RESEARCH DOSSIER
Property Insurance Exclusions and Anti-Concurrent Causation
Abstract. Property insurance exclusions allocate the financial consequences of hazards that can interact in time, space, and causal mechanism. Anti-concurrent-causation (ACC) provisions are especially consequential when a single catastrophe generates wind, water, fire, earth movement, structural failure, and regulatory rebuilding obligations. This monograph investigates the construction of exclusionary wording, efficient proximate cause, independent concurrent causes, sequential losses, mixed-damage allocation, judicial constraints on policy drafting, and the scientific quality of forensic evidence. Its comparative core is a close reading of original American judicial opinions from California, Texas, Mississippi, Florida, Washington, and Alaska, alongside United Kingdom legislation and appellate decisions. The inquiry distinguishes legal rules from physical causation, and source-reported results from illustrative counterfactual calculations. The cases establish a pattern of substantial variation: express ACC wording can govern inseparable mixed causes in some jurisdictions; statutory or judicial efficient-proximate-cause constraints matter elsewhere; the absence of applicable ACC language can make a concurrent-cause doctrine decisive; and genuinely separable damage retains its own causal identity. An original analytical synthesis links the unit of loss, the verified mechanism of damage, the contract's exact exclusion and exception, the governing law, and the evidence available for allocation. This synthesis is an interpretation of existing doctrine and documentary material, not a newly proposed theory. The principal conclusion is that the legally relevant question concerns an identified insured loss under a particular policy and governing law, rather than the undifferentiated label assigned to the catastrophe.
Keywords: Property insurance; anti-concurrent causation; exclusions; efficient proximate cause; concurrent causes; insurance contract interpretation; wind and flood; property damage; causation evidence; ensuing loss; mixed-peril allocation; comparative insurance law; catastrophe claims.
Introduction
A property policy translates a complex physical occurrence into a defined financial obligation. The translation requires several independent determinations: what insured interest exists; which property suffered which damage; whether the damage satisfies the initial coverage grant; which exclusion reaches the relevant cause; whether an exception restores coverage; and which monetary conditions govern payment. The word hurricane, for example, does not uniquely identify a covered or excluded event. The same storm may generate direct wind damage, storm-surge inundation, rain entering through a breached roof, freshwater flooding, and later mold. Each mechanism has different evidentiary and contractual consequences.
The recurring wind-versus-water disputes after Hurricane Katrina show why a general disaster narrative cannot substitute for a building-by-building causal account. Abraham's analysis of Katrina insurance claims locates the dispute at the intersection of standard homeowners exclusions, insurers' catastrophe exposure, and differing allocations of wind and water damage. Swisher's study of concurrent-causation disputes develops competing legal interpretations and the procedural defenses that make a seemingly plain ACC clause difficult to apply. Those works identify a problem of contractual interpretation and a problem of evidence that frequently appear together but must be answered separately.
This study uses comparative documentary analysis, not a representative statistical sample of all insurance claims. Primary materials consist of statutory texts, regulatory policy forms and reported judgments. Selected scholarship is used to situate disagreement and identify interpretive assumptions; the judgment itself controls the description of a case's actual holding. Court decisions are jurisdiction- and policy-specific. Their legal treatment may change with later authority, amendments, different endorsements, or statutory consumer protections. No inference of a universal rule is made from the selected decisions.
Throughout, four distinct quantities are kept separate: physical damage, economic consequences of that damage, contractually covered loss, and the insurer's amount payable after deductibles and limits. In catastrophe portfolios, there is also a fifth level: the insurer's aggregate cost across claims. The allocation of a peril at the asset level does not automatically determine the aggregate cost of catastrophe exposure, because limits, sublimits, exclusions, litigation outcomes, and policy mix intervene. The monograph concerns the contractual and scientific problem at the asset and claim levels while identifying broader market implications.
Definitions, Historical Context, and Primary Source Foundations
Coverage grant, exclusion, exception, and contractual hierarchy
The coverage grant defines the subject matter and triggering form of loss. A named-perils policy specifies hazards whose occurrence may establish coverage; a special-form or open-perils property policy generally begins with broadly described fortuitous direct physical loss and then applies exclusions and other conditions. Neither drafting approach makes the policy an unlimited guarantee. The decisive contract includes declarations, base forms, conditions, definitions, endorsements, and controlling statutory requirements. A peril description appearing in a schedule cannot be evaluated in isolation from a more specific endorsement.
An exclusion removes a defined category of loss otherwise potentially within the general grant, or specifies the boundary of that grant, depending on the contract's construction. An exception to an exclusion may restore a limited portion of coverage. An ensuing-loss or resulting-loss provision may insure distinct downstream damage that follows excluded defective work, while leaving the cost of correcting the defective work itself excluded. The Washington Supreme Court's Vision One decision illustrates how factual separation and an ensuing-loss clause change the outcome. A causal sequence is therefore read alongside the precise textual architecture of the policy.
Anti-concurrent-causation wording is typically attached to particular exclusions. Its practical aim is to prevent an excluded peril from becoming insured merely because another covered event contributed to the same loss. The clause often refers to direct or indirect causation and to other events contributing concurrently or in any sequence. Its domain must be read: some policies place ACC language before a list of specified hazards; others attach narrower language to a single peril. It is analytically unsound to assume that every exclusion in a policy inherits a global ACC rule.
From proximate cause to policy-specific causation
Insurance causation developed in dialogue with ordinary legal concepts of proximity, but its central purpose is the attribution of insured loss to contractual perils. The Marine Insurance Act 1906, section 55 articulates an English starting point: subject to the Act and contrary policy provision, the insured peril must be the proximate cause. In this context proximity refers to causal effectiveness, not necessarily the event immediately preceding damage. English law also recognizes that two effective proximate causes may operate together.
In California, Insurance Code section 530 addresses a peril insured against as proximate cause and an uninsured peril as remote cause. The California courts have treated the provision as foundational for efficient proximate cause in first-party property coverage. The operative question is which risk predominated in producing a particular loss, rather than merely which condition came first. The Howell opinion is important for its insistence that contractual language cannot simply erase the state-law causal attribution of damage efficiently caused by an insured fire, even when landslide follows.
