EASILY THE MOST OBSCURE & DIFFICULT TO UNDERSTAND INSURANCE COVERAGE DECISION EVER WRITTEN
RSUI INDEMNITY CO. v. THE LYND COMPANY 13-0080 (Tex. 2015)
*This is a more or less complete account of the themes and arguments of this case, together will my comments. Part I of this Blog published December 8, 2020, was an overview of one of the main themes to be found in this case the way occurrences are characterized and integrated into very complex property policies. (12/8/20) In this Part, I make a number of “Comments about the court’s opinion and the dissent, as they are presented. For better or for worse, these are in brackets.—MSQ
My overall view is that what makes this opinion so difficult is that it is interpreting a complex, complicated contract of insurance that was very poorly drafted. No skilled and thoughtful coverage lawyers, with a feel for the subtle, would let this happen. They would have seen in advance the types of problems manifest in this set of decisions and would have avoided them through a well-conceived and better-written policy.
This is a long essay about a very lengthy and extraordinarily complicated contract of insurance. It’s about how policy limits provision in first-party commercial contracts of insurance that are poorly drafted, and how insurance companies can try to measure the size of losses.
Before going any further, it would be a good idea to distinguish between what is often called scheduled policies and what are called blanket policies. Although these concepts apply to all sorts of first-property insurance policies, and not just to insurance the foundations of which are buildings or even widgets, these policy concerns buildings, some of their contents, and some loss of income.
Scheduled policies might cover several buildings by means of an all-in-one policy (aka contract of insurance), but each coverage for each building (etc.) would be independent of the others. Thus, the coverage of more than one of the buildings would never assent a screwy policy provision be used to calculate amounts of coverage for and/or policy limits on the others.
Blanket policies are, to some extent, the opposite. For multiple buildings, there will be circumstances where amounts of coverage and policy limits might be calculated together. It is possible if one of three buildings was damaged but the others or one of the others were not, all by the same event or occurrence, some of the policy limits from the undamaged building might be utilized to pay for damages to the one damaged. The same logic would apply if one of the buildings sustained serious damage, but the other two sustained lighter damages, and so on.
The central issue, in this case, was where the insurance policy at issue was a scheduled policy or a blanket policy. It is not the case that all blanket policies are for horrible weather. They could be for a lot of different events or occurrences.
Lynd Company was and is a “big-time” property management company. When Hurricane Rita came ashore in September 2005, fifteen (15) of “its” properties apartment buildings sustained physical damage.
RSUI was the excess property insurer. The primary carrier was Westchester Fire Insurance Company, and its policy limit was $20M per occurrence. RSUI agreed with Lynd that there was one occurrence, Hurricane Rita. RSUI’s limit was $480M above Westchester’s limit. RSUI’s policy was said to be a “follow form” policy, a proposition that does not appear to be quite true.
The RSUI policy is a very complex one. The policy does not specify by name and/or the description covered property. Instead, Lynd was required to identify the property to be insured. This is natural enough since different properties within Lynd’s jurisdiction will come and go. It is somewhat unusual when comparing and contrasting a huge range of property policies for this to be the way of establishing coverage size or policy limits, though this methodology is not impossible, incoherent, irrational, imprudent, or anything of the sort. One sees it fairly routinely in open lot inventory and other inventory first-party policies.
In any case, whether the policy was some sort of blanket policy or some sort of scheduled limits policy was the major issue in this case. This issue might be called arcane; however, the reasoning and semantics leading to a correct decision were and are not. The difficulties in this judicial decision are mostly the fault of the policy.
Back to the case at hand. There is an endorsement to the policy prescribing procedures for Lynd to put the property on the covered list. That endorsement is named “Scheduled Limit of Liability,” (SLL) and the list for covered properties are named “Statement of Values.” (SV) That statement set forth what Lynd or the actual property owners assigned as money values, and it did this in three separate categories: replacement value, the value of building contents, and the value of rental income for the specified property for one year.
