows. It explained that Citation failed to introduce evidence showing how the fan or its motor "deviated from their intended design." Id. at 313. It then determined that, accordingly, "no factfinder would be able to perform their duty to 'compare the propensities of the product as sold with those which the product's designer intended it to have.'" Id. (quoting Back, 378 N.E.2d at 970).

19

We fail to see how the malfunction theory undermines that reasoning. In the proceedings before the District Court, Citation chose to rely on the malfunction theory solely to show that there was a manufacturing defect. That choice necessarily invited the District Court to conclude that Citation had to show how, via circumstantial evidence, an inference could be drawn that a manufacturing defect caused the fire. Citation does not point to anything in the record, however, that supports that more particularized inference, even accounting for the malfunction theory.

After all, on Citation's own account of the malfunction theory, it does not, in and of itself, necessarily distinguish between manufacturing and design defects. So, even if the District Court had applied the malfunction theory to infer the existence of a defect, we fail to see how the application of that theory could have led the District Court to conclude that a factfinder could "compare the propensities of the product as sold with those which the product's designer intended it to have." Back, 378 N.E.2d at 970.

To be sure, we might conclude differently if Citation could show that the record reveals circumstantial evidence from which a fact finder could infer not only that a defect caused the fire but also that it was a manufacturing rather than a design defect. But Citation has not done so.

20

In its filing in opposition to summary judgment with respect to a manufacturing-defect-based theory of its claims, Citation identified three potential "deficiencies with the TCO at issue that could have ignited the fire":

(1) that the TCO was only rated for 90 degrees centigrade for an extended period of time and could degrade when exposed to higher temperatures, (2) that its eutectic fusible element can melt but remain in a pool such that it continues to conduct electricity, and (3) that bending the TCO in violation of the instructions that come with it can cause a fire.

It is far from clear, however, that a layperson would be equipped to conclude, absent expert evidence, that all those deficiencies would be due to a defect in manufacturing rather than design. For example, it is not apparent that the temperature rating of the TCO at issue here deviated from "identical products issued from the same manufacturer," Wasylow v. Glock, Inc., 975 F. Supp. 370, 377 (D. Mass. 1996), such that such a defect would be attributable to a manufacturing error.

We recognize that Citation argues on appeal that, under the malfunction theory, it need not identify a specific manufacturing or design defect so long as the circumstantial evidence supports an inference that the product at issue was generally defective. But, as we have already explained, Citation did not present that argument below. Instead, it chose to rely on expert evidence of reasonable alternative designs -- and not the

21

malfunction theory -- to advance a design-defect-based theory of its implied warranty claims. That choice invited the District Court to reject Citation's attempt to show there was a genuine issue of triable fact as to the existence of a design defect, for the very reason that Citation had failed to provide sufficient expert evidence to show that any defect was a design defect. See Citation Ins. Co., 793 F. Supp. 3d at 314.

In turn, Citation's choice to rely on the malfunction theory solely for its manufacturing-defect-based theory invited the District Court to conclude that Citation must show sufficient circumstantial evidence to allow a jury to infer the existence of a manufacturing defect. But, as we have explained, Citation has not explained how the District Court erred in concluding that no such showing had been made.

We acknowledge that it might appear odd to reject the application of the malfunction theory to Citation's manufacturing-defect-based claim on the ground that Citation failed to provide sufficient evidence specifically showing a manufacturing defect, given that such specificity is arguably not required under the malfunction theory. Nonetheless, having conducted the litigation as it did in the District Court and having chosen not to rely on the malfunction theory in support of a design-defect-based claim, Citation is stuck with the consequences of that approach. See Orenstein v. United States, 191 F.2d 184,

22

193 (1st Cir. 1951) ("An appellant will not ordinarily be permitted to complain of an error which he himself invited . . . ."); see also Box Pond Ass'n v. Energy Facilities Siting Bd., 758 N.E.2d 604, 615 n.14 (Mass. 2001) (same). We therefore cannot agree that the District Court erred in granting summary judgment to the defendants on Citation's implied warranty claims.

IV.

For the foregoing reasons, we affirm.