RESEARCH DOSSIER
Contingent Business Interruption and Supply Chain Dependency
Abstract. Contingent business interruption (CBI) insurance connects the economic consequences experienced by an insured enterprise to covered physical damage at a third-party location upon which that enterprise depends. This monograph examines the legal and operational architecture of that connection. It distinguishes supplier, customer, manufacturing, leader-location, and selected second-tier dependencies; explains the function of covered-peril, property-damage, location, territory, causal, waiting-period, mitigation, and limit conditions; and reconstructs interruption losses from production constraints and financial records. A detailed comparison of an Insurance Services Office (ISO) dependent-property endorsement, two U.S. federal appellate decisions, official supervisory materials, business-continuity standards, and peer-reviewed network-economics studies demonstrates that operational dependence, economic exposure, and insurance coverage are three separate quantities requiring separate evidence. The investigation develops transparent analytical procedures without proposing a new proprietary theory. Its central synthesis is that physical network propagation can be many-tiered while contractual coverage is defined by precisely delimited locations, relationships, perils, and periods. Empirical supply-chain research establishes that shocks cross firm boundaries and that input specificity and network position can magnify their economic effects; those findings cannot establish the existence of a particular insurance entitlement. Hypothetical worked examples show how inventory buffers, restoration timing, alternate production, and daily sublimits modify estimated losses. The monograph provides a source-to-claim audit protocol, sensitivity analysis, comparative legal interpretation, insurer accumulation considerations, and a reproducible evidence map. It specifies the remaining identification limits when full policy schedules, supplier contracts, loss-adjustment files, or proprietary transaction-level data are unavailable.
Keywords: Contingent business interruption; dependent-property insurance; supply chain dependency; commercial property insurance; supplier interruption; customer interruption; physical damage; business income; restoration period; contingent time element; causal attribution; network propagation; business continuity; catastrophe accumulation; forensic accounting; policy wording.
Introduction
An insured manufacturer may remain structurally undamaged while losing production because a specialized component supplier has burned down. A distributor may retain full warehouse capacity while an important customer ceases operations after a flood. A shopping business may lose customers after damage to a nearby attraction. These examples describe dependencies between separate economic actors and places. They create a potentially substantial business-interruption exposure even when the insured's own buildings and machinery remain intact. Whether that exposure results in an insured claim requires analysis of the actual contract as well as the physical event and the resulting financial loss.
The U.S. National Association of Insurance Commissioners (NAIC) describes contingent business interruption as an optional type of business-income protection against certain supply-chain disruptions in its business interruption guidance. The formulation is introductory: the coverage grant, exclusions, schedules, territory, deductibles, limits, and policy amendments determine the legal result. The underlying distinction between interruption of one's own property and interruption following damage to another party's property is elaborated in the published Property Insurance Science (Bogdanova, 2026a). Commercial insurance provides a broader framework for evaluating contracts written around a business's particular productive relationships, as developed in Commercial Insurance Science (Bogdanova, 2026b).
This dossier addresses three research questions. First, which damage events and supplier or customer relationships can activate a dependent-property extension? Second, how do the location, tier, substitutability, inventory, transport, and geographic attributes of a supply network determine the enterprise's actual exposure? Third, how can the economically caused loss be reconstructed and then reconciled with the narrower measure payable under a policy? The analytical unit alternates explicitly between a location, an interfirm link, a peril event, an insured enterprise, and an insurer's accumulation portfolio. Changes in unit are declared because a correlation observed in a production network is not, by itself, a causal finding about a particular insured claim.
The method is documentary and comparative. It examines the ISO CP 15 08 10 12 endorsement as one identifiable illustrative contractual form; judicial reasoning in Pentair (2005) and Millennium Inorganic Chemicals (2014); peer-reviewed studies of disaster transmission; official continuity, supervisory, and standards materials; and reproducible hypothetical arithmetic. Contract clauses are studied as clauses, judicial findings as jurisdiction- and wording-specific adjudications, scholarly estimates as estimates for their investigated populations, and scenarios as assumptions. No confidential insurer claims files or full firm-level transaction microdata were available for this study. The scope remains physical-damage-centered CBI; cyber-related non-damage interruptions, pandemic disease extensions, transport-only losses, ordinary trade-credit nonpayment, and supplier insolvency without an insured peril are treated as adjacent contractual problems.
Definitions, Historical Context, and Primary Source Foundations
The insured object and the contractual trigger
Business income is the economically relevant flow interrupted by an insured event, commonly reconstructed from net income that would have been earned and continuing normal operating expenses, with further adjustments dictated by the policy. Contingent business interruption denotes a time-element loss of the insured arising from a covered event at another property's location. In an ISO-form setting, the inquiry begins with the coverage form and the attached endorsement, proceeds through its schedule and limitations, and ends with the insured's proven suspension and time-element loss. The economic relation can exist without a claim; the claim can exist under an extension even though the insured's premises suffered no physical damage.