The policy specified how premiums would be calculated. The formula began with the values assigned in the “Statement of Values.” One then moved to the “Reporting Endorsement.” It specified that the premium rate would be a fraction of each $100 value; that fraction was $0.025 of the reported value.
The policy required quarterly updates, and Lynd complied with these requirements. It added several newly acquired properties by means of endorsements. Something peculiar about the policy, it seems to me, is that it provided that the policy would only cover losses “at the locations listed on the latest Statement of Values” and only if “a value is shown for [the] scheduled item.” It seems to me that one of the peculiarities of the contract is that it only requires that some value or other be listed on the latest SV; it does not require that it be accurate not even “more or less.”
The principal issue, in this case, was how to calculate how much RSUI owed on the loss Rita inflicted loss. This calculation was itself both unusual and complicated, and it was the methodology for it that was the principal issue in the case. This controversy was not a mere matter of disagreement about arithmetic. It was an issue as to what the contract of insurance required. The calculation in question, insofar as it was relevant to this case, hinged on whether the policy was a scheduled limits policy or a blanket policy
The parties agreed that Rita was a single “occurrence” that this single occurrence caused all the damages and that Lynd’s total loss after Westchester paid its limits was $24.5M. The use of the term “occurrence” is also an oddity in the policy, as I explained in Part I this blog entry was published on 12/8/20. Here, the court and apparently the parties utilized the “standard” meaning of “occurrence” which is inconsistent with the way that term is defined and used in the policy.
In any case, Lynd made a claim for $4.5M over Westchester’s $20M. RSUI refused to pay it and offered only $750,000. Its reasoning was based on a reading of the policy which was the heart of the dispute, and RSUI’s approach will be explained presently. It hinges on the idea of ambiguity in contracts and especially contracts of insurance, something also explained in an earlier post (“Interpreting Insurance Policies, (12/7/20) which was intended to be a preface to this post.)
RSUI has acted on its interpretation of the policy, to wit: it has refused to pay a whole claim based upon its reading. Lynd claims that this is an error and provided a different interpretation of the language of the policy. The court has to pick one of them if it determines that the policy language is genuinely ambiguous. And it must adopt the interpretation of the insured if it is reasonable.
The “Scheduled Limit of Liability Endorsement” is crucial to the court’s reasoning that Paragraph 1 must be spelled out at some length. It sets forth how the liability of RSUI is limited; it is to the least of three (3) alternatives for each “occurrence”:
“the actual adjusted amount of the loss, less applicable deductibles, and primary. . . limits; “115% of the individually stated value for each scheduled item of the property insured at the location which had the loss as shown in the latest Statement of Values in file with [RSUI], less applicable deductibles and primary. . . limits; “the Limit of Liability as shown on the Declarations Page of this policy. . . .”
In some sense, the case is really about 1b, as Justice Hecht noted in his dissenting opinion. The majority opinion holds that the language of 1b cannot be treated just by itself but must be considered the light of the rest of the wording to be found elsewhere in the policy. The dissenting opinion rejects this move because ends up with what–surely from the insurer’s point of view is an unrealistic and counter-intuitive result. For this reason, if no other, the language of 1b must not be taken by itself as to the relevant issues.—MSQ
RSUI and Lynd disagreed, said the court, “on how to compare and apply the three alternative limits when one occurrence causes losses at multiple locations.” RSUI determined what it thought least of them, namely 1.b, and applied it. However, observed Justice Boyd for the majority, RSUI applied it multiple times, once for each of the coverage items at each of the damaged locations.
So, this is what the case is about, and to some degree, it hinges on whether the policy is a scheduled limits policy covering multiple buildings or a blanket policy, also covering several buildings. To get a full grasp, it is necessary to look at an assortment of different things: the contract, more facts, the law as applied, and more, plus the dissenting opinion.
It is worth knowing a little bit about the history of the case: the trial court had decided the case in RSUI’s favor, but the court of appeals had decided in favor of Lynd. In the end, the Supreme Court affirmed the Court of Appeals and decided it in favor of Lynd.
