Summary
holding that the role of the resulting-loss exception is to prevent the insurer from "avoid[ing] coverage for losses remotely traceable to an excluded cause," but not to cover foreseeable results of the excluded cause
Summary of this case from Rocky Mountain Prestress, LLC v. Liberty Mut. Fire Ins. Co.Opinion
No. 09-1542.
Argued: April 20, 2010.
Decided and Filed: August 25, 2010. Rehearing and Rehearing En Banc Denied October 25, 2010.
Judge Cole would grant rehearing for the reasons stated in his dissent.
Appeal from the United States District Court for the Eastern District of Michigan, Robert H. Cleland, J.
ARGUED: Kenneth T. Brooks, Honigman, Miller, Lansing, Michigan, for Appellants. Megan K. Cavanagh, Garan Lucow Miller, P.C. Detroit, Michigan, for Appellee. ON BRIEF: Kenneth T. Brooks, Honigman, Miller, Lansing, Michigan, for Appellants. Megan K. Cavanagh, C. David Miller, Garan Lucow Miller, P.C., Detroit, Michigan, Mark E. Shreve, Garan Lucow Miller, P.C., Troy, Michigan, for Appellee.
Before GUY, COLE and SUTTON, Circuit Judges.
SUTTON, J., delivered the opinion of the court, in which GUY, J., joined. COLE, J. (pp. 580-86), delivered a separate dissenting opinion.
OPINION
In this insurance-coverage dispute, the insured (TMW) appeals the district court's grant of summary judgment in favor of the insurance company (Federal). Because a relevant exclusion to coverage applies, we affirm in part and reverse in part.
I.
On December 29, 2004, TMW, acting through its subsidiary, Shain Park Associates, bought a recently constructed condominium and retail building in Birmingham, Michigan. TMW insured the Birmingham property for $10 million with Federal Insurance.
The insurance policy — an "all-risk" policy — covers any "direct physical loss or damage" to the property unless "caused by or resulting from" an excluded peril. R.65-4 at 21. Among the policy exclusions is this one:
This insurance does not apply to loss or damage (including the costs of correcting or making good) caused by or resulting from any faulty, inadequate or defective:
. . .
— design, specifications, plans, workmanship, repair, construction, renovation, remodeling, grading, compaction;
. . .
of part or all of any property on or off the premises shown in the Declarations. This Planning, Design, Materials Or Maintenance exclusion does not apply to ensuing loss or damage caused by or resulting from a peril not otherwise excluded.
Id., at 35.
In May 2006, contractors hired by TMW began to renovate the building's exterior. They discovered that the original builder had improperly constructed the exterior walls, leaving them vulnerable to water infiltration. And they observed that water had indeed entered the facility, "weakening the structural integrity of the building" by corroding its steel structure. R.69-9 at 1. After further investigation, the contractors confirmed that similar construction and deterioration problems plagued all four exterior walls. They warned TMW that, without proper repair, the building faced "potential mold growth" and "catastrophic" structural failure due to moisture exposure. Id. The repairs required TMW to remove the building's undamaged exterior masonry. TMW estimates that it has spent $3.9 million so far to repair the building.
TMW notified Federal of the damage to the building. After performing its own inspection of the property, Federal denied TMW's claim. Federal attributed the damage to "construction defects" and "wear and tear," both of which the policy excluded. R.69-16 at 2. TMW filed this lawsuit in federal court, seeking a declaratory judgment that it was entitled to insurance coverage and money damages. After both parties moved for summary judgment, the district court ruled in favor of Federal, deciding that TMW was not entitled to coverage. TMW asks us to take a second look.
II.
This is a diversity case — TMW hails from Delaware, with its principal place of business in Michigan, and Federal hails from Indiana, with its principal place of business in New Jersey — and Michigan law governs our interpretation of the contract. See Shields v. Gov't Employees Hosp. Ass'n, Inc., 490 F.3d 511, 515 (6th Cir. 2007). All of the traditional rules for assessing the factual record in a summary judgment motion apply, see Fed.R.Civ.P. 56(c), but so far they have made little difference. No material factual disputes were raised below. What we have instead is a matter of contract interpretation: What does the coverage exclusion mean?
The insurance policy provides that all physical loss is covered unless an exclusion applies. The key exclusion bars coverage for "loss or damage . . . caused by or resulting from . . . faulty . . . workmanship . . . [or] construction." R.65-12 at 35. Both parties' investigators agree that the original builder improperly constructed the building's exterior. The faulty construction included "haphazardly installed" Tyvek building wrap, missing or improper flashing along the siding's edges and joints and a lack of "weep holes," which "allow moisture in the wall cavity to drain out." R.65-12 at 3. The investigators also agree, and the parties do not dispute, that this faulty workmanship caused the water infiltration that damaged the building's structural steel, fireproofing and insulation. Put in terms of the insurance exclusion, the undisputed facts establish that the claimed damage was "caused by or resulting from . . . faulty . . . workmanship . . . [or] construction." R.65-4 at 35. So far as Federal and the district court are concerned, that is all there is to this dispute: The insurance contract contains an exclusion that directly applies to the loss claim, and accordingly Federal acted well within its rights in denying coverage.