Historical evolution cannot be reconstructed solely from modern storms. Disputes over fire and explosion, rain and landslide, negligent construction and soil movement, or insured marine perils and vessel defects all expose the same tension. The choice of which event counts as a peril, and the scale at which causes are individuated, changes the legal analysis. The later insertion of ACC language was a contractual response to that uncertainty, rather than an empirical discovery about the physics of catastrophes.
The scientific unit of loss
Physical causation is studied at a defined unit. For an insured building, appropriate units may be a roof section, a wall assembly, electrical equipment, a first-floor inventory group, or the entire structure when it has failed indivisibly. For a business-interruption claim the unit may be a time interval or revenue stream, subject to specialized coverage provisions. Two perils can affect the same location at different times and generate separate damage, or act simultaneously on one assembly. These cases require different factual descriptions even when the policy's ACC wording is identical.
Let the set of physically damaged components be indexed by i. Define D_i as the cost to restore component i under a stated valuation basis; its causal attribution is a separate finding. A loss may be described by indicators W_i for wind damage, F_i for flood damage, and M_i for an inseparable wind-flood contribution. These indicators are evidence-dependent classifications, not independent dollar amounts automatically additive across overlapping destruction. If a wall was damaged by wind and subsequently destroyed by flood, summing full replacement values for each cause would double-count the same asset. Loss reconstruction must track pre-event state, subsequent damage and residual value.
Material scope of exclusions
Common property exclusions concern flood or surface water, earth movement, ordinance or law, deterioration, corrosion, wear and tear, latent defect, faulty workmanship, intentional acts, nuclear hazards and other specified causes. The title of an exclusion does not determine all its terms. Coverage may be restored through purchased flood insurance, ordinance-or-law endorsements, equipment breakdown provisions, or tailored manuscript forms. FEMA's Standard Flood Insurance Policy defines a distinct federal flood coverage arrangement with its own exclusions, conditions, and covered-property requirements. Its presence illustrates a separate risk-financing contract, not an implied change to a private homeowners policy.
The NAIC flood insurance overview explains the broad American division between ordinary homeowners and separate flood coverage. That descriptive market fact is different from a judicial construction of any particular contract. A comprehensive analysis must therefore retain three documentary levels: original policy wording; legal authority deciding how it operates; and market or regulatory evidence describing why the coverage distinction matters.
Method and authority boundaries
The research compares a purposive collection of original judicial decisions. It records the court and year, material factual mechanism, relevant policy language, doctrine applied, and resolution of the disputed claim. Federal diversity decisions are identified as federal courts applying or predicting state law, while state supreme-court decisions state that court's resolution within its jurisdiction. Later state decisions can clarify or displace earlier federal predictions. The source set is sufficient to demonstrate material doctrinal variation, not to calculate national prevalence or current litigation success rates.
Academic writing is secondary evidence about legal debates. Smith and Simpson's Texas first-party causation analysis distinguishes multiple causal structures and allocation difficulties. Altemose's Hurricane Ike commentary presents a critical policyholder perspective on ACC exclusions. Vaughan's wind-versus-water analysis articulates a contrasting case for enforcing written hazard boundaries. This monograph evaluates those positions against primary decisions rather than adopting one as a universal rule.
Causation and Concurrent Perils
Four causal configurations
A sequential dependent chain has one event materially setting another in motion: heavy rainfall saturates a slope; the slope fails; debris destroys a house. A separate independent-cause event consists of two forces capable of causing distinct damage: wind removes roof tiles in the morning; floodwater later destroys ground-floor contents. A joint or convergent event occurs when covered and excluded forces simultaneously damage the same component, as when wind-driven surge and wind pressure together collapse a wall. An overdetermined event occurs where two sufficiently powerful forces each might have produced the same outcome, though the actual historical process involved both. These are descriptions of possible mechanisms; their legal treatment follows policy wording and governing law.
Time order alone cannot settle the distinction. A wind-driven surge may occur after wind damage while remaining physically connected to the meteorological event. But a policy exclusion using the phrase in any sequence still requires a causal relationship within its textual scope; a remote coincidence somewhere in the storm's chronology is not necessarily a cause of every damaged component. In Corban v. USAA, the Mississippi Supreme Court drew particular attention to separately caused sequential wind and water damage and rejected an interpretation that would conflate those independent losses. Its discussion of simultaneous indivisible forces should be read with the actual policy and procedural posture.
Efficient proximate cause
Under an efficient-proximate-cause inquiry, the finder of fact identifies the predominant causal force that set the relevant loss-producing mechanism in operation, subject to applicable authority. In Kish v. Insurance Co. of North America, the Washington Supreme Court explained the doctrine's application when two independent forces operate and a covered peril initiates an unbroken chain resulting in loss. Its opinion also cautioned against relabeling a single excluded cause as if it were two independent forces. That caution matters in flood claims where rainfall, runoff, overflow, surge and river stage may be used interchangeably despite different physical meanings.
California's Julian ruling makes the causal individuation issue especially clear. Rain inducing a landslide was treated as an intelligible specific peril that could be excluded, while the court also recognized limitations on drafting that would make coverage illusory. The holding addressed the particular mechanism and precise exclusion before the court. It should not be paraphrased as an unlimited permission to exclude a loss efficiently caused by any covered risk, nor as a prohibition on all specific weather-related exclusions.
An efficient-proximate-cause decision requires at least a plausible factual distinction between cause and consequence. Engineering records may distinguish fire-induced vegetation loss from precipitation as causal antecedents in later debris flow, but the legal test then asks which of these is efficient under the applicable doctrine. A geological model demonstrating that precipitation was necessary is informative without itself deciding whether the fire or rain predominated legally.
Independent concurrent causes and the absence of ACC wording
In Sebo v. American Home Assurance, the Florida Supreme Court applied the concurrent cause doctrine to independent perils combining in an all-risk property loss where the pertinent policy language did not preclude that recovery. Rain, wind and construction defects jointly caused damage; the court rejected the insistence that a single efficient cause always had to be identified. The decision was a construction of a specific policy, and the absence of effective anti-concurrent wording in the relevant exclusion was material to the result.