Another important feature of this case, which has already been mentioned in Part I is the “Ambiguity Rule” as to the proper interpretation of insurance contract language applies. That rule is this: if a court finds a term to be actually or genuinely ambiguous, as opposed to a word with respect to which there is a disagreement between opposing parties, the court must find use the interpretation of the party that did not draft the contract even if its interpretation is the less reasonable.
This doctrine applies with special force to insurance policies. This proposition is probably never actually said explicitly by courts, but it is true for two reasons. First, the insurer is the draftsman of the contract of insurance in all but very unusual cases. Second, there is a “special relationship” between insurance companies and their insureds making many courts inclined toward protecting the interests of the policyholder.
The language of the endorsement is crucial. The court’s discussion is divided up into several sections, so that organization shall be followed:
A. The Language of the Endorsement
This case is mainly about the endorsement in the policy that is named “Scheduled Limit of Liability.” The language of it is therefore central, though not completely decisive. The court examines several components of that language.
1. “The Title”
The first major section is entitled “A. The Language of the Endorsement[.]” It has several subsections. The majority of the court, i.e., the court, “conclude[d] that the endorsement reasonably can be read to support either party’s proposed construction and is therefore ambiguous.”
The title of the endorsement is “Scheduled Limits of Liability. (The emphasis is that of the court.) The term “Scheduled” is not defined in the policy, but RSUI argues that its meaning is “widely accepted within the insurance industry that a ‘scheduled’ policy provides ‘scheduled coverage’ meaning coverage is limited on an item-by-item basis. RSUI argued, said the court, that “scheduled” necessarily limits coverage on an item-by-item basis.
The phrase widely accepted within the insurance industry is a key focus of the arguments in this opinion and the dissenting opinion. The use of the methodology expressed in the phrase changed dramatically the traditional method of contract interpretation.—MSQ
The court’s reaction to this argument includes the following: “When construing undefined contractual terms, courts may consider the terms’ commonly accepted meaning within relevant industries. Those meanings are not necessarily determinative, however, and may not apply when the language and its context demonstrate that the parties intended a different meaning.” And the parties certainly disagreed in this case.
The court indicates that titles, like any other portion of a contract, are relevant to determining to mean; “headings and titles provide context and can inform the meaning of the sections they label. [Citations omitted.] Generally speaking, courts should construe contractual provisions in a manner consistent with the labels the parties have given them.”
Hence, while RSUI’s view that the title provides some support for RSUI’s view of “limited on an item-by-item basis” that is not the whole story. “‘[A]lthough “‘courts may consider the title of a contract provision or section to interpret a contract, ‘the greater weight must be given to the operative contractual clauses of the agreement.’ Thus, titles and headings are not determinative, and when they are inconsistent with the plain meaning of the provision’s operative language, [courts shall] afford greater weight to the operative language.”
Shouldn’t testimony be required as to this matter? To be sure, the long-established view in the legal profession is that courts interpret language, but the question under discussion isn’t like that. It is about the empirical fact as to whether specified word choices are used in a given industry. This is an empirical fact as to which testimony is relevant. “Yes, we do use that phrase in our industry, and here’s how we do it.” If someone contradicts that view–(1) “No we don’t use it,” or (2) “Yes, we do use that phrase in our industry, but not that way.”–then the judicial inquiry and analysis begin there.]—MSQ
2. The Cautionary Language
The cautionary language appears twice in the endorsement, immediately above the title and immediately below it. Here is the language: “This Endorsement changes The Policy. Please Read It Carefully,” and “This endorsement modifies insurance provided under. . . ALL COVERAGE PARTS.”
According to the Court, the cautionary language is relevant, but not in a way to affect this case. The RSUI excess policy is a “follow form” policy the parties agree to this. This means that the policy is “generally subject to the terms and conditions of Westchester’s primary policy except were RSUI’s policy expressly modifies those terms.”