The four-page ISO CP 15 08 broad form, bearing form identifier CP 15 08 10 12, explicitly modifies the Business Income (And Extra Expense) or Business Income (Without Extra Expense) Coverage Form. Its schedule distinguishes contributing, secondary contributing, recipient, secondary recipient, manufacturing, and leader locations. Under the principal coverage clause, a necessary suspension of the insured's operations must be caused by direct physical loss or damage at a scheduled dependent property resulting from a covered cause of loss. A central consequence of this drafting is that a damaged outside property, a covered cause, and a caused suspension are all necessary findings within the scope of the wording.
The ISO broad form is a source of concrete comparison rather than a universal commercial-insurance code. Insurance contracts may employ negotiated manuscript clauses, different ISO editions, separate limited forms, different territorial extensions, or different jurisdictional interpretive rules. A reference to ISO wording does not establish the terms of an unidentified insured's policy. The broad form's exclusions and qualifications are also consequential: its text excludes loss where the only damaged subject is electronic data, and separately addresses electronic data at secondary locations. The defined property classifications exclude certain utilities from the category of contributing location in that form. These qualifications explain why a physical business dependency discovered by engineers can remain outside a particular coverage grant.
An institutional account of interruption and continuity
Contemporary supply chains distribute production, fulfillment, processing, marketing, logistics, and customer access across companies and territories. Specialized intermediate goods and limited substitution increase exposure to failures that originate beyond an insured's premises. Hendricks and Singhal (2005), using 885 announced glitches involving publicly traded firms, documented significant adverse associations with subsequent operational performance in their supply-chain performance study. The sample and announcement-based design are important limits: the findings concern observed public-company disruptions and their performance, not the prevalence of CBI insurance or the probability that any particular incident was covered.
At the macroeconomic scale, Acemoglu, Carvalho, Ozdaglar, and Tahbaz-Salehi (2012) showed how production-network structure affects the propagation of individual shocks into aggregate fluctuations in their network-origins analysis. Barrot and Sauvagnat (2016) found stronger downstream transmission from disrupted suppliers when inputs were more specific in their input-specificity study. These economic studies explain why indirect links matter physically. Their mathematical objects are industry or firm production networks, not insurance policies. They therefore inform exposure identification and accumulation analysis but cannot rewrite a contractual definition of a covered dependent property.
Operational continuity scholarship supplies a separate evidence source. The official Ready Business impact worksheet asks enterprises to distinguish operational from financial impacts and to consider their timing and duration, including lost income and increased costs. The published ISO 22301:2019 overview describes requirements for a business continuity management system; its detailed normative requirements require access to the standard itself. NIST's contingency planning guidance provides a public methodological reference for business-impact analysis and recovery priorities, particularly for information systems. Together, these sources justify a repeatable identification of critical operations and restoration constraints; they do not supply an insurance coverage determination.
Five layers of evidence
The documentary inquiry separates five layers that are often collapsed in commercial narratives. The first is physical: the location, property, damage, peril, and repair sequence. The second is network-operational: who supplies whom, through which routes, with which inventories, technical qualifications, and alternate capacities. The third is enterprise-economic: the insured's lost production or sales, avoided variable costs, continuing expenses, additional expenditure, and recovery trajectory. The fourth is contractual: covered causes, dependent-property categories, named sites or qualifying unnamed sites, period of restoration, waiting periods, conditions, exclusions, and limits. The fifth is institutional: timely notice, investigation, adjustment, dispute, settlement, and actual payment.
Evidence at one layer cannot be silently substituted for evidence at another. In particular, an invoice proves a trading relationship but not necessarily the contractual category of a dependent location; physical damage proves a loss at an outside site but not necessarily suspension of the insured's own operations; falling revenue proves a financial observation but not that the insured event caused its entire decline. The review uses this separation throughout the case comparisons and financial reconstructions.
Supplier and Customer Triggers
Contributing, recipient, manufacturing, and leader locations
A contributing location supplies materials or qualifying services to the insured or on the insured's account. A recipient location accepts the insured's outputs. A manufacturing location produces goods contractually for delivery to the insured's customers. A leader location attracts customers to the insured, as might a landmark or major commercial attraction. These distinctions, reflected in the ISO dependent-property definitions, change the direction and identity of the dependent transaction. In each category, the business interruption is measured at the insured enterprise, while the specified physical loss occurs elsewhere.
Supplier interruption is typically transmitted downstream. An outside producer is damaged; an input becomes unavailable or delayed; the insured depletes stocks or cannot qualify replacements; its own operations slow or stop. Customer or recipient interruption moves upstream: a buyer's damaged premises prevent acceptance of output, which may result in lost production, forced inventory accumulation, or reduced insured gross earnings. Manufacturing-location exposure can arise when a contract manufacturer loses its plant and fails to deliver directly to the insured's customers. Leader-location interruption differs from supply flows: the mechanism is impaired customer attraction, so revenue causation can be harder to distinguish from other local demand changes.
