NATIONAL UNION FIRE INS. OF PITTS. v. PUGET PLASTICS

Puget-manufactured chambers leaked, they were simply replaced with new Puget-manufactured chambers, and the Seisco water heaters were restored to full capacity. See, e.g., Transcript of Depo. of Michael Carr at 137; Seitz, 2 C.T. 448; Seitz, 3 C.T. 546; Seitz, 7 C.T. 1318; Transcript of Depo. of William Uecker, III at 46. Thus, in many instances, when a chamber leaked in a Seisco water heater, the water heater was, by definition, (1) tangible property, (2) other than Puget's product, (3) that could not be used because it incorporated a Puget chamber that was known to be defective, (4) but could be restored to use by the replacement of the Puget chamber. See The Policy. In these instances, the impaired property exclusion applies and, assuming no exception applies, there is no coverage.

Similar to its method of citation in its original opinion, the Court will cite to the coverage trial using the following format: Name, X C.T.Y, with "Name" representing the name of the witness testifying, "X" representing the volume of the coverage trial, and "Y" representing the page within that volume. The state-court trial will be cited using the following format: Name, X T.T. Y, with "Name" representing the witness testifying, "X" representing the volume of the state-court trial, and "Y" representing the page within that volume.

Puget contests this conclusion, arguing that because the Seisco water heaters could not in fact be "restored to use" by the simple replacement of the Puget chambers, the impaired property exclusion does not apply. According to Puget, the chambers were altered by Microtherm after they left the Puget factory and, therefore, were no longer Puget's product when they were installed into the Seisco water heaters. See Transcript of Hearing on Mot. for New Trial at 43. If this assertion were true, Puget argues, the impaired property exclusion would not be applicable and all damages caused by the leaky Puget chambers would be covered under the Policy.

Puget bases its argument that the chambers were not Puget's products on Fireman's Insurance Co. v. Bauer Dental Studio, Inc., 805 F.2d 324 (8th Cir. 1986). In that case, a dental corporation purchased crowns from Bauer, the insured party. Dentists at the dental corporation ground and polished the crowns to achieve the correct fit and bite. After customers complained of the crowns cracking and fracturing, Bauer replaced the defective crowns, then turned to its insurance carrier to reimburse the replacement costs. Id. at 325. The policy in Bauer covered damages resulting from property damage, but excluded damages to "the insured's work product." Id. Based on this exclusion, the court held that Bauer's carrier had no duty to indemnify because once the dentists ground and polished the crowns, the crowns were no longer Bauer's work product, but a composite product. Id. Puget argues this case is applicable to the present case because the Puget chambers underwent many modifications after leaving Puget's factory but before being installed into the Seisco water heaters. See Transcript of Hearing on Mot. for New Trial at 41-42.

Perhaps the facts of Bauer were such that the holding was appropriate in that case, but the Court declines to apply the same reasoning here. Additionally, it seems that Puget's decision to argue that Bauer should apply forces Puget to argue that the chambers, as they were installed in the Seisco water heater, were no longer Puget's product. This argument is contrary to the facts adduced both in this Court and in the state-court trial. Further, this position is at odds with, and has never been argued during, the last eight years of litigation in this case. Microtherm has consistently argued that the chambers were Puget's and has never claimed that it modified them so radically as to change the nature of the product. Puget has not disclaimed responsibility for the chambers it manufactured, but instead asserted that the chambers were either not faulty or, in the alternative, that any damage they caused was property damage covered under the National Union Policy. For Puget to now argue that the chambers were not actually Puget's product, but a composite product, is disingenuous and therefore an argument this Court declines to seriously address. Indeed, if this were true, Puget might not be liable at all, and its insurer would positively be free from liability. The Court reaffirms its decision that the impaired property exclusion applies to at least a portion of the damages recovered in the underlying trial.

2. The "Sudden and Accidental Physical Injury Exception" Does Not Apply.

In addition to arguing that the impaired property exclusion does not apply at all, Puget argues in the alternative that an exception to that exclusion applies. As cited above, the Policy excepted from the impaired property exclusion "the loss of use of other property arising out of sudden and accidental physical injury to [Puget's] product or [Puget's] Work after it has been put to its intended use." The Policy at 7. Dr. Maureen Reitmann testified, before this Court in the coverage trial, that cracks in the Puget-manufactured chambers were present at the time they were manufactured and simply grew over time until the chambers finally began to leak. See Reitmann, 5 C.T. 930-31, 1002 (testifying that the leaks "weren't, you know, an immediate event"). Based in part on this testimony, the Court concluded that the damage caused by Puget-manufactured chambers was not sudden and accidental.