The Westchester policy was not a “scheduled” policy; the parties agree about this. The cautionary language and the wording of the title make it clear that the RSUI-Lynd policy is a “scheduled” policy and so is different from the Westchester policy. Thus, the cautionary language makes it clear that this policy is not, in fact, a “follow-form policy,” even though relevant people might have called it that, and even though the phrase “follow-form policy” might have occurred elsewhere in the policy, and/or relevant documents. Thus, the endorsement changed the policy. The parties agreed on all this.
But said the court, the real question is how the endorsement changed the policy. The fact that there was some change does not solve or even address the crucial issue in this case. The methods of determining policy limits are different, but that difference does not resolve the question as to how they differ.
3. Paragraph 1’s Introductory Statement
The opening sentence of the endorsement three different alternative possible coverage limits “‘in the event of [a covered] loss.'” Which of the three alternatives is used in a given case will depend on the one which most substantially limits the exposure of the insurer, Elsewhere in the policy, the term “loss” is defined; it means “a loss or series of losses arising out of one event or occurrence.”
I believe that the entire case collapses at least in part because of this definitional malpractice.]
The definitions in the next paragraph are just as bad, although perhaps less catastrophic.]—MSQ
The policy specifies that policy contains three alternative methods of specifying policy limits “‘in anyone ‘occurrence.'” The policy defines the term “‘occurrence'” to “‘mean any one loss, disaster, casualty or series of losses, disasters, or casualties, arising out of one event.'” The language goes this way: “when the occurrence is a hurricane or other significant peril, one event” includes ‘all losses arising during a 72-hour period.’ [Court’s emphasis.]
Thus [said the court], at least for the purposes of a significant peril, the policy defines the word ‘occurrence’ essentially the same way it defines the word ‘loss’: an “occurrence” is one hurricane taken as a whole, including causes of it, and any or losses arising within it. Most importantly to interpret this policy the notion of a series of losses is incorporated into the single term “loss” or the term “one loss” both individual losses and series of losses.
In any case, there are three possible ways to calculate policy limits.
Part 1a Limit
The first one”Part 1a’s Limit” in the relevant endorsement is this: “[t]he actual adjusted amount of the loss, less applicable deductibles, less applicable deductibles, and primary and underlying excess limits.”
RSUI took the term “loss” to be singular–after all, the word itself was singular and not plural. In contrast, Lynd understood the word to be plural reaching a series of losses and well as each separate individual loss, since that’s the way the definition worked. So, said the court, the term was consistent with the views of both parties. This fact rendered the term ambiguous, something in favor of the insured.
Not that it makes any real difference in the end, but this reasoning of the court is dead wrong. It is not ambiguous. It is just as Lynn interpreted it. In this policy the term “loss” did not have an ordinary usage meaning; it had only a technically defined meaning, and this was not ambiguous. The term “loss” was defined so that it includes losses. This fact is a consequence of simple semantics. If the district court had recognized this obvious fact, thousands of thousands of dollars in legal fees might have been avoided.—MSQ
Part 1b Limit
The second of the three possible criteria specified in the policy the carrier might use to determine limits on the payable amount of the insured’s claim, i.e., determine its policy limits is this: “115% of the individually stated value for each scheduled item of the property insured at the location which has the loss as shown in the latest Statement of Values…, less applicable deductible and primary and underlying excess limits.”
Lynd attempted to aggregate locations by suggesting that in the context of this case the word “individual” should be understood as all of the locations taken together. The court rejects this view. It also noted that “each” scheduled item could not be used to group different items together. “Each” means each; it does not mean “all”; it does not mean “taking-them-together.”
The opinion points out that this criterion does not rest on the amount of the “loss” but the amount of the “stated value” at the location that “had the loss.” In part, because of the “the” before the word “location,” neither losses nor location gets aggregated. The phrase “the location” in this context does not mean all locations.