The later Florida appellate Tower Hill v. Bermudez decision emphasizes placement. It recognized that an earth-movement ACC clause would matter if earth movement caused the damage, but that it did not automatically govern other covered/excluded combinations. The jury's factual finding concerning blasting-induced shock waves therefore mattered. This is a strong demonstration that broad descriptions such as the policy has ACC language are insufficient. The questions are where the clause appears and whether the factfinder identified a cause within the enumerated exclusion.
Identifiability and the counterfactual problem
For a component i, imagine a hypothetical comparison between D_i(W=1,F=0), D_i(W=0,F=1), and D_i(W=1,F=1), where W and F represent wind and flood exposure under otherwise fixed conditions. The comparisons can reveal whether damage is additive, substitutive, or interactive. They are counterfactual constructs, not observable claims facts. Only the realized history D_i(W=1,F=1) is directly observed in a mixed event; the other outcomes require a credible physical model, comparable undamaged structures, or admissible expert inference.
If two causes jointly exceed a failure threshold, each may have had a substantial causal role even though neither alone would have produced the loss. Conversely, if wind already destroyed a roof before flood arrived, later flooding cannot automatically be credited with that earlier roof damage. A catastrophe model's loss fraction is not legal attribution by itself. The admissible evidentiary question is whether sufficient observations and engineering assumptions allow the claim's material loss units to be reconstructed to the required standard of proof.
Burden of proof and causal segregation
The insured generally must establish an initial covered loss under applicable policy and procedural rules; the insurer may carry the burden of establishing an exclusion. Allocation requirements depend on the jurisdiction and contract. In Wallis v. USAA, the Texas court required a reasonable evidentiary basis for allocating foundation damage to plumbing leaks rather than other excluded causes. Expert testimony that leaks contributed, without a supportable estimate of the amount attributable to them, was insufficient for the damages percentage found by the jury.
This exposes a distinction between existence evidence and magnitude evidence. A photograph of broken shingles can support the presence of wind damage; it does not necessarily establish the full replacement cost of every roof component. An expert's statement that two perils were contributing factors can support mixed causation; it does not quantify covered damage. Insurers' exclusion defenses and insureds' loss allocations must be evaluated through their distinct evidentiary burdens.
Double causation and nonadditive damages
A component-level accounting identity can be written as D_total = D_separable + D_joint + D_unattributed, provided the three terms represent nonoverlapping physical restoration costs on one consistent valuation basis. This is merely an accounting partition. Legal coverage is not obtained by assigning D_joint to an insurer mechanically: the jurisdiction's ACC language, concurrent-cause rule, ensuing-loss provision, and proof requirement control its treatment. Unattributed damage is an epistemic category, not automatically an excluded peril.
The insurer's payment also differs from loss. For an illustration with $100,000 of proven covered repairs and a $5,000 ordinary deductible, an assumed dollar-deductible policy with adequate limits would pay $95,000. If only $35,000 of those repairs are established as covered, the same illustrative contract yields $30,000. Such arithmetic demonstrates sensitivity to allocation, while actual deductibles can be percentage-based, per occurrence, per peril, or affected by other terms. No probability or national payment rate is inferred from the example.
Anti-Concurrent Clauses and Legal Effect
How the clause changes the coverage question
ACC wording attempts to specify that a listed excluded cause defeats coverage for an affected loss even if a covered cause contributes simultaneously or in a causal sequence. The clause addresses contractual attribution. It does not prove that an excluded peril was present; the causal predicate remains a factual question. Nor does it decide whether an exception or endorsement specifically restores coverage. The full contract must be applied to an established loss, not the clause alone.
The Fifth Circuit's Leonard v. Nationwide judgment is a central example from Hurricane Katrina. The court enforced flood-related exclusion and ACC language under its reading of Mississippi law, while preserving coverage for damage shown to have resulted solely from wind. It rejected the proposition that storm surge was outside the water exclusion merely because wind propelled it. The insureds' proof of a larger undifferentiated repair estimate did not establish that the full amount was wind damage.
In Tuepker v. State Farm, another Fifth Circuit Katrina case, the court evaluated the interaction of hurricane wind and storm surge under a different policy. The judgments are evidence of federal courts' contemporary reading of the forms and state law, not a substitute for the Mississippi Supreme Court's later, more granular discussion in Corban. A scientifically controlled comparative account records the difference in courts, dates, form language and causal facts before comparing the results.
Separately damaged property and indivisible loss
A clause directed to loss caused by flood does not make an entire hurricane one undifferentiated excluded loss. The relevant property item may have suffered independent wind damage that can be reconstructed even if other parts of the same house were flooded. In the Corban opinion, the court distinguished the timing and separateness of loss when construing the ACC language. The strongest general inference is procedural: loss segmentation must occur before treating a multi-peril catastrophe as an indivisible event.
Indivisibility is a conclusion requiring evidence. A structure's collapse may have physically combined forces that forensic observation cannot adequately separate, or the available records may simply be too poor to disentangle otherwise separable events. The first possibility concerns the damage mechanism; the second concerns epistemic uncertainty. Treating lack of evidence as proof of indivisible physical causation can change contractual outcomes without scientific warrant. A careful claims file states which problem exists.
Texas: independent damage, combined causes, and contingent obligations
The Texas Supreme Court's JAW The Pointe decision concerned ordinance-compliance costs after Hurricane Ike. Wind damage was covered and flood damage excluded. The city-required rebuilding obligation depended on substantial damage; the policy's ordinance endorsement required a covered cause to trigger that obligation. The court found that both wind and flood damage in fact contributed to the city's enforcement decision and applied the ACC provision to deny the disputed ordinance-compliance claim.
JAW is not simply a judgment that every item struck by both wind and flood is excluded. The precise claimed loss was the expense of complying with city ordinances. The court examined the historical causal basis of the city's action, including the insured's combined permit estimate and the city's notice. It rejected an alternative argument that wind damage hypothetically could have been sufficient by itself, because the actual enforcement evidence showed combined causation. The case is therefore especially useful for analyzing a secondary legal obligation whose trigger depends on an underlying physical-loss calculation.
Texas allocation doctrine also has a different history under policies without decisive ACC text. In Wallis, inability to prove a reasonable amount of solely covered damage defeated recovery on the relevant proof. These cases address different questions: evidence of a separable covered dollar amount, and whether an ACC clause excludes an indivisible downstream ordinance obligation. Neither should be reduced to a universal formula for all Texas property disputes.