Not so fast, TMW responds. There is more to the faulty workmanship exclusion than the "caused by or resulting from . . . faulty . . . workmanship . . . [or] construction" language. At the end of that exclusion and the other three that go with it, the contract says that all four exclusions do "not apply to ensuing loss or damage caused by or resulting from a peril not otherwise excluded." R.65-12 at 35. It may be, as TMW acknowledges, that faulty workmanship allowed water to seep into the walls. But the intruding water, it argues, nonetheless amounts to a "peril not otherwise excluded" because the water caused some of the damage, and water-related damage is not otherwise specifically excluded — making it an "ensuing loss" and thus a covered loss.
Instead of carving out an exception to this exclusion, this theory of interpretation would create a virtual, if not complete, exclusion of the exclusion. When a policy excludes "loss or damages . . . caused by or resulting from . . . faulty . . . workmanship . . . [or] construction" of a building, it should come as no surprise that the botched construction will permit the elements — water, air, dirt — to enter the structure and inside of the building and eventually cause damage to both. TMW's chain of reasoning — that water technically was the final causative agent of the damage, as opposed to the faulty construction, that "water damage" is not specifically excluded from the policy, that coverage accordingly applies — essentially undoes the exclusion.
As an "all-risk" policy, this insurance policy basically covers everything unless specifically excluded. That means the number of possibilities for last-in-time "but for" causes of damage are limited only by the imagination of the reader. What if a roof contains a flawed design (think Frank Lloyd Wright, see Essex Ins. Co. v. Fidelity Guar. Ins. Underwriters, Inc., 282 Fed.Appx. 406, 409 (6th Cir. 2008)), and it leaks water into the house, which ruins one of the floors? But for the water, no damage to the floor would have occurred. Yet the contract does not exclude damages caused by "water." Coverage? What if faulty construction allows humid summer air to enter the building, which rusts metal fixtures? But for the exposure to the summer air, no damage to the fixtures would have occurred. Yet the contract does not exclude damages caused by "air." Coverage? What if a poorly constructed ceiling beam falls, smashing the floor below? But for the force of gravity, no damage to the floor would have occurred. Yet the contract does not exclude damages caused by "gravity." Coverage? As in each of these examples, so too here: The very risk raised by the flawed construction of a building came to pass. To say that the risk was not covered because other elements or natural forces were the last causative agents of the damage, though to be sure utterly foreseeable causes of the damages, is to eliminate the exclusion. It is exceedingly strange to "think that a single phenomenon that is clearly an excluded risk under the policy was meant to become compensable because in a philosophical sense it can also be classified as water damage." Aetna Cas. Sur. Co. v. Yates, 344 F.2d 939, 941 (5th Cir. 1965) (Friendly, J., sitting by designation).
But if the "ensuing loss" clause does not apply here, TMW asks, doesn't that make it superfluous, empty words with no independent function? No. There are two possible functions served by the clause, both sufficient to defeat TMW's argument. One is this: The clause means simply that what is not excluded is covered. The four exclusions spell out when coverage does not apply, and the caveat at the end reminds us that if an exclusion does not apply, then coverage exists. The contract repeats this coverage-off, coverage-on language or slight variations on it in seven other places in the contract, suggesting that it flows from the kind of systematic and formulaic precision, sometimes overdone precision, in which lawyers often seem to take great pleasure. No doubt, as a matter of sheer logic, the coverage-on language is never technically necessary to establish which of two binary options applies. Nor for that matter do traffic lights need a green signal. But the inclusion of the coverage-on language still serves to remind the readers that there are two sides to every coverage issue. Just as all light switches contain an "on" and "off designation, so it is useful to spell out when coverage applies and does not apply. All contract drafting that involves belts (certain damages are excluded) and suspenders (all damages not excluded are covered) has this quality. And "[n]othing prevents the parties from using a belt and suspenders approach in drafting the exclusions, in order to be doubly sure." In re SRC Holding Corp., 545 F.3d 661, 670 (8th Cir. 2008) (quotation marks omitted).
