Introduction
In the summer of 1996, Alexandra was molested repeatedly at the hands of a local priest at and around her family’s church in Queens, New York.1 This horrific alleged abuse continued for nearly six years, including in her own home and in the priest’s car.2 While Alexandra was eventually awarded a two-million-dollar settlement, the Roman Catholic Diocese turned around and sued one of its insurers, National Union, for failing to reimburse the Diocese for the settlement.3 The Diocese contended that all of the alleged abuse constituted only one “occurrence,” which would maximize its coverage, while National Union argued that the alleged abuse actually constituted multiple occurrences.4 This was the first time New York addressed the definition of “occurrence” in the context of claims of multiple instances of sexual abuse of a minor by a priest spanning several policy periods.5
In so doing, the court chose to adopt the “unfortunate event” test, which required it to identify whether the cause and result of an action or event were linked in a close temporal or spatial relationship, without intervening factors.6 Ultimately, the court held in favor of National Union, concluding that the several instances of sexual abuse constituted multiple occurrences.7 National Union was thus only required to provide coverage during the policy periods where it was the primary insurer, a number much lower than what it would have had to cover had the court held that there was only one occurrence.8 Confusingly, in a variety of insurance cases that have since followed National Union, New York courts have applied the “unfortunate event” test inconsistently.9 With the advent of new legislation designed to give victims of child sexual abuse more time to file claims, this inconsistency is especially concerning.
Allegations of child sexual abuse have unfortunately affected so many of the institutions that are bastions of community life in their respective neighborhoods and cities across the country.10 From schools to religious organizations,11 Boy Scout troops to hospitals, seemingly nowhere remains unaffected.12
This will no doubt reveal abuses that we’ve never heard of before. It’s not just the church. It’s not just some Jewish institutions. It’s not just the Boy Scouts. . . . Although nothing should surprise us because this kind of child sexual abuse has been so endemic to our society.
Julie Zauzmer & Sarah Pulliam Bailey, New York Braces for a Flood of Lawsuits, as One-Year Window Opens for Child Sexual Abuse Victims to Bring Cases, Wash. Post (Aug. 14, 2019, 12:26 PM), https://www.washingtonpost.com/religion/2019/08/14/new-york-braces-flood-lawsuits-one-year-window-opens-child-sex-abuse-victims-bring-cases [https://perma.cc/KEQ2-662W]. In New York, after a long struggle that spanned more than a decade to increase protections for victims of child sexual abuse, proponents of reforming the statute of limitations governing child sexual abuse claims were finally rewarded with the passage of the Child Victims Act (CVA).13
Through the passage of the CVA, which was officially signed into law by Governor Andrew Cuomo on February 14, 2019,14 New York follows a growing number of states that have passed similar legislation to amend and reform their statute of limitations with regards to child sexual abuse claims.15 The passing of this legislation, which originally floundered and failed for thirteen years in the then Republican-controlled New York Senate,16 marks a historic moment for child sexual abuse survivors, their supporters, and activists. For civil claims of child sexual abuse, the CVA extends the statute of limitations for individuals to file litigation from twenty-three years of age to fifty-five years of age.17 Additionally, and perhaps most importantly, the CVA provides a one-year look-back window wherein previously time-barred civil claims alleging child sexual abuse can now be litigated with no statute of limitations restrictions.18 The look-back window, which opened on August 14, 2019, and was slated to end on the same date in 2020, has already ushered in a “tidal wave” of litigation.19 As of October 31, 2019, 975 CVA cases had been filed since the look-back window opened.20 Marci Hamilton, founder and CEO of the Philadelphia nonprofit Child USA that advocates to end child abuse, stated that by the end of the look-back period, there may be as many as two thousand to three thousand cases filed.21 The actual number of cases filed is likely to be even higher than previous estimates, given that Governor Cuomo announced on May 8, 2020 that due to the ongoing COVID-19 pandemic the look-back window would be extended until January 14, 2021.22 Shortly thereafter, on May 27, 2020, the New York State Senate and Assembly voted to extend the look-back window to August 14, 2021.23
While New York’s CVA is long overdue for some, not everyone is lauding the bill passage, and the subsequent rush of litigation, as a win. For many of the large institutions around the state that have served children in some capacity, this new legislation means bracing for a wave of complicated litigation as “revived CPLR 214-g”24 lawsuits continue to pour in, mostly alleging some kind of negligence on the part of institutions.25 Likewise, insurance companies who insure these institutions are scrambling to make certain that they are not saddled with the full liability of their policyholders. Insurance companies are also padding their claims reserves in anticipation of the coming onslaught of litigation.26
This Note advances a critique to challenge New York’s current “unfortunate event” test, which has been litigated most commonly in the asbestos and construction context,27 and suggests that the test is both unwieldy in practice and largely inconsistently applied.28 With the CVA already producing a “tidal wave” of litigation29 that will continue for many years after the look-back window closes, New York needs an alternate test for all future insurance disputes that simplifies the question currently faced by courts in determining insurance liability and coverage: whether there have been one or multiple “occurrences” when the policy language is unclear.