California: mandatory attribution and intelligible exclusions
California's efficient-proximate-cause line is often described as a constraint on attempts to contract around statutory causation. The Howell analysis involved a fire, subsequent rains and landslide damage; the covered fire was argued to have been the efficient cause. The court resisted treating a broadly worded earth-movement ACC exclusion as decisive independently of the efficient-cause inquiry.
The California Supreme Court's later Julian decision qualified any overbroad reading of that proposition. The court upheld a specific exclusion for rain inducing a landslide on the facts before it and maintained that insurers can define genuine compound perils clearly, while warning against manipulative drafting that renders coverage illusory. The underlying principle is not that exclusionary drafting is prohibited. It is that the contract's delineation of a real insured hazard, the state-law causation requirement, and the objectively understood peril must cohere.
Alaska: enforcement of contractual ACC allocation
The Alaska Supreme Court's Bongen decision upheld an earth-movement exclusion framed to apply irrespective of concurrent or sequential causes in a mudslide claim involving alleged upstream construction activity. Alaska's court permitted an insurer to define the scope of coverage by that language and declined the lower court's conclusion that the exclusion was unenforceable. Bongen is a counterexample to any assertion that a single nationwide public-policy rule invalidates ACC clauses.
The case also shows the difference between two scientifically distinct assertions: that human land disturbance helped cause a landslide, and that the insured policy covered a loss when landslide was excluded irrespective of other contributing causes. An engineering report can establish the first while the text and applicable law decide the second. Conflating them produces an argument that is causally plausible and contractually incomplete.
Endorsements and ensuing losses
A water exclusion can coexist with added sewer-backup coverage, a limited flood endorsement, or a separate flood contract. An ordinance-or-law exclusion may coexist with additional coverage activated only by damage from a covered cause. Faulty workmanship can remain excluded while ensuing damage to other property is covered. The question is not merely whether an endorsement exists, but what factual predicate and monetary limit it creates.
In Vision One, the Washington Supreme Court treated the resulting-loss clause in a construction collapse as material and considered the insurer's chosen causal grounds for denial. The decision illustrates the importance of distinguishing costs of fixing defective shoring from damage to different completed building components. Such provisions prevent the broad word defect from being treated as a conclusive label over all subsequent physical consequences.
Legality, consumer protection and claim handling
The freedom to specify excluded risks is not identical across jurisdictions or policyholder classes. Mandatory statutes, rules on ambiguity, requirements for clear and conspicuous exclusions, market-conduct obligations, and public policy may all affect the enforceable scope. The United Kingdom's FCA claims-handling rules require claims to be handled promptly and fairly and constrain unreasonable rejection. These rules concern conduct standards and cannot alone determine how an American state court would read an ACC clause.
Within the United Kingdom, Insurance Act 2015, section 11 addresses terms designed to reduce risk and circumstances where noncompliance could not have increased the risk of the loss. Its scope excludes terms defining the risk as a whole. It is a valuable comparative example of legislative attention to causal relevance, but it is not a statutory invalidation of all insured-peril exclusions or a direct transposition of U.S. ACC doctrine. Song's analysis of section 11 causation distinguishes the risk-increasing potential of a breach from proof of actual causal contribution, which reinforces the importance of identifying the legal question being decided.
Comparative Coverage Litigation
Mississippi after Katrina: wording, timing and proof
The Katrina sequence generated an unusual concentration of cases involving a familiar pattern: wind was a covered peril, flood or storm surge was generally excluded under homeowners wording, and catastrophic destruction impaired later observation. In Leonard, the Fifth Circuit addressed the enforceability of ACC wording and the distinction between wind-only damage and water damage. In Tuepker, the court considered another insurer's wording. In Corban, the state supreme court held that separate sequential wind and flood damage could not be treated as an indivisible concurrent loss on the record before it.
The resulting picture is more precise than a binary pro-insurer or pro-insured classification. Flood exclusion language can exclude flood damage while allowing provably distinct wind damage. A simultaneous inseparable mixture can engage ACC wording. A later state-court opinion can refine the meaning of the very words a federal diversity court had construed earlier. Swisher (2014) and Abraham (2007) frame the stakes of the disagreement; the cases define its adjudicated details. Conclusions drawn from the litigation must be assigned to the decided contracts and records, not projected without qualification to every future policy.
Florida: concurrent cause with a narrow contractual predicate
Florida's Sebo holding concerns the combination of independent perils under the specific all-risk wording. It does not convert all accidental physical damage into covered loss; exclusions and their causal limits remain part of the contract. The Tower Hill opinion demonstrates the dispositive placement of anti-concurrent language in an earth-movement clause and the need for evidence that earth movement actually occurred. It also supplies later authority showing that Sebo continued to be applied conditionally, with explicit attention to ACC wording.
This pair of cases makes policy-form comparison necessary. A study that extracts a state's concurrent-causation rule without recording each actual exclusion and the insured physical loss will misclassify outcomes. A study that records only ACC presence will also misclassify outcomes when its location and the causal mechanism matter more than its mere presence.
California, Washington and Alaska: three distinct answers
California's Howell and Julian decisions address efficient-cause protection and the legal possibility of defining a genuine compound exclusion. Washington's Kish and Vision One examine independent forces, resulting-loss language and the specific causal theory advanced by the insurer. Alaska's Bongen demonstrates contractual enforcement of a strong earth-movement ACC provision.
These jurisdictions share the underlying physical problem while employing different legal premises and textual analyses. Whether a particular cause is covered or excluded must first be defined. Whether the loss was caused by that event is a factual matter. Whether an ACC clause may govern despite a covered efficient cause is a distinct legal determination. Any comparative claim that merges the three is likely to be overgeneralized.
Texas: two different causal objects
Wallis concerned allocating damage among plumbing leaks and soil-related causes for foundation repairs. JAW The Pointe concerned increased costs imposed by a city's rebuilding ordinance after wind and flood damage. The cases differ in what the insured was attempting to recover, the form language and the required causal proof. If one characterizes both merely as mixed-peril claims, the legally decisive distinction disappears.