But even if we choose to label this type of drafting a form of redundancy, which we do not think it is, that label surely is not a fatal one when it comes to insurance contracts, including this contract, where redundancies abound. In just this one provision of the 80-page insurance contract, there are at least three truly redundant phrases, as opposed to logically redundant phrases: (1) "loss or damage"; (2) "caused by or resulting from"; and (3) "faulty, inadequate or defective." As in so many insurance contracts, iteration is afoot throughout — from an exclusion for "war and military action" to one for "fraudulent, dishonest or criminal acts or omissions" to one for flooding of "lakes, reservoirs, ponds, brooks, rivers, streams, harbors, oceans or any other body of water or watercourse" to numerous others.
Not just insurance lawyers frequently say two (or more) things when one will do or say two things as a way of emphasizing one point. Courts themselves frequently apply "arbitrary and capricious" review in administrative law cases. See, e.g., F.C.C. v. Fox Television Stations, Inc., ___ U.S. ___, 129 S.Ct. 1800, 1811, 173 L.Ed.2d 738 (2009). But no one, I suspect, has ever seen agency action that was "arbitrary" but not "capricious." Emblazoned on the front of the United States Supreme Court are the words "Equal Justice Under Law." But who would defend something as "Justice Under Law" if it did not apply equally to all citizens? See In re Ocwen Loan Servicing, LLC Mortg. Servicing Litig., 491 F.3d 638, 646 (7th Cir. 2007) (Posner, J.) ("The full name of the duty, both in the complaint and in the cases — `duty of good faith and fair dealing' — could be thought ominously open-ended. But the full name is merely what is called a `doublet,' a form of redundancy in which lawyers delight, as in `cease and desist' and `free and clear.'" (quoting Bryan A. Garner, The Redbook: A Manual on Legal Style § 11.2(f) (2d ed. 2006))).
All of this helps to reveal the limits of the interpretive canon upon which TMW stakes this appeal: that courts must avoid interpreting contracts to contain superfluous words. The canon is one among many tools for dealing with ambiguity, not a tool for creating ambiguity in the first place. "Where there are two ways to read the text" — and the one that avoids surplusage makes the text ambiguous — "applying the rule against surplusage is, absent other indications, inappropriate." Lamie v. U.S. Trustee, 540 U.S. 526, 536, 124 S.Ct. 1023, 157 L.Ed.2d 1024 (2004); see Chickasaw Nation v. United States, 534 U.S. 84, 94, 122 S.Ct. 528, 151 L.Ed.2d 474 (2001). It may be that "interpreting an insurance policy to give a reasonable meaning to all provisions is preferable to interpreting the policy in a way that creates surplusage or leaves a portion of the policy useless or inexplicable," but "surplusage alone does not make an insurance policy ambiguous." Michigan Twp. Participating Plan v. Pavolich, 232 Mich.App. 378, 591 N.W.2d 325, 330 (1998). When "the policy is clear as written, we are bound by the specific language, and will not construe the policy to cover defendant simply to avoid a finding that there is surplus language in the contract." Id. at 331. (Our colleague suggests that Federal never addressed this interpretation during the appeal. That may or may not be true with respect to the appellate briefs, though both parties addressed the point at one level of generality — the role (or lack of a role) of potential contract redundancy, see TMW's Br. at 47-49; Federal's Br. at 31 — but it clearly is not true with respect to the oral argument, where both parties addressed this precise point. See Oral Arg. at 7:30, 21:35, 25:00.)
This, however, is just one way of looking at the clause. There is another. The "ensuing loss" clause also fairly could be construed as a causation-in-fact-breaking link in coverage exclusions, establishing that independent, non-foreseeable losses caused by faulty construction are covered. Cf. Prudential Prop. Cas. Ins. Co. v. Lillard-Roberts, No. CV-01-1362-ST, 2002 WL 31495830, at *19 (D.Or. June 18, 2002) ("[T]he `ensuing loss' clause [applies] in those rare cases where the reasonable damage expected to be caused by faulty workmanship leads to another peril that causes damage beyond that normally expected."); Montefiore Med. Ctr. v. Am. Prot. Ins. Co., 226 F.Supp.2d 470, 479 (S.D.N.Y. 2002) ("An ensuing loss provision does not cover loss caused by the excluded peril, but rather covers loss caused to other property wholly separate from the defective property itself.").