Ironically, Puget's earlier argument that the chambers, as they were installed into Seisco water heaters, were not Puget's product would deem this exception to the impaired property exclusion inapplicable. Needless to say, Puget/Microtherm does not argue this position.

Although Puget challenges Reitmann's testimony as "pure speculation and contrary to all other evidence," Puget Mot. at 27, the Court declines to agree. Having viewed her testimony originally and having reviewed the testimony concerning the "sudden and accidental" nature of the chamber leaks, the Court reaffirms its decision that the "sudden and accidental" exception to the impaired property exclusion does not apply. This finding, in conjunction with the finding that at least some of the damage was caused by instances that fall into the impaired property exclusion, requires Puget to allocate the damages recovered in the underlying trial.

C. Puget Failed to Allocate Between the Property Damage Covered Under the Policy and the Property Damage Not Covered Under the Policy.

Having reviewed Texas case law requiring an insured party to allocate between covered and non-covered damages, and having concluded that the state-court verdict in this case was based on both covered and non-covered damages, the Court finds that Puget failed to satisfy its duty of allocation. Puget was required to (1) distinguish which portion of damages assessed against Puget were the result of covered property damage, and (2) distinguish the damages assessed against Puget from those assessed against the other defendants. Puget failed to do either.

The only jury question from the state-court trial that related to property damage concerned "parts provided or work performed by" the defendants, and this Court has been affirmed by the Fifth Circuit in holding that such damages are not covered under the Policy. See Pl.'s Opposition to Wasau's Mot. to Dismiss, Ex. A (State Court Judgment) (Docket No. 14); see also Puget I, 532 F.3d at 403. Thus, the source of damages still at issue and which might be covered under the Policy are the past and future lost profits and the damage to the value of Microtherm that resulted from covered property damage.

With respect to how much of Micro therm's damages resulted from chamber failures which are covered under the Policy, Puget argues that this Court, in holding that there was no reliable evidence of allocation, erred in not picking "a percentage number with which it is most comfortable" within the range of testimony provided at either the underlying trial or the coverage trial. Puget Mot. at 32 (citing State Farm Fire and Cas. Co. v. Rodriguez, 88 S.W.3d 313 (Tex. App. — San Antonio 2002, pet. denied)). As stated above, only those instances in which a Puget-manufactured chamber leaked and caused damage either to a circuit board or a homeowner's property are covered under the Policy. Having reviewed the relevant testimony again, the Court stands by its original holding that there is "simply no reasonable, reliable, non-arbitrary basis supported by a preponderance of the evidence in the record upon which to allocate economic damages between covered and non-covered risks." Puget II, 649 F. Supp. 2d at 652.

David Seitz, the President of Microtherm, testified at different times that when a Puget chamber leaked (1) it shorted the circuit board ninety percent of the time, see Transcript of Depo. of David Seitz (2009), at 45 (emphasis added); (2) it shorted the circuit board "well over 95% of the time," Seitz, 2 C.T. 372 (emphasis added); (3) "it oftentimes will take the board and short it out, too," Seitz, 14 T.T. 58 (emphasis added); (4) it "had" to get the board wet, Seitz, 2 C.T. 372

(emphasis added); and (5) it damaged the circuit board in almost every case, Seitz, 3 C.T. 488 (emphasis added). William Uecker, co-owner of a company that marketed Seisco water heaters, testified that (1) he saw a few leaky chambers that did not short out a circuit board, see Transcript of Depo. of William Uecker, III, at 46 (emphasis added); (2) when there was a leak, it shorted out the circuit board, id. at 13; and (3) there were a tremendous amount of leaks at the neck of Puget chambers, and those leaks would short out the circuit board, id. at 20 (emphasis added). Michael Drabczyk, a former employee of Microtherm, testified that when there was a chamber failure, water sooner or later hit the circuit board. See Transcript of Depo. of Michael Drabczyk, at 12 (emphasis added).

