DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE

______ Law Review
VOLUME __ _____ 2008 NUMBER _

DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE CONTRACTS AUTHOR† What causes an accident? It is rarely a simple answer. Answering whether or not one has insurance for the loss is often even more complex a question. It is all the more so when multiple causes combine together to produce a loss. This is concurrent causation. When hurricane Katrina struck and thousands lost their homes, was the loss caused by rain, winds, a storm surge, ineffective water barriers, or flooding? The answer to the question drives whether or not a person’s property insurance covers the loss. Some causes of a loss are covered by insurance policies. Other causes are specifically excluded from coverage. What happens when causes come together concurrently to produce an ultimate end result? If some causes are covered by insurance and others excluded, does the insurance policy pay? Does the insured get nothing? Complexity heightens when one speaks of coverage under liability insurance. Liability insurance provides coverage for negligent behaviour. Assume the driver of a pick-up truck, while burning brush all day on his acreage, shovels up the pile of now-burned ashes into the back of his truck in an attempt to clean up.1 He thinks the fire is out. He speeds around town and

Copyright © 2009 by author † author information 1 The facts in this hypothetical are similar, though not identical, to Waseca Mutual Ins. Co. v. Noska, 331 N.W.2d 917 (Minn. 1983).

DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE

2

_______ LAW REVIEW

[Vol. __:_

swerves while he drives, in an attempt to impress his passenger. While negligently driving dangerously close to a building, some embers relight in the back of his truck, and fly out, causing a fire to the building. Is the fire loss caused by the negligent driving, in which case his automobile insurance policy will provide coverage? Or is the fire loss caused by the negligent ash cleanup, in which case his homeowners insurance policy will provide coverage? What if his homeowners policy excludes coverage for negligent acts involving an automobile? Do both policies respond? Neither? What if someone was hurt by the ensuing fire? Do they have no source of compensation if they sue the driver and he is held to have no insurance coverage for the accident? These questions of concurrent causation are not only difficult to answer. They are expensive to answer. Different common law jurisdictions have different doctrinal rules for solving these insurance disputes. The problem is that the current rules produce unpredictable results. Insureds and insurers cannot predict whether or not a concurrently caused loss will be covered by a certain insurance policy. Most jurisdictions ask a fact-finder to determine a “proximate” or “dominant” cause of the loss.2 If that cause is covered under the insurance contract’s coverage provisions, the insurance policy pays the insured for the loss. If that cause is excluded from coverage, the insured is left without indemnification. Yet what is the dominant cause in the hurricane loss example above? The wind? The storm surge? The rain? What is the dominant cause in the pick-up truck example? The negligent driving? The negligent shoveling of smouldering ashes? Different courts may arrive at different results even though the same doctrinal test is applied.
2

Also called “efficient” or “effective” cause. This article adopts the term “dominant” cause to demarcate the difference between causation in insurance and the tort concept of proximate cause. See Jeffrey Stempel, STEMPEL ON INSURANCE at 7-15 [“Stempel”], Robert H. Jerry II, UNDERSTANDING INSURANCE LAW, 4th ed. (2007) [“Jerry”] at 560566, and Mark D. Wuerfel & Mark Koop, “Efficient Proximate Causation” in the Context of Property Insurance Claims (1998) 65 Def. Counsel J. 400 [“Wuerfel & Koop”] (approximately three-fifths of American states utilize the dominant cause approach to concurrent causation).

DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE

2008]

Default Rules for Concurrent Causation

3

Other jurisdictions use either a liberal3 or conservative approach to solving concurrent causation insurance disputes. Under a liberal approach, if one cause in a causal chain of events is covered by the insurance policy, the entire loss is covered, even though other, non-covered or excluded causes might have combined to produce the loss. Under a conservative approach, if one cause in a causal chain of events is excluded, the entire loss is excluded, even if other covered causes combined to produce the loss. The liberal approach is pro-coverage, pro-insured. The conservative approach is pro-exclusion, pro-insurer. Both provide all-or-nothing answers to a concurrent causation dispute. Could the various competing causes in a loss scenario be apportioned somehow, to avoid an all-or-nothing approach? No courts have yet applied this methodology, but perhaps it is a solution which best does justice to both the coverage and exclusionary language of insurance policies. This article takes account of the varying approaches to concurrent causation in insurance and proposes two default rules4 for efficiently solving these disputes. Concurrent causation situations can be grouped into two distinct factual matrices which necessarily require two different default rules as
Robert H. Jerry II was the first to categorize three possible approaches to concurrent causation in insurance. He dubbed them the dominant cause, conservative, and liberal approaches. This article adopts his nomenclature as it is a helpful functional descriptor for the three approaches, and adds a new fourth approach: apportionment. See Robert H. Jerry II, UNDERSTANDING INSURANCE LAW, 4th ed. (2007) at 567-592. See also Wuerfel and Koop, id. for a state-by-state catalogue of approaches to concurrent causation in insurance. 4 A default rule is a legal rule which is followed when there is no contractual term in a contractual setting to direct how the contract is to be applied or interpreted. See i.e. Ian Ayres and Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules (1989) 99 Yale L.J. 87. Kenneth Abraham notes that insurance law doctrine contains few default rules beyond contractual interpretation rules, because standard form insurance contracts provide internal default rules within the contract. Kenneth S. Abraham, Peril and Fortuity in Property and Liability Insurance (2001) Tort and Ins. L.J. 7877. This article argues that default rules for concurrent causation are necessary to add to current insurance law doctrine because insurance law contracts do not provide for efficient and effective solutions for resolving disputes about concurrent causation.
3

English courts do not interpret an insurance policy contra proferentem unless the policy is ambiguous. Insurance Law Rights at Variance with Policy Provision. They appear in other common law jurisdictions as well. Most American courts construe insurance contracts contra proferentem. St. L. Why Are Insurance Contracts Subject to Special Rules of Interpretation? Text Versus Context (1992) 24 Ariz. 2d ed. (2005) at 233-234. 995. it should not affect the adoption of a helpful framework for concurrent causation. L. While this may have an effect on how a court deciphers the meanings of words within an insurance contract. Fischer. L. English courts also do not apply the reasonable expectations doctrine. Kenneth S. nor do they interpret coverage and exclusion clauses differently.e. a portable methodology workable in all three jurisdictions is the goal. A few American jurisdictions apply some form of the reasonable expectations doctrine. the article draws from comparisons to British and Canadian approaches to concurrent causation. In the end. 1037 and Judicial Interpretations of Insurance Contract Disputes: Toward a Realistic Middle Ground Approach (1996) 57 Ohio St. A Theory of Insurance Policy Interpretation (1996) 95 Mich.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 4 _______ LAW REVIEW [Vol. as that analysis comes after the interpretive exercise that searches for contractual meaning. Therefore. Keeton. Rev. INSURANCE LAW: DOCTRINES AND PRINCIPLES. which interprets policies so as to give effect to the reasonable expectations of an insured (see i. Abraham. J. unlike many American courts.5 It is important to keep in mind throughout this article that courts in different jurisdictions employ varying interpretive rules when assessing the meaning of insurance contracts. Part I (1970) 83 Harvard L. 310 at 317-318 (English courts adhere to an insurance contract as “bargain” theory of interpretation. Rev. Judicial Rationales in Insurance Law: Dusting Off the Formal for the Function (1991) 52 Ohio St. Canadian courts fall between American and British courts in terms of focus on strict adherence to a contractual approach to interpretation and a balancing of rights of insured over the 5 . 543). Concurrent causation issues in insurance law are not unique to the United States. Proximate Causation in Insurance Law. where helpful. with some regularity. __:_ modification to the current positive law. less protective of the insured than most American states) and John Lowry and Philip Rawlings. Robert E. See John Lowry and Philip Rawlings. the article takes a comparative approach. In solving this puzzle.J. (2005) 68 Modern L. James M. as against the insurer. Most also interpret coverage clauses broadly and exclusion clauses narrowly. L. However. Rev. 961.J. Peter Nash Swisher. 531. All three common law jurisdictions have taken differing approaches to the same issues.

The first type of dispute involves coverage triggers. in turn. It examines how those differences drive the problems with the current approaches to concurrent causation. but one policy excludes coverage for a particular cause in a concurrently caused loss. How does an insurance policy respond when a loss is caused by two or more possible causes and one cause is covered by the policy and one not covered? This is an expensive. Cases about insurance coverage and concurrent causation generally fall into insurer. and if payout is conditional upon a covered factor being the cause. Part III of this article discusses the important differences between causation in tort and causation in insurance law. ambiguity is the gatekeeper to the more insured-friendly interpretive tools found in many American states. See i. This Canadian hybrid approach to insurance contract interpretation first calls for solving the interpretive question using standard contractual principles. Craig Brown. Part IV classifies causal events into two workable categories: serial and parallel causation scenarios. exclusions narrowly. and the other policy grants coverage for that excluded cause but may. what if it is only “a” cause in a panoply of possible causes? This Part concludes by proposing that some predictable.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 5 Parts I and II of this article define concurrent causation in insurance law. If the contract is ambiguous. . coverage clauses are interpreted broadly. It also sets an important baseline for constructing workable default rule solutions for assessing concurrently caused insurance losses. time-consuming type of dispute that creates a great deal of litigation. In Canada. GENERAL PRINCIPLES OF CANADIAN INSURANCE LAW (2008) at 136-145.e. exclude a different concurrent cause? If a loss is caused by more than one causal factor. efficient default rules are the best solution to cleaning up this remarkably messy and unpredictable area of insurance law. INSURANCE LAW IN CANADA (looseleaf) and Barbara Billingsley. The second type of dispute involves loss distribution among insurers. then the court looks to contra proferentem and the reasonable expectations of both insured and insurer alike in determining meaning. However. These Parts expose two pervasive types of disputes regarding concurrent causation which result in disputes in common law jurisdictions. How does an insurance policy respond if two or more overlapping insurance policies may be potentially called upon to pay a loss.

concurrent causation occurs when a loss is brought about by two or more potential causes. The benefits and problems of each approach are discussed in an attempt to unveil the inherent inefficiencies of each. the policy insures against “all risks” to the property. The insurance indemnifies the property owner once the property is harmed in a . and the resulting need for predictable.e. The cause. theft or water damage) or. As will be noted. or Canada could use to resolve these insurance cases. The causation question is complicated in situations of concurrent causation because one cause of the loss may be covered by an insurance policy and the other cause excluded from coverage. Britain. Yet. of a loss dictate whether or not a person who has an insurance policy (an insured) in fact has insurance coverage for that specific loss. It can be thought of as “external force” insurance. I. The distinct nature of losses caused by serial causation versus those caused by parallel causation necessitates a default rule for each type of causal scenario in order to maximize efficiency of results and. simple default rules. as is more common for standard homeowners insurance. the categories are unique enough that each requires a distinctive approach when creating default rules for efficiently assessing insurance coverage questions.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 6 _______ LAW REVIEW [Vol. specified perils like fire. Finally. What Is Concurrent Causation in Insurance? In insurance law. Part VI proposes two separate default rules that any court in the United States. fairness to insured and insurer alike. __:_ one of the two categories. ultimately. each model produces different results for both serial and parallel causation cases. Property insurance policies generally insure against damage to the property resulting from either exposure to enumerated events (i. the liberal approach. or causes. Part V of this article identifies and explores four possible models for crafting an efficient default rule for concurrent causation in insurance contracts: the dominant cause approach. Each also creates differing sets of concerns depending upon the underlying type of insurance dispute: coverage or loss distribution. and the apportionment approach. the conservative approach. Coverage disputes call into question different efficiency issues than loss distribution disputes.

and various other uninsured causes act together to produce the total loss. St. Rev. This is supposed to ease dispute resolution between insured and insurer.e. Trebilcock. the insurer simply pays the total insured value of the building.e. Scott Edwards. 6 . and Org. when applied to a concurrent causation situation. However. See i. The Role of Insurance Considerations in the Choice of Efficient Civil Liability Rules (1988) 4 J. U. inflates the extent of her insurance claim after a loss. creating incentives for low risk insureds to exit the pool. To control the insurer’s liability to indemnify the insured to only those risks the insurer wants to cover. for The difficulties in resolving these types of property insurance disputes are often not made simpler in instances where concurrent causation causes total property destruction in jurisdictions which have valued policy laws. The factual matrix comprising a concurrently caused loss may include multiple covered. It thus becomes difficult to determine whether or not the insurance policy pays for the loss when the loss is caused by multiple causal factors. 243 at 244. Michael J. Econ.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 7 specific fashion. overlapping coverage and exclusion clauses make the application of valued policy laws problematic when a small portion of the loss may be caused by one covered cause. Adverse selection occurs when the insurance risk pools are drawn too broadly. are tailored to catch behaviour and thereby combat moral hazard. Exclusions to coverage. L. 7 Insurers face the twin problems of adverse selection and moral hazard. Property insurance is “first party” insurance – it indemnifies the property owner directly for resulting losses to the property. Moral hazard occurs when a fully insured individual acts less carefully because she is insured or. the policy includes specific exclusions from coverage. and excluded causes of a single.7 Automobile liability insurance.6 Liability insurance policies generally provide broad insurance coverage for damages an insured may be called upon to pay a third party as a result of the insured’s actions. A valued policy law requires a property insurer to pay the full value of the policy limits under the insurance contract in the event of the total loss of an insured building. such as an exclusion for intentional or criminal conduct. non-covered. L. 541 (demonstrating that a state’s valued policy law. The insurer is required by statute to pay the cost of the entire insured property. alternatively. The Wind and the Waves: The Evolution of Florida Property Insurance Law in Response to Multiple-Causation Hurricane Damage (2007) 34 Fla. See i. can create unfair results to insurers and insureds in large catastrophic events like hurricanes which cause total losses through multiple causal factors).

L. 569 at 570 [“McDowell”]. v. liability insurance is third party insurance – it is a source of compensation for an injured third party. [2001] 1 S. Often. liability insurance indemnifies the insured so that the insured can then pay the entity. There are also innumerable causal combinations of behaviours which could cause a loss to a third party. 2008) and Sansalone v. Rev.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 8 _______ LAW REVIEW [Vol. unpredictable.. For this reason. Liability insurance covers an insured for any behaviour which causes harm to a third party. Banks McDowell.3d 752 (7th Cir. Wawanesa Mutual Ins. Corp. Liability insurance is often thought of as “tort” or “behaviour” insurance for the insured. The scope of available coverage under liability insurance is markedly broader than that which is usually available under property insurance. There are innumerable ways that this could come about. Co. provides indemnity to a driver if the driver’s negligence brings about harm to a third party and the driver becomes liable to pay the third party some form of compensation. 9 Banks McDowell notes that insurance law disputes about causation usually involve issues of either concurrent causation or “dual coverage. National Union Fire Ins. It becomes a difficult. the standard homeowners liability insurance coverage grant insures the insured for any liability or damage incurred as a result of his or her actions or an occurrence “anywhere in the world. Causation in Contracts and Insurance (1988) 20 Conn. causes act in concert. and expensive question to determine whether or not a certain concurrently caused loss is covered by a liability insurance policy. __:_ example. Types of Insurance Disputes There are two main types of insurance disputes which feature concurrent causation issues and which require some coherent doctrinal approach to enhance efficient dispute resolution. of Pittsburgh. If the insured commits a tort and has to pay an entity because of her conduct.8 The reason is simple. 8 .R.e.” See i. 544 F.C. this author chooses the terms “coverage dispute” and “loss distribution dispute” instead. 627. Rockwell Automation.9 The first and most common dispute is a simple coverage question: does an insurance policy provide coverage for a loss caused by two or more concurrent causes? These For example. Inc.” Because both coverage disputes and loss distribution disputes encompass concurrent causation concepts.