Because of the difficulty of determining how close in time and space a series of incidents has to fall to be considered along the same “causal continuum” when utilizing the “unfortunate event” test,30 and the courts’ unwillingness to draw a bright line rule with regard to that determination,31 holdings have become inconsistent and broadly results-oriented, leaving insurance companies uncertain about the type of policy language necessary to demonstrate an intent to aggregate incidents.32 Instead of the cumbersome and confusing test currently applied, this Note suggests the adoption of an alternative test currently utilized in Minnesota.
With the advent of the CVA, instead of utilizing the “unfortunate event” test, New York’s judiciary should seek to utilize Minnesota’s “actual-injury” test.33 Minnesota’s “actual-injury” test provides a much more practical method where each instance of abuse determines which insurance policies will be triggered, and where the number of people injured determines the number of occurrences.34 This approach to counting occurrences greatly simplifies the analysis, providing a clear direction for both insurers and policyholders. Minnesota’s test also strikes a happy medium between ensuring alleged victims get proper compensation, while taking into account the difficulty for insurers and institutions to shoulder the entire financial burden of a perpetrator’s actions.35 By spreading out the liability more equitably between insurers and the institutions they cover, this would also work to protect all New York taxpayers from a drastic increase in school taxes and various insurance premiums, since it is ultimately the average taxpayer that ends up paying the price for an institutional employee’s misdeeds.36
Furthermore, this Note argues that a continued benefit of the “actual-injury” test is the ability for insurers to share liability when multiple policy periods are implicated for the same occurrences, ensuring no one insurer remains solely liable.37
This Note proceeds in three parts. Part I of this Note begins with a background on respondeat superior and other common law tort theories that allow institutions to be held responsible for the conduct of their agents, then proceeds with the effects of potential insurer liability on the general public and a detailed analysis of how the CVA could affect insurance claims broadly.38 Next, Part I attempts to explain the age-old problem of the one and the many in Pre-Socratic Philosophy39 and its emergence in other areas of law, analyzing this problem specifically in the context of insurance claims relating to the 2001 World Trade Center Attacks. Finally, Part I concludes with a summary of the “occurrence” problem as it relates to Commercial General Liability (CGL) policies, briefly explaining the three main tests used to determine occurrences in CGL policies.
Part II begins with an analysis of the problems with applying New York’s current “unfortunate event” test, both in general, and in light of the recent passage of the CVA. Part II also briefly introduces Minnesota’s “actual-injury” test. Finally, Part III outlines why Minnesota’s test is one way to ensure more judicial certainty for policyholders and insurers alike, creating a way for alleged victims to have their day in court without financially overburdening the institutions and insurers whose livelihoods are crucial to community functioning.
I. Background
Very few of the revived CPLR 214-g lawsuits have been filed against individuals or alleged perpetrators.40 Instead, the vast majority of these lawsuits have been filed against institutions and civic organizations.41 This poses a potentially grave problem to insurers of those institutions. Because of the huge influx of CVA claims against religious organizations, school districts, and other large institutions that insurance companies will likely have to defend,42 insurance rates could increase in a variety of different areas, with far-reaching consequences for the entire public.
A. Respondeat Superior, Negligence, and Underlying Tort Theories
To really get to the heart of why the current rule for determining “occurrences” employed in New York is insufficient to deal with the onslaught of insurance claims that have already lead to a mountain of litigation, it is crucial to understand the underlying tort theories that allow institutions to be held responsible for the conduct of their employees and volunteers, as well as the theories that require an insurance company to defend or indemnify.
Respondeat superior, a term that means “let the superior make answer” in Latin, is a doctrine that holds an employer liable for their employee or agent’s actions committed during the scope of their employment.43 At first glance, this doctrine would seem the obvious choice for use by those filing CVA claims against institutions in order to make an institution liable for the alleged perpetrator’s actions.44 However, respondeat superior does not generally apply to sexual abuse claims.45 Under this doctrine, employers cannot be held vicariously liable for torts committed by an employee or agent when the wrongful acts are committed solely based on personal motives unrelated to the “furtherance of the employer’s business.”46 In general, an act of sexual abuse or assault by an employee or agent is considered a “clear departure” from their scope of employment and committed entirely based on personal motives, and thus, cannot be said to be related to the furtherance of an employer’s business.47
Instead, many plaintiffs who file child sexual abuse claims utilize a negligence framework, since institutions can be held liable for negligent supervision, negligent hiring, or simply common law negligence.48 Additionally, negligence is the most common underlying tort theory utilized in these types of claims, since New York is a pure comparative negligence state.49 As a comparative negligence state, New York adopts the view that the claimant’s negligence, no matter how great, will not bar recovery, though damages will be reduced in proportion to negligence.50 This is beneficial to claimants, because in order for defendants to avoid paying damages, they must prove by a preponderance of the evidence51 that they were not negligent at all.52