Naming and classifying the actual locations matters. A multinational company may use one legal supplier with five production sites; damage at an unnamed plant is not necessarily equivalent to damage at the scheduled site. Conversely, a form can provide specified limited protection for qualifying unscheduled locations. Corporate supplier master data alone seldom contains the precise address, occupancy, equipment, hazard profile, or edition-specific insurance classification needed for a coverage analysis. It should be linked to site-level records, contractual performance obligations, shipment records, and insurer schedules.
Direct and indirect suppliers: two appellate tests
In Pentair, Inc. v. American Guarantee and Liability Insurance Co., 400 F.3d 613 (8th Cir. 2005), an earthquake in Taiwan damaged a power substation serving factories that supplied Pentair. Those factories lost electricity and interrupted production. Pentair incurred airfreight costs after production resumed, but its appellate claim under a contingent time-element clause failed. The court found that the substation did not supply goods or services to Pentair within the relevant provision, and that the supplier factories' loss of power alone did not establish the required physical damage to their property under the policy wording and governing interpretation. The Pentair appellate opinion (2005) therefore demonstrates two independent evidentiary questions: identifying a qualifying supplier and locating the required damage. It does not establish a global rule that power-outage losses are always excluded under all CBI contracts.
A separate dispute concerned a gas processing facility on Varanus Island. In Millennium Inorganic Chemicals Ltd. v. National Union Fire Insurance Co., 744 F.3d 279 (4th Cir. 2014), an explosion disrupted gas production by Apache. Millennium purchased gas from Alinta, which obtained gas from several producers. The parties' policies contained CBI endorsements referring to direct suppliers, and the schedules and binders played a decisive role. The Fourth Circuit reversed a judgment for Millennium and directed judgment for the insurers, concluding that the direct-supplier restriction excluded Apache in the examined contractual arrangement. The court's Millennium appellate opinion also records that direct named and unnamed contributing properties carried different sublimits. The operative distinction was contractual coverage of the direct relationship, not a denial that a physical upstream explosion had economically affected Millennium.
The two decisions should be compared carefully. Pentair addresses a material-damage condition and the identity of a property allegedly qualifying as the insured's supplier. Millennium addresses the meaning and documentary implementation of direct-supplier restrictions in a network containing an intermediary. Neither decision endorses a simple equivalence between economic reachability in a supply chain and insured reachability. Both show that policy-specific words, applicable law, contemporaneous binders and schedules, and the actual physical and commercial paths must be reconstructed before conclusions about entitlement.
Secondary dependencies and the depth of the network
Some policy forms expressly extend coverage beyond one transaction tier. The ISO CP 15 08 10 12 endorsement makes secondary contributing and secondary recipient coverage conditional on the schedule showing its applicability; the underlying damage must cause disruption of the named first-tier location's supply or acceptance and thereby suspend the insured's operations. The clauses also specify restrictions where damage consists solely of electronic data. Under the same form, the secondary provisions do not add a separate business-income limit above the applicable limit. These details are evident in the secondary-dependency clauses.
In network terms, the number of economically important tiers has no natural ceiling. A semiconductor assembly plant may rely on a chip designer, a wafer foundry, a specialist chemical producer, a power utility, and a geographically concentrated transport node. Each can be crucial for production. Insurance contracts frequently draw a boundary, however, around recognized property categories, territories, tier counts, named or unnamed locations, and certain excluded services. A robust underwriting file must therefore maintain two linked representations: the observed dependency network and the subset eligible under the actual policy. The difference between those representations is an empirically investigable coverage gap, not an error in network analysis.
Peril, territory, exclusion, and suspension
The proper event sequence requires (a) damage at the qualifying site; (b) a cause that falls within the covered causes of loss; (c) an operative dependency through which the site affects the insured; (d) a necessary interruption or reduction of insured operations; and (e) an economic loss within the period and limits described by the contract. These are analytical checkpoints, not independent universal legal tests: actual contracts may add or revise terms. A wildfire at a supplier may pass a property-damage test yet fall outside a policy's geographic territory; a storm may be physically destructive but be excluded by a particular flood or windstorm provision; a manufacturer may temporarily halt delivery while the insured's buffer inventory prevents any current suspension.
The distinction also separates physical-damage CBI from non-damage BI products. The U.K. Financial Conduct Authority's business interruption test-case record concerns disease, prevention-of-access, and related clauses in selected policies, and shows why insured wording and judicial context matter. Its outcome is not a general determination that ordinary dependent-property physical-damage endorsements respond to a pandemic. Technology and cyber outages similarly require inspection of their own triggers and exclusions. NIST's supply-chain security guidance helps assess cyber dependency risk but supplies no replacement for the property policy's electronic-data and physical-damage conditions.
Dependent Location Exposures
From supplier names to site-specific dependency maps
A dependency map is a directed representation of economic relationships augmented by geography and operational constraints. Nodes can represent production facilities, transport terminals, specialist service assets, customer sites, or a final insured operation. Edges indicate transfer of materials, goods, processing capability, or customer acceptance. For insurance purposes each node also needs a legal operator, location identifier, physical attributes, relevant peril and territory, and possible contractual status. An edge records who supplies whom and under what contractual or operational conditions. Multiple facilities of one supplier are recorded as different sites because they may have different catastrophe exposure and repair times.