The major efficiency problems with contemporary approaches to concurrent causation include issues of jurisprudential consistency. inefficient. in fact. The Need for Efficient Default Rules for Concurrent Causation Insurance Disputes A new set of default rules is necessary to efficiently solve issues about concurrent causation in insurance. various approaches provide little doctrinal guidance to courts. . and insureds faced with determining coverage questions about concurrently caused losses. Even within jurisdictions. It is difficult for insurers and insureds alike to arrange their insurance and indeed.” are often exceedingly expensive to answer because they involve multiple insurers with multiple competing interests. Settlement is therefore highly unlikely. at its heart. These are the most costly. what happens if more than one insurance policy may potentially provide coverage for the loss? What if one cause is covered by one policy and another cause covered by another policy – which policy ultimately pays for the loss? Both? Neither? One or the other? These overlapping insurance disputes. inefficient pleading problems. around shifting standards for resolving these disputes. insurers. Different jurisdictions approach the problem of concurrent causation in different ways. tortured and unpredictable of insurance cases. excluded from coverage by the insurance policy. or “loss distribution disputes. owing to the unpredictable nature of the outcomes. They also appear remarkably frequently in the litigation system. The second type of dispute is. It adds a level of complexity to the same coverage question: if a loss is caused by two or more concurrent causes. inefficient counsel involvement. a loss distribution dispute among multiple insurers. II. and a potential offloading of liability to a secondary insurance market: the insurance brokers.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 9 “coverage disputes” become complicated when one of the potential concurrent causes is a covered loss and the other potential concurrent cause is either not covered or. their very conduct.

4th 747.2d 123 (Cal. In 2005. 11 In 1973. 1973). Rev. the Katrina Disaster. The Doctrine of Efficient Proximate Cause. even if other. even though the facts and policy wording are similar case to case). 292 (Cal. First. and even within jurisdictions. Joseph Lavitt. 2005). Prosser’s Folly. Rev 1 [“Lavitt”] (proximate cause analysis produces different results in different jurisdictions. the Court further narrowed the application of concurrent causation in a property insurance case by upholding an exclusion clause to oust coverage in a concurrently caused loss scenario when a covered cause combined with an excluded cause to produce a loss.. The Cost of Inconsistency There is no question among courts and academics that concurrent causation insurance cases across jurisdictions. 35 Cal. The dominant or proximate cause approach was adopted for property insurance cases involving concurrent causation. not whether or not there even exist factual elements which bring the insurance contract at all into play. The dispute is then a simple one.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 10 _______ LAW REVIEW [Vol. and the Third Restatement of Torts: Cracking the Conundrum (2008) 54 Loyola L. 569 at 585 (noting the predictability problems and increased litigation costs of insurance causation disputes). as compared to cases where the cause of the loss is a single cause. there are increased information costs in litigating concurrent causation disputes in insurance.11 This creates serious cost consequences in a variety of ways. 1989). Co. excluded causes combined to produce the loss. 10 . one about interpretation of the coverage provisions of the insurance contract. Co. State Farm Fire and Casualty Co.. 12 In cases where the cause of the loss is a single cause. produce wildly unpredictable results and are jurisprudentially inconsistent. v.3d 903 (Cal. The Court restricted the liberal approach to only cases involving liability insurance in 1989 with Garvey v.10 For example. – is the loss resulting from the cause a loss that is covered under the insurance policy. Hartford Underwriters Ins. the California Supreme Court adopted a liberal approach to concurrent causation in State Farm Mutual Automobile Ins. 110 P. the California Supreme Court has shifted to three perceptibly different doctrinal approaches to concurrent causation from 1973 to 2005. L. 514 P.12 Information costs Banks McDowell. Julian v. Partridge. __:_ A. Rptr. the jurisprudential inquiry is one of contract interpretation only. Causation in Contracts and Insurance (1988) 20 Conn. 257 Cal. where the Court held that a loss was covered by an insurance policy.

Econ. Someone has to get the evidence about the causal story. there are increased administration costs. Third. Such costs include those additional costs that are beyond mere informational or fact-finding costs. Second. Complexity costs are those costs which are incurred due to the fact that a certain dispute is more complex to process. The interaction of the fact pattern with the insurance policy coverage language and concomitant exclusionary language makes for a more complicated dispute. Often. experts will have to disentangle causal factors in forensic fashion. and Org. Because the legal rules for concurrent causation lead to unpredictable end results. They also include the cost of determining in a loss distribution dispute which of one or more competing insurance policies will cover an insured loss.13 It is far more costly to push complex insurance litigation through the See i. Trebilcock. It makes for protracted litigation. They include the cost in a coverage dispute for discerning meaning in an insurance contract when faced with a concurrent causation factual pattern and dueling possibilities for coverage or exclusion of the loss. The more complex the story. the more costly it is to get the evidence and sort through it. Fact-finding becomes a serious cost in these cases because there is a marked incentive to search through the chain of causal events to find either a covered (for the insured) or excluded (for the insurer) cause. litigants therefore are incentivized to lead evidence of long chains of complex casual events and multiple hypothetical causal scenarios (some plausible. 243 (calling for administrative efficiency in insurance liability rules to contain decision and error costs). The Role of Insurance Considerations in the Choice of Efficient Civil Liability Rules (1988) 4 J. there are increased complexity costs in litigating such cases.e. flowing from the complexity and information cost increases.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 11 here comprise the costs of obtaining the facts to determine the causation situation in dispute. in order to determine the cause and effect of various causal factors acting together to produce the loss. Expert evidence often becomes necessary. some implausible) in the hopes of either tagging or avoiding coverage under an insurance policy. Michael J. If one is litigating in a jurisdiction which requires litigants to choose a dominant cause. there is often a costly inquiry to determine that question as well. L. over and above the norm for solving a standard insurance contract dispute. 13 . This is expensive.

or at best. the consistency costs. on a systemic level. __:_ civil justice system when litigants are unable to predict outcomes. Courts cannot utilize haphazard legal principles to come to internally consistent decisions. Insurers. Case law consistency is mutable with the wind. Different pressure is leveraged on an insured who is at the cusp of litigation or who cannot even predict whether or not her policy will cover a concurrently caused loss. and inefficient litigation. increased resources to move the cases through the system. and increased procedural wrangling. It surely must be difficult for national and international insurers to settle cases or even set premiums with regard to concurrent causation cases when the doctrinal rules to resolve these disputes produce inconsistent results. Cases about concurrent causation therefore stay in the justice system and sap resources precisely because it is impossible for litigants to predict outcomes. risky. it seems. Pleading Problems One of the inefficiencies of the current approaches to concurrent causation across jurisdictions is the fact that insureds suing for coverage are unaware of how to most effectively plead their lawsuits. B. The cases stay in the system and get to court. Add to this the confusion created by mixing faultbased tort principles of causation with principles of contract law and a simple coverage case becomes a quagmire of mixed up doctrine. cannot be predicted. Even repeat players such as insurers and insurance lawyers cannot enjoy heightened informational positions when predictability of results is so low. There is every incentive to do so. insureds. The incentive to settle is low because the players do not have concrete information about how a court will assess a concurrent causation scenario. Not being able to consistently predict how the coverage or loss distribution dispute will unfold puts the insured in a difficult informational position with respect to litigation decisions. This results in costly delay. and their counsel cannot reliably predict a court’s response to coverage in a concurrent causation situation. Finally. increased court time. Gaps in insurance coverage can result. The overflowing inefficient result is surely felt in the insurance market. This results in “overpleading” in an attempt to . The current doctrinal rules utilized by courts permit different courts to see similar causal situations differently.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 12 _______ LAW REVIEW [Vol. The result: wasteful. are highly inefficient.

it may result in misdirected pleading. 1395 and Ellen S. as insurance protection may be denied the insured. D. If a loss resulting from concurrent causes potentially may invoke more than one insurance policy. Claims Stories. That insured may then be required to obtain separate counsel from the insurers to protect the insured’s financial interests.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 13 catch one or more insurance policies which may be called upon to respond to the concurrently caused loss. Pryor. L.14 In addition. depending upon the regime for deciding concurrent causation in the insured’s jurisdiction. 1721. The result: an inefficient web of pleading behaviour which increases complexity and administration costs to courts and litigants alike. Constructing the Insurance Relationship: Sales Stores. If there is the risk of one or more exclusions being triggered among the various policies in play. Insureds may be incentivized to characterize their losses in non-concurrent terms to avoid exclusionary language. Counsel Inefficiencies Because concurrent causation case results in many jurisdictions are unpredictable. be incentivized to cast the factual scenario as one of concurrent causation in order to trigger the same exclusionary language. and Insurance Contract Damages (1994) 72 Tex. This all results in an inefficient web of potentially duplicative lawyering that increases complexity and administrative costs. lawsuits may require multiple coverage counsel for various policies at risk of being triggered. L. each insurer will have its own lawyer to protect its interests. Offloading Indemnity Cost to Broker Insurance Markets In addition. Insurers may. Rev. any cost of gaps in insurance coverage created in a case by a jurisdiction’s approach to concurrent causation may actually instead be off-loaded to insurance This phenomenon of “overpleading” by insurance lawyers to tell a story with the intention to tag or avoid insurance coverage has been thoughtfully documented by Tom Baker and Ellen S. This is especially likely in loss distribution disputes. The Stories We Tell: Intentional Harm and the Quest for Insurance Funding (1997) 75 Tex. See Tom Baker. 14 . in liability insurance contexts there is an additional risk of the insured facing personal exposure to the cost of liability. by contrast. Rev. C. Pryor.

even though a covered loss was partly the cause of the loss. the applicable legal rule denies coverage for losses caused by a concurrent.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 14 _______ LAW REVIEW [Vol. if the broker is found negligent in procuring suitable insurance. that state does not consistently uphold anti-concurrent cause clauses. Alternatively. using specifically worded exclusion clauses. The insured subsequently suffers a loss resulting from concurrent causes. Bragg notes that this is precisely what California insurers did after the 1983 earthquake. This is likely something the insured did not expect. The insured expected “full coverage. may have been partly covered by an insurance policy. The legal rule applicable in the insured’s particular home jurisdiction happens to deem the excluded cause the dominant cause of the loss and deny coverage. Anti-Concurrent Cause Clauses Unpredictably Enforced A deceptively simple solution to concurrent causation may be to just contract out of concurrently caused losses entirely. E. Michael E.” The loss initially insured by the commercial liability insurer (who collected a premium for underwriting at least part of this risk) then becomes the subject of a suit against the broker. Assume an insured purchases a commercial liability insurance policy from an insurance broker and thinks she has complete coverage for all liability surrounding her business enterprise. excluded loss. The broker’s own liability insurance is therefore triggered to respond to the “gap” in coverage. __:_ brokers who sold the policies to the insured consumer. However. but the cause of the lack of indemnity for the loss. but for the fact of concurrent causation. The insured is left with no coverage. the other cause is excluded.15 Courts. One cause of the loss is covered by the policy. This may place greater strain on the brokers’ insurance market with a perhaps corresponding inefficient enrichment of the primary insured’s insurer who sold a policy and took a premium for a loss that. A simple example will illustrate. 15 . There is a gap in coverage for concurrently caused losses. Scenarios such as this result in a double inefficiency: a gap in available coverage plus the saddling of liability to a party who was not the primary cause of the loss. The insured then sues the broker in negligence for failing to provide adequate insurance or advise the insured of a potential gap in coverage. as will be described later below.

18 Usually including “directly or indirectly” as well as “caused by. State Farm Fire & Casualty Co. 773 P. L. have only sporadically enforced insurers’ contractual attempts to circumvent concurrently caused losses.” “contributed to. These courts cite to strict adherence of contractual language plus freedom of contract principles as reasons for allowing insurers to circumvent the prevailing concurrent cause legal rule in the jurisdiction. 16 See i. United Pac. 20 See i.2d 116 (Wash. 1996). 17 See i. Hirschmann. Ins. Findlay v. 112 Wash. 1989) (anti-concurrent cause clause did not oust . The language.16 This adds to the unpredictability of the dispute. for example.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 15 however.20 If a jurisdiction. Some courts uphold the language as ousting coverage in a concurrently caused loss scenario.. 6610 N. (2006) at 116-117 (noting that it is unclear what type of wording is required for an insurer to circumvent the prevailing legal rule for insurance causation).” “aggravated by.2d 954 (Mass. Jussim v. Dak. COLINVAUX’S LAW OF INSURANCE.. Safeco Ins. Bay Ins. deeming it a contractual attempt to dodge the doctrinal rules for assessing concurrently caused insurance losses in that jurisdiction. and Alf v. regardless of any other cause or event contributing concurrently or in any sequence to the loss….2d 621. Concurrent Causation and the Art of Policy Drafting: New Perils for Property Insurers (1984) 20 Forum 385. But Saying No to Contracting Out of It (2003) 79 N. 1993).e. Co. Co. 917 P.e. Robert Merkin. v. 561 (explaining that California. converts the particular causal inquiry into the conservative approach to concurrent causation.e.17 A typical anti-concurrent cause exclusionary clause reads as follows: “We do not insure for losses caused directly or indirectly by any of the following. the effect of such exclusions in a concurrent causation situation is questionable.2d 413 (Wash. Rev.18 However. Case Comment: Insurance Law – Property Insurance: Adopting the Efficient Proximate Cause Doctrine. Mass. Passa.E.e. Paulson. 850 P. 756 P.2d 1272 (Utah 1993). 1988). and Washington preclude parties from contracting out of the efficient proximate cause doctrine). North Dakota.2d 764 (Wyo. Julie A..” 19 See i. State Farm Fire & Casualty Co. Co. has adopted a See Michael E.” There are a variety of permutations of anti-concurrent cause language. in effect. and frankly defeats the purpose of such language in the vast majority of cases.19 Most courts instead invalidate an anti-concurrent causation clause. of America v. 8th ed.” or “resulting from. Bragg.

in Washington. Saunders notes that courts would likely read down any anti-concurrent cause clauses (like “directly or indirectly”). which distinguished Hirschmann. 267 Ca. 509 S. App. id.23 This difference needs to be recognized when crafting coverage for concurrently caused loss caused partly by earth movement – case distinguished and limited in 1996 by Findlay. 21 I. 140 at 162. courts are often hesitant to allow insurers to sidestep the rule.. The end result is unpredictability of enforcement of such clauses. In Canada.E. 23 Indeed. Rev. See Anthony J. ex ante. Proximate Cause in Insurance Law – Before and After Derksen.. reasonable expectations doctrine. id.e. (2006) 32 Advocates Q. Rptr. 1998). . 569 at 576-577. Causation in Contracts and Insurance (1988) 20 Conn. interpreting coverage clauses broadly and exclusions narrowly. Tort causation he calls “liability causation” because the inquiry focuses on fault-based concepts of liability. Banks McDowell differentiates between tort and insurance concepts of causation. State Farm Fire & Casualty Co. predict when the clause will convert a jurisdiction’s concurrent causation legal rule into a conservative approach.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 16 _______ LAW REVIEW [Vol. and Howell v. This article therefore treats anti-concurrent cause clauses in an ambivalent fashion and instead focuses on crafting default rules which assist courts whether or not such causes are upheld at any point in time in any common law jurisdiction. III. supra note 14. State Farm Fire & Casualty Co. 1st Dist. L. 708 (Cal. Anthony J. See also Jeffrey Stempel. Banks McDowell. 22 See ie. __:_ dominant cause approach to concurrent causation as a default rule of legal doctrine. There is something insidiously unfair about the clauses in that an insured often cannot. and contra proferentem. to an insured’s detriment. Contract causation he calls “damage causation” because the inquiry focuses on whether or not a particular contractual indemnity payout is triggered.Va. The pro-insured interpretive rules to be applied when construing insurance contracts in American and Canadian jurisdictions21 bolster courts’ conclusions to invalidate anti-concurrent causation clauses.2d 1 (W. Purpose of Causation in Insurance Causation in insurance law is different than causation in tort law. Findlay.22 That decreases the very efficiency the jurisdiction’s causation default rule was likely designed to augment. Saunders. 1990). even among the same courts in the same state. Murray v.).