Under a negligence schema, alleged abusers are rarely afforded coverage by insurance companies because of the intentional nature of the acts.53 Insurance policies generally do not provide coverage for intentional acts, because that violates the central idea of insurance law: fortuity.54 The whole point of insurance is that it protects people from an accident: something harmful that is unforeseen by the insured.55 Likewise, in tort law, negligence can only be argued if the act was not intentional.56 Coverage for intentional acts thus cuts against the very heart of both insurance law and tort law’s negligence schema.57 In contrast, institutions are much more likely to be afforded coverage under this schema, since intentionality is usually lacking; most institutions do not ask the perpetrators to commit intentional acts of physical harm as part of their job description.58
On the insurer’s side, when an allegation within the four corners of a complaint by an alleged victim even potentially gives rise to a covered claim, or where an insurer has knowledge of facts establishing the reasonable possibility of coverage, an insurer has a “duty to defend” its policyholder.59 A “duty to defend” is the insurer’s obligation to provide a defense to their policyholder from claims made under a liability insurance policy.60 While CGL policies contain duty-to-defend clauses that are triggered when there are any third-party suits against the policyholder seeking damages because of bodily injury or personal injury,61 it is unclear whether or not CVA claims fall within this scope of protection, especially regarding older policies without sexual molestation and abuse exclusions.62 In turn, the lack of clarity creates an additional burden for insurance companies, which could end up liable for thousands in defense costs for even potentially covered claims.63
The only way an insurer may be relieved of this duty to defend is if it can show that the obligation to defend is outside of the scope of the insurance policy.64 In order to be relieved of the duty to defend, an insurer bears the burden of showing that the allegations detailed in the complaint lie entirely within the policy exclusion implicated, and that it is the only way to interpret the exclusion, making it impossible to argue any basis for which the insurer might eventually be required to indemnify its insured.65 Of course, as discussed earlier, the majority of older insurance policies do not contain the sexual abuse or sexual molestation exclusions that post-1986 CGL policies contain, making this another issue that courts are facing with the onslaught of revived CPLR 214-g claims.66
In contrast to the duty to defend, the “duty to indemnify” is much narrower,67 only applying if the injury is actually covered by the insurance policy.68 This indemnification duty is the obligation of the insurer to compensate the policyholder for damages if liability is established against them in litigation.69 In relation to CVA claims, when a policyholder knew or should have known that the employee’s sexual misconduct was substantially probable, many courts have found that the insurer has no resulting indemnity obligation.70 However, what is important to note in the scope of this duty is that when an insurer does in fact have a duty to indemnify its policyholder for damages, these damages must be because of bodily injury or property damage that is caused by an “occurrence.”71 What makes this especially challenging though, is that the definition of “occurrence” in New York is nebulous and unclear, and the impact of whether multiple instances of child sexual abuse across multiple policy periods are one “occurrence” or many has been the subject of much litigation over the past several years that has led to inconsistent outcomes.72
B. Why Insurer Liability Matters
Unlike the handful of states that have also passed similar laws suspending the statute of limitations for child sexual abuse claims for a certain window of time, New York’s CVA allows lawsuits against both public and private institutions.73 This means an increase in institutions that are likely to be sued. And while insurance claims and coverage for large institutions may seem at first glance to have little to do with the general public’s bottom line, if insurance rates increase, people’s taxes are likely to increase as well.74
One area in particular that remains particularly vulnerable is New York’s nearly seven hundred school districts.75 Because of the close proximity teachers, administrators, and staff have to children every day, public (and private) schools are already facing an onslaught of revived CPLR 214-g lawsuits. While school districts in New York currently have plenty of coverage,76 insurance companies are warning that the CVA is likely to increase school district insurance rates.77 This anticipated increase could then result in a subsequent increase in individual family school taxes as well,78 forcing New York’s taxpayers to saddle the cost of increasing rates caused by allegations of abuse.79
School districts are not the only institutions facing allegations of child sexual abuse that may result in raised taxes for the public.80 The CVA also impacts public hospitals, city and state agencies, and even some public universities, and the full impact to the public of which is yet to be determined.81
Besides the potential tax implications the CVA could pose to the public, the financial burden that litigation will put on many of the institutions that are implicated will be immense.82 This, in turn, could affect not only the compensation of alleged victims who come forward later,83 but also the services these institutions offer the public.84 With mounting jury awards, institutions that have to file for bankruptcy may have to limit services to the public or increase their service costs.85 Additionally, institutions that file for bankruptcy, and which have to rearrange to restructure their debts, may have to sell properties or lay off workers.86 This restructuring could affect both the jobs and the services that institutions are then able to provide to the surrounding community.87
The CVA is also likely to have a significant impact on the cost and availability of insurance coverage for all kinds of organizations, including religious organizations and nonprofits, who remain particularly vulnerable.88 This increase in cost ultimately affects the people that these organizations serve. Because of all these concerns, finding a way to quickly and clearly resolve litigation between insurance companies and institutions is crucial for all parties involved.