While the faulty workmanship exclusion applies to loss or damage "caused by or resulting from" the construction defect, the "ensuing loss" provision clarifies that the insurance company could not use the exclusion to avoid coverage for losses remotely traceable to an excluded cause. Among the definitions of "ensuing" is the meaning that something "follow[s] in chronological succession." Webster's Third New International Dictionary 756 (2002). The clause establishes that chronologically later-in-time damages "caused" by "a peril not otherwise excluded" remain covered. If, say, the water leak in this case had shorted an electrical socket and started a fire, any fire damage would be covered in light of the "ensuing loss" clause. Thus, if, on the one hand, the damage came "natural[ly] and continuous[ly]" from the faulty workmanship, "unbroken by any new, independent cause," Michigan Sugar Co. v. Employers Mut. Liability Ins. Co. of Wise, 107 Mich.App. 9, 308 N.W.2d 684, 686 (1981), the exclusion applies and the ensuing loss provision does not. See Berger v. Travelers Ins. Co., 379 Mich. 51, 149 N.W.2d 441, 442 (1967) (contract language referring to the "sole cause" of something means "the efficient, dominant, proximate cause"); Ely v. Wolverine Mut. Ins. Co., No. 225584, 2002 WL 362836, at *2 (Mich. Ct.App. Mar. 1, 2002) ("Michigan courts have long applied the `proximate cause' standard to determine if' a loss was "caused by" or "resulting from" "a particular event."); 7 Couch on Ins. § 101:39 ("The rule of causation as applied in the context of insurance coverage is `proximate cause.'"). But if, on the other hand, the later-in-time loss flows from a non-foreseeable and non-excluded cause, it is covered. In this instance, because defective wall construction naturally and foreseeably leads to water infiltration, the language of the exclusion, not the exception to the exclusion, ought to apply.
That leaves one lingering consideration. As the district court viewed the dispute, the identification of the faulty workman-ship exclusion, together with undisputed factual evidence that there was a "but for" causal relationship between the damages and this exclusion, meant that summary judgment for Federal was in order. In view of the reality that the "ensuing loss" clause, under either way of looking at it, does not permit Federal to deny coverage for losses not proximately caused by faulty workmanship, and in view of the fact that TMW did not appear to have an opportunity to seek coverage for such losses, we think TMW should be given an opportunity to do so on remand.
A few respectful words in response to our colleague: First, we are skeptical of an interpretation that invokes the anti-redundancy canon with one breath, see Dissent at 15, then closes its lips to the redundancy problems that arise with TMWs interpretation, see Dissent at 13 n. 1. If the ensuing loss clause provides coverage for the water damage to TMW's walls, the only role left for the faulty workmanship exclusion is to block coverage for remedying the defects that allowed water to infiltrate in the first place. But this reading makes repetitive the exclusion's separate provisions for "the costs of correcting or making good" construction defects and "loss or damage . . . caused by or resulting from" those defects. All of the examples offered by our colleague to show the exclusion's effect deal only with "correcting" or replacing faulty construction, not with damage "caused by or resulting from" the excluded faulty workmanship. TMWs interpretation thus makes any mention of causation "nugatory," Klapp v. United Ins. Group Agency, Inc., 468 Mich. 459, 663 N.W.2d 447, 453 (2003), falling into the same trap that it tries to push Federal into. Nor does TMW or the dissent even begin to explain away the countless other redundancies that fill this contract and others like it, which is why "applying the rule against surplusage is" often overrated, why "absent other indications" it is "inappropriate" to apply, Lamie, 540 U.S. at 536, 124 S.Ct. 1023, and why it is particularly inappropriate to apply here where it purports to resolve one set of redundancies while creating several others.
Second, the canon requiring us to construe ambiguous insurance contracts against the drafter makes no difference here for several reasons: (1) a disagreement among three judges about whether a contractual provision is ambiguous does not establish that it is ambiguous; (2) a case, see Blaine Constr. Corp. v. Ins. Co. of N. Am., 171 F.3d 343 (6th Cir. 1999), arising under another State's laws, involving differing contractual provisions and above all not including a comparable exclusion does not establish contractual ambiguity; (3) the anti-redundancy canon by itself cannot create contractual ambiguity, and that is all TMW has to establish the threshold requirement of contractual ambiguity mandated by the rule; (4) TMW must establish two reasonable, independently supportable interpretations, and it has not done so because the key premise of its interpretation (resolving contractual redundancies) is a false one, and a premise at any rate with which it cannot comply; and (5) our colleague ultimately takes the view that his interpretation is the right one, whether thumbs are placed on the scales or not, see Dissent at 15.
III.
For these reasons, we affirm in part and reverse in part.
DISSENT
"If [a contract's] plain language is clear, there can be only one reasonable interpretation of its meaning and, therefore, only one meaning the parties could reasonably expect to apply. If the language is ambiguous, longstanding principles of contract law require that the ambiguous provision be construed against the drafter." Wilkie v. Auto-Owners his. Co., 469 Mich. 41, 664 N.W.2d 776, 787 (2003) (internal quotation marks omitted); see also Cole v. Auto-Owners Ins. Co., 272 Mich.App. 50, 723 N.W.2d 922, 924 (2006) ("A contract is ambiguous when its words may be reasonably understood in different ways."). Because the language of the relevant exclusion in the insurance contract at issue is ambiguous, as indicated by our precedent in Blaine Construction Corp. v. Insurance Co. of North America, 171 F.3d 343 (6th Cir. 1999), the provision should be construed in TMWs favor, and against Federal.