. and Richard Wright. (2007) 92 Cornell L. Posner. Hart and Tony Honoré. Tort Law and Moral Luck. or blame. THE ECONOMIC STRUCTURE OF TORT LAW (1987). J. 26 For corrective justice standpoints on the law of torts.26 STEMPEL ON INSURANCE CONTRACTS. 403. 463. tort causation “encompasses socially imposed regulation of relationships. Rev.L. There is a remarkable temptation when assessing issues of insurance causation to reach for tools from tort law. (1985) 14 J. and John Goldberg & Benjamin Zipursky. Another is the very notion that liability insurance uses the tort system as a filter through which liability insurance coverage is triggered.e. it is to determine when the operative terms of a contractual bargain come into play. A major source of temptation stems from the fact that the word ‘causation’ is the same as a fundamental step in tort law’s negligence analysis – the determinations of cause-in-fact and proximate cause. (1989) 34 McGill L. 24 Jeffrey Stempel notes that.e. unlike insurance causation. (2003) 68 Brooklyn L. The Special Morality of Tort Law. The tort system is thus integral to the operation of the insurance system.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 17 default rules for solving insurance causation disputes. 1031.24 It operates in a fault-based system where causation is the bridge between the tortfeasor’s act and the fault. Steven Shavell.A. H. Yet the two systems examine the causation question for fundamentally different purposes. (1985) 28 J. (1985). id.e. In insurance.” See Stempel at 7-6. Legal Stud. 435. L. THE IDEA OF PRIVATE LAW (1995). 2d ed. at 7-3 – 7-9 and Jerry. Uncertainty Over Causation and the Determination of Civil Liability. Steven Shavell. How one adopts tools for solving disputes about concurrent causation in insurance ultimately first depends upon what causation in insurance law means. Landes & Richard A. Ernest A. & Econ.25 Regardless of one’s positivist viewpoint on the tort system. Rev. Weinrib. An Analysis of Causation and the Scope of Liability in the Law of Torts. The Fall and Rise of Blame in American Tort Law. Weinrib. Probabilistic Linkage: The Bane of Economic Analysis. Actual Causation vs. Causation in tort ties the responsibility of the tortfeasor’s sub-standard behaviour to the actual loss suffered by the injured plaintiff. see i. 1123. Ernest A. For law and economics standpoints on the law of torts. see i. Anthony Sebok. supra at 579-580. Legal Stud. 25 See i. 587. it is to assess fault for wrongdoing. Weinrib. and William M. (1980) 9 J.. CAUSATION IN THE LAW. In tort.

and Richard W. Risk and Remoteness – Causation Worse Confounded? (1964) Mod. Causal Contribution. Rev. 1401. contractual breach is traditionally conceived of as a strict liability damages system. for example. P. 1225. L. Proximate cause analysis is designed precisely to determine whether or not the isolated cause-in-fact will actually be deemed the legally responsible cause.27 Insurance law. Once More into the Bramble Bush: Duty. The tort system’s twin goals of compensation and deterrence drive the fault-based inquiry. The causation inquiry is only important as it relates to insurance coverage. and arguably much simpler. Even if a cause of an insured loss is human behaviour – a breach of the standard of care via the tort system – the causation question is not asked to determine fault of the tortfeasor but instead merely asked to determine whether or not insurance coverage is triggered within the language of the policy. it will not be a legally responsible cause. Cohen has. In insurance law. and the Extent of Legal Responsibility (2001) 54 Vanderbilt Law Review 1. Foreseeability in Negligence Law (1961) 61 Columbia L. See i. task. L. unpredictable legal doctrine. 28 Indeed. 27 . 344. Goldberg. Rev. however. uses causation for a different. a property insurance claim for weather damage to a house.e. __:_ causation attempts to link blameworthy conduct with responsibility for loss. as is the case with.” Insurance causation therefore bears no resemblance to the policy-laden proximate cause analysis of tort law. It is essentially a question that merely asks whether or not the “topic” of a The literature about the inherent mystery of proximate cause is vast. Gerald Dworkin. Cause-in-fact analysis asks whether or not a defendant’s breach of the standard of care is a de facto cause of the plaintiff’s harm. fault of “the cause” is irrelevant. George M.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 18 _______ LAW REVIEW [Vol. The Fault Lines in Contract Damages (1994) Va. Comment: Rethinking Injury and Proximate Cause (2003) 40 San Diego L. Courts and scholars soundly agree that proximate cause analysis in tort is a murky. 1315. Rev. provided a unique opposing proposition: a fault-based economic theory of contract damages designed to illuminate why contractual damages are often uncompensatory and how various traditional damages interests operate. Rev. Leon Green. John C. The question is not “who is to blame and why” but merely “what happened.28 The cause may often not even be human behaviour. Wright. See George M. If the cause is too far down the causal chain of events. however. Cohen.

the causation analysis in insurance is almost the reverse of tort. Stempel. Rev 1 at 32-33 (though Lavitt eventually advocates a return to a ‘but for’ tort-like analysis for concurrent causation). Co. Finally. St. Next one looks at causation. (1950) 340 US 54: “the subtleties and sophistries of tort liability for negligence are not to be applied in construing the covenants of [an insurance] policy.e. Insurance Causation Issues: The Legacy of Bird v. it is not the tort system.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 19 contractual right is going to be in play. and the Third Restatement of Torts: Cracking the Conundrum (2008) 54 Loyola L. 86 (N.E. see Peter N. 659 N. United States. Smiley. analogies to the tort system’s conceptions and jurisprudence of cause-in-fact and proximate cause are unhelpful and often extremely misleading in an insurance law context. largely because the very contractual language of the insurance policy itself acts as a limiter on causation. Paul Fire & Marine Ins. supra at 576. As Banks McDowell notes. 120 N. 276 Ill. It is a contractually driven lossspreading mechanism.App. 2 Nev. the Katrina Disaster.Y. 31 This concern was echoed by the United States Supreme Court in the marine insurance case of Standard Oil Co.). 32 See i. 30 McDowell.. While insurance may be the backbone of the tort system. Courts31 and commentators32 have cautioned against borrowing tort concepts of causation when dealing with Jeffrey Stempel notes that insurance causation is a much narrower type of analysis than tort causation. but as only an occurrence as related to insurance coverage (and not a fault question).J. one looks at the scope of coverage (or extent of liability).30 One first asks if the loss is the type for which the insurer agreed to indemnify. Prosser’s Folly.E. 2 Dist. a case about marine insurance and a fire. see i.e. Allstate Ins. supra at 7-7. L. 1995. St. Stempel at 7-7 and Joseph Lavitt. not a “liability” question..App. Co. Cardozo’s causation analysis called for a recognition of the reasonable expectations of the parties to the contract.2d 1345 (Ill. Paul Fire & Marine Ins. To that end. 351 (2002).3d 971. Swisher. Co.” An early juridical example of the “contractual” view of causation is Justice Cardozo’s decision in Bird v. 1918). v.29 It is a “payout” question. v. The thinking in this decision was likely a pre-cursor to the evolution of the reasonable expectations doctrine as it exists in insurance law today. For a helpful contextual analysis of the importance of this decision to insurance law. 29 . The Doctrine of Efficient Proximate Cause. For a recent example of courts recognizing the need to separate tort from insurance causation.

to the cause with the greatest relative blameworthiness. See Daniel J.34 The proximate or dominant cause approach to insurance causation. __:_ insurance disputes as the concepts. the Daniel J. implicit or explicit. U. L. Econ. Michael Trebilcock argues that the current insurance system is a “random form of judicially administered wealth redistribution” which does little or nothing to foster efficiency. in cases of concurrent breach of contract. discussed in full later. Q. Coverage decisions then get made with reference. L. advocates choosing the most “responsible” cause. No clause grants an insured coverage rights based on which loss trigger was most at fault in the moral sense of the word “fault. 2) tort injury addresses primarily personal injury and property destruction. 3) tort law serves an inherent distributive function that contract law does not). Liability for Concurrent Breach of Contract (1995) 73 Wash. some courts inappropriately import tort concepts of joint and several liability into contract law in order to solve contractual disputes. It borrows heavily from proximate cause analysis in tort. 35 One very interesting and elegantly simple solution to causation in insurance. The Role of Insurance Considerations in the Choice of Efficient Civil Liability Rules (1988) 4 J.35 Often. and Org. The tools of tort were created specifically for fault-based inquiry. 97 at 107-120 (citing the following reasons that such behaviour is inappropriate: 1) unlike tortfeasors and victims.” The policies grant coverage based on the mere existence of a causal event that brought about a “happening” in reality. create more problems than they often solve. That is as high as “cause” is put in the insurance world. while contract addresses economic loss tied to performance of the contract. there are few if any tools to which courts can turn. Read any insurance policy. when transported.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 20 _______ LAW REVIEW [Vol. Using those same tools in insurance settings changes the contractual analysis in a fundamental way. Trebilcock. Morality creeps in. parties in a contract choose each other. Bussel notes that. 243 at 258. Michael J. corrective justice or distributive justice concerns. articulated by Malcolm Clarke.33 It opens the door for morality-based decision patterns which produce illogical and unpredictable results in a contractual sphere. is to limit the causation inquiry to only those possible causes which are actually mentioned by 33 . 34 In fact. Unfortunately. There is a marked tendency in cases that adopt a dominant cause approach to implicitly assess relative “blame” or “fault” to a certain cause of a loss in a way other than as one of a faultless series of coverage triggers. Bussel. in cases of competing concurrent causes.

) as a default rule. The reasons may be that the prior academic frameworks likely tried to do too much with too little.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 21 result is an importation of proximate cause principles of tort. (1961) 59 Mich. Categorizing Concurrent Causation A. POLICIES AND PERCEPTIONS OF INSURANCE LAW IN THE TWENTY-FIRST CENTURY (Oxford: 2005) at 184.37 And when courts import the tort principles into insurance. The only causes in play are those mentioned by the contract of insurance. even in the presence of an alternate. Past Academic Attempts at Categorizing Concurrent Causation Past attempts at categorical approaches to solving concurrent causation issues in insurance have been largely unhelpful as anything other than ex post case descriptors. This approach. supra at 574-577. 36 See i. in practice.38 IV. and certainly have never been applied in practice in crafting an ideal solution to the voluminous insurance litigation about the topic. and the underlying morality issues that come with that. RESTATEMENT (SECOND) OF CONTRACTS §§ 351 and 352. This is likely why past academic attempts to provide some tools for insurance causation analysis have repeatedly led to an adoption of proximate cause principles borrowed from tort law. the results are unpredictable and costly to litigate. Causation principles in contract law generally are relegated to questions about the link between breach of the contract and reasonably foreseeable loss.) at 226-71. Because only the covered cause is in play in the analysis.e. By categorizing broad series the policy in either coverage or exclusion clauses. Brewer. ESSENTIALS OF INSURANCE LAW (1957) (2d ed. See Malcolm Clarke. Concurrent Causation in Insurance Contracts. Any other causes are outside of the spectrum of inquiry. competing cause. when a covered cause combines to produce a loss with a non-covered but not excluded cause. appears to lead to a liberal approach to causation (see Part V. 38 McDowell. William C.. Section C. L.36 There is little to fill the gap between tort and contract causation. simply because the covered cause is stipulated in the policy. Rev. 1141. Clarke anchors this idea in the notion that insurance causation is a contractual inquiry only. Patterson. . and Jerry. coverage must attach. Jr. 37 See Part IV and the works of Edwin W. supra at 585.

39 A “cause” sets an “event” in motion that causes a loss or “result. 40 For a critical approach to Patterson’s framework.40 By that point. Causation. Patterson. An “exception” clause deals with the cause of the insured event itself. to ex ante determine the legal effect of the clause when presented with a potential causal scenario. there is no predictive value to the framework. the frameworks instead cut too close to tort law’s unsatisfying explanations of proximate cause. after the loss has occurred and the court has rendered a decision. from the mere wording of a coverage clause. __:_ of discrete causal events and attempting to translate that into ex ante interpretive frameworks for coverage clauses or the events themselves. The problem with this methodology is that it is often difficult. uncontroverted sequence in the factual matrix of the loss. The difficulty with the Patterson framework likely flows from the fact that the Edwin W. This is most problematic in determining whether a clause limiting coverage is an exception or exclusion clause. Exclusion clauses negate coverage.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 22 _______ LAW REVIEW [Vol. 39 .) at 226-71. ESSENTIALS OF INSURANCE LAW (1957) (2d ed. Exception clauses define not what is excluded but instead what is just not covered by the insurance. Even ex post. in crafting an efficient default rule with some predictive value for concurrent causation disputes. Instead. Edwin W. it might be a more efficient approach to categorize causes without reference to their relative “responsibility” but instead to their actual. An “exclusion” clause involves the event. Patterson described a framework for identifying three types of causal events from which one could divine the legal effect of a coverage clause and resolve concurrent causation disputes. as it drives insurance coverage determinations.” This chain-like analogy follows through with labels for the legal effect of three types of insurance policy coverage clauses. This is likely the fundamental flaw in past attempts to derive some predictable legal rules for insurance causation. and not as something more. with all the unpredictability trappings of a fault-based analysis. it is still sometimes debatable as to which clause has done what legal work. And a “consequence” clause deals with the result that flows from the loss. should be related to its role as a contractual indemnity trigger only. see Jerry supra at 571-574.