Contrary to the semi-consistent testimony of these three witnesses is the testimony of Michael Carr, a customer service employee of Microtherm. This Court previously noted that Carr testified that seventy-five percent or more of the chamber leaks did not damage a circuit board. See Puget II, 649 F. Supp. 2d at 651; Transcript of Depo. of Michael Carr, at 139. Carr's testimony was certainly not crucial to the Court's finding, but because Puget has contested the Court's interpretation of this testimony, the Court will address this testimony in more detail.

The portion of Carr's testimony at issue is the following: Q. All right. Now, out of the hundred field trips that you made, I think you testified earlier that at least 20 of those trips were made for bad thermistors? A. Yes. Q. Out of the — the remaining 80, how many of those were attributed to failures of the plastic in one way or another? A. Percentage-wise, or number-wise? I don't know the actual number. But it was the majority of them. Probably in the 60s, the 70s range. Q. What other types of failures for the balance of the — the field trips that you had? A. Sometimes they were just to change out a circuit board because of a previous leak or because of a — you know, possibly a plumbing leak or a lightening strike. But changing out the circuit boards would be the secondmost one. Sometimes they were damaged by the installer. I had a couple of those where I'd have to go out and basically correct an installer's mistake. Id. at 139-40. From this testimony, the Court concluded that of 100 "field trips" that Carr took to service Seisco water heaters, 20 were related to faulty thermistors. Id. When asked how many of the remaining 80 trips had to do with faulty chambers, Carr responded, "Percentage-wise or number-wise? I don't know the actual number. But it was the majority of them. Probably in the 60s, the 70s range." Id. It is unclear whether Carr intended for "the 60s, the 70s range" to be interpreted as percentages or as raw numbers. Finally, when asked "What other types of failures for the balance of — the field trips that you had?", Carr responded that the majority were to change out circuit boards that had been damaged by a previous leak, a plumbing leak, a lightening strike, or an installation error. Id.

The Court's reading of this testimony, that 75% of the leaky chambers did not result in a shorted circuit board, is certainly plausible. Eighty of Carr's 100 field trips did not involve thermistors. If only 20 (80 — 60 = 20) of those 80 non-thermistor-related trips were made in order to replace shorted circuit boards, then no more than 25% (20 out of 80) of leaky chambers could have shorted circuit boards, and that number is reached by assuming that all 80 of Carr's non-thermistor-related trips were made to repair leaky chambers, an assumption not necessarily grounded in Carr's testimony, but certainly favorable to Puget. That means that 75% (80 — 60) of the trips were for plastic failures without circuit board involvement. Puget contests this interpretation, but fails to offer an alternative interpretation of this testimony. It is important to realize that Carr did not testify live.

One could have conceivably interpreted this testimony to say that none of the plastics problems caused circuit board problems if the first 20% included only thermistors, and the last 20% were circuit board replacement problems due to problems other than chamber leaks (e.g. lightning). This would leave only the remaining 60% of service trips that were related to plastics failure, none of which, according to Carr's testimony, would have included circuit board failures. This testimony could also be interpreted to mean that all of the 60 field trips included circuit boards and therefore 100% of the chamber leaks shorted circuit boards. Finally, one could conclude that Carr's testimony has no relevance at all. In other words, Carr's testimony could be interpreted to mean anything from 0% of chamber leaks shorted circuit boards to 100% of chamber leaks shorted circuit boards.

His deposition was submitted to the Court pursuant to an agreement of the parties, and the Court did not have the benefit of being able to ask clarifying or follow-up questions.

Setting aside whatever interpretation is placed on Carr's testimony, the Court finds it difficult to rely on the testimony of Seitz, Uecker, Drabczyk. Seitz testified that his estimations, cited above, were based roughly on the warranty records produced by Puget at the coverage trial. See Seitz, 3 C.T. 473 (Seitz). Uecker testified as well that he would rely on these records for information. See Transcript of Depo. of William Uecker, III at 45 (responding "Yes" to the question, ". . . [W]ould you rely upon the warranty records made by Microtherm of these phone calls if they still exist?"). Michael Carr was one of the Microtherm employees who actually collected and documented the information from customer phone calls that eventually became the warranty records. See Transcript of Depo. of Michael Carr at 13.