As the majority describes, the relevant exclusion, entitled "Planning, Design, Materials Or Maintenance" (the "Faulty Workmanship exclusion"), reads as follows:
This insurance does not apply to loss or damage (including the costs of correcting or making good) caused by or resulting from any faulty, inadequate or defective:
• planning, zoning, development, surveying, siting;
• design, specifications, plans, workman-ship, repair, construction, renovation, remodeling, grading, compaction;
• material used in repair, constructions, renovation or remodelling; or
• maintenance,
of part or all of any property on or off the premises shown in the Declarations. This Planning, Design, Materials Or Maintenance exclusion does not apply to ensuing loss or damage caused by or resulting from a peril not otherwise excluded.
The parties' dispute centers on the meaning of the "Ensuing Loss clause," which states that the Faulty Workmanship exclusion "does not apply to ensuing loss or damage caused by or resulting from a peril not otherwise excluded." The policy does not define "ensuing," "caused" or "resulting," but the ordinary meaning of each of these words is largely the same: "to ensue" may be defined as "to follow as a chance, likely, or necessary consequence" or "to follow in chronological succession;" "to cause" as "to serve as cause or occasion of or to "bring into existence;" and "to result" as "to proceed, spring, or arise as a consequence, effect, or conclusion." Webster's Third New International Dictionary (2002); see also Cole, 723 N.W.2d at 924 ("Unless otherwise defined, contractual language is given its plain and ordinary meaning."). Thus, TMW argues, the plain language of the Ensuing Loss clause carves from the scope of the Faulty Workmanship exclusion any loss or damage that is brought about by a subsequent, nonexcluded peril, even if a listed excluded peril is present earlier in the causal chain.
This interpretation of the policy wording is not perfect. But it is reasonable. It certainly does not, as the majority suggests, eviscerate the Faulty Workmanship exclusion. To the contrary, one readily may imagine a host of losses that would be excluded under TMW's reading. Redesigning and installing a new heating system because the original ventilation plans were faulty, for example. Shoring up a foundation resting on loose soil, not discovered earlier because of inadequate surveying and siting, for another. A third: replacing defective fixtures that mistakenly were installed by an inattentive contractor during a renovation. None of these losses involves a subsequent, non-excluded peril, but all would involve considerable expense that an insurer in Federal's position might well want to exclude. Similarly, in TMW's view, Federal would be liable only for the costs of remedying the water damage that occurred here, but not for any of the costs involved in rectifying the construction defects that permitted this damage to happen, including removing the building's exterior walls, properly reinstalling the moisture barrier, fixing and adding flashing in the cavity between the barrier and the exterior walls, replacing these walls, and drilling weep holes in the walls to permit moisture in the cavity to escape. Given that the very case before us features losses that would be excluded under TMW's interpretation — and given the near infinite number of similar conceivable losses — the majority's claim that TMW's reading is unreasonable because it "would create a virtual, if not complete, exclusion to the exclusion," thereby "essentially undo[ing]" it, seems plainly incorrect.
Federal argues that TMW's interpretation also is unreasonable because it violates the principle that an exception to an exclusion cannot create coverage for losses that the exclusion otherwise removes. However, the case that Federal cites for this proposition, Hawkeye-Security Insurance Co. v. Vector Construction Co., 185 Mich.App. 369, 460 N.W.2d 329 (1990), establishes no such principle. The court in Hawkeye-Security held only that exclusions cannot create coverage for losses that an insurance policy does not otherwise include, since "exclusionary clauses limit the scope of coverage provided under the insurance contract; they do not grant coverage." Id. at 336-37. This holding cannot readily be extended to exceptions to exclusions, as the purpose of such exceptions is to protect coverage already offered under the policy. Cf. id. at 336 ("The exception to [the] exclusion . . . preserves [the coverage offered by the insuring agreement] and brings it back within the broad all-risk coverage of the insuring agreement.") (internal quotation marks omitted).