Brewer’s methodology does not comport with the notion that insurance causation is best left to a contractuallycentric approach. which in turn results in little predictive utility. one then applies William C.44 Finally. L.43 Jerry’s first proposed step is to assess all the possible causal factors that could have brought about the loss in question. If that cause is covered. An “actual cause” is the inciting event which links together events and results. through a proximate cause sort of analysis. See Malcolm Clarke. Brewer.. similar to Brewer’s framework. 1141.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 23 categorical approach to labeling the legal effect of coverage clauses stems from the labeling of the causal events themselves. This pushes one toward a tort-like proximate cause analysis of describing ‘causes.41 Causes in chains of potentially insured events are labeled by Brewer according to their legal role. see Jerry.’ and ‘results. Finally. Concurrent Causation in Insurance Contracts. supra at 574. Rev.’ The framework proposed by William Brewer falls into a similarly problematic vein which relies heavily on a proximate cause tort analysis at its heart. identical to a proximate cause in tort law. Both fall into the trap of tort law. The court rests on one cause as being “the” cause of the loss. coverage attaches. 43 Jerry supra at 574-577. A “responsible cause” is a substantial cause to which the loss should rightly be attributed. Subsequently. focusing on triggers of contractual indemnity. The fact-finder must determine the most “responsible” cause.’ ‘events. the insurer is not liable for the loss. 42 For a critical description of Brewer’s framework. 44 Contrast this approach with Malcolm Clarke’s proposal to restrict consideration to only those causes that are specifically mentioned in the insurance policy language. A “substantial cause” is an actual cause that is deemed not too remote. (1961) 59 Mich. Jerry would have a court apply a tort-like remoteness filter to determine how far back in the causal chain of events one can rightly look in that given scenario. Jr. Like Patterson’s framework. Jerry proposes a dual-filtered approach to solving concurrent causation in insurance coverage disputes. also similar to Brewer. If not. POLICIES AND 41 . Robert H.42 Brewer’s framework essentially advocates a dominant cause analysis after determining all potentially possible causes.

See Part VI of this article for a complete description of each approach. For proof. From there. particularly in instances of concurrently caused losses. a return to causation analysis from tort law imports a number of serious consistency problems in insurance law. As indicated in the beginning of this article. The disadvantage. or conservative approach. What is a legally “responsible” cause to which a court can look in order to direct the coverage question? The answer likely lies in something akin to a proximate cause analysis. supra at 577. Rather than add to the existing doctrine. Such a move would also likely create more litigation. It may be counter-productive to attempt to apply the hoary law around proximate cause in tort to determine which causes of a panoply of potential concurrent causes a court could consider as being relevant to the insurance coverage assessment. The labels do not work because they rely upon the inherent “responsibility” for each categorized cause. PERCEPTIONS OF INSURANCE LAW IN THE TWENTY-FIRST CENTURY (Oxford: 2005) at 184. An analysis completely removed from causal “responsibility” might help to avoid the costly questions akin to a fault-based proximate cause inquiry.” That puts information and complexity costs out of reach of most insured litigants.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 24 _______ LAW REVIEW [Vol.46 Consistency of results is thus problematic. __:_ whatever concurrent causation approach is in vogue in one’s jurisdiction. the prior proposed frameworks depend on the answers to those initial questions. as Jerry appears to admit. is that remoteness of competing causes is often in the eye of the beholder as between insured and insurer.45 The advantage to Jerry’s methodology is that it is jurisdictionally portable. 45 Either the dominant cause. like the frameworks proposed by Patterson and Brewer. To perform the first branch of Jerry’s dual-filtered approach. 46 See Jerry. liberal. Instead of debating whether or not a concurrent cause is covered. one must first have some sort of structure for the filter. the dispute is displaced to an unproductive initial inquiry of “which causes in the chain make up the start and end of the chain. one only has to look to the plethora of endlessly inconsistent cases attempting to divine the “proximate cause” of an insured loss. from case to case. . past methodologies create more costly confusion.

Disputes involving first party property insurance and concurrent causation typically involve a certain causal sequence of events. rather than its individual components. Rather than attempt to group the causal factors and categorize their effect on the property. Only then can a consistent default rule be constructed to provide predictable dispute resolution results. cumulatively producing some effect on the insured property.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 25 B. it is therefore helpful to categorize the two .. causal sequence of events. Disputes involving third party liability insurance and concurrent causation also typically involve a certain. or some external force). recurring factual patterns in which concurrent causation is central to the dispute. in insurance law. it may in fact be more assistive to reverse the analysis and discuss cause in relation to its sequence upon the loss. Instead. weather. One does not ask whether a breach of the standard of care is a cause. Instead. burglary. The contest for coverage in these cases is generally about the combination of multiple causal factors acting in concert to produce a loss. A glance at the general trends in the case law reveals two central. Predictably solving both coverage and loss distribution disputes about concurrent causation necessitates finding a recognizable pattern to the cases. Serial and Parallel Causation Scenarios Instead of constructing labels for the constituent parts of any possible causation situation and then attempting to define its pattern and outcome. The end loss is the result of this sequence. though different. as stated above. one starts with the loss and works backward to determine if a cause of the loss is a contractually covered cause. as in tort. perhaps it is simpler to recognize the pattern or sequence of causes as being a category itself which drives a certain. When assessing solutions for concurrent causation issues with insurance. This makes more sense simply because the causal analysis in insurance law is the reverse of a causal analysis in tort law. in sequence. The contest for coverage is generally about multiple external causal factors (i. Again. it may be simpler to refer to this brand of concurrent causation dispute by its causal pattern.e. it is even more difficult to attempt to label causal factors in liability-related coverage disputes because the precise interaction of the factors themselves can be impossibly intertwined. The key difference is that the causal factors here are negligence-based behavioural factors that combine either together or in some sequence to produce a loss. predictable loss result.

110 P. Graham v. Ill.4th 747. Consecutive causes are not “true” concurrent causes but separated in time.2d 1077 (1983) (mudslide from Mt. really more akin to singular-cause cases. This is a loss caused by two or more causes occurring simultaneously. 527. At 833.3d 346 (5th Cir. The second scenario is a loss caused by parallel causation. U. See i. Julian v. L. and should be treated as such.3d 903 (Cal. most notorious of which are the property insurance cases in the wake of natural disasters such as earthquakes. acting on its own. and not as to how that operation could relate to contractual coverage issues.Concurrent Causation: Examination of Alternative Approaches (1985) S. Helens volcano eruption). See Anthony J.48 Each cause. Unfortunately. 2007) (flood water from hurricane Katrina). Hartford Underwriters Ins. 2005) (rain-induced landslide). 48 Examples include a wide variety of cases. 35 Cal. these attempts have not resulted in any greater predictability in case outcomes. __:_ generalized types of concurrent causation scenarios that present in the cases. State Farm Fire & Casualty Co. acting on its own. hurricanes. Fierce also noted distinctions between “dependent” and “independent” causes in Insurance Law . Proximate Cause in Insurance Law – Before and After Derksen. Serial causation cases exhibit independent. Malcolm A. St. Malcolm Clarke notes a difference between causes which are consecutive and causes which are concurrent. For example. 656 P.. Co. The reason largely stems from the fact that the categories are descriptive as to causal operation only. and floods. Independent causes are similar to serial causes in that each cause. most often prevalent in first party property disputes. Richard A.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 26 _______ LAW REVIEW [Vol. and 47 . Interdependent causes are similar to parallel causes in that the causes must combine together to produce the loss. (2006) 32 Advocates Q..e.. The Law of Insurance Contracts (2006) at 837.47 The first scenario is a loss caused by serial causation. Co. most often prevalent in third party liability disputes. Anthony Saunders uses the terms “serial” and “concurrent” to describe the same phenomenon in insurance causation. individually sufficient causes.2d 533. Clarke also divides causes into two classes: interdependent and independent causes. together. 98 Wash. Public Employees Mut. Ins. Clarke.J. Tuepker v. 507 F. 140 at 145. This is a loss caused by two or more causes occurring in a sequence. Saunders. cumulatively. produces some type of loss. will cause There have been other academic and judicial attempts to categorize causal patterns in insurance cases and build legal rules around such categorization.

exposing the house contents to the elements. Covering or excluding the entire loss somehow seems unfair to insurer or insured.W. The rain then turns to hail. Assume a house is insured by a standard “all risks” property insurance policy. 49 Examples of parallel cases abound. parallel causation cases exhibit interdependent.49 Without both Shelter Mut. 2d 810 (Mo. in serial fashion. The coverage dispute centred around which of two competing causes caused the loss – negligence in planning to remove the tree or negligence in operating the truck. Mo. not necessarily loss distribution disputes. Carthon v. But each acted cumulatively. Co. This type of causal event comprises causes acting one after the other. 965 F. High winds blow parts of the roof off. 2002) (flood in home followed by mould). v Aetna Life & . State Farm Fire & Cas. then. Co. Rain then pours into the house. does one assess coverage for the loss? Serial causation presents an analytic problem because it becomes tempting to try to divide the causes based on some sort of notion of causal responsibility.Supp. Noble Broadcast. Finally. Co. though cumulative in end result. See i. soaking the furniture and the internal structure of the house. For the balance of this article.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 27 some portion of the loss.) See also Gulf Ins. 309 F. 936 S.e.. Assume the insurance policy covers loss caused by wind and rain. v. but not hail. Maples. see also Travelers Ins. The wind alone caused some damage. but not the total cumulative loss.3d 1068 (8th Cir. 1997) (negligent crowd control during a parade combined with negligent automobile operation to produce injury – commercial general liability policy and automobile liability policy potentially implicated by concurrent causes). as did the rain and hail alone. individually insufficient causes. Ins. By contrast. which smashes breakable items within the house. hereinafter called the “Weathered House” scenario. Serial causation most often occurs in a first party property loss context. How. 1262 (W. They are also usually the subject of coverage disputes. in sequence to produce the ultimate loss. Co. 1997) (coverage contest between a homeowners liability policy and automobile liability policy where negligent tree removal in the insured’s backyard with a truck caused the death of a bystander. The entire loss presents a serial causation issue. it is most instructive to refer to a consistent factual example of a standard serial causation insurance dispute. v.D. The house is damaged in a large storm. because the causes are separated in time and effect from each other.

The causes must occur in parallel fashion. Parallel causation most often occurs in a third party liability insurance context. use or operation” of an automobile.51 Instead. He then moved the compressor with his vehicle. The contest was between both an automobile liability insurer and a commercial liability insurer. 1991) (adult day care centre client in a wheelchair fell down porch steps while being transported to the centre by van .” In that case. Both causes acted simultaneously to produce the loss. The auto policy covered only for losses arising out of the “ownership. [2001] 3 S. or tort insurance. but excluded coverage for losses caused by “ownership. into an oncoming school bus. The insured did not properly load a heavy construction sign base into his vehicle. Assume an insured has coverage for work-related accidents under a commercial general liability policy. hereinafter called the “Derksen scenario. does one assess coverage for the loss?52 Part of the loss was caused by a covered cause (negligent clean-up) and part was caused by an excluded cause (negligent automobile operation). 539938 Ontario Ltd. there were two potential causes of the loss: negligent work-site clean-up and negligent operation of an automobile. how.. at the same time. he rested it on an air compressor that was being towed by the insured’s vehicle.C.contest between automobile liability coverage and commercial general liability coverage). 52 The Derksen case was. at heart. The sign base somehow became dislodged and flew through the air. __:_ causes acting together. For the remainder of this article. Tragically. it will be used as a helpful example of a parallel causation scenario. non-auto related cause of the accident. 51 Thus the act was deemed a separate. 398 [“Derksen”]. the loss could not have occurred. Co. then. . The Canadian case of Derksen50 is a classic example of a concurrently caused loss resulting from parallel causes. one child was killed and three others were seriously injured. use or operation” of an automobile. 571 NE. The commercial general liability policy covered losses resulting from an insured’s negligence. Cas. 50 Derksen v.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 28 _______ LAW REVIEW [Vol.R. Parallel causation also presents an analytic problem because it is nearly impossible to articulate what caused a loss without reference to both causes operating in parallel fashion. If the policy excludes coverage for use and operation of a vehicle (which nearly all commercial liability policies do). a loss distribution case.2d 1383 (Mass.

there would be no loss.53 as well as Britain.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 29 There is no one “cause. The negligent driving alone would have caused no loss. 438 Mich. there would be no loss if there were only the negligent work-site clean-up without the negligent vehicle operation. The “Conservative” Approach The purest form of the conservative approach to concurrent causation holds that. if one cause in a causal chain is excluded from insurance coverage. This is created by insurance law’s all-or-nothing exercise of often having to choose between a covered and excluded cause.” Both become necessary and sufficient factors for the accident to have even taken place.. v.e. Each approach will be discussed in turn. In parallel causation cases.54 generally apply the 53 See i. 463. 475 N. when both were necessary to produce the loss. In other words. V. which required vehicle operation. This is a different consideration than the sequential analysis required when scrutinizing losses resulting from serially caused losses. the exclusion takes precedence and the loss is not covered. A minority of states. This discussion will form the basis for the default rules constructed to deal with concurrent causation disputes across common law jurisdictions. as there would be no improperly placed sign-base to rocket through the bus window. Four Approaches to Concurrent Causation There are four possible approaches to solving concurrent causation disputes in insurance. exploring the benefits and challenges of each. Clarke. Allstate Ins. Co.W. The efficient solution to concurrent causation therefore requires two different tools for resolving disputes based on the causation scenario: serial or parallel. 1) Michigan – Vanguard Ins. there often exists a certain paralysis about determining coverage issues. Co. In the Derksen scenario. even if other causes may be covered by the policy. without the negligent work-site clean-up. 262 . the entire loss must be excluded from coverage. followed by Hayley v. A.2d 48 (1991). By the same token. The sign base needed to be in motion to propel it through the air. if there are two concurrent causes and one cause is covered by the policy and one excluded.

07-2458. 187 N. The Employers’ Liability Assurance Co.. 110 P. v.A. followed by Wayne Tank and Pump Co.. Ins. 533 A.2d 4 (N. one of which was excluded (wear and tear). N. ed (2005) (the adoption of the conservative approach in British courts can be explained by the differences between American and British insurance contract interpretation principles). Co. Ltd. Norwich Union Fire Ins. Ins.1993) (concurrent causes “arising out of” excluded event not independent). (2006) at 118-119. 69 Wash. v. Mut.D.R.. of North America. and Midland Mainline Ltd. See also Robert Merkin. See also North Dakota’s statutory response to insurance causation under section 26. 703. 526. 35 Cal. 54 Leyland Shipping Co.2d 329 (Neb. 4) Nebraska – Lydick v. [1973] 3 All E. Hartford Underwriters Ins. § 530: “An insurer is liable for a loss of which a peril insured against was the proximate cause. Co..App. drowning caused by ice collapse. but he is not liable for a loss of which the peril insured against was only a remote cause. Insurance Co. 571. 643 N.2d 682 (Md.. Ltd. Eagle Star Ins. of Nebraska. 1971) (cattle driven off by windstorm fell through ice and drowned.L..App. Civ. Ltd.” (codifying the dominant cause approach as a default rule if the insurance contract is silent as to concurrent causation). 850 P.C.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 30 _______ LAW REVIEW [Vol.W. Inc.2d 533 (Wash. Inc.A.” Also. Unigard Sec. INSURANCE LAW: DOCTRINES AND PRINCIPLES. Ins. General Star Indem. of America v. Co.2d 273 (Mich. an excluded cause) but see Curtis O. App. Insurance: The Proximate Cause in English Law (1981) 40 Cambridge L.4th 747.C.W. Co.D. COLINVAUX’S LAW OF INSURANCE. although a peril not contemplated by the contract may have been a remote cause of the loss. and see especially John Lowry and Philip Rawlings. 1987).2d 602 (Neb. 284..App..W.) (business interruption caused by two concurrent causes.). . 2004) and Iroquois on the Beach.J. 247 Neb.) Dec 23. Inc.3d 903 (Cal. 2) Washington – Krempl v.). [2004] E. 3) Maryland – Northern Assur. 686 N. 2005) (rain-induced landslide loss excluded by broad weather-related exclusion) but see also Cal. University of North Dakota. 2008) (NO.1-32-03. a loss which would not have occurred but for that peril is excepted although the immediate cause of the loss was a peril which was not excepted. Code. 528 N.. Div. Ltd. 311 Md. 2d. Farm Bureau Ins. 1. Griess & Sons. [1918] All E.C. Malcolm Clarke. 1995) (windstorm-borne virus infects pigs and windstorm held to be the “proximate cause” of the harm). Co. 8th ed. Society. v.A.W. v. v. EDP Floors. v.R. 2008 WL 5334031 (6th Cir. 5) California (for first party property insurance only) – Julian v. 217. which provides: “When a peril is excepted specially in an insurance contract. __:_ Mich. Co. 1042 (C. Ltd. 07-2495).(Mich.W. 2002). 443 (H. Co. and entire loss therefore excluded). see Western Nat. 825 (C.