A reproducible map should retain the source of each relation. A purchase order documents a transactional connection, a supplier declaration may identify the actual plant, a bill of lading or manufacturing batch can reveal the route taken, and a site survey can identify bottleneck equipment. The date matters: a supplier relationship valid last year may not have been in force when the insured loss occurred. Ownership and contract privity are useful descriptors, but Millennium illustrates why the operative language must determine which relationship is relevant. The Fourth Circuit's documentary analysis includes competing accounts of physical supply, legal title and directness.
Input criticality and substitution
Supplier exposure depends on the technical role of the input, not simply its purchasing value. A low-cost chemical catalyst can be indispensable if there is no qualified substitute; a high-cost commodity can be replaceable from multiple providers. Barrot and Sauvagnat (2016) find economically meaningful stronger propagation from disruptions to suppliers producing specific inputs, as presented in the peer-reviewed input-specificity evidence. Boehm, Flaaen, and Pandalai-Nayar (2019), exploiting the 2011 Japanese earthquake, document complementary production inputs and cross-border transmission through Japanese affiliates in the United States in their firm-level input-linkages study. Neither study supplies a single insurance-wide conversion factor from input shortage to covered loss.
For a given insured production line, a practical constraint representation is output Q(t) = min over essential inputs i of [a_i(t)/b_i], subject to labor, equipment, transport, and demand constraints. Here a_i(t) is the usable quantity of input i during period t and b_i is its units required per finished unit; the expression is a deliberately simplified fixed-proportions production identity for the illustrative scenario. More flexible substitutions would be modeled with separate technical relationships and observed changeover costs. An interruption at a critical supplier changes a_i(t), but the insured may continue producing for several days using inventory. This timing can place the observed production loss later than the date of the supplier's damage and alter the overlap with a contractual waiting period and restoration window.
Geographic concentration and common-cause events
The structure of exposure is also spatial. Two nominally independent suppliers may share a river floodplain, a grid substation, a port, an industrial park, a specialized foundry, or a logistics corridor. A single catastrophe can disable several locations at once. Firm-level transaction studies of the 2011 Great East Japan Earthquake show both upstream and downstream effects and meaningful wider economic propagation. Carvalho, Nirei, Saito, and Tahbaz-Salehi (2021) estimate a 0.47 percentage-point reduction in Japan's real GDP growth in the year following the earthquake under their production-network model; see the Great East Japan study. That figure is a country-level model estimate for that disaster, not an estimate of insurance losses or a direct forecast for an unrelated portfolio.
A portfolio accumulation calculation must combine locations' exposed values, policy coverage status, peril footprints, repair time, business dependencies, and insured limits. Insured enterprises can share suppliers without recognizing them as shared systemic concentrations. The accumulation is particularly important for highly concentrated inputs and transport or utility infrastructure. The OECD supply-chain resilience review illustrates that localized, redundant, and internationally distributed sourcing choices involve economic tradeoffs; simple geographic relocation should not automatically be equated with higher resilience.
Restoration as a path rather than an event date
Restoration requires several processes: physical repair at the affected site, procurement of replacement equipment, requalification of machinery and material, reopening of logistics, replenishment of buffers, restart of the insured's line, and recovery of sales. These processes can have different clocks. In a material-damage dependent-property form, the covered restoration period may be defined primarily by reasonable repair or replacement of the dependent location, not by the time needed for the insured to regain every lost customer or rebuild inventory indefinitely. The ISO restoration-period definition begins 72 hours after the event and ends when the dependent property should be repaired, rebuilt or replaced with reasonable speed and similar quality, subject to stated qualifications.
The operationally observed recovery time may thus differ from both the required physical restoration time and the insured indemnity period. Multiple damaged suppliers can create overlapping recovery clocks, while alternate sourcing can restore the insured's operations before the original supplier reopens. The relevant claim period must be reconstructed at daily or weekly granularity, using policy definitions and evidence of engineering repair and commercially available mitigation.
Causal and Economic Loss Attribution
Counterfactual revenue and avoided costs
The foundational valuation question asks what the insured business would probably have earned absent the covered event, given the real trading environment and other non-event influences. Let R0(t) be counterfactual revenue, R1(t) observed revenue, V0(t) and V1(t) the corresponding avoidable variable costs, and F(t) any continuing expense treatment specified by the operative policy. A simple gross-earnings-style illustrative interruption measure over periods t is the sum of [(R0(t) - V0(t)) - (R1(t) - V1(t))], modified for savings, extra expenses, and the actual contractual formula. This identity is not presented as a universal policy loss definition. Revenue is not profit; gross sales cannot be used as a stand-in for a covered margin without identifying saved purchasing, production, distribution and other noncontinuing costs.
A sound counterfactual uses pre-loss volumes and prices, orders, customer records, seasonal patterns, scheduled maintenance, industry demand, comparable unaffected units, and post-event changes attributable to external conditions. Supply and demand events must be separated. A supplier fire and an unrelated industry-wide collapse in demand can coincide; attributing the entire decline to the supplier fire would overstate economically caused interruption. Conversely, if the insured could sell every unit produced despite a supplier loss, the loss may follow the quantity constraint rather than demand decline.