Admittedly, while the countless losses associated with the "costs of correcting or making good" faulty, inadequate, or defective "design, specifications, plans, workmanship, repair, construction, renovation, remodeling, grading, [or] compaction," would be excluded under TMW's interpretation, it is difficult to imagine any other losses that would not necessarily involve a later-in-time, non-excluded cause (and so would be covered by the policy under TMW's reading). If this is the case, then the Faulty Workmanship exclusion's reference in parentheses to losses " including the costs of correcting or making good" is redundant, violating the principle that insurance contracts be construed "so as to give effect to every word, clause, and phrase," and avoid rendering "any part of the contract surplusage or nugatory." Royal Prop. Group, L.L.C. v. Prime Ins. Syndicate, Inc., 267 Mich.App. 708, 706 N.W.2d 426, 432 (2005) (citing Klapp v. United Ins. Group Agency, Inc., 468 Mich. 459, 663 N.W.2d 447, 453 (2003)). But this problem is not sufficiently grave to reject TMW's interpretation when both approaches suggested by the majority create equally (if not more) substantial redundancies, as well as other significant interpretive problems that the majority is at a loss to explain.
Moreover, the majority's argument that none of the contingencies excluded under TMW's interpretation involves loss or damage "caused by or resulting from" faulty workmanship — thereby rendering "any mention of causation nugatory" — considerably overstates the magnitude of the redundancy under this reading. The majority's argument seems to be that TMW's reading eliminates the causal relationship between faulty workmanship and "loss" or "damage," since, for TMW, the faulty workmanship appears to be the loss or damage. But the majority's argument is incorrect: Under TMW's interpretation, these terms refer to the financial loss or damage resulting from faulty workmanship, not the faulty workmanship itself. Thus, this reading certainly does not render "any mention of causation nugatory." Rather — as illustrated by the examples offered above in support of TMWs interpretation — the causal relationship is between the faulty workmanship and the costs of correcting or making it good, which are excluded under TMW's reading of the contract.
While the policy does not expressly define "loss" or "damage," these terms encompass financial as well as physical deprivation. Cf. Webster's Third New International Dictionary (2002) (defining "loss" as a "decrease in amount, magnitude, or degree," or, more specifically, "the amount of an insured's financial detriment due to the occurrence of a stipulated contingent event," and damage as "injury or harm to person, property, or reputation," or, more specifically, "the estimated reparation in money for detriment or injur)' sustained").
Because the interpretation TMW offers is reasonable, even if an equally or more plausible interpretation of the contract existed that favored Federal, the policy should be construed in TMW's favor, since the ambiguity created by two reasonable interpretations of a contract must be construed against its drafter. See Wilkie, 664 N.W.2d at 787; Cole, 723 N.W.2d at 924. In fact, though, the two readings of the contract offered in Federal's favor appear less plausible than TMWs.
The first interpretation offered by the majority suggests that the Ensuing Loss clause simply performs a "belt and suspenders" function by reminding the reader that, "if an exclusion does not apply, then coverage exists." In support of this interpretation, the majority argues that the contract "repeats this coverage-off, coverage-on language or slight variations on it in seven other places in the contract, suggesting that it flows from the kind of systematic and formulaic precision, sometimes overdone precision, in which lawyers often seem to take great pleasure." This argument, however, ignores that the policy also includes, not just slight, but materially different variations on the Ensuing Loss clause. For instance, from eight other exclusions, only "specified perils" — defined by the contract as aircraft or self-propelled missiles, explosion, fire, leakage from fire protection equipment, lightning, mine subsidence, riots or civil commotion, sinkhole collapse, smoke, vandalism, vehicles, volcanic action, and windstorm or hail — are carved out, not all "perils not otherwise excluded." Still other exclusions, like that for losses due to war and military action, carve out neither "specified perils" nor all "perils not otherwise excluded," stipulating instead that they apply "regardless of any other cause or event that directly or indirectly . . . contributes in any sequence to . . . the loss or damage, even if such other cause or event would otherwise be covered." The differing treatment of non-excluded causes under the different exclusions indicates that Federal intended the Ensuing Loss clause to do more than merely provide a formulaic reminder that coverage exists only when not excluded.
Even overlooking this problem, the belt-and-suspenders interpretation remains implausible. It seems doubtful that a purchaser of a commercial insurance policy would require a single reminder that it covers only those losses that are not excluded when the clause providing coverage — the first in the policy form — states clearly: "We will pay for direct physical loss or damage . . . caused by or resulting from a peril not otherwise excluded." But, according to the reading posited by the majority, Federal drafted the policy to include eight such reminders. Particularly in light of the principle that insurance contracts be read "so as to give effect to every word, clause, and phrase," and avoid rendering "any part of the contract surplusage or nugatory," Royal Prop. Group, L.L.C., 706 N.W.2d at 432, this seems a singularly strange reading of the policy, which perhaps is why Federal has not endorsed it.