Id. is to render the exclusion meaningless. The English appellate court held that.56 the exclusion would trump the covered peril because a specifically excluded cause acted concurrently with a covered cause. Lord Denning M. Perhaps the exclusion may be seen as defining the limits of coverage. if both causes were seen as co-dominant. Exclusions therefore trump coverage under the conservative approach. Therefore. there is a covered cause plus a non-covered but non-excluded cause.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 31 conservative approach in solving concurrent causation insurance disputes. even if a dominant cause were determined. held that the dominant cause of the loss was the defective equipment. The conservative approach gets more confusing if. as best adhering to the “insurance contract as bargain” nature of insurance law. See Malcolm Clarke. use a dominant cause approach. an excluded peril. or deem a non-covered Id. held. 58 See Malcolm Clark. id.J. among the concurrent causes. supra. at 833. The insured purchased the policy with that exclusion and assented to it. all three judges agreed the exclusion cause would still take effect and oust coverage. as noted in note 5.R. This is likely synonymous with the British tradition in interpreting insurance policies as contracts. A conservative approach to concurrent causation is generally synonymous with a view that insurance contracts are contracts bargained for by insurer and insured. especially in a policy with a broad coverage grant like liability insurance. and Roskill L. The classic example of this approach is the British case of Wayne Tank.57 This approach makes the exclusion paramount. POLICIES AND PERCEPTIONS OF INSURANCE LAW IN THE TWENTY-FIRST CENTURY (Oxford: 2005) at 185.55 A factory that used liquid wax suffered a fire resulting from two possible causes.58 The insurer drafted an exclusion to define what was not an accepted risk. the exclusion should be upheld because. apportion the payout for each cause. to do otherwise. However. 57 Malcolm Clarke appears to prefer the conservative approach to concurrent causation.J. as Cairns L. The approach then has to decide whether to grant coverage. failure to install proper equipment (an excluded cause) and employee negligence in leaving the factory unattended (a covered cause). 56 55 .

apparently no complexity costs beyond the initial determination of the causal elements. 527 at 542. according to Richard A.). it is a remarkably predictable low-cost default rule to apply. __:_ cause equivalent to an excluded cause.60 If any one event in a causal chain is excluded by the policy.Concurrent Causation: Examination of Alternative Approaches (1985) S. Insurance Law . a cornerstone interpretive concept in many American jurisdictions.61 The insured contracted for coverage. First. There are. for all the same contractually-based reasons that that jurisdiction would deny a loss resulting from a covered plus an excluded cause. 60 The value of this approach lies in its predictability. if any. Ltd.e. The insurance contracts are written as either allowing no recovery or total recovery – an “all or nothing” contract. All a court must do is examine the causal elements in question and determine which. However. (adverse sea conditions (a covered cause) combined with defective boat design (policy silent as to coverage or exclusion) to damage a ship – coverage allowed because no explicit exclusion excepted coverage). [1987] 1 Lloyd’s Rep. U. on its face. v. Regardless. are excluded from coverage. The insurer See i. Yet there feels something inherently wrong with this approach. Consistency costs are also lowest under this rule. (The ‘Miss Jay Jay’). Britain. Yet the problem with this argument is that the policy also did not denote the insurer’s agreement to cover the non-covered cause either. The most logically consistent approach in a jurisdiction which applies a conservative approach to concurrent causation would be to also deny coverage for a loss resulting from a covered plus a noncovered cause. See Peter Nash 59 . Fierce. oust coverage for the entire loss. JJ Lloyd Instruments Ltd. Ill. a jurisdiction with a conservative approach to concurrent causation. 61 Peter Nash Swisher argues that the conservative approach goes against the reasonable expectations of the insured.A.J. because the insurance contract did not specifically exclude the non-covered concurrent cause. does not do this. 32 (C. There are low information and administrative costs in applying this rule. Perhaps the allor-nothing analysis of insurance coverage boxes the court into this anomalous result. L. Northern Star Ins.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 32 _______ LAW REVIEW [Vol. Co. there are some obvious efficiencies to the conservative approach to concurrent causation.59 It instead takes a pro-coverage approach in these instances.

Insurance Causation Issues: The Legacy of Bird v. obviously. Recall that the coverage grant for both all-risks property and liability policies is generally remarkably broad. it is often “all risks” insurance. L. administrative. Similarly. 351 at 363.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 33 contracted to provide coverage. particularly in serial causation cases. with only specific. liability insurance coverage typically insures against liability stemming from the insured’s actions anywhere in the world. Finally. The conservative approach raises the spectre of causal remoteness. which hearkens dangerously back to the baggage of proximate cause analysis in tort law. It is most often “negligence” insurance. and certainly consistency costs into insurance law. As indicated in the beginning of this article.62 Which causes would be deemed concurrent? How far back can one trace causal events. Co. but also to avoid coverage in certain specified events. . or tort insurance. On a practical level. application of this approach will create gaps in insurance coverage that no current product on the market fills. the conservative approach has three fundamental additional problems. in order to deem an excluded cause “the” cause which ousts coverage? What is a legally “responsible” cause to which a court can look in order to direct the coverage question? The answer likely lies in something akin to tort law’s proximate cause analysis. infra. For standard property insurance. this approach errs on the side of nothing. all risks save and except those which the insurer thought to exclude. enumerated exclusions. The coverage grant provides a legal defence and indemnity for Swisher.J. a return to causation analysis from tort law imports a number of information. Paul Fire & Marine Ins. Why should an insurer benefit from the fact that a normally covered event will now not be covered because an excluded cause is operating somehow in concert with that typically covered cause? The application of this legal rule digs at one’s sense of fairness because. “All risks” coverage is comprehensive property loss coverage for. St. The approach has the real potential to frustrate coverage language within the policy: it denies coverage in all parallel causation situations. 62 See Part III. in an all-or-nothing equation of insurance coverage. These additional inefficiencies eclipse the costsavings of a seemingly simple default rule to apply. (2002) 2 Nev. complexity.

DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 34 _______ LAW REVIEW [Vol. First.63 In fact. with a corresponding premium increase. 63 . because concurrent causation scenarios are difficult for insurance consumers to reliably predict. Most factual scenarios leading to a loss have multiple causal factors. increase costly litigation behaviour as it incentivizes insurers to search for a concurrent excluded (or perhaps non-covered) cause in the hopes of ousting coverage. no insurer markets concurrent causation insurance. Like all risks property coverage. Insurers may be escaping contractual liability for losses they actually agreed to pay because of the fortunate happenstance of a second non-covered cause in the factual matrix. The exclusions in a standard liability policy carve out behavioural exceptions for which the insurer will not provide indemnity. Disputes about concurrent causation only exist because one potentially excluded cause is in the mix. applying the conservative approach to causation robs the insured of this broad coverage every time. Such an approach may. the insured always loses. If. like intentional acts or criminal behaviour. The gap issue is more problematic in liability insurance markets than in property markets. __:_ any losses resulting from the insured’s actions for which the insured becomes liable to pay a third party. So the cost savings in the conservative approach being a simple. liability insurance coverage projects the general assumption that coverage is granted. The all-or-nothing methodology of the strict conservative approach means that. That fact is surely not brought to the insured’s attention in any insurance policy wording anywhere. predictable default rule may be lost not only in increased litigation costs but in an unfair pocketing of premiums by insurers who are essentially then writing contracts which are only enforceable in non-concurrent causation situations. the very purpose of the broad coverage grant is frustrated. coverage is denied. Insureds may be misled as to the scope of their coverage. in disputes about concurrent causation. Anti-concurrent cause clauses are really just contractual embodiments of the Though there is certainly no reason why an endorsement for such could not be marketed. every time that occurs. in fact. For what could be broader than “all risks” or indemnity for anything anyone does in the world? Yet. as has already been described. in a concurrently caused loss. The conservative approach therefore creates serious market gaps in insurance products. most policies are rife with anti-concurrent causation language.

Assume an insured is covered by a standard commercial liability policy which covers for all negligent acts but excludes those losses caused by negligent use and operation of an automobile. rain. is caused by two concurrent parallel The complexity and information costs are increased when the scenario adds more than two possible concurrent causes. The insured cannot purchase anything to circumvent this. a serial causation example using property insurance. the injury and death of the schoolchildren on a bus. However. Take the Derksen scenario. The covered risk of rain in the “all risks” coverage has actually been converted to a covered risk of “rain only. A parallel causation example using liability insurance further demonstrates that there are particularly troubling gap problems with so-called “tort insurance” and the conservative approach. During the night. 64 . the insured has no coverage. is a covered peril. further damaging the home and its contents to a greater degree.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 35 conservative approach to concurrent causation.64 The insured paid a premium and is absolutely insured for rain damage but because that cause combined in sequence with an excluded cause. the insurer can deny coverage. However. This makes it difficult for the insured to arrange its insurance accordingly. an insurance dispute about loss distribution but with imbedded coverage issues as well. like the Weathered House example (wind plus rain plus hail). the insurer pockets a premium even though the bulk of the risk that caused the loss. because ties go to the insurer for concurrently caused losses. The resulting loss in this example. The insurer agrees. A pair of simple examples will demonstrate how the conservative approach leads to uninsured gaps which create market inefficiency through losses for which insurer liability for risk is initially accepted but subsequently dodged. Imagine a house is insured under a property policy for all risks save earth movement. This allows rain to get into the home. ex ante. an earth tremor causes damage to the house such that portions of its roof become unsecured. because the excluded cause – earth movement – combined with the covered cause – rain – to produce the loss. The exclusion trumps coverage.” It is an extra-contractual modification. The insured thought she had coverage for rain damage. Under the conservative approach to concurrent causation. First. the insurer can argue that it chose not to cover anything related to earth movement.

These scenarios are extremely difficult to predict under what potential circumstances there might be a cause which might trigger an exclusion. but. This creates an inefficient doubling of the insurance market as the brokers are forced to take up the secondary slack. the insurer again has agreed to cover one of the two competing causes of the loss. the schoolchildren) is now left with nothing because of the application of the conservative approach in a concurrently caused loss scenario. The insurer pockets the premium for the very loss that was partially caused by the covered cause. In this instance. The insured loses coverage because a covered cause combined with an excluded cause to produce the loss. the victim injured by the insured is also left with no source of compensation. . Neither operating alone is sufficient. __:_ causes: negligent automobile operation and negligent work-site clean-up. The liability context adds an additional inefficiency to the scene: whatever third party was expecting compensation from the insured’s indemnity proceeds (in Derksen. because of the all-or-nothing conservative approach which favours exclusions over coverage. The insured in this parallel causation situation paid for broad. Predictability and stability suffer. The insurance market itself is distorted by the frustration of promised-for coverage. the wealth distribution mechanisms of insurance in the tort system fail to be deployed.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 36 _______ LAW REVIEW [Vol. then. No coverage results. The exclusion trumps the coverage clause. comprehensive tort insurance but had the misfortune of having one cause of the loss characterized as an excluded cause. The unpredictable gaps created by the increased incidence of coverage denial may eventually offload some or all of the risk of concurrently caused losses to not insurers but insurance brokers instead. to the entire insurance system in both parallel and serial concurrent causation cases outweigh the benefits of a simple default rule like the conservative approach. But the loss could not have even been brought about without the second excluded cause operating in conjunction. The costs. Again. So aside from the wealth-protection mechanism that is arguably a secondary role of liability insurance. there is no market product for insureds who wish to protect themselves in the wake of concurrently caused losses. The problem with liability insurance is that loss caused by negligent conduct is very often a combination of parallel concurrent causes.

v. University of North Dakota. 65 . or proximate cause.D. when one turns to the automobile policy. The “Dominant Cause” Approach The most prevalent approach to deciding concurrent causation insurance disputes is the dominant cause. there was a contest between an auto and commercial liability policy. 643 N.66 Yet this approach is also the most responsible See Robert Merkin. and Joseph Lavitt. Mut. the insured had two separate policies both of which could have potentially been triggered.e. If an insured has two potentially overlapping insurance policies. The simple defence to these types of suits is that no product currently exists on the market to fill the gaps. Rev 1 at 16. then the loss is excluded under the commercial liability policy. the conservative approach essentially cancels out both policies when a loss is resulting from concurrent causes. B. and the Third Restatement of Torts: Cracking the Conundrum (2008) 54 Loyola L. That may not be enough to contain the litigation costs of insureds who asked their brokers for “full coverage” only to have no coverage in a concurrently caused loss scenario. as that policy only covers automobile-related losses. Ins. Co. Take the Derksen example. Stempel at 7-14. the Katrina Disaster. Finally. In that case. The Doctrine of Efficient Proximate Cause. in fact. COLINVAUX’S LAW OF INSURANCE. If negligent work-site clean-up is covered and automobile negligence excluded by the commercial liability policy.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 37 Brokers may face liability lawsuits for selling insurance products with gaps to unsuspecting insureds. using the conservative approach. 2002) at 12. the greatest difficulty with the conservative approach to concurrent causation is that it never works for loss distribution disputes for both serial and parallel causation cases. Western Nat. The auto exclusion trumps the non-auto coverage. based on that logic.W. Prosser’s Folly. This reductionist approach leaves the insured with a perverse result: no liability coverage when. there would be no coverage under that policy either. approach. 8th ed.65 The conservative approach therefore leads to inefficient results. However. (2006) at 119 (warning of this very absurdity in result).2d 4 (N. Efficiency gains made in consistency and complexity are lost in market spillovers. potential elimination of important victim compensation. and impossible loss distribution frameworks. 66 See i.