So, Boyd’s opinion states that so far the correct reading of “Part 1b Limit” supports RSUI’s interpretation, the same as Justice Hecht points out in his dissenting opinion. But that is not the whole story, says Justice Boyd, something Justice Hecht rejects. To do more, one must examine the required subtraction of “applicable deductibles and primary and underlying excess limits.”
RSUI then applies the $25,000 deductible to each of the individual items in its item-by-item counting process. This requires that the $25k number be applied again and again. But deductibles are determined based on the occurrence, and the term “occurrence” is defined in terms of interconnected series of losses unless there is just one individual loss included within the occurrence. The trouble is that each occurrence is conceptually linked to a series of losses.
There is a presumption in legal interpretation, however, to the effect that if a word in a contract means x in one place it is presumed to mean the same thing in another. The opinion calls this “a natural presumption.” Atl. Cleaners & Dyers v. United States, 286 U.S. 427, 433 (1932). This presumption is not “rigid,” however, so it is not conclusive; sometimes a word can mean x in one place of the contract but y in another where x and y are not both be true. If the opposite of the “natural presumption” is true here, then RSUI’s position would not be undermined; otherwise, it is.
Another way to save RSUI’s approach is to apply the deductibles on a “pro-rata” basis, but nothing like that is suggested in the policy’s language, so this approach won’t work. Apparently, this is one of the points with respect to which there is the most intense disagreement between Justice Boyd and Justice Hecht.
Consequently, although Lynd’s view does not conform to the language of Part 1b Limit, its overall view is more consistent with the basic ideas of the policy.
The Part 1c Limit
The court interpreted this limit but rejected its applicability to this case, so it need not be considered here.
B. Is Lynd’s Proposed Construction Reasonable
A word or phrase is ambiguous in a contract (or other text) if and only if each proposed interpretation is reasonable. The court virtually starts from the position that RSUI’s interpretation is reasonable. The issue of ambiguity, therefore, is determined whether Lynd’s interpretation is also reasonable. Justice Boyd concludes that it is.
RSUI argues that there is a commonly accepted meaning in the insurance industry as to the meaning of “scheduled” so that the meaning of the phrase “Scheduled Limit of Liability” is clear and undeniable.
Lynd rejects this view. It argues that there are two different recognized types of insurance policies having this general function to be found in the insurance industry. One of them to be sure is the “scheduled” coverage type, while the other type is “blanket” coverage policies.
RSUI more or less admits this but distinguishes them sharply. According to it, one of them insures all buildings (entities) together for various occurrences, while the other insures several buildings (entities) in one policy, but insures them separately or individually. The “specific” type provides different, separate coverage limits while the blanket policy provides one limit. The latter types aggregate the insured buildings as to coverage; the former does not but treat them separately.
The fact that two entities are insured in the same policy and aggregated together in that way does not entail that their coverage and therefore limits are aggregated. Those who have auto coverage as to two or more cars insured under the same policy, for example, a family policy, all know this. Such people know that even if both vehicles are damaged or destroyed in the same accident, each insured auto has its own policy limits.—MSQ
However, said Justice Boyd, a court cannot start with the proposition if of one of these sorts and then apply limits. A court must first look at the policy language and draw inferences as to which type a given policy is. Not even a given title will make this choice certain. [Quinn’s Meta-Comment: One might keep in mind that there is such a thing as “fraud by title,” although this phrase is my invention.]
On this matter, RSUI is correct, and Lynd admits. The question is not what kinds of policies exist, but what kind of policy this one is. Of course, it is true that scheduled policies are a popular way of insuring several buildings at once. But it is the language of the policy which determines what kind of policy the contract is.
This very important observation indeed, a pillar of contract interpretation sets up the remainder of the logic of the majority argument.
The court’s argument works this way. The opinion lists all the considerations RSUI gives in support of its position and rejects each of them as determinative. Since Lynd’s position is reasonable, irrespective of whose is the more reasonable, the terms of the policy are ambiguous. Hence, if RSUI’s various premises individually or together do not entail its position, Lynd prevails. So, the rest of the opinion is spent considering RSUI’s various positions, i.e., premises.