The majority notes that the parties addressed this interpretation at oral argument. It certainly is true that this interpretation was discussed — not surprisingly, since the majority repeatedly suggested it as a plausible reading of the contract. ( See Oral Argument at 7:25, 8:20, 10:50, 24:35.) Even with this evident support for the belt-and-suspenders approach, however, Federal did not adopt it, instead arguing consistently only for the other interpretation the majority has offered. ( See, e.g., id. at 13:05, 13:50, 15:45, 18:00, 21:10, 22:40.) Indeed, at one point during argument, the panel recognized Federal's reluctance to adopt the belt-and-suspenders approach. ( Id. at 24:35 ("I really think your interpretation comes to nothing more than a belt and suspenders, that it truly accomplishes nothing. I guess it's probably hard for you to say that, and I can understand your resisting the point, but it does kind of seem that way.") (Sutton, J.).) Following this comment, Federal's counsel stated that she was not aware of any cases adopting the beltand-suspenders approach in interpreting ensuing loss clauses (or of cases prohibiting insurers from drafting a contract in this manner), but that, "given that construction, it gives effect to the exclusion as a whole, as opposed to what I think [TMW] is focusing on, which is merely the last sentence of that exclusion. And I think that that is fairly clear under Michigan law: that [TMW's approach] is not how they [i.e., the Michigan courts] would interpret it [i.e., the contract]." ( Id. at 25:00.) In other words, while counsel was willing to countenance the majority's reading, she immediately returned to a line of attack she thought better supported by Michigan case law. Particularly when viewed in light of the remainder of Federal's argument, focusing exclusively on the second interpretation the majority offers, counsel's response does not suggest endorsement of the belt-and-suspenders reading; to the contrary, it indicates the opposite.
The majority goes to considerable length to avoid the conclusion that this interpretation renders surplus the eight clauses addressing ensuing losses, arguing instead that the law abounds with such redundancies and suggesting that there is a difference between "true" and "logical" superfluousness. Whether these clauses are in fact redundant is largely secondary to the more fundamental problem with the belt-and-suspenders interpretation; namely, that it ignores the strong indication, given the different treatment non-excluded causes receive under different exclusions in the policy, that Federal intended the "ensuing loss" language to be more than a formulaic reminder that losses are covered, except when they are not. Thus, the conflict between this interpretation of the contract and TMW's would remain even if these redundancies did not exist. Nonetheless, it bears noting that the principle that insurance contracts be read so as to avoid surplusage is well-established in Michigan law, and we are not free to ignore it. By contrast, the majority offers no support for its purported distinction between "true" and "logical" redundancies, and it is difficult to imagine a better illustration of surplusage than the eight unnecessary reminders that the majority would have us read into the contract. Indeed, the very idiom the majority has chosen to explain this interpretation suggests that a second, superfluous tool (suspenders) is being used where the first (the belt) would do. Cf. Oxford Dictionary of English Idioms (2005) (defining this idiom as "providing double security by using two means to achieve the same end"). The majority's argument that these reminders are not redundant thus does little to increase the plausibility of this interpretation of the contract.
The second interpretation suggested by the majority — this one also offered by Federal — is equally problematic. Under this reading, the scope of the Faulty Workmanship exclusion extends to the natural, probable, and foreseeable consequences of faulty construction and the other perils listed in the exclusion; conversely, the Ensuing Loss clause covers only separate and independent, non-foreseeable losses caused by non-excluded perils. Reaching this result, however, requires reading quite a lot into the plain language of the contract. As plainly written, the Ensuing Loss clause does not cover only "ensuing loss or damage caused by or resulting from a peril not otherwise excluded that is not the natural and foreseeable consequence of a peril listed above," nor does it cover only " separate and independent ensuing loss or damage caused by or resulting from a peril not otherwise excluded." Rather, the plain language of the clause, which we are compelled to follow, see Cole, 723 N.W.2d at 924, suggests that any loss "caused by or resulting from a peril not otherwise excluded" is covered. If Federal had intended the Ensuing Loss clause to preserve coverage only for separate and independent losses, it could have drafted contractual language to say so clearly, just as in the several exclusions that apply "regardless of any other cause or event that directly or indirectly . . . contributes in any sequence to . . . the loss or damage, even if such other cause or event would otherwise be covered," it employed language that plainly excludes such losses.
This interpretation also falls afoul of the principle that insurance contracts be read to give effect to all language and render no language surplusage or nugatory. See Royal Prop. Group, L.L.C., 706 N.W.2d at 432. Again, under this reading of the contract, the Faulty Workmanship exclusion extends to the natural, probable, and foreseeable consequences of the listed perils, and the Ensuing Loss clause preserves coverage only for separate and independent losses caused by non-excluded perils. But if the exclusion extends to the natural, probable, and foreseeable consequences of the listed perils, the Ensuing Loss clause does no work, as the limit it supposedly provides already is inherent in the scope of the exclusion. The Ensuing Loss clause, therefore, is surplusage or nugatory.