68 Jerry and Stempel.J. where the House of Lords held at 369 per Lord Shaw that the proximate cause is the cause which is “proximate in efficiency. v. __:_ for the haphazard. Fierce. is better than the unpredictable “hocus pocus” of the dominant cause approach.e.” 69 Wayen Tank. It is also used in Britain. Ltd. If that cause is an excluded or non-covered cause.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 38 _______ LAW REVIEW [Vol. Stempel at 7-10 – 7-18. If that cause is a covered cause. The dominant cause approach borrows heavily from tort law’s doctrine of proximate cause. Nevertheless.U.Concurrent Causation: Examination of Alternative Approaches (1985) S. A.68 It is not necessarily the last cause in the chain.. even the conservative approach with all its problems. POLICIES AND PERCEPTIONS OF INSURANCE LAW IN THE TWENTY-FIRST CENTURY (Oxford: 2005) at 185-6 (the unpredictability of the “common sense” approach to choosing “a” cause leads to haphazard results and too much “judging”).67 The reason is simple: choosing one cause in a causal chain of events as “the” dominant cause often invites equity-based “justice” concerns to creep in and affect predictability of the result. 527 at 544-545 advocates that any approach. It is the most “responsible” cause. Prosser’s Folly. Forcing a court to choose the most “responsible” cause prompts a court to confuse causation principles in tort and contract law. and the Third Restatement of Torts: Cracking the Conundrum (2008) 54 Loyola L. 350 (H. 67 . Most often. Norwich Union Fire Ins. the Katrina Disaster.C. the dominant cause approach leads to the most “responsible” cause being deemed based on who the court wants to be “responsible” for bearing the loss. L. the loss is covered. id. not necessarily the factual cause of the loss. Jerry. this is the predominant approach to concurrent causation in the United States. Ill. Richard A.L. The approach stems from the 1917 British decision of Leyland Shipping Co. the loss is not covered.). and Malcolm Clarke. The dominant cause is said to be the cause which had the greatest effect on bringing about the loss.69 in See i. supra. Insurance Law . supra at 560-566 and 578-580. unpredictable jurisprudence surrounding concurrent cause. Society Ltd. Joseph Lavitt argues against the dominant cause approach in The Doctrine of Efficient Proximate Cause. It requires a court to search among the possible concurrent causes and choose a dominant or proximate cause. Rev 1 (efficient proximate cause doctrine produces different results in different jurisdictions for property insurance disputes).

DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE

2008]

Default Rules for Concurrent Causation

39

combination with the conservative approach. It featured largely in Canadian insurance jurisprudence until the Supreme Court of Canada adopted the liberal approach in Derksen70 in 2001. The dominant cause approach is likely so deceptively attractive to courts because it offers a sense of decisive choice with a dash of equity: choose the cause most “at fault” for the loss.71 There is a perception that only the insurer who has agreed to underwrite the main causative risk will be saddled with the cost of covering the loss. Being able to determine coverage disputes by saying “that is the cause that most brought about the loss” carries with it a profound sense of finality and fairness. But its application has been anything but fair on a systemic level. The reason is simple: courts are not consistent in choosing a dominant cause. Answers to the dominant cause question differ based on how the factual story leading up to the loss is told. “Dominance” is often in the eye of the beholder. There is little consistency among jurisdictions and even among the same courts with similar fact patterns. This has created a tortured pattern of litigation because litigants cannot reliably predict coverage in concurrent causation cases where the dominant cause approach will be applied.
Derksen, supra. Peter Nash Swisher argues that the dominant cause approach is probably the best middle ground approach to concurrent causation in insurance, rather than a liberal or conservative rule, as neither is consistent with the reasonable expectations of the parties. The dominant cause approach, according to Swisher, is the best option to apply if a fact-finder can indeed isolate a dominant cause. If no cause can be isolated the dominant cause, Swisher argues for a pro-coverage liberal rule to be applied. See Peter Nash Swisher, Insurance Causation Issues: The Legacy of Bird v. St. Paul Fire & Marine Ins. Co. (2002) 2 Nev. L.J. 351. This concept is similar to Robert H. Jerry’s dualapproach strategy. Robert H. Jerry argues for a dominant cause approach as a default rule where the dominant cause is discernable. If it is impossible to choose a dominant cause, Jerry advocates switching to the liberal approach. The problem with leading with a dominant cause approach, as Swisher and Jerry propose, is that it continues to invite the hair-splitting causal inquiry that creates such inefficiencies in the first place. Perhaps there will be few times when a court would admit that it is “impossible” to find a dominant cause in a given factual scenario. See Jerry at 568.
71 70

DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE

40

_______ LAW REVIEW

[Vol. __:_

In parallel concurrent cause situations, it is actually impossible to find a dominant cause. The test does not work for this subset of cases. Because both causes are necessary to bring about the loss, because without both the loss would not have occurred. Choosing “dominance” in an impossible to answer situation therefore becomes an exercise of something other than application of a legal rule. Therein lies the source of inconsistency among parallel concurrent cause situations. Justice McLachlin of the Supreme Court of Canada cautioned against the very unproductive “metaphysical debate” that surrounds the dominant cause approach to concurrent causation.72 In the Derksen scenario, how can the “dominant cause” be negligent automobile operation when, without the negligent work-site clean-up, there would be no loss at all? The vehicle would have harmlessly passed by the school bus without the deadly sign base plate resting on the towed air compressor. Similarly, without the negligent work-site clean-up, the sign base plate would not have ended up on the cross-member of the compressor. That alone is not sufficient to produce the loss. The vehicle needed to be in operation in order to propel the sign base at the school bus. These causes are, in effect, co-dominant. In serial concurrent cause situations, it is at least more theoretically possible to find a dominant cause, but the information, complexity and administrative costs in litigating that question are unusually high. This leads to extraordinary consistency costs. The causation question usually becomes one of proportionality: dominance equating to responsibility for largest proportion of the damage. That may not be an obvious answer in most cases. It may certainly not even be a consistent answer among courts adjudicating the same fact pattern. In the “Weathered House” example, what is the dominant cause? Is it the wind which took the roof off and allowed the rain to come in? Is it the rain which soaked a greater proportion of the insured’s property? Or is it the hail, which is an excluded cause but which also smashed the insured’s property, thereby causing the greatest dollar amount of damage? Regardless of the answer, the route to get there is expensive in information and complexity costs and lacks predictability. Tortured litigation results.

CCR Fishing Ltd. v, British Reserve Ins. Co., [1990] 1 S.C.R. 814 at 823.

72

DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE

2008]

Default Rules for Concurrent Causation

41

The costs of error in using the dominant cause approach to arrive at an inefficient answer are remarkably high. In coverage disputes, the all-or-nothing approach of choosing a dominant cause forces an insurer to perhaps cover all of a concurrently caused loss when only perhaps one cause was agreed upon in the policy to be covered. Alternatively, the result could be that the insurer pockets a premium for a potentially covered loss from which the insurer escapes liability. Because the loss was a concurrently caused loss, the deemed dominant cause may not be covered, but the secondary cause contributing to the loss would normally be covered, if operating alone. No coverage would result in this situation. This result is particularly problematic in catastrophic, widespread loss cases caused by a complex causal factor, like an earthquake, volcano, flood, or hurricane.73 If the dominant cause of the loss is held to be an excluded cause within the widespread loss event itself, large groups of insureds would likely have no insurance coverage for the single biggest loss of their lives. This tips the balance against those who are in most desperate need of insurance coverage in favour of stable insurance markets. That may not be an efficient market response if the government is then left with providing some compensation to the destitute left with nothing after a widespread loss disaster. In loss distribution disputes, the insurer who specifically did not agree to underwrite additional concurrent risks can easily end up bearing the entire cost of the insured loss if the deemed dominant cause happens to be a covered cause. There is no proportional recovery under the dominant cause approach. The same logic operates to either grant the insured a windfall or a total loss in the event of a concurrently caused loss. The dominant cause approach therefore has a high potential to do violence to the very language of coverage grants and exclusions in insurance policies. Finally, like the conservative approach, the dominant cause approach creates serious market gaps in coverage for
See i.e. Joseph Lavitt, The Doctrine of Efficient Proximate Cause, the Katrina Disaster, Prosser’s Folly, and the Third Restatement of Torts: Cracking the Conundrum (2008) 54 Loyola L. Rev 1 at 2-7 and Rob Risley, Comment: Landslide Peril and Homeowners’ Insurance in California (1992) 40 U.C.L.A. L. Rev. 1145 (advocating a liberal approach to causation for earthquake damage cases).
73

which does cover negligent work-site clean-up. even though the contractual language of the policy indicates coverage is likely for a secondary concurrent but non-dominant cause. the dominant cause approach creates these gaps in a more unpredictable and often contradictory manner. limits issues might also cap recovery in an arbitrary fashion. The Derksen example demonstrates this problem. If the dominant cause of the loss is deemed to be negligent automobile operation. It depends on a court’s view of which cause is “responsible. are outside the inquiry. __:_ which there is currently no product to fill. 74 . Even though both policies stipulated to cover some aspect of the ensuing loss. it specifically did not contract to cover for negligent work-site clean-up. This insurance policy is immediately taken out of play. There is no guarantee from case to case that a scenario with similar facts may end up with the same cause being deemed the dominant cause. the automobile policy covers the loss. Alternatively. choosing a dominant cause forces an insured into a coverage box – coverage questions are only answered with reference to a single cause in a concurrently caused loss. while the auto policy insurer agreed to indemnify for losses caused by automobile negligence. choosing a dominant cause locks the insured into centreing the inquiry around that one cause. Only those policies which cover the deemed dominant cause can be triggered. the insured (and through the insured. the commercial liability policy. If there were an operative exclusion in the auto policy which might somehow deny coverage to the insured. the insured is locked into the auto policy and its corresponding exclusions and limits.74 Imagine if the auto policy’s limits were substantially less than the commercial liability policy. Second. To that end.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 42 _______ LAW REVIEW [Vol. The result depends on the story-telling of the causal scenario.” and all the error bias that that inquiry entails. or do not cover the loss. It depends on the pleading of the factual matrix making up the loss. Potentially. with multiple policies potentially covering the loss. Those policies which exclude. However. the insured is left with nothing. has an exclusion for auto negligence. If an insured is in a loss distribution dispute. This is problematic on its face because. the injured accident victims awaiting Ontario’s no-fault automobile insurance scheme at the time of the accident did not provide coverage for non-pecuniary losses.

The dominant cause approach is. The end result is a perceived unfairness in operating this particular doctrinal rule. Choosing one cause over another in a parallel causation situation is akin to being wrong one hundred percent of the time. nondominant cause will dictate that coverage is ousted.” It is in the eye of the beholder. This has resulted in increased information expenses. The all-or-nothing nature of the dominant cause approach creates such insurance gaps. the concurrent secondary. C. No one can predict what the fact-finder will deem the “dominant cause. If one cause in the causal chain is a covered cause. for all these reasons. Therefore. if a court gives effect to the anticoncurrent cause language. even though the insured purchased two separate insurance policies designed to cover part of the concurrently caused loss. The benefits of overlapping coverage disappear. But the dyadic nature of the “one or the other” effect of the dominant cause approach ignores the whole problem of a concurrently caused loss: there is more than one cause at work. The dominant cause approach therefore creates unpredictability in application and inefficient market effects on insurance. the entire loss is covered under the liberal approach. This rule operates whether the dispute is a coverage . The rule does provide artificial solace in that choosing a ‘responsible’ cause as the dominant cause can create the perception that justice is being done. even though the dominant cause may be covered by the policy. increased administrative and complexity costs. These gaps become even more problematic if one or more policies have anti-concurrent cause clauses which are upheld as valid. the insured wins. The “Liberal” Approach The liberal approach to concurrent causation is the reverse of the conservative approach: in the case of a contest among causes. the most inefficient of the four possible approaches to concurrent causation in insurance. and very high error costs. The insured will not be able to turn to the second competing insurance policy in a loss distribution dispute because the dominant cause has been set as triggering the first insurance policy.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 43 compensation) could end up with nothing or a substantially capped amount of recovery.

[2001] 3 S. 1989). 1262. 6) New Jersey – Salem Group v. Co. Co. 2) Texas – Guaranty Nat.4th 747. 110 P. Co. Co. Dist.J.E. State Farm Fire and Casualty Co. in a parallel causation case called Partridge. That state applies a dominant cause approach for first party property insurance cases. North River Ins.R..C.357 magnum pistol so it would The following states have adopted some permutation of the liberal approach to concurrent causation: 1) Minnesota – Waseca Mutual Ins. 35 Cal. App. 2d 1386. v. v. 2005. 398.2d 514 (Wis. Partridge. Co. 1. 909 F. Maxey.D. but only where the concurrent causes are completely independent of one another. v. 514 P.2d 883 (Tenn. each capable of causing some loss: 1) Florida – Guideone Elite Ins. 166 Vt. 4) Vermont (Vermont State Farm Mutual Auto Ins. Ctr. & Guar. State Farm Fire & Cas. Co. Noska.3d 203. but for third party liability insurance cases only (State Farm Mutual Automobile Ins.2d 123 (Cal. 539938 Ontario Ltd.D. Supp.W. Only a minority of states75 and Canada76 adhere to the liberal approach to concurrent causation. 420 F. 331 N. 2003. St.2d 917 (Minn. 1990) (Tex.2d 63 (Minn.. 1988). 140 (1992). Hartford Underwriters Ins. California also applies the liberal approach. 257 Cal.). 514 P.2d 138. supra. Co. Corp. v.. 3) Ohio – United States Fid. 2) Wisconsin – Lawver v. arguably it is not. 292 (Cal. The Supreme Court of Canada adopted the liberal approach for Canadian insurance cases in Derksen v. 452 (Vt. Like the conservative approach. 5. App. 129 Ohio App. Tenn. 1983). 527 So. Paul Fire & Marine Ins.. it is a simple default rule to apply. 1973).) and Bituminous Cas. 1205-06 (1998). 965 F. 716 N. Co. 51-52. The question remains as to whether it is a costly legal rule to maintain.3d 1317. 3d 45. The liberal approach was first invoked in California. 137 (5th Cir. Roberts. 1330 (11th Cir. 3) Tennessee – Allstate Insurance Company v. 1997).W.. v. Co. 238 N. v.W. St.2d 1201.3d 903 (Cal. Co.2d 123 (Cal. Supp. 1973). Rptr. 481 N. Elizabeth Med. v. v. 1976). 4) Missouri – Cawthon v. Old Cutler Presbyterian Church.). had filed down the trigger of his .W.. 2d 901 (W. 1992). v. 2001). Seefeld. __:_ dispute or a loss distribution dispute. 136 F.77 The insured. 1387-88 (Fla. followed by Davidson Hotel Co. 5) Florida – Wallach v. Rosenberg. Partridge.1991). Co. 811 S. 215 (Tex. followed by State Farm Ins. 77 State Farm Mutual Automobile Ins. Garvey v..). Co. 110 S. and Julian v.2d 133. Ct. 607 A. On a systemic level. Partridge. Inc. Oliver.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 44 _______ LAW REVIEW [Vol. 76 Derksen.. 1997). v. 128 N. 1269 (W. Other states recognize the liberal approach to concurrent causation. Mo. Watts. 2005) (Fla. Ins. Boling. 75 .W.