For the ordinance example, three quantities must be distinguished: (1) the damage to a building; (2) the authority's determination that a rebuilding threshold was crossed; and (3) the additional cost of legal compliance. A covered wind loss may be paid independently, yet the third quantity may still be excluded when the second was in fact triggered by combined damage under policy language like that in JAW. This separation has implications for adjuster files: municipal permit data, valuation of the property and contemporaneous repair estimates can be as causally important as meteorological observations.
English law: insured, uninsured and excluded proximate causes
The United Kingdom provides a comparative reference beyond American ACC drafting. The Marine Insurance Act's proximate-cause provision has a default subject to contrary policy terms. In FCA v. Arch, the UK Supreme Court examined overlapping effective causes and distinguished the situation where a proximate uninsured cause is merely outside the insured description from one where the cause is expressly excluded. The judgment revisited the marine decision The Miss Jay Jay while analyzing causation in pandemic business-interruption wording. It is used here to illuminate general insurance causation reasoning, not to assert that pandemic interruption is physically equivalent to hurricane damage.
The English Court of Appeal's London International Exhibition Centre judgment restated that where one of two equally effective proximate causes is insured and the other is neither insured nor excluded, cover can follow, whereas an express exclusion can alter the result. This distinction is often obscured by treating all not-covered events as equally excluded. A peril outside an affirmative grant, an express exclusion, and an excluded risk subject to an exception are contractually different categories.
Zhao's English and Chinese comparative study situates the problem within another legal system and makes the case for careful separation of the insured, uninsured and excluded categories. Its proposals remain the scholarly author's argument, not a declaration of current uniform law in England or China. This monograph's primary comparative conclusions rest on the English statutes and reported judgments actually examined.
Cross-jurisdictional comparison
| Jurisdiction and principal authority | Causation issue | Decisive contract or legal feature | Supported conclusion |
|---|---|---|---|
| California, Howell and Julian | Fire/rain/landslide and compound weather perils | Section 530, efficient proximate cause, genuine peril definition | Broad avoidance language faces limits; specifically defined rain-induced landslide was enforced on Julian's facts |
| Washington, Kish and Vision One | Independent forces and ensuing collapse damage | Efficient-cause analysis, resulting-loss clause, denial grounds | Specific loss mechanism and exception language determine treatment |
| Alaska, Bongen | Human activity, mudslide and excluded earth movement | Express earth-movement ACC wording | Clause enforced on the dispute before the court |
| Mississippi, Leonard, Tuepker and Corban | Wind, surge and possible sequential loss | Water exclusions, ACC scope, independent damage | Proven separate wind damage must be distinguished from flood or indivisible mixed damage |
| Florida, Sebo and Tower Hill | Independent convergence; blasting/earth movement | Concurrent-cause doctrine absent applicable ACC; placement of ACC | A clause for one exclusion does not automatically extend to another peril |
| Texas, Wallis and JAW | Foundation allocation and ordinance trigger | Evidentiary segregation; ACC on downstream obligation | Different loss objects entail different proof and coverage results |
| United Kingdom, FCA v. Arch and London International Exhibition Centre | Multiple effective proximate causes | Insured versus merely uninsured versus expressly excluded risk | Express exclusion changes concurrent-proximate-cause analysis |
This table is a descriptive comparison of selected authorities. It is not a frequency survey of decisions or a current fifty-state legal memorandum.
Empirical Verification, Uncertainty, and Competing Explanations
Documentary and evidentiary design
A reproducible causation review starts by retaining the exact policy edition and endorsement schedule effective on the loss date. The next records are pre-loss property condition, event chronology, damage photographs with metadata, engineering inspection, hazard observations, municipal determinations, contractor quantity surveys, repair invoices, and payment records. The last layer is the authoritative legal record for the policy's jurisdiction. Each layer answers a different question. A weather observation supports event intensity; it does not alone prove that a particular wall failed from wind pressure. A repair invoice supports cost; it does not establish that the repaired damage is a covered loss.
In large events, claim investigators can observe waterlines, debris patterns, roof anchorage failures, direction of displaced materials, building elevations, local gauges, and photographs of adjacent properties. The evidentiary value of each depends on timing and alternate explanations. A waterline shows inundation occurred; it does not exclude earlier roof damage. An absent roof after a storm does not uniquely distinguish wind uplift, floating debris impact, or demolition during recovery. Evidentiary integrity is therefore a scientific constraint on both policyholder and insurer inferences.
Controlled reconstruction of a mixed-loss example
Consider an entirely hypothetical warehouse with undamaged roof and electrical panels before a storm. A contemporaneous early photograph shows $60,000 of wind-broken roofing. Hours later, a documented flood damages electrical panels by $90,000. Engineers also identify $40,000 of wall damage from simultaneous wind and hydrodynamic pressure. Assume a special-form policy covers wind, excludes flood by applicable ACC wording, and contains no pertinent endorsement. Total nonoverlapping restoration cost is $190,000.
Under an assumed factual finding that the first $60,000 was caused solely by wind, that amount is a covered-loss candidate. The $90,000 caused solely by flood is excluded. The $40,000 joint component depends on whether the court enforces the ACC provision on the identified mixed damage. If it does, the candidate covered loss is $60,000. If a controlling nonwaivable causation rule or a more specific exception changes attribution, the result must be recalculated. With a hypothetical $10,000 applicable ordinary deductible and sufficient limits, a $60,000 covered loss produces an illustrative $50,000 payment. This is a transparent arithmetic example, not a prediction of an actual case.
Suppose instead that the early roof photograph is absent and experts can only say the roof was damaged sometime during the event. The physical event has not changed, but the evidence quality has. The demonstrated covered amount might fall below $60,000, or additional reliable evidence might recover that inference. The proper record is an allocation interval or contested estimate. A model should not silently transform missing information into proof that flooding caused the entire loss.
Reconstruction of an ordinance trigger
Assume a municipality requires code upgrades when covered and excluded property damage combined exceeds 50% of a reference value V. In a hypothetical case, let V = $2,000,000, proven wind repair cost = $700,000, proven flood repair cost = $500,000, and combined qualifying damage = $1,200,000, subject to the municipality's actual measurement rules. The combined percentage is 60%; wind alone is 35%. If the upgrade is triggered by combined damage and an endorsement covers it only when a covered cause triggers the ordinance, the policy's ACC provision becomes relevant to the additional code-compliance cost. The 50% rule here is a stipulated example; its presence, basis and valuation method must be confirmed in any real jurisdiction.