In sum, neither interpretation of the contract offered in favor of Federal seems any more plausible than the one TMW offers: Both create equally (if not more) substantial redundancies and additional interpretive problems that the majority cannot explain. At best, these readings are reasonable alternatives to TMWs interpretation, in which case the resulting contractual ambiguity still must be resolved against Federal. See Wilkie, 664 N.W.2d at 787; Cole, 723 N.W.2d at 924. In reaching the opposite conclusion, the majority misconstrues the role that the canons regarding contractual redundancy and ambiguity play under this analysis. Contra the majority's suggestion, the principle that contracts be interpreted to avoid surplusage does not create the ambiguity here. Rather, the ambiguity exists because TMW has offered a reasonable interpretation of the insurance policy, which the majority erroneously dismisses as "undoing" the Faulty Workmanship Exclusion. Because TMWs reading is a reasonable one, the contract would have to be construed in its favor even if an equally or more plausible interpretation existed that favored Federal. That the readings favoring Federal create considerable interpretive problems — including, but by no means limited to, problems of redundancy — simply demonstrates that these interpretations are, if anything, less plausible than TMWs. In other words, the majority is incorrect in stating that the anti-redundancy canon is "the key premise" "upon which TMW stakes this appeal," and is "all TMW has to establish . . . contractual ambiguity": Even if these redundancies did not exist, the contract would remain ambiguous. By the same token, the majority's skepticism of an approach "that invokes the anti-redundancy canon with one breath . . . then closes its lips to the redundancy problems that arise with TMWs interpretation," is misplaced. That TMWs reading violates this canon is readily conceded. But when the majority's proposed readings do the same, while also creating additional significant interpretive problems, the resulting contractual ambiguity must be resolved against Federal.
This conclusion is firmly supported by our precedent in Blaine Construction Corp. v. Insurance Co. of North America. The insurance policy at issue in that case excluded "[e]rrors in design, errors in processing, [and] faulty workmanship or faulty materials, unless loss or damage from an insured Peril ensues and then only for such ensuing loss or damage." 171 F.3d at 346. The insurer argued, as Federal does here, that "an ensuing loss, to be covered, must be the result of a new, separate and independent peril from the peril that is excluded, rather than a loss that follows naturally and ordinarily from an excluded peril." Id. at 350 (internal quotation marks omitted). We rejected this argument: Relying on Farmers Chemical Ass'n v. Maryland Casualty Co., 421 F.2d 319 (6th Cir. 1970), we determined that it was ambiguous whether the loss (like here, water damage caused by construction defects) was excluded under the first clause of the exclusion or preserved under the second, and so resolved the ambiguity in favor of the insured. Id.; see also Eckstein v. Cincinnati Ins. Co., 469 F.Supp.2d 455, 461-62 (W.D.Ky. 2007) (considering similar policy provisions and holding that "[t]here is nothing . . . to indicate that an ensuing loss must be the result of a separate cause from the excluded loss. To the contrary, the policies are clear that faulty construction losses are excluded, but losses taking place afterward . . . are covered."). While the insurance policies in Blaine Construction and Farmers Chemical were governed by Tennessee law, the relevant principles of contractual interpretation are the same as in Michigan. See, e.g., Hollis v. Doerflinger, 137 S.W.3d 625, 629 (Tenn.Ct.App. 2003) ("When an insurance contract is susceptible to more than one reasonable interpretation, it is considered ambiguous. Tata [v. Nichols], 848 S.W.2d [649,] 650 [(Tenn. 1993)]. In such a case, the language must be construed in favor of the insured. See, e.g., Allstate Ins. Co. v. Watts, 811 S.W.2d 883, 886 (Tenn. 1991)."). The majority's decision today creates the strange result that an Ensuing Loss clause identical to the one the majority now construes in favor of Federal would be construed in favor of TMW in Tennessee, despite the States' common principles governing contractual interpretation.
As Federal concedes, "various courts throughout the country" have disagreed, in interpreting ensuing loss clauses, regarding "whether [the] ensuing loss must be a loss separate and independent from the excluded peril . . . or whether coverage is given back for the natural, probable and foreseeable consequences of the excluded peril." (Appellee Br. 49.) The very fact that reasonable jurists (including the members of this panel) have disagreed regarding the meaning of these clauses cuts in favor of finding against their drafters. For this reason and those presented above, I would not remand this case for the district court to determine whether the losses TMW suffered were proximately caused by faulty construction — a question that is relevant only if we construe the contract in Federal's favor. Instead, I would reverse the district court's grant of summary judgment for Federal and find in favor of TMW.
Accordingly, I respectfully dissent.