The Ready Business impact-analysis categories support a broad enterprise accounting of lost income, delayed cash, overtime, expediting, and potential penalties. A property policy typically indemnifies a narrower defined set, and particular categories may be excluded or handled under separate coverage. The calculation must therefore distinguish enterprise loss from potentially eligible insurance loss and from cash eventually paid. A later payment figure may reflect a negotiated compromise or claim handling as much as the physical event's economic magnitude.
A transparent inventory-and-recovery example
Consider a hypothetical manufacturer with normal daily sales of $100,000 and variable costs equal to 60% of sales. Its daily contribution margin is therefore $40,000. A named specialized supplier suffers covered physical fire damage and cannot ship for 30 days. The insured has sufficient usable input inventory for five production days, then experiences 25 days of full production loss. Assume no other cause of lost sales and that the insured can neither recover production later nor substitute input. On this strictly hypothetical basis, the event-caused contribution loss is 25 × $40,000 = $1,000,000. The supplier's physical restoration takes 30 days; the insured's production interruption lasts 25 days because inventory absorbs the first five.
Now suppose an equivalent alternate input becomes available after day 18 following the fire, and qualification allows normal production from day 19. The interruption runs from day 6 through day 18 inclusive, or 13 days; the contribution loss becomes 13 × $40,000 = $520,000. Assume additional expediting costs of $80,000 and that these costs are necessary, documented, and potentially within the relevant extra-expense clause. The combined economic amount is $600,000 before any policy-based restrictions, and the difference from the no-substitution case is $400,000. Whether the $80,000 is recoverable or treated as expense to reduce loss depends on wording and efficient mitigation assessment. Those assumptions must be visible in every claim calculation.
If the particular covered restoration period begins 72 hours after supplier damage, inventory may already delay the loss beyond that boundary; a 72-hour attachment would then not mechanically subtract three more days from the 13 days. Conversely, where interruption starts immediately and a waiting period is measured by calendar time after physical damage, early lost days may be outside the compensable period. The form's actual definition and the observed loss calendar must be overlaid before allocating days. This example is a mathematical demonstration, not an actual claim or a general insurer settlement rule.
Contract limits, daily caps, and secondary locations
The ISO broad form includes a miscellaneous-location additional coverage with a daily maximum of 0.03% of the applicable business-income limit for interruption arising from one qualifying unscheduled location. Suppose the business-income limit were $2,000,000 and an otherwise eligible miscellaneous-location loss lasted 20 days after all applicable period conditions. The daily maximum would be $600, and 20 days would yield a calculated maximum of $12,000 for that location, subject to the text's other conditions and overarching limit. This is a direct arithmetic application of the ISO miscellaneous-location clause; it illustrates why an economic loss of several hundred thousand dollars does not imply a comparably sized policy payment.
A named dependent property may be treated under a different limit structure; a manuscript insurer program might use scheduled sublimits, percentage conditions, deductibles, aggregated event limits, or tier-based restrictions. In Millennium, for example, the appellate record describes named and unnamed direct supplier sublimits under the relevant manuscript contracts. Those numbers describe the litigated policies, not a generic benchmark. Claims analysis must distinguish a location limit from a per-occurrence aggregate and recognize that one physical event could impair multiple dependent properties. The fact that several limits are described in a schedule does not establish that all are additive for one occurrence.
Causation, mitigation, and double counting
The appropriate causation inquiry proceeds from property damage to a demonstrated operational constraint, then to an economic deviation relative to an explicit counterfactual. It considers whether and when the insured could reasonably use alternative materials, routes, customers or plants. The ISO endorsement's resumption-of-operations condition provides for a reduction in business-income loss to the extent operations can be resumed through other available materials or outlets. This makes a mitigation record a component of contractual measurement, not merely a discretionary managerial account.
Forensic reconciliation should exclude duplicate measurement of the same lost margin under several suppliers' claims. An interruption attributed to a shared upstream event may appear in multiple affected nodes, but the insured produces one consolidated profit-and-loss result. If two suppliers are disabled and either loss alone would have halted production, the contribution of each event is not identified merely by subtracting separately calculated claim amounts. The investigator should reconstruct alternative worlds under a single consistent causal model and apply the coverage terms governing multiple causes or locations. Claims and insured enterprise profitability should be compared at matching periods, accounting conventions, and consolidation perimeters.
Empirical Verification, Uncertainty, and Competing Explanations
Evidence hierarchy and observable records
The documentary hierarchy begins with the policy contract and all effective endorsements, declarations, schedule entries and applicable binders. The next layer comprises the claimed event's public or private loss records: property inspection, adjuster findings, engineering repair timeline, peril evidence, and official hazard records. The third layer identifies business relations through contracts, invoices, bills of lading, supplier attestations, production bills of materials, and accepted orders. The fourth includes the insured's production logs, inventory balance, finance ledgers, sales bookings, and shipping capacity. The final layer comprises counterfactual comparisons, independent industry information, and claim settlement materials. Each document should record version, legal entity, location, effective date, custodian, and the specific proposition it supports.