Testimony concerning the accuracy of the warranty records called into question any projections that might be made based upon those records. For instance, during cross-examination, Seitz testified that he would not rely on the "user-defined" section of the records except for the most basic of information that does not pertain to customer complaints. See Seitz, 6 C.T. 1217, 1224. He further testified to the likelihood of errors occurring in the "notes" section of the records, Seitz, 6 C.T. 1225, the portion of the records from which most of the customers' complaints could be gathered.

In addition to the possibility that errors occurred in creating these records, Michael Carr testified that these records were created based in large part on customer diagnosis, not the diagnosis of a Microtherm technician. See, e.g., Transcript of Depo. of Michael Carr at 50 (testifying that on at least one occasion, no Microtherm employee verified the diagnosis of a Microtherm customer). Carr also testified that it would be possible for a customer to complain of leaky chambers when the leak in question was actually coming from a mechanical joint. See id. at 53.

With the testimony of Seitz, Uecker, and Carr tied to the warranty records, the Court conducted its own independent review of the warranty records and was unable to reach the same conclusions as those key witnesses. Making every assumption in favor of Puget, the Court concluded from the warranty records that the number of leaky chambers which also damaged circuit boards or other property was somewhere between 55% and 85%, a range which is inconsistent with the testimony of Seitz, Uecker, or Drabczyk.

The Court estimates that approximately 50% to 55% of the records clearly reflect the fact that a leaky chamber had shorted a circuit board. On the other hand, about 75% of the warranty records reflected both a leaky chamber and the fact that a circuit board had been replaced. In calculating this second category of records, the Court took no note of whether the leaky chamber was in any way related to the replaced circuit board. Finally, the Court added to this last category of records those instances in which a customer purchased an upgrade for his/her water heater which, according to testimony, would include a new circuit board. In this final category, much like the second, the Court took no notice of which upgrades seemed to be related to leaky chambers, reaching an approximate percentage of 85%.

Puget contends that, given this range of testimony and evidence, this Court is required to "choose a percentage with which it is most comfortable" and erred in not doing so. Puget cites Lyons v. Millers Casualty Insurance Co., 866 S.W.2d 597 (Tex. 1993), and State Farm Fire and Casualty Co. v. Rodriguez, 88 S.W.3d 313 (Tex. App.-San Antonio 2002, pet. denied), for this proposition. Having reviewed Lyons and Rodriguez closely, however, the Court is not persuaded that those cases place a mandatory duty on this Court to "choose a percentage with which it is most comfortable." Instead, those cases indicate that had this Court decided to allocate a certain percentage of damages to covered property damage, the Fifth Circuit likely would have affirmed that decision so long as the percentage chosen was supported by testimony provided to the Court. Such a determination must still, however, be based on credible evidence and "must afford a reasonable basis for estimating the amount of damage or the proportionate part of damage caused by a risk covered by the insurance policy." Travelers Indemnity Co. v. McKillip, 469 S.W.2d 160, 163 (Tex. 1971). What allocation evidence there is in this case is totally insufficient and amounts to rank speculation and guesswork. What little evidence there is not only varies greatly from one source to another, but in some cases the same witness gave divergent testimony. This Court finds such "evidence" to not be credible and to not be based on any credible or believable analysis. Finally, to suggest that this Court rely on this unsubstantiated speculation is to ask this Court to engage in an exercise of sheer guesswork that would not even equate to a "ball-park" guess or even fall into the category of "close enough for government work."

As seen above, depending on the interpretation, the evidence can range from 25% to 100% depending upon the witness and the interpretation put on the testimony. Thus, under this argument — setting aside any award based on additional damages, prejudgment interest, or attorneys' fees — this Court could award a verdict as small as approximately $5.6 million or as large as approximately $22.5 million, based on actual damages alone. While courts have not required exact precision when allocating, this level of imprecision is totally unacceptable.