C. The CVA’s Impact on Insurance Claims
Looking beyond the potential impact that the CVA could have on the general public, the new legislation’s impact on the insurance industry, and insurance claims in particular, is astronomical, and not just because many organizations have misplaced their original policies for the periods in question.89
First, the passage of the CVA has already brought about a “tidal wave” of litigation, as people rush to file claims before the one-year look-back window closes.90 And as the last few months have already demonstrated, the huge influx in litigation is only continuing to grow. 91 With the CVA extending the statute of limitations to fifty-five years of age to file a civil claim,92 litigation for child sexual abuse claims is bound to continue to increase, even long after the end of the extended look-back window.93
For insurance claims, the increase in litigation also means an increase in the amount of previously time-barred claims, which introduces a huge amount of uncertainty.94 Unlike the previous statute of limitations for child sexual abuse claims,95 the claims being brought as a result of the new legislation are predominantly being brought by much older plaintiffs.96 Not only are the plaintiffs who are bringing the claims much older, their alleged perpetrators––who were adults when the abuse took place––are often no longer alive.97 As a result, plaintiffs filing revived CPLR 214-g claims98 are more likely to sue the institutions who employed their alleged perpetrators, increasing their likelihood of obtaining big individual settlements, at least for those who are able to settle early.99 Additionally, because so many of the claims are being brought against the same organizations (i.e. Boy Scouts of America,100 Roman Catholic Diocese,101 and Rockefeller Hospital,102 to name a few), hundreds of millions of dollars are at stake.103 The huge amount of money involved, in combination with the potentially dire financial situation faced by the institutions on the other end of these lawsuits, ensures that organizations will be seeking coverage from their insurers.104
Second, litigation is likely to be complicated with regard to these insurance claims, since older insurance policies from the 1950s, 1960s, and 1970s usually do not provide exclusions for sexual abuse.105 These exclusions are typically written provisions within the policy that outline specific hazards, perils, circumstances, or property not covered by the policy.106 Most relevant policies of the time were written without sexual abuse exclusions, perhaps because those who wrote the policies simply did not contemplate a large number of child sexual abuse claims.107 It was not until the 1980s that the majority of general liability policies began to contain sexual misconduct exclusions.108 Current insurance policies now also usually contain language that excludes any claims resulting from an intentional act.109 However, since many of the claims being brought under the CVA were previously time-barred under the old statute of limitations, many of the insurance policies implicated in these cases are not current ones.110
In addition to the difficulty of ascertaining the parties’ intents, since the majority of these old policies do not contain sexual misconduct exclusions, a major issue many organizations face is how to deal with misplaced or damaged original insurance documents for the policy periods in question.111 In some cases, it has been nearly seventy years since the alleged abuse happened, which requires looking back to insurance policies from as far back as the 1950s.112
Because of how the CVA impacts insurance claims, there needs to be a rule that is both practical and efficient in this context to deal with the large volume of pending litigation and ensure that victims receive compensation. The rule must further attempt to ensure that the financial burden rests equally on the institutions and their insurers, without institutions passing off all of the burden onto insurers and without the insurers wholly refusing coverage.
D. The Problem of the One and the Many
The question of whether an “occurrence” for the purposes of insurance liability is one or multiple events is not the first time that the problem of the one and the many arises, nor is it its only iteration in the law. In fact, dating as far back as the fifth and sixth centuries BCE, pre-Socratic philosophers struggled with this age-old problem in their musings on the nature of the universe.113 Whether the world was composed of many things or whether there was a unifying force behind everything, this question of the one and the many permeated these philosophers’ writings, arguments, and politics as they attempted to make sense of the world around them.114 Some pre-Socratic philosophers like Parmenides sought out a single source or explanation of all that exists, in contrast to pluralist pre-Socratic philosophers like Anaxagoras and Empedocles, who took the view that many kinds of things exist.115 Anaxagoras and Empedocles believed that there were several basic forces or substances that all else could be reduced to.116
In the legal context, this problem of the one and the many arises in a host of different areas as well, including in criminal law. For example, if a person punches another person several times, do they commit only one crime or several?117 Temporally and spatially, does it matter how much time passes between the acts, or if they took place in the same location or in different places?118 The problem of “multiple punishment” is a continually recurring one that has stumped everyone from legal philosophers to Supreme Court Justices.119 Similarly to insurance law, criminal law is widely unclear in this area, forcing prosecutors and defense attorneys to determine “on a case-by-case basis” how they will respectively charge and defend a series of closely-related actions.120 How prosecutors decide to charge these closely-related actions then affects every aspect of a criminal prosecution, from determining whether there is heightened pressure for the defendant to plead guilty to how lengthy a sentence will be.121 Much like determining “occurrences” in insurance law, there are multiple approaches and tests that courts have employed to deal with the issue, with little consensus across the board as to the best approach to be utilized.122 And similar to New York’s approach to determining occurrences, a critique of the various methods utilized to deal with closely-related actions in criminal law is the ambiguity and inconsistency that occurs when results-oriented people apply different standards.123