The treatment of uncertainty must match the evidence. Official policy wording is strong evidence for the terms it contains, but not for whether a given insured purchased the endorsement. A reported court ruling is direct evidence of that court's disposition, not universal national or international law. Published scholarly estimates can demonstrate statistical mechanisms for specified samples; they cannot independently prove a particular insured's causal loss. In the present dossier, the Great East Japan earthquake study and the supplier-specificity analysis are reviewed at their published abstract and bibliographic evidence level; the paper does not claim to have re-estimated either study's underlying microdata.
Competing mechanisms and identification
A decline in insured output following a supplier catastrophe may reflect the dependent site or may coincide with diminished demand, internal breakdown, labor shortages, adverse price movements, regulatory restrictions, or lost transportation access. A sound causal reconstruction explicitly identifies plausible simultaneous causes. It measures observed interruption separately from the portion explained by the event under the hypothesized production constraints. Multiple methods may be available: comparisons with unaffected product lines; difference-in-differences among matched units with aligned trends; plant-level engineering capacity simulations; order backlog and production batch reconstruction; or a documented direct customer-level lost-sales analysis. The evidentiary strength of each varies with the design and records available.
Comparative disaster studies provide a useful warning against simplistic identification. Carvalho and colleagues (2021) distinguish shock transmission from propagation through firms not directly hit by the earthquake. Boehm, Flaaen, and Pandalai-Nayar (2019) estimate cross-border production effects through firm input relationships. The input-linkages evidence concerns a particular natural experiment and affiliated firms; applying its average elasticity to a different industry requires evidence about production technology and sourcing. Hendricks and Singhal's public-firm glitch sample likewise establishes associations conditional on announcement and control selection, and is subject to selection and measurement limitations. This dossier adopts mechanisms from these studies while refraining from inventing an aggregate insurance loss multiplier.
Sensitivity analysis and audit procedures
At least five uncertain inputs should be tested across a plausible range: counterfactual demand, inventory availability, alternative supplier lead time, physical restoration time, and marginal contribution per unit. A sixth concerns the relevant policy schedule and location category. Consider the previous hypothetical case: with a $40,000 daily contribution, every additional lost production day adds $40,000 before coverage restrictions; a two-day error in the interruption duration changes the economic estimate by $80,000. If the realized margin were 35% rather than 40% of daily revenue, a 13-day full interruption would imply $455,000 rather than $520,000 in lost contribution. Neither variation demonstrates statistical confidence intervals; each is an explicit scenario sensitivity.
Auditing a proposed claim should reconstruct the timeline at the most granular practical unit. Mark the site-damage date, immediate physical impairment, supplier shipment cutoff, the insured's depletion of inventory, actual production reduction, alternate supply start, restoration milestones, and final resumption. Overlay contractual waiting periods and indemnity definitions. For each claimed monetary component, identify its ledger account, its operational cause, its supporting invoice or finance report, its treatment in the counterfactual, and its corresponding contract clause. Reconcile the sum of line items to financial statements and remove duplicative measures of revenue, margin and expense.
The workflow is also appropriate for insurer concentration studies: geocode dependent locations, assign peril and vulnerability assumptions, connect facilities to insured legal entities and policies, then identify shared nodes and time-dependent aggregation. Yet proprietary portfolios differ in how precisely supplier addresses and secondary dependencies are recorded. Unknown suppliers must be flagged as missing location data rather than represented as known nonexposures. A point estimate generated from an incomplete map must be accompanied by a coverage and dependency ascertainment sensitivity. The economic importance of shared suppliers makes this a material model-risk issue in the sense developed by Systemic and Emerging Risks Science (Bogdanova, 2026c).
Comparative boundaries across jurisdictions and products
An ISO endorsement is one contractual architecture. UK, continental European, Asian, Latin American and multinational insurance placements may operate under substantially different forms, legal systems, and underlying loss definitions. The FCA policy-checker explanation emphasizes that individual insured wordings and schedules are necessary to evaluate even policies affected by a common judicial test case. The broader implication is methodological: transfer of coverage conclusions across jurisdictions requires demonstrating sufficiently similar wording, facts, and governing law. This dossier's two U.S. appellate decisions remain specifically U.S. case-law evidence.
A second boundary concerns operating resilience rather than indemnity. The OECD's 2024 supply-chain preparedness study distinguishes ordinary disruptions from rare extreme disruptions in assessing resilience interventions. Insurance is one possible financing mechanism within a larger response involving supplier diversification, inventory, alternate transport, qualified production capacity, and public emergency systems. The design of protection should account for the total critical path of recovery, with a separate, contract-specific calculation of the insured part.