A different hypothetical scenario with wind cost $1,100,000 and flood cost $100,000 makes wind independently sufficient to cross 50%. Even then, the legal effect depends on evidence of how the authority actually decided to enforce its rule and what the contract says. The JAW court's actual-trigger analysis explicitly shows why a theoretical sufficiency argument must not replace the historical causal record. The arithmetic is an audit device; the judgment itself is the authority for the doctrine described.
Sensitivity analysis and error propagation
Suppose the disputed wall-repair component of $40,000 in the warehouse example has a plausible expert range from $25,000 to $55,000, with the other quantities unchanged. The physical total ranges from $175,000 to $205,000. If the entire joint component is excluded under a valid ACC provision, the covered wind candidate remains $60,000 throughout the range. If the joint component is coverable, the covered candidate ranges from $85,000 to $115,000. The range is driven by a legal classification as well as by estimation error. Replacing it with a single expected value conceals the most important discontinuity in the dispute.
An uncertain model's output should indicate whether uncertainty is about property cost, peril occurrence, temporal sequence, causal sufficiency, legal classification, or documentary completeness. These are different uncertainties and should not be collapsed into one confidence score. Cost inflation affects the repair estimate. A missing policy endorsement affects contractual scope. A disputed flood height affects physical attribution. A later appellate decision affects the legal rule. Each requires a separate corrective action.
Comparative research validity
This study uses purposive legal authorities selected for doctrinal contrast, not a random sample. Therefore it cannot estimate the proportion of U.S. claims excluded by ACC clauses or the financial effect on households from these cases. Such claims would require policy-form exposure denominators, claim counts, peril-specific repair estimates, geography, time, litigation selection effects, insurer decisions, and a coherent sampling strategy. Published appellate cases are highly selected: they reflect disputes that survived negotiation, settlement, judicial filtering and appeal. The pattern of reported opinions cannot be interpreted as a representative claims dataset.
Peer-reviewed and university-hosted scholarship identifies contested explanations. Knutsen's study of catastrophic insurance causation describes instability in causation approaches and argues for alternative legal rules; that is a proposed academic solution, not an adopted universal doctrine. Swisher (2014) emphasizes interpretive and policyholder defenses; Vaughan (2008) emphasizes contractual certainty and insurer pricing capacity. Both perspectives illuminate mechanisms, but neither replaces the original case law. This monograph's synthesis draws only conclusions supportable by the verified record.
Exclusion economics and distribution
An insurer that excludes flood risk from a homeowners policy has shifted the financing need for that hazard to the owner, another insurer, a flood program, or residual public and household balance sheets. The exclusion may help define insurable capacity and pricing, but a household's understanding of the resulting gap is a separate empirical variable. The federal flood coverage explanation makes explicit that ordinary homeowners insurance and flood insurance are typically separate arrangements. The existence of a private property contract therefore cannot be used as evidence that all economically catastrophic natural hazards are funded.
Ambiguous damage allocation also carries costs: engineering investigations, delayed rebuilding, expert disagreement, and litigation. These costs need not be assigned automatically to either party; the evidence of their magnitude is absent from this dossier. Nonetheless the direction of the mechanism is clear. A clause that is precise about peril and unit of loss can reduce some forms of uncertainty; a sweeping clause applied to poorly segregated damage can increase factual controversy. The scientific objective is to reveal where the uncertainty lies and which evidence can resolve it.
Alternative explanations and falsification checks
A roof breach attributed to wind may instead arise from preexisting defects; the claim file must distinguish pre-event deterioration from storm-created breakage. Moisture intrusion attributed to flood may arise from rain through an opening or ground seepage; the physical path should be measured. A collapsed wall attributed to storm surge may have failed earlier from wind pressure; dated imagery or structurally consistent deformation may test the hypotheses. A purported ACC outcome may reflect not the clause's legal reach but the claimant's failure to prove a separate covered dollar amount. Every explanation implies a different observable fact or documentary requirement.
For judicial comparison, a competing explanation is that an apparent jurisdictional difference simply reflects wording differences. The test is to compare the full exclusion, the coverage grant and the factual unit across cases. Another competing explanation is temporal: a federal court predicted law later clarified by a state supreme court. The test is to order judgments by date and source of authority. A third is procedural posture: a motion to dismiss, summary judgment, jury verdict and appeal each permit different factual assumptions and review standards. A careful causal/legal comparison records these distinctions rather than treating every holding as a controlled experiment.
Integrative Comparative Scientific Synthesis
The loss precedes legal attribution
The corpus supports a consistent analytical sequence. First, identify the exact property interest and the physically damaged component. Second, establish the actual loss-producing mechanism and chronology. Third, read the coverage grant and the relevant exclusion or exception at that scale. Fourth, determine whether the applicable jurisdiction applies ordinary efficient proximate cause, an independent concurrent-cause principle, allocation of separately covered damage, or enforceable ACC wording. Fifth, quantify the covered amount under the governing valuation, deductible and limit terms. This is a practical synthesis of existing legal and scientific methods, not an instituted new causal theory.
The ordering matters because otherwise the legal label begins to dictate the supposed physical finding. The statement flooding is excluded cannot show that a roof was flooded. The statement this was a hurricane cannot show that wind alone damaged every component. The statement construction was defective cannot establish whether excluded defects and covered ensuing collapse have identical insured subjects. The scientific interpretation therefore starts with asset-level facts, while acknowledging that a contract may define some compound events as a distinct excluded peril.
Four comparisons that change outcomes
The first comparison is independent earlier wind damage versus later flood destruction. Corban shows the importance of not eliminating separate historical losses by merging them into one event. The second is concurrent independent causes under policy wording that lacks an applicable ACC clause. Sebo demonstrates why the concurrent-cause inquiry can then matter. The third is an expressly excluded cause contributing to the same downstream obligation. JAW The Pointe illustrates the contractual and evidentiary consequences. The fourth is a genuine compound peril within a jurisdiction emphasizing efficient cause. Julian demonstrates the importance of precise hazard definition.