Within insurance law itself, the problem of the one and the many, specifically as it relates to determining the number of “occurrences,” arises in a variety of situations from automobile accidents124 to employment discrimination.125 Perhaps most famously, the issue reared its head in the aftermath of the September 11 terrorist attacks, as the lessees of the buildings and insurers tried to figure out who was obligated to cover the damage.126 The financial ability to rebuild after the World Trade Center attacks depended on whether the attacks on, and subsequent destruction of, the Twin Towers were one occurrence or two.127 The problem was a particularly complicated one, and not just because it was set against a backdrop of heightened tensions as the whole nation watched how the litigation would unfold.128 Because there were two hijacked planes and two buildings that were destroyed in one complex during one attack, the question of whether the destruction of the Towers was one occurrence or two puzzled the courts.129 Additionally, while there were several insurers who had bound themselves to policy forms that had a clear definition of occurrence that indicated that all losses stemming from “one cause or one series of similar causes” were to be considered one occurrence, there were also a few insurers who had forms that either failed to define “occurrence” or incorporated a less clear definition.130 Ultimately, the jury found that the majority of insurers of the property were bound to a form that treated the attacks as one occurrence, while a minority of insurers were bound to a form that treated it as two occurrences.131 The result meant that from some of the insurers, the lessees were entitled to only the three-and-a-half billion dollars that the complex was insured under for “any one occurrence,” while from others, they were entitled to a two-occurrence outcome, a whopping seven billion dollars.132 And while many thought that the World Trade Center problem was the occurrence problem to end all occurrence problems, coverage disputes continue to arise in a variety of different areas because of the difficulty of anticipating the impact policy language will have in the face of unimaginable incidents and losses.133
E. The Occurrence Problem under CGL Policies
Much like the World Trade Center insurance question, determining occurrences when the injury is multiple incidents of child sexual abuse occurring over multiple policy periods is particularly complex, and it inevitably erupts in a series of coverage disputes. Because of the rapidly increasing amounts of revived CPLR 214-g claims being filed before the look-back window closes, New York needs a more efficient way than its current method to deal with claims that will signal to insurers and insureds alike what to expect from coming litigation.
So why is the definition of “occurrence” so vague in CGL policies, and why is this so important in the context of the CVA? In most standard CGL policies from the pre-1986 era, “occurrence” is defined as “an accident, including continuous or repeated exposure to conditions, which results, during the policy period, in bodily injury or property damage neither expected nor intended from the standpoint of the insured.”134 Immediately, the lack of clarity in this definition is evident. One reason for that lack of clarity might be the fact that most relevant policies of the period were written without taking the possibility of child sexual abuse into account, a major reason many of these policies failed to have exclusions for sexual abuse and molestation in the first place.135
As a result of the general lack of awareness of many instances of child sexual abuse, the language in standard CGL policies from the period is immensely unclear in relation to sexual abuse claims and the potential negligence of the institutions that employed alleged perpetrators, prompting much debate about what insurance will cover.136
Additionally, not only do these early policies lack exclusions for sexual abuse, they usually have lower per-occurrence limits than do current CGL policies, in part due to inflation.137 Thus, determining the number of occurrences directly impacts the number of limits available under a policy, especially since pre-1980s policies generally do not contain aggregate limits.138 Aggregate limits, which refer to the maximum amount that an insurer will pay per policy period,139 were implemented into CGL policies starting in the 1980s, the insurance industry realized how much liability insurers were subjecting themselves to as policyholder attorneys continued to develop new and creative methods to magnify coverage, far beyond what was initially anticipated when the policy was issued.140
Because the CVA implicates many pre-1980s CGL policies, the lack of aggregate limits could prove disastrous for insurance companies. Depending on how the number of occurrences is determined, it could result in an extensive amount of insurance coverage for organizations.141 Insurers could be required to pay the maximum per-occurrence limit several times over if it is determined there were multiple occurrences during the policy period.142
F. Three Major Tests for Determining Occurrence in CGL Policies
As a result of the limitations of old CGL policies that are the predominant policies in question for CVA claims, how one defines occurrence becomes a crucial question in determining who will be liable for what under a given policy.143 As is becoming increasingly evident, the “jurisprudence of occurrence-counting” is neither a model of consistency nor one of coherence.144 Not only are there several different tests employed by different jurisdictions to count occurrences, courts within New York employ their “unfortunate event” test inconsistently as well.