Integrative Comparative Scientific Synthesis
Economic reach and contractual reach
The combined evidence supports a structural distinction. Production networks transmit shocks through direct, indirect and common-cause relationships; CBI contracts define conditions under which the insured enterprise's resulting loss becomes a covered monetary obligation. The economic system can register a loss several tiers away from the physically damaged asset. The contract may cover only a scheduled first-tier property, may extend to selected secondary locations, or may provide a limited miscellaneous-location benefit. The ISO secondary-location provisions show an explicit contractual expansion beyond a simple one-tier condition, while Millennium shows an effective restriction of direct-supplier eligibility in its particular negotiated program.
For scientific comparison, four quantities should be reported separately: the total economic loss causally linked to an outside event; the loss arising from qualifying locations and perils; the amount calculated under applicable time-element provisions and limits; and the actual insurer payment. Differences between these quantities may result from policy design, insufficient documentation, legal disagreement, mitigation, delayed claims handling, or uncovered business dependencies. The analytical usefulness lies in retaining these differences, not in folding them into a single summary percentage described as insurance effectiveness.
Insurance as a partial map of productive dependence
A firm-level supply graph that lists every economically critical location is not equivalent to an insurance schedule. Both objects contain useful information and should remain linked. Product engineers identify critical machines and qualified inputs; procurement records identify supplier and customer relations; insurance schedules define the eligible dependent properties and sublimits; claims accountants determine the earnings effect. Coverage review is most reliable when all four descriptions use persistent entity and location identifiers and record the date on which each association is valid. This is an information-governance conclusion derived from the documentary comparison, not a claim that a novel proprietary conceptual system has been established.
Three predictable gaps follow. The first is a discovery gap, where a critical lower-tier plant is unknown to the insured. The second is a wording gap, where a known plant is not a qualifying or covered dependent property. The third is a valuation gap, where an event may be covered but the impact cannot be substantiated with sufficiently granular records. Each gap demands a different intervention: supplier-site discovery, contract design, or stronger operational-financial recordkeeping. This distinction helps avoid a common false solution of treating more expensive insurance limits as the sole answer to an omitted peril or an unlisted dependency.
Network propagation and insurer aggregation
Barrot and Sauvagnat (2016) and Carvalho and colleagues (2021) establish different empirical routes by which geographically localized shocks propagate through firm networks. Acemoglu and colleagues (2012) provide a formal explanation for why important nodes and higher-order links matter at the aggregate scale. The network origins study supports concentration analysis of economically central suppliers; the earthquake production study demonstrates observed disaster propagation in a particular national network. An insurer can face several correlated insured losses arising from one shared supplier event, but a robust portfolio forecast requires the distribution of covered dependent sites, policy terms, hazard exposures, economic durations, and limits. The macroeconomic estimate cannot be converted directly into an insurance expected-loss figure.
Risk management also has an efficiency dimension. Geographic diversification can reduce dependence on one hazard footprint while increasing exposure to long-distance transport, regulation, or supplier qualification delay. The OECD resilience review reports modeled costs of large-scale relocalization and notes that it does not consistently increase resilience. A scientifically defensible CBI assessment compares actual alternate source reliability, total response time, and policy coverage, rather than treating local sourcing or global dispersion as inherently superior.
Implications for research and underwriting
The research supports a practical sequence: identify the named insured and insured operation; obtain actual policy and amendments; inventory critical supplier, customer, manufacturing and leader locations; geolocate material lower-tier dependencies; identify possible common peril events; model alternative production and inventory buffers; translate production effects into verified business-income variables; and apply time periods and policy limits. The sequence can be independently audited because each step has a different documentary foundation. It also allows a research analyst to report credible partial findings where some site data are missing rather than assert unsupported certainty.
Future empirical work could compare declared dependent-property schedules with supplier transaction networks over several underwriting years, evaluate how limits correspond to true network criticality, estimate restoration-time distributions for specialized equipment, or test whether exposure ascertainment predicts claim disputes and uninsured interruption. Such studies would require lawfully obtained original policy and claims data, systematic sampling, and an identifiable causal design. The present paper claims original cross-disciplinary documentary synthesis and transparent example calculations, not invention of a new scientific doctrine or demonstration of a new population-wide causal effect.
Conclusion
Contingent business interruption insurance is a contractual response to specified consequences of damage at third-party properties connected to an insured's production and commercial activity. The verified ISO CP 15 08 10 12 form reveals differentiated supplier, recipient, manufacturing, leader and optional secondary-location coverage; a physical-damage and covered-peril condition; a defined period of restoration; qualifying miscellaneous-location limitations; and mitigation provisions. The appellate decisions in Pentair and Millennium Inorganic Chemicals demonstrate that the identity of the damaged property, the meaning of the direct supplier relationship, and the location of physical damage can be outcome-determinative under the specific contracts examined.
Empirical production-network studies demonstrate that economic losses can propagate upstream and downstream and can be amplified by input specificity and concentrated network positions. These observations support broader supply-chain exposure mapping and insurer accumulation research. They also underscore why network-wide economic harm and insured business-income liability must be measured separately. A complete claim reconstruction identifies the damaged location, the covered relationship, the causal production path, the counterfactual earnings effect, reasonable mitigation, the contractual period and limits, and the actual claim outcome.