However. The contest for payment of the loss was between two third party liability insurers: Partridge’s homeowners insurance policy and an automobile policy. There are some obvious benefits to the pure liberal approach. 35 Cal. California is unique in that that state’s courts only apply the liberal approach to concurrent causation in California for liability insurance disputes. First. subsequently. in which one or another cause can more often be separated out as to its “responsible” effect.. and administrative costs. First party insurance. complexity. Partridge was holding the pistol in his lap while riding as a passenger in a truck.4th 747. It is often impossible to separate out the causes. Hartford Underwriters Ins. Richard Fierce. The reasoning as to why the California Supreme Court created this split are informative for defining a set of default rules for concurrent causation. Concurrently caused losses in property insurance are most often serially caused losses. See i. The truck struck a bump and the pistol discharged.e. Therefore. 257 Cal.e. 292 (Cal. The inquiry stops. Because the loss was held to be concurrently caused both by Partridge’s negligence in filing down his pistol trigger and the negligence in driving. The nature of property insurance is such that it covers losses which produce an effect on a tangible thing (i. both insurance policies were called upon to cover the loss.79 One only has to examine the various causes at work. injuring the driver.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 45 fire at the slightest touch. 79 78 .. Rptr. There are strong predictability reasons for such a default rule. Julian v. the loss is covered. allows the insurer and insured to more surgically control both coverage and exclusionary language. 110 P. weather damage to a home). unlike third party insurance. Co. State Farm Fire and Casualty Co. 2005). concurrently caused losses in the liability insurance context are more complex. supra at 544 (liberal approach is more expensive but has greater certainty costs). If one concurrent cause is covered. its application enjoys comparatively low information. The California Supreme has since relegated the liberal approach to concurrent causation to only cases of third party liability insurance in which concurrent causation is featured.78 First party property insurance losses follow the dominant cause approach. There are virtually no complexity costs beyond ascertaining the various See Garvey v.3d 903 (Cal. 1989) and.

This removes the dispute from an inquiry about interpreting responsibility among causes to instead one of true loss distribution. __:_ causes in play in a factual scenario. barring any contractual language indicating otherwise.82 If an insurer has explicitly excluded a certain loss. It may be too pro-coverage. 81 In addition. This approach may fit more reasonably in those American jurisdictions and Canada which apply insurance contract interpretation rules that are built to counter the Jeffrey Stempel notes that. multiple overlapping insurance policies may be triggered under this approach.80 This is a more efficient dispute. the jurisdiction’s legal rules about overlapping insurance coverage take effect to resolve the issue. as many insurance contracts either have provisions for loss sharing or. the liberal approach does create an efficiency and fairness issue. there is no reason why both insurers should not pro-rate the cost of indemnity. but between insurers themselves as to how to ratably share the loss to be indemnified. 82 Peter Nash Swisher notes that the insurer probably never contractually intended to provide such broad coverage as a liberal approach to concurrent causation “creates”.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 46 _______ LAW REVIEW [Vol. However. Insurance Causation Issues: The Legacy of Bird v. St. the liberal approach avoids any issues with gaps in coverage that the conservative and dominant cause approaches create. Jerry notes that the liberal approach invites a view that causes are somehow acting independently. 351 at 366. The liberal approach stops the “metaphysical debate” about causation. 80 . The dispute is then not between insured and insurers. even if they are not. In loss distribution disputes.J. Paul Fire & Marine Ins.1. In effect. Robert H. L. the loss is covered. this is a pro-coverage approach. the liberal approach may still require that insurer to cover the loss. Most important. alternatively. independent. See Jerry at 982.81 The approach to concurrent causation has the potential to frustrate exclusionary language in a policy. A division between serial and parallel causes solves this problem. coverage trumps exclusionary language. See Stempel at 22-86. In fact. Co. As long as one cause in the causal chain is covered. (2002) 2 Nev. if a tort arises from two concurrent causes. which are most expensive and unpredictable to solve. This is a marked savings particularly for parallel causation cases. See Peter Nash Swisher. in fact.

L. For example. where two necessary concurrent causes produce the loss and neither on its own is sufficient to cause some harm. Fierce. and consistency cost reductions are instead off-loaded into the insurance marketplace in terms of increased payouts for insurers and corresponding increases in insurance premium costs for insureds. rather than have those costs wasted in decreased systemic efficiency. apportionment. Ill. both of which favour an insured in interpreting insurance policies. 527. complexity. the liberal approach to concurrent causation gives true credence to this maxim. The liberal approach is not perfect. There is a strong argument that the best place to solve that issue is to have those costs funneled into the insurance marketplace to fund creative loss distribution efforts. there may still be coverage because a secondary or tertiary sequential cause. Richard A. if that cause is an excluded cause. Where there may be one cause among the sequential causes that is seen as more “responsible” for the damage. has not yet been adopted in any jurisdiction. but has been suggested in scholarly literature. only to have that excluded loss occur. This result smacks of unfairness for the insurer who articulated a desire not to cover a loss. and still have to indemnify the insured. Insurance Law . Fierce first suggested an apportionment approach.U. The “Apportionment” Approach The final possible approach to concurrent causation in insurance.Concurrent Causation: Examination of Alternative Approaches (1985) S.J.83 This Richard A. D. The ultimate question is whether or not the efficiencies gained in information. borrowed from comparative negligence analysis. The notion was again posited by 83 . The approach also best fits with the contra proferentem and reasonable expectations doctrines.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 47 perceived imbalance between the insurer who drafted the policy and the insured who purchased the policy. The liberal approach perhaps promulgates a fairer result in instances of parallel causation. one can arrive at the opposite result in instances of serial causation. unable to bargain over its terms. administration. seen as a “minor” or “less responsible” cause. if a jurisdiction interprets coverage clauses broadly and exclusion clauses narrowly. will be covered under the policy. However.

The coverage question can then be asked on a “percause” basis. Robertson. at 544. 423 at 453. This is particularly common in loss distribution disputes. Rev. Proximate Cause in Insurance Law – Before and After Derksen. In the Weathered House example.e. L. __:_ approach involves apportioning percentage responsibility to each cause in the causal scenario that had some role in bringing about the loss. 977. Apportionment is akin to a comparative negligence analysis in tort. Rev. What Will Policyholders Expect? (1989) 29 Santa Clara L. A contest among “other insurance” clauses often ensues. so the insured should rightfully not be indemnified by the insurer for that proportion of the loss. .86 It therefore begs the question: why not move the approach to the front end of the dispute and use it to assist in solving concurrent cause cases? The benefit to an apportionment approach is that it is capable of acknowledging both coverage and exclusionary Mark Umeda in Concurrent Proximate Causes in Insurance Disputes: After Garvey. Various loss distribution principles have been created in insurance law to deal with this losssharing among competing insurers. (2006) 32 Advocates Q. Commonly called “other insurance” clauses. and in Canada by Anthony J. If wind and rain are covered causes. supra. Saunders.85 Such an analysis in the insurance law sphere is routinely undertaken to determine payout contribution among competing insurers but after coverage determinations have taken place. 175 (1998) and Ellen M. Bublick. and hail is an excluded cause. 86 There is often language in insurance policies which dictate costsharing priorities. assume wind is responsible for 20% of the ultimate loss. rain responsible for 20% of the ultimate loss. 85 See i. David W. Fact-finders must then determine which insurers pay how much. Love and Fury: Recent Radical Revisions to the Law of Comparative Fault. 59 La.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 48 _______ LAW REVIEW [Vol. Comparative Fault to the Limits (2003) 56 Vanderbilt L. most of these clauses deem the policy “secondary” insurance and require other insurers to pay out the loss first on a “primary” basis. 84 See Richard Fierce. and hail responsible for 60% of the ultimate loss.84 where various tortfeasors are assessed percentage contributions of fault. 140 at 166. Rev. Insureds are then indemnified according to the percentage responsibility of each concurrent cause in the factual matrix. Sixty percent of the loss was caused by an excluded peril. the insured is indemnified for 40% of the ultimate loss by the property insurer.

87 . but the initial run of cases would prove unpredictable. What See Richard A. apportionment avoids the problem of insurers indemnifying for losses which they specifically agreed not to indemnify.J. Apportionment is a “split the difference” methodology. Insurers indemnify insureds only for what the parties contracted. there will be complexity costs in determining proportionate responsibility among the various causes that happen in sequence.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 49 language in insurance policies. over time. For serial causation especially. L. even though hail did more monetary damage?) This numerical exercise will also likely be completely different for each case.Concurrent Causation: Examination of Alternative Approaches (1985) S. thereby avoiding complete gaps in coverage. Unlike the all-or-nothing qualities of the conservative and dominant cause approaches. but outweighed in tradeoff for greater recognition of reasonable expectations of both insured and insurer) and Mark Umeda. it is likely impossible to assign anything but an equally ratable share of responsibility on a per cause basis. did rain touch more of the property than the hail. Fierce. Perhaps.e.87 This will require additional evidence and court time in order to sort out. and administrative costs in assigning percentage responsibility for concurrent causes. complexity. case precedents will. by analogy. varying with the facts. Ill. There will likely be disputes about the various proportions themselves. consistency. To do otherwise is too costly to answer. apportionment allows for at least some recovery in concurrent causation situations. How does one arrive at a number in the Weathered House example? What numerical percentage corresponds to each cause? Does one use the estimated dollar value damage amount as the guidepost in a serial causation situation? Or perhaps instead the proportion of the cause which affected a greater amount of the property? (i. supra at 454-455 (potential for approach to create frivolous claims if causal thresholds for ‘responsibility’ not clearly set). Unlike the liberal approach. prove helpful. neither pro-insurer nor pro-insured in aggregate result. There are comparatively moderate information. U. 527 at 544 (higher uncertainty and litigation costs with apportionment approach. There are some problems with the apportionment approach. For parallel causation cases. Insurance Law . and on a rationally proportional basis.

The Efficient Solution to Concurrent Causation The efficient solution to concurrent causation among common law jurisdictions is to develop not one but two default rules for the two separate types of causation scenarios. VI. For serial causation cases. Proximate Causation in Insurance Law. supra. See i. For parallel causation cases. So there is a definite cost-savings. 310 at 319. Garvey. The numerical exercise of assigning percentage responsibility is costly to perform. according to John Lowry and Philip Rawlings. Apportionment is therefore likely inapplicable for parallel causation cases. The comparative causal factors can at least be notionally separated in sequence. one would expect in that case that a fact-finder would take the most efficient route and split the concurrent cause responsibility down the middle: 50% to each of the two causes. particularly for those disputes about loss distribution. Rev. a concept repeatedly echoed by courts88 and scholars. Instead. 89 88 .89 See ie.e. but likely results in the fairest result to insurer and insured alike. Using separate default rules tooled to the differences between serial and parallel causation is largely akin to treating first party property insurance as different from third party liability insurance in the causation analysis. The one benefit to apportionment in loss distribution cases is that applying apportionment at the coverage determination stage also answers the loss distribution question among competing insurers. fact-finders should use the liberal approach for both coverage and loss distribution cases. (2005) 68 Modern L. McDowell. the most sensible approach is to use apportionment. British courts have not yet addressed the distinction between property and liability insurance in causation analyses.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 50 _______ LAW REVIEW [Vol. Apportionment is thus perhaps best suited to serial causation cases and particularly those dispute involving loss distribution questions. __:_ is the percentage of negligent driving responsible for the ultimate loss in the Derksen scenario? Thirty percent? Forty percent? Imagine the myriad of evidentiary permutations that would be required to exactly answer that question. The cost is too great to put a numerical value on a parallel cause. supra at 591-592.

90 . where the policy requires the loss to have been caused by a specific peril).” like a hurricane. and no insurer is guaranteed a chance to escape scott-free even though coverage is clearly provided. But that investment is far less costly in the short run than the cost to the overall insurance market in the scenarios using the all-or-nothing dominant cause and conservative approaches. (2005) at 233-234 (advocating different treatment of concurrent causation in property insurance than liability insurance.e.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 51 A. John Lowry and Philip Rawlings. no insured or third party loses the opportunity of at least some compensation. Property insurance policies allow for more surgically tailored coverage and exclusion clauses. This is especially important in catastrophic loss events. the exclusionary language is often frustrated. Therefore. 2d ed. McDowell. supra. complexity and administrative costs on the front end. An Apportionment Default Rule for Serial Causation Cases Apportionment is the most effective approach in concurrent causation cases where the loss is caused by serial causation because it produces the most cost-efficient and indeed fair answer among the four possible approaches. at 592 (first party property insurance causation analysis more closely tied to strict contractual analysis than tort analysis). The insured often loses coverage which she paid for and expected in just such a loss. Recall that serial causation cases are most often first party property insurance cases. and for which the insurer pocketed a premium and underwrote a portion of the very loss that occurred. See i. INSURANCE LAW: DOCTRINES AND PRINCIPLES. tornado. or earthquake. There is also no elimination of the possibility of multiple policies being triggered in loss distribution disputes. parties would invest some information. Furthermore. particularly for “all risks” property coverage. By assigning a percentage contribution for each sequential cause in the serial causal chain. because property insurance coverage is defined in terms of causation. Causes have acted concurrently in a sequential nature to produce an end loss.90 The insurer has greater linguistic control over the underwritten risk. Insurance brokers will not be called upon to indemnify insureds. which typically have multiple causes acting together in one conceptual “event. gaps in coverage are eliminated. In those instances. thereby avoiding that market inefficiency. as opposed to liability insurance where the insurer’s exposure is as wide as tort law.