While New York currently utilizes the “unfortunate event” test, there are a number of tests other courts apply. The three main tests that courts employ to determine the number of occurrences in CGL policies145 are (1) the “cause” test, (2) the “effects” test, and (3) the “unfortunate event” test.146
The “cause” test, which a majority of jurisdictions use, looks to the cause of the injuries or the property damage. 147 If there is more than one event or action that caused the multiple injuries, courts adhering to this test will likely find more than one occurrence.148 In Missouri, a state that utilizes the “cause” test, Fellowship of Christian Athletes v. AXIS Insurance Co. exemplifies one way in which courts have dealt with this test.149 The court there held that there was only one occurrence when two boys drowned in a pool while at Christian camp, noting that the underlying cause of the accident was the Fellowship of Christian Athletes’ negligence in allowing the boys to attend a pool party while knowing that the boys could not swim and in failing to properly train and supervise the camp counselors.150 The court also made clear that it was undisputed that the boys arrived at the Aquatic Center at the same time, swam in the pool during the same one-hour period, and were discovered drowned at the bottom of the pool at the same time.151
In contrast, a minority of jurisdictions use the “effects” test, or a variation thereof, which looks to the effects that an event had.152 Essentially, this test employed by a minority of states looks at the number of injuries to determine the number of occurrences.153 An example of the “effects” test in action, American Modern Select Insurance Co. v. Humphrey was a case in which a jogger sustained 147 wounds after seven dogs attacked her.154 Applying the “effects” test, the court held that the twenty-minute attack represented only one occurrence because the time between the bites was closely linked, and the connection between the attack and the jogger’s wounds was close enough to represent one injury or “effect.”155
The “unfortunate event” test––the current rule adopted by New York––is a modification of the “cause” test that determines the number of occurrences by examining whether a cause and result are closely linked in time and space.156 It was first utilized in Arthur A. Johnson Corp. v. Indemnity Insurance Co. in 1959, a third-party construction liability case where unprecedented rainfall breached two walls on a job site, resulting in multiple claims of flood damage.157 The court ultimately held that the collapse of two different walls in two different buildings on one job site amounted to two different occurrences, since the breach of the individual walls––and not the heavy rainfall––was the cause of the flooding.158 Since then, courts have typically applied a two-part inquiry when determining the number of occurrences.159 The first inquiry is whether the events giving rise to the injury are closely linked in time and space.160 Next, the court must ask whether the incidents can be viewed as part of the same “causal continuum,” without any intervening factors.161 In theory, the “unfortunate event” test works as a framework within which to determine whether or not separate acts that caused the injuries were sufficiently related enough to be considered a single occurrence.162
II. Analysis: The Problems in the Application of New York’s Unfortunate Event Test
While some of the rationale behind the “unfortunate event” test is certainly appealing and speaks to the complexities of when multiple closely-related incidents are involved, the test in practice is unwieldy and difficult to apply.163 Because of the judicial uncertainty that alleged victims, organizations, and insurers will face as courts try to figure out how to apply this test to CVA claims, New York also risks hurting judicial efficiency through a pile-up of complicated litigation if a new test is not utilized.
In addition to concerns about judicial economy that courts cannot afford to ignore with the huge influx in litigation from revived CPLR 214-g claims, the “unfortunate event” test is antiquated and hard for courts to apply. It was first utilized in 1959 in the construction liability context,164 and since then, unimagined and increasingly complicated liability cases have arisen that pull at the fabric of the test’s viability.165 Furthermore, it has become increasingly difficult for judges to articulate what a “singular causal continuum” is for the purpose of determining an occurrence when there are many injuries to multiple persons at play in the analysis.166 Under the “unfortunate event” test, courts tend to seek to maximize coverage for policyholders, focusing on the desired result instead of the intent of the parties.167 Ambiguity in policy language and the contractual maxim of contra proferentem, which mandates that a document be interpreted against the drafter,168 has led to a somewhat strained analysis and increasingly inconsistent application.169
This strained analysis and inconsistent application is demonstrated in a series of cases that have utilized the “unfortunate event” test. In Stonewall Insurance Co. v. National Gypsum Co., an asbestos action where plaintiffs sued the policyholder claiming property damage as a result of the policyholder’s asbestos-based products, the court held that there was only one occurrence.170 The court opined that the single occurrence was the decision of the policyholder to manufacture and sell asbestos-containing products, not the individual property damage to each building caused by the asbestos, as the insurance company had argued.171 Although the court alleged that they were utilizing the “unfortunate event” test and that neither party gave much in the way of extrinsic evidence, their decision seemed to hinge primarily on simply ensuring the maximum coverage for the policyholders in accordance with contra proferentem.172 Interestingly, on appeal, the Second Circuit reversed this portion of the lower court’s decision, holding that each installation of the asbestos products constituted a separate occurrence, requiring the application of another deductible.173
In 2007, in an action seeking coverage for a lawsuit involving supposedly toxic chemicals the policyholder had used in their microwavable popcorn butter, a court held that each employee’s claim was a separate injury.174 Because the employees sustained their injuries based on deliveries of the flavoring over a period of several years, each becoming exposed to the chemicals at different times and for different periods of duration, the court stated that each separate injury constituted a distinct occurrence.175
In contrast to the first two cases, in a third case that involved an automobile crash pile up, a different court held that there was only one occurrence.176 The policyholder’s automobile struck one approaching vehicle, ricocheted off that vehicle, and then struck a second one over one hundred feet away.177 The court there held that the multiple collisions constituted a single occurrence because they transpired only a few moments apart.178
Similarly, in a case where two infant tenants suffered serious injury as a result of exposure to lead in the policyholder’s apartment building, a court held that there was only one occurrence.179 Since the injuries arose from continuous exposure to essentially the same conditions, this court reasoned that there was only one occurrence, despite the fact that the injured infants may have ingested the lead at different times and may have ingested different amounts.180