These comparisons collectively defeat two overstatements: that any covered contributor always produces payment, and that any excluded contributor always defeats payment. Both disregard policy and jurisdiction. The verified conclusion is narrower and more useful: the relationship among particular causes, loss units, clauses and legal constraints determines coverage.
Implications for insurance product design
A well-specified property form should state whether exclusions govern particular components or the entire indivisible event, whether sequential causes matter, how exceptions and resulting-loss provisions interact, and how the policy handles physically separable damage. The design should distinguish building damage, contents damage, debris removal, ordinance-compliance cost and interruption of income. Those distinctions can reduce inadvertent inconsistencies between a broad ACC preamble and an endorsement apparently purchased to restore a portion of the excluded exposure.
Disclosure should explain an ordinary loss scenario where covered and excluded causes interact. An owner may understand that flood is excluded yet remain uncertain about a wind-destroyed roof, a flood-destroyed ground floor, or code upgrades triggered by both. Clear examples reveal the actual risk transfer more effectively than a bare list of peril names. They also improve the quality of the record available to regulators examining product design and fair handling.
Implications for adjustment and scientific reproducibility
A robust claim investigation preserves the sequence of observations before major reconstruction changes the site. It maintains the exact policy wording, ties each image or expert conclusion to a component and time, records allocation assumptions, identifies disputed causal alternatives, and explains whether uncertainty concerns observability or a genuinely joint physical mechanism. The FCA fair-claims requirements are one regulatory expression of the importance of prompt, reasoned handling, though their jurisdictional scope is specific to the United Kingdom.
Reproducibility is strengthened by a ledger containing every material causal claim, the supporting observation, the contract term to which the claim matters, and the authoritative legal source used in interpretation. In an actual claim, confidentiality and privacy obligations would govern the records. This monograph uses public decisions and hypothetical examples and contains no private insured claims dataset. Its evidence ledger supports documentary audit of statements rather than reanalysis of confidential loss files.
Research implications
Future observational work could compare similar catastrophe losses across jurisdictions and policy wording, but reliable causal inference would require controlling for hazard severity, construction, insured value, endorsement take-up, and the selection of disputed claims. It could also study how detailed claim segmentation affects settlement duration, the frequency of reopening decisions, and the cost of disputed engineering evidence. Such projects would require new data access and an explicit design; they have not been conducted here.
Comparative legal research should include statutes and subsequent cases as of a defined cutoff, because policy wording and state doctrine evolve. Research on cross-border commercial risk should separately classify English proximate-cause doctrine, statutory conditions on risk-mitigation terms and contractual exclusions. Zhao's comparison of English and Chinese law is a useful research agenda, while the present primary-source collection establishes only the comparative points explicitly examined.
Final scientific interpretation
The most consequential analytical error in multi-peril claims is to choose the legal category before identifying the loss. A catastrophe is an environmental occurrence; a policy covers defined losses to defined interests; an ACC clause regulates the treatment of excluded causes within the relevant wording; and a court applies that wording within governing law. The relationships across these levels explain why apparently similar storms produce differing outcomes and why apparently clear exclusions can still require extensive expert investigation.
There is also a practical asymmetry. Scientific uncertainty is usually gradual, expressed through competing mechanisms and ranges. Contractual coverage can be discontinuous: establishing one excluded contributing cause within an enforceable ACC provision may change the treatment of an entire joint component or downstream obligation. That discontinuity makes precision in physical loss segmentation and documentary causation especially valuable. It is an inference from the reviewed legal and physical structure, not an estimated empirical effect or new doctrinal rule.
Conclusion
Property insurance exclusions and anti-concurrent-causation provisions govern the intersection of peril interaction, loss measurement and legal attribution. The reviewed primary authorities confirm that neither the event name nor the mere presence of an excluded hazard resolves the claim. The central object is the particular physical or consequential loss, including whether causes produced independent, successive, joint or genuinely inseparable damage.
The comparison establishes several conclusions. An express ACC clause can govern mixed damage under applicable law, as shown in significant Alaska, Texas and federal Katrina litigation. The recoverability of separable wind damage remains conceptually distinct from flood damage, as the Mississippi decisions demonstrate. Florida's concurrent-cause analysis is conditional on the pertinent policy wording; the location of an ACC clause can be decisive. California's efficient-proximate-cause framework places significant emphasis on statutory attribution while permitting the exclusion of intelligibly defined compound perils on appropriate facts. Washington decisions emphasize independent forces, resulting losses and the insurer's articulated causal ground. English decisions further clarify the essential distinction between a merely uninsured proximate cause and a positively excluded one.
The evidence establishes the value of component-level physical reconstruction, preservation of event chronology, inspection of the precise policy edition, and separate proof of cause and amount. A municipal rebuilding trigger, an ensuing-loss exception or a special endorsement can produce a separate legal question from the direct physical destruction that preceded it. Insured physical damage, economically consequential loss, contractually covered loss and actual payment therefore remain distinct analytical categories.
The original contribution of this research is a source-grounded comparative synthesis of these distinctions and two fully identified hypothetical sensitivity exercises. No novel legal theory, universal insurer payment formula, original statistical estimate or cross-jurisdiction empirical claim is proposed. The dossier is designed to support further insurance science research through transparent source attribution, a clear statement of doctrinal limits, and a reproducible evidence record.