Multiple Properties and Multiple Locations: RSUI’s policy is a follow-form policy, it says, but the primary policy is a blanket policy. These multiples prove nothing.
Required List of Properties and Values: The presence of such lists does not prove that a contract is a schedule policy. The list may be there for other reasons, e.g., to set premiums. Several courts around the country have agreed with this.
Reliance on Values to Determine Risk and Premium: RSUI averaged the premium amounts for various different items to determine the premium. This fact is usually associated with the [“]blanketness[“] of a policy.
Various courts have disagreed about precisely (or close to) the issues before the court in this case, and their decisions are mixed. Therefore, this court need not be influenced by established patterns of court decisions.
Conclusion of Majority
The court held “that the Scheduled Limit of Liability endorsement, in this case, is reasonably subject to both parties’ constructions and that the endorsement is therefore ambiguous. Because our rules require us to construe an insurance policy’s ambiguous coverage in favor of coverage for the insured, we affirm the court of appeals’ judgment adopting Lynd’s proposed construction.”
Adjustment (aka claims handling) problems are barely mentioned in this opinion, although it is clear that they were central to the claim, its processing, and the ultimate dispute. The reason that claims handling is important in this case is that it involved understanding the “accounting” aspects of the claim, what the policy permitted and demanded, the investigation to Lynd’s own accounting, and it involved understanding how insurance law worked as applied to the limits found in the policy. It is common that adjusters are required to understand the applicable law. The demand here, however, was extraordinary is what it demanded. This was an amazingly difficult case, and so it would have been for the adjusters. Who thought that adjustment was an easy profession?
The dissenting opinion to which we now turn said as little about the adjustment problems inherent in the case as did the opinion of the court.
Justice Hecht wrote a short dissenting opinion. I, and many other lawyers, have immense respect for Justice Hecht, but in this case, a number of lawyers have been harshly critical of his opinion, particularly as to tone. I have no view about this; I do not regard it as my place.
However, I find myself in agreement with the majority opinion on several grounds.
First, it is a fundamental rule of contract interpretation that the language of the contract is to be analyzed taken as a whole. Thus, without an explicit definition, a term may not be used in different senses without there is ambiguity. Meaning consistency and uniformity is fundamental provision of contract interpretation. If this makes the contract economically or prudentially unreasonable, so be it; at least the drafting party was in charge.
Second, Justice Hecht is critical of the majority opinion for an overly detailed analysis of language. He criticized justices in the majority for relying on their judicial intuitions as to semantics. Well, as a matter of law, that is how language in such documents as contracts are supposed to be understood. This is a strong and deep tradition. Also, it is a good thing to remember that this insurance policy was a mess, so linguistics and semantics were very important. [I myself have doubts about this tradition since it lets what is really an empirically based decision, “How do most reasonable people think about the meaning of this word with another empirically based answer “Is my intuition about how reasonable people think about the meaning of this word Arguably, all empirical factual matters are to go the trier of fact, not into the semantic mind of the judge. My doubts are not an issue here, obviously.]
Third, Justice Hecht uses the term reasonable in two different senses. He seems to say that an interpretation is unreasonable if it ends up interpreting a contract document in a way that a reasonable party would not adopt upon reflection, e.g. because it is (or leads to) a financial disaster for it.
Fourth, Justice Hecht seems to think that if both parties have similar and reasonable but somehow differing analyses of the unambiguous terms of a contract, then the disagreement must be resolved in favor of the insurer, at least in surplus lines markets. Of course, he would no doubt agree, the opposite is true if the relevant terms are genuinely and actually ambiguous, then the ambiguity must be resolved in favor of the insured, even if the insurers’ account is the more plausible. [Of course, he would say, what counts as a genuine and actual ambiguity depends upon widely accepted usage in a relevant market, which is semantically stable.]