As evidenced in the above examples, cases that could have been adjudicated similarly often end up with vastly different outcomes. Whereas a car crash that injured multiple people and lead that poisoned multiple babies each represented only one occurrence, the toxic asbestos in the buildings and chemicals in the microwavable popcorn butter that also injured multiple people represented multiple occurrences.181 Beyond almost instantaneous multiple injuries, the courts have been unclear about whether events are close enough in time and space to count as one occurrence.182
Because of the difficulty with applying the test and the lack of clarity involved, the analysis of the rule often strains the concept of causation.183 In situations of diffused responsibility, such as with child sexual abuse cases, it then becomes a practice in arbitrary decision-making to decide both how many occurrences there are and to whom liability and coverage should ultimately be assigned.184 Strictly speaking, if a drunk driver hits a pedestrian with their car, the car manufacturer who built the car and the designer of the car are “but-for” causes of the accident.185 Of course, if the car did not malfunction, it would be an absurdity to hold the car manufacturer and the designer liable for the drunk driver’s actions. Proximate cause provides a way to limit liability of an actor from the far-flung consequences of his actions, using notions of responsibility that are related to both foreseeability and risk creation.186 In the context of determining occurrences under the “unfortunate event” test, this problem of causation arises when courts arbitrarily decide to attribute to one cause the assignment of responsibility that arguably belongs to more than one cause or to another cause. Contributing or concurrent causes, and the difficulty with determining how close in time and space these causes have to be to count as a single occurrence, act to confuse the causation analysis in occurrence determinations.187
With the huge swarm of litigation that the CVA brings, courts need a test for occurrence-counting that is practical, easily applied, and consistent in its results, not one that debates how to define “cause,” much less the determination of a chain of events in time and space.188 Because the “unfortunate event” test asks whether cause and result are linked in time and space, it becomes a complicated balancing analysis of what determines a “single causal continuum” that simply cannot be applied consistently across the board.189
[1] Arthur A. Johnson Corp. v. Indem. Ins. Co. of N. Am., 164 N.E.2d 704 (N.Y. 1959). Heightening this inconsistency is the fact that although courts identify the “unfortunate event” test as part of its analysis, they often subsequently describe factors that are more consistent with the application of an entirely different test.190
In addition to being applied inconsistently, the “unfortunate event” test creates judicial uncertainty. Litigants often have no idea what to expect and no inkling about how judges will rule in these cases because the “unfortunate event” test could be applied in so many different ways.191 Using this test, New York courts have sometimes fallen into the trap of being “result-oriented”192 instead of focused on the language in the policy and the intention of both of the parties in drafting it. Courts have sometimes expressed a desire to interpret insurance contracts in ways that specifically benefit policyholders or the alleged victim, believing that these policies are contracts in which the parties are not similarly situated in terms of sophistication and understanding.193 However, since no preference for a particular mode of interpretation of “occurrence” is stated either, the judicial desire to benefit the injured parties often results in vastly different outcomes.194 This type of judicial uncertainty and results-oriented analysis will not be sufficient to deal with the onslaught of CVA litigation. Policyholders and insurers alike need a test that will be consistent, so that they can know what to expect going into litigation.
III. Proposal: Minnesota’s Actual-Injury Test and its Practical Possibilities for CVA Claims
In 2013, the Minnesota Legislature enacted their Child Victims Act, opening up a three-year look back window for victims whose claims would otherwise be time barred to file.195 The test that Minnesota’s courts chose to utilize was known as the “actual-injury” test.196 The test––much like New York’s––is a hybrid test that combines multiple elements of both cause and effect. The “actual-injury” test establishes an occurrence at the time the complaining party was actually damaged, not at the time the wrongful act was committed.197 In cases of sexual abuse, there is an occurrence triggered for policy purposes when the sexual abuse occurs.198 Each victim is then counted as a separate occurrence.199
In the seminal case on the issue, In re Diocese of Duluth, the court clarified that if a victim was injured by two priests during one policy period, that would be considered two occurrences, but if they were injured repeatedly by the same priest during one policy period, that would be one occurrence.200 In re Diocese of Duluth lays out a standard for occurrence-counting for child sexual abuse cases that is clear and easy to apply, utilizing the “actual-injury” test that has long been applied to all CGL policies in different insurance contexts.201 Because it potentially implicates multiple policy periods and several layers of coverage, it also means multiple insurers can share liability and policyholders may be able to access their excess insurance.202
With the advent of the CVA and its one-year look-back window, the New York judiciary should seek to utilize Minnesota’s “actual-injury” test that provides a much more concrete method for determining occurrence than New York’s current rule. Additionally, utilizing Minnesota’s test would ensure that courts move away from an exclusively results-oriented analysis. Not only will insurance companies be able to share proportionate liability when a claim implicates multiple policy periods,203 alleged victims—who may have suffered excruciating abuse—will be able to have their day in court. Meanwhile, the organizations that are implicated in these cases will have some semblance of judicial certainty about how the court will decide and will no longer have to worry about courts making decisions based solely on maximizing coverage for alleged victims.
A. Utilize Minnesota’s “Actual-Injury” Test over the “Unfortunate Event” Test to Avoid Judicial Uncertainty
In contrast to New York’s rule, Minnesota’s “actual-injury” rule provides an easy-to-apply blueprint for determining the meaning of “occurrence” when policy language is unclear, which provides more consistent results across the board.204 Although it may still be difficult to determine which policy periods are triggered, because alleged abuse victims are older and may not remember all the details of their abuse, determining the number of occurrences once the policy periods implicated can be ascertained will be simple, since the number of people injured will determine the number of occurrences.205 Furthermore, limiting occurrences to one per child per policy year strikes a happy medium between ensuring alleged victims get proper compensation, while taking into account the difficulty for insurers and institutions of shouldering the entire financial burden of a perpetrator’s actions.206 With Minnesota’s rule, there is no longer a need to complicate the analysis by figuring out whether actions are sufficiently closely related in time and space. Instead, the analysis becomes an easy counting job.