References
Abraham, K. S. (2007). The hurricane Katrina insurance claims. Virginia Law Review In Brief, 93, 173–180. https://virginialawreview.org/articles/hurricane-katrina-insurance-claims/
Alaska Supreme Court. (1996). State Farm Fire & Casualty Co. v. Bongen, 925 P.2d 1042 (Alaska 1996). https://law.justia.com/cases/alaska/supreme-court/1996/s-6526-1.html
Altemose, A. L. (2010). The anti-concurrent clause and its impact on Texas residents after Hurricane Ike. Texas Wesleyan Law Review, 16, 201–223. https://scholarship.law.tamu.edu/txwes-lr/vol16/iss2/4/
California Court of Appeal. (1990). Howell v. State Farm Fire & Casualty Co., 218 Cal. App. 3d 1446. https://law.justia.com/cases/california/court-of-appeal/3d/218/1446.html
California Legislature. (2025). California Insurance Code § 530 (2025 codification). https://law.justia.com/codes/california/code-ins/division-1/part-1/chapter-6/article-2/section-530/
California Supreme Court. (2005). Julian v. Hartford Underwriters Insurance Co., 35 Cal. 4th 747. https://law.justia.com/cases/california/supreme-court/2005/s109735a.html
Financial Conduct Authority. (2026, accessed November 16). Insurance: Conduct of Business Sourcebook, ICOBS 8, claims handling. https://handbook.fca.org.uk/handbook/icobs8
Florida District Court of Appeal, Third District. (2023). Tower Hill Prime Insurance Co. v. Bermudez, No. 3D22-0828. https://law.justia.com/cases/florida/third-district-court-of-appeal/2023/3d22-0828.html
Florida Supreme Court. (2016). Sebo v. American Home Assurance Co., 208 So. 3d 694 (Fla. 2016). https://law.justia.com/cases/florida/supreme-court/2016/sc14-897.html
Knutsen, E. S. (2010). Confusion about causation in insurance: Solutions for catastrophic losses. Alabama Law Review, 61, 957–1022. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1322862
Mississippi Supreme Court. (2009). Corban v. United Services Automobile Association, 20 So. 3d 601 (Miss. 2009). https://law.justia.com/cases/mississippi/supreme-court/2009/co58302.html
National Association of Insurance Commissioners. (2026). Flood insurance: Background and consumer information (updated June 23, 2026). https://content.naic.org/insurance-topics/flood-insurance
National Flood Insurance Program. (n.d., accessed November 16, 2026). Buy a flood insurance policy. https://www.floodsmart.gov/get-insured/buy-a-policy
Song, M. (2023). Causation or correlation: The chimera in section 11 of the Insurance Act 2015. Legal Studies, 43(2), 278–294. https://www.cambridge.org/core/journals/legal-studies/article/causation-or-correlation-the-chimera-in-section-11-of-the-insurance-act-2015/C2D616A19244BE6752A5CED7EC11C31D
Smith, R. L., & Simpson, F. A. (2006). Insurance causation: Texas law on first-party claims. Texas Wesleyan Law Review, 13, 63–93. https://scholarship.law.tamu.edu/txwes-lr/vol13/iss1/4/
Swisher, P. N. (2014). “Why won't my homeowners insurance cover my loss?”: Reassessing property insurance concurrent causation coverage disputes. Tulane Law Review, 88, 515–572. https://scholarship.richmond.edu/law-faculty-publications/569/
Texas Court of Appeals, Fourth District. (1999). Wallis v. United Services Automobile Association, No. 04-97-00803-CV. https://law.justia.com/cases/texas/fourth-court-of-appeals/1999/6929.html
Texas Supreme Court. (2015). JAW The Pointe, LLC v. Lexington Insurance Co., 460 S.W.3d 597 (Tex. 2015). https://law.justia.com/cases/texas/supreme-court/2015/13-0711.html
United Kingdom Court of Appeal. (2024). London International Exhibition Centre plc v. Allianz Insurance plc and others, [2024] EWCA Civ 1026. https://www.bailii.org/cgi-bin/format.cgi?doc=%2Few%2Fcases%2FEWCA%2FCiv%2F2024%2F1026.html
United Kingdom Parliament. (1906). Marine Insurance Act 1906, section 55, included and excluded losses. https://www.legislation.gov.uk/ukpga/Edw7/6/41/section/55
United Kingdom Parliament. (2015). Insurance Act 2015, section 11, terms not relevant to the actual loss (in force since August 12, 2016). https://www.legislation.gov.uk/ukpga/2015/4/section/11/enacted?view=extent
United Kingdom Supreme Court. (2021). The Financial Conduct Authority v. Arch Insurance (UK) Ltd and others, [2021] UKSC 1. https://www.bailii.org/cgi-bin/format.cgi?doc=%2Fuk%2Fcases%2FUKSC%2F2021%2F1.html
United States Court of Appeals, Fifth Circuit. (2007a). Leonard v. Nationwide Mutual Insurance Co., 499 F.3d 419 (5th Cir. 2007). https://law.justia.com/cases/federal/appellate-courts/ca5/06-61130/06-61130-cv0.wpd-2011-02-25.html
United States Court of Appeals, Fifth Circuit. (2007b). Tuepker v. State Farm Fire & Casualty Co., 507 F.3d 346 (5th Cir. 2007). https://law.justia.com/cases/federal/appellate-courts/ca5/06-61075/06-61075-cv0.wpd-2011-02-25.html
United States Federal Emergency Management Agency. (2021, as codified). Standard Flood Insurance Policy, Dwelling Form, 44 C.F.R. pt. 61 app. A(1). https://www.law.cornell.edu/cfr/text/44/appendix-A%281%29_to_part_61
Vaughan, B. R. (2008). Watered down: Are insurance companies getting hosed in the wind vs. water controversy? University of Illinois Law Review, 2008, 777–810. https://illinoislawreview.org/print/volume-2008-issue-2/watered-down-are-insurance-companies-getting-hosed-in-the-wind-vs-water-controversy/
Washington Supreme Court. (1992). McDonald v. State Farm Fire & Casualty Co., 119 Wash. 2d 724 (1992). https://law.justia.com/cases/washington/supreme-court/1992/58078-2-1.html
Washington Supreme Court. (1994). Kish v. Insurance Co. of North America, 125 Wash. 2d 164 (1994). https://law.justia.com/cases/washington/supreme-court/1994/60228-0-1.html
Washington Supreme Court. (2012). Vision One, LLC v. Philadelphia Indemnity Insurance Co., 174 Wash. 2d 501 (2012). https://law.justia.com/cases/washington/supreme-court/2012/85350-9.html
Zhao, L. (2022). Insurer's liability under concurrent causation: English law and Chinese law compared. Legal Studies, 42(1), 120–136. https://www.cambridge.org/core/journals/legal-studies/article/insurers-liability-under-concurrent-causation-english-law-and-chinese-law-compared/7B3AEDB91BA02547B786C3BED83A9A6C