There may be some incentive to plead so as to attempt to tag more insurance policies in a loss distribution case. 91 . Insurance Law . __:_ This produces the most fair and predictable result. The incentive to overplead will be reduced in a coverage case. Litigation costs will likely be significantly reduced in the long run. as there will no longer be a concern that characterization of one specific concurrent cause as “the” covered concurrent cause will solely dictate coverage.e. and away from the muddying effect of importing tort principles. The removal of the complete avoidance of coverage as incentive for litigation behaviour may alter the dynamic enough to promote settlement of these cases. which are designed to operate on fault-based principles. as best as the proportional inquiry can achieve.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 52 _______ LAW REVIEW [Vol. little incentive for an insurer to take a chance in court at escaping indemnifying the insured altogether under a proportional recovery scheme for concurrent causation.Concurrent Causation: Examination of Alternative Approaches (1985) S. The apportionment question should focus on the endresult dollar value of the damage caused by each of the concurrent causes. U. There is. a dominant cause approach. though this would likely not increase in incidence than exists already under.91 This avoids complicating analogies to tort law’s comparative negligence concepts. If the insurer covered one of the concurrent causes. 527 at 533 (suggesting that causation analysis should examine the sufficiency of damage as determinative. Keeping the inquiry to damage costs only on a per-cause basis forces courts and litigants away from the temptation to assess apportionment based on some causal “responsibility” other than indemnity alone. in the hopes of increasing the available aggregate proportional recovery among various policies. There may also be a corresponding need for additional coverage counsel in loss distribution disputes. L. the insurer will be paying something. Ill. rather than have the dispute be about coverage or no coverage at all. Perhaps then counsel can more easily agree on the proportion. say. The incentive to litigate is reduced because each party will know ex ante that the policy See i.J. however. That keeps the case grounded in insurance law contract principles. An ideal proportional assessment of concurrent causation is really then ideally based on the proportion of the risk underwritten by the insurer. Richard Fierce. Each party gets what each party agreed to. in addition to causal interaction).

to produce the loss. simultaneously. as wide as tort law.e. An apportionment default rule for serial causation cases therefore promotes the greatest efficiency in information. The loss would not have resulted without both causes acting together. 92 .DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 53 must pay something. 2d ed. Rev. The court system’s initial systemic investment in determining proportional percentages on a case by case basis will likely soon be made up in the precedential value of various fact patterns in cases that can assist litigants in charting their insurance litigation. other than by limiting the dollar value of the indemnity. The question then just becomes “how much. Liability insurance offers a remarkably wide indemnity coverage. Proximate Causation in Insurance Law.93 Banks McDowell notes that the liability insurer largely cannot control its indemnity responsibility within the insurance contract itself. complexity. (2005) at 233-234 (advocating different treatment of concurrent causation in property insurance than liability insurance – split the causes equally in liability insurance cases). Recall that parallel causation cases are most often cases involving third party liability insurance. Causes act concurrently in concert. John Lowry and Philip Rawlings. so parallel causation losses should be covered as long as one cause is covered). It is certainly a cheaper question to answer on a systemic basis than the tortured dominant cause litigation that proliferates. See also John Lowry and Philip Rawlings. B. 93 See i.” That is likely an easier question to settle upon than an all-or-nothing coverage question. This approach avoids the inefficient and unpredictable “all or nothing” methodology of both the conservative and dominant cause approaches. A Liberal Default Rule for Parallel Causation Cases The liberal approach to concurrent causation is the most effective legal rule for solving parallel causation cases. administrative. INSURANCE LAW: DOCTRINES AND PRINCIPLES. supra at 591.92 Exclusions are difficult to draft because it is challenging to predict the myriad permutations of negligence that may come together to produce a loss. (2005) 68 Modern L. 310 at 317-318 (liability insurers offer wider indemnity than property insurers. and consistency costs. See McDowell. not unlike what occurred when jurisdictions moved to comparative fault regimes in tort.

the liberal approach allows insureds and insurers to plan their insurance affairs. Unlike the apportionment rule. but such is done only after the coverage question is efficiently answered in the affirmative. as noted in the previous section. Those problematic leanings toward tort concepts of causation are completely avoided by this default rule. L. the answer to which then controls available coverage.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 54 _______ LAW REVIEW [Vol. The Role of Insurance Considerations in the Choice of Efficient Civil Liability Rules (1988) 4 J. and Org. There is no forced and artificial determination of a single dominant cause to an accident. Michael Trebilcock notes that third party liability insurance questions are consistently more expensive questions to answer than first party property insurance questions. Trebilcock. It has the greatest savings in complexity costs among the other approaches for parallel causation cases. and consistency costs unlike any of the other approaches. predictable default rule that saves in information. It is inefficient to use an apportionment approach in parallel causation cases because apportionment is nearly always impossible or completely unpredictable and inefficient to discover. administrative. the liberal default rule for parallel causation also reduces or destroys the incentive to litigate about concurrent causation. Econ. There is nothing stopping an insurer in a loss distribution dispute from seeking contribution from the additional insurers on a ratable basis. There is no unpredictability and no temptation to mix tort “responsibility” with a dominant cause’s “responsibility” for the loss. Like the apportionment default rule above for serial causation. The liberal default rule creates no dangerous gaps in coverage. Insurers and insureds will know that a loss caused by parallel causes will be covered. the question does not even then get to “how much?” It is a much simpler analysis: there is indemnity from the policy. This sidesteps all the litigation costs of dominant cause cases. This is an extremely important feature because liability insurance is a direct source of Michael J.94 By removing the assessment of responsibility for multiple causes. The liberal approach avoids the inefficiency of a dominant cause analysis. 94 . 243 at 253. Both know there will be coverage for a concurrently caused loss caused by parallel causation. __:_ The liberal approach is a simple.

The insurer underwrote part of a risk and so therefore must indemnify. the insured wins. The certainty in that. Insurers at least know where they stand under a liberal default rule. The strong policy reasons for compensation for the injured which. If an insured is denied insurance coverage under the conservative or dominant cause approach. It also completely avoids the problem of insurance brokers becoming implicated in insurance gap cases. Better yet. There will no longer be an incentive to tag more than one policy. supra. but by the injured third party victim seeking to recover in tort from the insured. Overpleading will likely cease in loss distribution cases in which various policies overlap and cancel each other out with coverage and exclusion clauses. There is some indirect recouping of the loss to insurers and the corresponding violence done to exclusionary language in instances where insurers are called upon to indemnify when a covered cause combines with an excluded cause. while prompting greater incidences of payouts by insurers. The insured’s expectation of coverage is acknowledged by the fact that the insurer has to pay for a covered cause. In a contest among concurrent causes. lead nearly all jurisdictions to adopt some form of state-run mandatory automobile insurance. 95 . Actuarial evidence of concurrently caused losses under a liberal default rule can actually help insurers better predict and underwrite such risks.95 With the real risk that a court may not hold the clause enforceable.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 55 compensation for injured third parties. and then charge corresponding premiums. See Part III. save and except for cases where insurance coverage limits are at issue. results in an overall cost savings for the insurers who are repeat players in the insurance system. This is recouped in system-wide dispute resolution savings. A liberal default rule for parallel concurrent causation cases produces the most fair result for insured and insurer. incidentally. This will also reduce the necessity for multiple coverage counsel if only the policy truly at risk of being called upon to pay is involved in the dispute. perhaps rather than attempting to contract out of concurrent causation. point to the adoption of a liberal rule of concurrent causation for parallel causation in order to maintain the efficacy of the insurance system as being a rough and ready compensation system for the injured. that loss is eventually usually borne not by the insured.

While admittedly some insurers offer discounts for purchasing both policies with the same company. Kenneth Abraham has warned against the fragmenting of modern insurance coverage as creating potential insurance gaps and less than optimal insurance in Kenneth S. predictable default rule could bring about. adopting these two default rules achieves more than just moving money previously spent on insurer litigation to money spent instead on insurer payouts to insureds. insurers should have enough predictive information in Banks McDowell suggests insurers sell combination liability policies. In fact. McDowell’s thinking underlines the incentives for market creativity in insurance that a properly crafted. The move to the dual default rules increases the ability of insurers to reliably predict loss payouts and set premium rates accordingly. 85 at 105-107. and the possibility of litigation about a concurrently caused loss. particularly for parallel causation cases involving tort. See McDowell. This often occurs when the negligence of multiple. Adopting the two default rules increases an insurer’s ability to rate payout experiences for the loss-causing factual matrices that comprise concurrent causation losses. L. The move also drastically reduces spillover effects into other insurance markets. Indeed. 96 .96 C. Abraham. to avoid loss distribution disputes among competing insurers. Relatively quickly. such as a homeowners-automobile insurance policy. at present.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 56 _______ LAW REVIEW [Vol. insurers should be able to accurately develop actuarial information about just how often insurers are required to make payouts for parallel concurrently caused losses. separately insured parties combines to produce a concurrently caused loss. The Rise and Fall of Commercial Liability Insurance (2001) 87 Va. __:_ an insurer could invent a concurrent causation endorsement and charge an additional premium based on the reasonable increase in possibility that such indemnity provisions would be triggered. based on better experience rating for two crucial events: the loss-causing factual matrices which demand payout. Systemic Efficiencies On a systemic level. Rev. supra at 589. McDowell’s suggestion would go a long way to avoiding loss distribution disputes among liability insurers in the event two potentially overlapping policies are not underwritten by the same insurer.

as will be mentioned below. it is such an aggregate improvement from the unpredictable dominant cause approach that savings garnered with serially caused losses could be used to initially fund whatever perceived short-term losses insurers initially face in the move to a liberal approach to parallel causation. The inefficiencies of an all-or-nothing regime are completely avoided. how can that be a savings when the rate of payout probably increases? The answer lies in the loss distribution mechanism inherent in insurance itself: the power of more certain aggregate loss information. is a tremendous savings and aids in an actuarially predictable premium-payout ratio. through avoidance of insurance market spillovers. For serially caused losses. That second type of fortuity has often become the cause of much inefficiency that was never meant to be. More desirably. insurers have to pay for only the losses rationally linked to premiums charged for expected enumerated experiences. Moving to the twin default rules also allows insurers to predict the frequency of litigation about concurrently caused . say. At first blush. insurers will benefit from the move from an all-or-nothing payout regime to one of proportional payouts directly related to the risk underwritten in the first place. Premiums can be set to properly reflect the risk underwritten. not the fortuity of having to indemnify the insured. insurers will be better able to predict the percentage payouts resulting from serially caused losses in various scenarios. In short. and by enjoying drastically decreased litigation expenses.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 57 hand about how often they would be required to pay if a jurisdiction moved from. insurers should instead find the savings elsewhere through a new regime for serially caused losses. That. a dominant cause approach to a liberal approach for parallel causation losses. For serially caused losses under an apportionment regime. an insurer can then reliably increase premiums to what a more certain market would bear. The incidence of payouts and denials on all parallel concurrently caused losses would. the default rule for parallel causation takes some of the uncertainty out of what is often a fortuitous question. Once insurers have a greater certainty of experience for payouts. In fact. Over time. even if the payouts are aggregately higher. under a liberal approach to concurrent causation. switch to payouts only. The fortuity that is left is rightly only in the event causing the loss. in itself.

The litigation costs spent in disputing coverage and loss distribution can instead be spent on payouts which are certain. if anything other than a liberal rule for concurrent causation is adopted. Alternatively. But there is real potential in parallel causation cases to have spillovers into other insurance markets as well. The potential effect of gaps in coverage on insurance brokers’ liability insurance have already been discussed. This is a much less resource-intensive litigation fight to have. most cases should quickly arrive at some relative scope about which a sensible settlement discussion is the more rational alternative to unpredictable and costly litigation. insurers should be able to actuarially predict parallel causation payout incidences and have competitive premiums based on actual experience ratings under the new regime. With the litigation subject moving away from allor-nothing coverage to instead only the proportion of coverage attributable to the insurer. __:_ losses. The dispute. take the corresponding savings in serial causation causes to fund whatever initial cost is required for liability insurance cases. insureds and insurers face greater incentives to settle. insurers can raise premiums based on rational payout incidence or.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 58 _______ LAW REVIEW [Vol. If. the dollar figure of the damage attributable to that rain alone is likely quantifiable into some reasonable range. Inefficient spillover effects into other insurance markets are drastically reduced. rain damages a home in a hurricane. Litigation about parallel causation cases should be largely eliminated. even if expert assistance is required to set the reasonable ranges. for example. instead. Insurers are incentivized to settle because liability to indemnify is no longer questionable merely because the loss is concurrently caused. . Insurers are free to use the savings from eliminated litigation to fund the payout increases. then. Finally. will be about the bounds of that range and not about whether or not the loss is covered at all. an insurance world which adopts the two default rules for parallel and concurrent causation is one which allocates losses according to the parties who undertook the risks ensured. The dispute is about “how much.” not about full coverage or absence of coverage altogether. The insurer pays in a concurrently caused loss. Once liability insurance payout patterns become discernible. if litigation ensues at all. Litigation about serial causation cases will be limited to only a debate about the percentage contribution of a certain insured loss. While there is certainly a lot of room to dispute within that range of possible proportions.

or instead to societal benefits such as social security or the public welfare social safety net. in this unpredictable system. can rely on a predictable source of compensation and not be occasionally unpredictably forced to seek first party benefits or state welfare. losses are not borne by the parties who charge a premium to underwrite a certain risk. If a third party accident victim is injured in a jurisdiction which has an insurance loss distribution system with an unpredictable legal rule for concurrent causation. In short. Insureds can then also rely on insurance coverage when those insureds are called upon to pay for losses stemming from their tortious liability. unpredictable. who are not parties to these insurance contracts at all. And dangerous gaps in insurance . the injured third party accident victim may often be denied compensation because the underlying tortfeasor is denied liability insurance coverage in a concurrent causation scenario. Insurers can then rely on the certainty of the system to set rates that the market can bear. then those parties profit while the resultant loss – the compensation sought by the accident victim – falls to be redressed elsewhere in the system. The spillover is avoided. the one who agreed to bear the loss (the insurer) is now not predictably bearing the loss under similar repeat circumstances. and incredibly inefficient to resolve. Creative lawyers attempt to seek indemnity for their clients by implicating various policies. There will be no insurance gaps. VII.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 59 Recall that liability insurance is the rough and ready backbone of the compensatory aspect of the tort system.Conclusion Concurrent causation disputes are expensive. like the all-or-nothing dominant cause approach. as insureds grasp for coverage in the wake of large-scale disasters and rising costs of accidents. That “elsewhere” is often either to other insurance sources.” A liberal approach to concurrent causation in parallel causation causes at least ensures that the insurer who took the premium for some of the loss bears the burden of payment. such as the injured victim’s first party insurance benefits like long term disability insurance or first party health insurance. And accident victims. If. because of the fortuity of a concurrent uninsured event being deemed the “dominant cause. They are also becoming increasingly prevalent.

531 at 555-556 and Daniel Schwarcz. __:_ coverage expose not only an insured’s assets but remove a source of compensation for injured third party victims. 1389 (describing reasons why insurance policies are inherently inefficient and why insurers maintain such inefficiencies). applying an apportionment default rule to solve the concurrent causation dispute leads to greater cost savings for insureds. and the justice system. If a loss is caused by a covered cause. L. Insurers at least have the bank of court precedents to assist in predicting how policy provisions will be interpreted. This stops the inefficient all-ornothing approach to coverage and assists with loss sharing in loss distribution disputes. and Mary L. the loss is covered. there are large systemic incentives for insurers to persist with current insurance policy language and products. A Products Liability Theory for the Judicial Regulation of Insurance Policies (2007) 48 Wm. This approach is deceptively simple to apply. By creating two separate default rules for two different kinds of concurrent causation scenarios. They are simple to apply in coverage and loss distribution disputes. Finally. even though an excluded or noncovered cause may have combined together to produce the loss. Rev. It also creates market opportunities for insurers to produce an insurance product which actually insures for concurrently caused losses. See Kenneth S. applying a liberal default rule also is the most efficient response.97 Both default rules provide serious disincentives to litigation. even if it is inefficient in the short run to do so. Whatever increase in liability for insurers is created in its application is more than made up in the increased predictability savings on a system-wide level. 97 . This approach also preserves the unique compensation aspect of liability insurance that is so important to injured third parties. courts in common law jurisdictions can solve many of the inefficiencies that currently plague these types of cases. Proportional responsibility for the loss is aggregated on a per-cause basis.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 60 _______ LAW REVIEW [Vol. Abraham. A Theory of Insurance Policy Interpretation (1996) 95 Mich. insurers and insureds can reasonably predict the coverage outcomes of a Of course. For those liability insurance disputes where the loss is caused by parallel causation. which is a significant cost-savings. For those property insurance cases where the loss is caused by serial causation. insurers. Rev.

a worthy cause.DRAFT – DEFAULT RULES FOR CONCURRENT CAUSATION IN INSURANCE 2008] Default Rules for Concurrent Causation 61 dispute. . in itself. Adopting these two default rules in common law jurisdictions for that single reason is.

Sign up to vote on this title
UsefulNot useful