The central research conclusion is that the adequacy of CBI protection depends jointly on knowledge of critical dependencies, the contract's representation of those dependencies, and the quality of operational and financial proof after loss. A documented gap between any of these elements changes what a firm can demonstrate or recover. Further scientific progress depends on linked location-level policy, transaction, engineering, and claims data with explicit legal and causal boundaries. All numerical examples in this monograph are hypothetical; comparative judicial holdings and scholarly estimates retain their original geographic, temporal, contractual and methodological scope.
References
Acemoglu, D., Carvalho, V. M., Ozdaglar, A., & Tahbaz-Salehi, A. (2012). The network origins of aggregate fluctuations. Econometrica, 80(5), 1977–2016. https://doi.org/10.3982/ECTA9623
Barrot, J.-N., & Sauvagnat, J. (2016). Input specificity and the propagation of idiosyncratic shocks in production networks. The Quarterly Journal of Economics, 131(3), 1543–1592. https://doi.org/10.1093/qje/qjw018
Boehm, C. E., Flaaen, A., & Pandalai-Nayar, N. (2019). Input linkages and the transmission of shocks: Firm-level evidence from the 2011 Tōhoku earthquake. The Review of Economics and Statistics, 101(1), 60–75. https://ideas.repec.org/a/tpr/restat/v101y2019i1p60-75.html
Bogdanova, A. (2026a). Property Insurance. Insurance by Bogdanova. Published by Aisentica Development. https://insurancebybogdanova.com/science/property-insurance/
Bogdanova, A. (2026b). Commercial Insurance. Insurance by Bogdanova. Published by Aisentica Development. https://insurancebybogdanova.com/science/commercial-insurance/
Bogdanova, A. (2026c). Systemic and Emerging Risks. Insurance by Bogdanova. Published by Aisentica Development. https://insurancebybogdanova.com/science/systemic-and-emerging-risks/
Carvalho, V. M., Nirei, M., Saito, Y. U., & Tahbaz-Salehi, A. (2021). Supply chain disruptions: Evidence from the Great East Japan earthquake. The Quarterly Journal of Economics, 136(2), 1255–1321. https://doi.org/10.1093/qje/qjaa044
Financial Conduct Authority. (2026). Business interruption insurance [Regulatory information and test-case updates]. https://www.fca.org.uk/firms/business-interruption-insurance
Financial Conduct Authority. (2025). Business interruption insurance — policy checker [Updated December 5, 2025]. https://www.fca.org.uk/firms/business-interruption-insurance/policy-checker
Hendricks, K. B., & Singhal, V. R. (2005). Association between supply chain glitches and operating performance. Management Science, 51(5), 695–711. https://doi.org/10.1287/mnsc.1040.0353
Insurance Services Office, Inc. (2011). Business income from dependent properties — broad form (CP 15 08 10 12) [Insurance endorsement]. https://assets.touchpointmarkets.com/6c/c7/717708804960ba1fa78969b2c58a/cp-15-08-10-12-business-income-from-dependent-properties-broad-form.pdf
International Organization for Standardization. (2019). ISO 22301:2019: Security and resilience — Business continuity management systems — Requirements [Official standard overview; amendment listed for 2024]. https://www.iso.org/standard/75106.html
Millennium Inorganic Chemicals Ltd. v. National Union Fire Insurance Co., 744 F.3d 279 (4th Cir. 2014) [Published appellate opinion]. https://www.ca4.uscourts.gov/opinions/published/131194.p.pdf
National Association of Insurance Commissioners. (2026). Business interruption and business owner policy [Consumer and industry overview, updated June 25, 2026]. https://content.naic.org/insurance-topics/business-interruption-and-business-owner-policy
National Institute of Standards and Technology. (2010). Contingency planning guide for federal information systems (SP 800-34 Rev. 1, updated reference page). https://csrc.nist.gov/pubs/sp/800/34/r1/upd1/final
National Institute of Standards and Technology. (2022, updated 2024). Cybersecurity supply chain risk management practices for systems and organizations (SP 800-161 Rev. 1, Update 1). https://csrc.nist.gov/pubs/sp/800/161/r1/upd1/final
Organisation for Economic Co-operation and Development. (2025). OECD supply chain resilience review: Navigating risks [Report overview]. https://www.oecd.org/en/publications/2025/06/oecd-supply-chain-resilience-review_9930d256.html
Pentair, Inc. v. American Guarantee and Liability Insurance Co., 400 F.3d 613 (8th Cir. 2005) [Published appellate opinion]. https://law.justia.com/cases/federal/appellate-courts/F3/400/613/606363/
Ready Business / Federal Emergency Management Agency. (2020). Business impact analysis worksheet [Official worksheet]. https://www.ready.gov/sites/default/files/2020-07/business-impact-analysis-worksheet.pdf
Thakur-Weigold, B., & Miroudot, S. (2024). Promoting resilience and preparedness in supply chains (OECD Trade Policy Papers No. 286). OECD Publishing. https://www.oecd.org/en/publications/promoting-resilience-and-preparedness-in-supply-chains_be692d01-en.html