D. Puget Failed to Connect the Property Damage to the Verdict in the State-Court Trial.

Even assuming Puget had presented this Court with credible, reliable evidence upon which it could properly allocate between those leaky chambers which caused damage to circuit boards or other third-party property and those leaky chambers that did not, the fact remains that Puget failed

to allocate damages on the macro level, that is, among the various defendants found liable at the state-court trial. On this point, Puget argues that the jury verdict in the underlying trial, because it assessed separate damages to each of the defendants, allocated the damages on the macro level. See Transcript of Hearing on Mot. for New Trial at 13. According to Puget, the state-court jury verdict takes care of the allocation issue. While the argument that the allocation was made by the state-court jury is certainly persuasive, this Court cannot simply accept that verdict as representing the proper allocation for insurance coverage purposes, but must conduct its own review of the entire record of both the underlying trial and the coverage trial in order to reach this conclusion based on actual evidence.

Upon review of the state-court record, this Court was unable to find any evidence apportioning the damages among the various defendants, or any evidence connecting the damages awarded to any covered losses. In fact, Dennis Arnie, Microtherm's lost profit expert, testified during the underlying trial that any attempt at allocation would be speculative based on the records and information he had available to him. Arnie, 34 T.T. 31 ("[B]ased on what I was hearing, I made a professional determination that there was not an adequate basis to do that."). When asked if he had been able to attribute any of Microtherm's damages to the chambers alone, he replied, "[n]ot separate and distinct." Id. at 35; see also id. at 60. Based on this evidence, this Court concluded in its judgment, and now reaffirms, that there was no reliable, non-speculative evidence upon which to allocate damages in the state-court trial among the various defendants.

This Court's opinion in this regard is buttressed by the recent court of appeals of Texas decision on appeal from this case's state-court judgment. See Dana Corp. v. Microtherm, No. 13-05-00281-CV, 2010 WL 196939 (Tex. App. — Corpus Christi-Edinburg Jan. 21, 2010, pet. filed). That court, when asked whether Microtherm had offered evidence of divisible damages with respect to another defendant in the underlying trial, held that "the evidence was legally insufficient to support the divisible lost profit damages and the divisible lost value damages awarded against [that defendant]." Id. at *19. Although that opinion was issued with respect to a defendant other than Puget, the same holds true for Puget. The parties in the underlying trial simply did not introduce evidence upon which any court can allocate damages among the different defendants, and they certainly did not allocate between covered and uncovered losses under the National Union Policy.

The Court's review of the state-court testimony should not be perceived as allowing National Union to bring a collateral attack on the state-court judgment. Nevertheless, the Court is required to review the evidence of the underlying trial in order to determine whether Puget has fulfilled its burden of allocating damages per the Policy, an issue properly before this Court. Therefore, its conclusion concerning allocation, although arguably inconsistent with that reached by the state-court jury, has no impact on the validity of the state-court judgment.

Having conducted a review of the state-court trial and coverage trial records, and having considered the state-court verdict in this case, the Court reaffirms its holding that there is no reliable, non-arbitrary basis supported by a preponderance of the evidence upon which this Court can allocate lost profits or lost value to Microtherm between covered and non-covered losses in the state-court trial.

E. Conclusion Concerning Allocation of Damages.

Having reviewed this Court's prior holding in this case, the contentions of the parties, the evidence, and the testimony, this Court finds that, with respect to allocation of damages, it is not "reasonably clear that prejudicial error has crept into the record or that substantial justice has not been done." Sibley, 184 F.3d at 487 (quoting Del Rio Distrib., Inc., 589 F.2d at 179 n. 3). Furthermore, Puget has failed to "clearly establish either a manifest error of law or fact" or "present newly discovered evidence." Rosenzweig, 332 F.3d at 863 (quoting Simon, 891 F.2d at 1159) (internal quotation marks omitted). Puget simply failed to carry its burden of allocation. Thus, its motion for new trial or, in the alternative, motion to alter or amend the judgment, so far as it pertains to Puget's burden of allocation, is DENIED.

V. CONCLUSION

Based on this Court's reading of the Fifth Circuit interlocutory opinion which it followed in this case and its own independent assessment of Texas insurance law concerning what constitutes an "accident" for insurance purpose, this Court reaffirms its holding that no "occurrence" took place in this case. Additionally, the Court finds that Puget needed to allocate between covered and non-covered losses and that it failed to satisfy its burden of allocation in this case, having presented no reliable evidence upon which this Court can apportion damages between covered and non-covered damages. The Court therefore denies Puget's motion for a new trial or to alter or amend the judgment. Any point of error raised by Puget, but not expressly addressed in this opinion, is also denied.