B. A Response to Legal and Practical Oppositions to Using Minnesota’s Rule
Despite the appeal of an easily applied rule like the “actual-injury” test, there are a series of critiques that have been advanced in opposition to the Minnesota rule for counting occurrences. One argument that has been touted is that using Minnesota’s rule would be unwise, since after their look-back window closed, five out of six of the Catholic dioceses in the state filed for Chapter 11 bankruptcy in order to settle hundreds of claims of sexual abuse at the hands of clergy.207 While the bankruptcies allowed the dioceses to settle mounting claims of sexual abuse without going before a jury, victims received a fraction of what juries might have awarded them, and these bankruptcies stripped these victims of their day in court.208
In response to this critique, it is important to note that Minnesota’s look-back window was significantly longer than New York’s look-back window (three years instead of just one).209 This created a much longer timeline for previously time-barred claims to be filed. Additionally, not one diocese actually ceased to function because of Minnesota’s CVA.210 In fact, under New York’s current rule, the Diocese of Rochester has already filed for Chapter 11 bankruptcy in an attempt to settle claims in bankruptcy court,211 so having a clear-cut rule is more important than ever for the quick adjudication of claims regardless of whether organizations file for bankruptcy.
Furthermore, unlike Minnesota, many institutions in New York have been anticipating the passage of the CVA and have enacted victim compensation funds to protect themselves from the prospect of big jury awards.212 The benefits of victim compensation funds are that victims receive some compensation without having to go through lengthy and costly litigation, while organizations also avoid having a jury decide settlement awards.213
Others have argued that with all the difficulty that comes with defining “occurrence” when policy language is unclear, this determination of whether or not to change the test is best left up to the legislature to remedy. In Arkansas, for example, the legislature enacted a statute that required a CGL policy offered for sale in Arkansas to define an “occurrence” to include “[p]roperty damage or bodily injury resulting from faulty workmanship.”214 In South Carolina, the legislature passed a statute that provided that the definition of “occurrence” in CGL policies had to include “(1) an accident, including continuous or repeated exposure to substantially the same general harmful conditions; and (2) property damage or bodily injury resulting from faulty workmanship, exclusive of the faulty workmanship itself.”215 In Hawaii, there is a statute that states that “the meaning of the term ‘occurrence’ shall be construed in accordance with the law as it existed at the time that the insurance policy was issued.”216
However, while legislation may indeed be one way to remedy the judicial uncertainty attached to this problem, there have not been any statutes that specifically codify which test should be used to determine occurrence, just as there has not been a policy that truly brings clarity to the meaning of the word “occurrence.”217 While it is unclear why exactly this is the case, one thought is that the inquiry is extremely context specific, and there will always be losses that were previously unimaginable. Whatever the case may be though, it is clear through the lack of legislation on the issue that the legislature has deferred to the courts to decide what test to utilize.
Others have argued that the Minnesota “actual-injury” test fundamentally works in the same way as New York’s “unfortunate event” test, so it would be repetitive and unnecessary to adopt Minnesota’s rule.218 However, while each test may end up producing similar results in certain case contexts, the Minnesota test is much easier to apply and is consistent in a way the “unfortunate event” test cannot be.
A final critique that has been advanced in opposition to Minnesota’s test is that although the current rule in New York is impractical and produces an immense amount of judicial uncertainty that leads to inconsistent results, it is still the most practical of the main three tests “because it corresponds most with what the average person anticipates when he buys insurance and reads the ‘accident’ limitation in the policy.”219 This reasoning, however, is countered by the fact that judges have applied the “unfortunate event” test inconsistently,220 often using a results-driven analysis that leaves both parties to the litigation confused about what to expect. This is neither practical nor what the “average insured anticipates” when reading the “‘accident’ limitation in the policy.”221 Furthermore, child sexual abuse does not fit neatly into this framework, since it was not until the 1980s that it was properly anticipated as a possibility for excluding coverage.
Conclusion
Humans have long struggled with the problem of the one and the many, iterations of this affecting how people understand and analyze the world around them. In insurance law, the jurisprudence of counting occurrences is well-documented, perhaps most famously in the 2001 World Trade Center Attack insurance litigation. Recently, with the passage of the CVA, courts are once again faced with this complex issue.
As litigation continues to pile up with the passage of the CVA and its one-year look-back window, it is imperative that New York is able to quickly and consistently handle these cases. The need to adjudicate these cases fairly and efficiently is crucial not just for alleged victims, organizations, and insurance companies; it is also important for the average taxpayer who may see their taxes rise as a result of increasing insurance rates. While New York has designated forty-five judges to oversee CVA claims and has trained them on issues that commonly arise in such claims,222 there is no way to be truly consistent with the “unfortunate event” test that leaves so much up for individual analysis. With such a large portion of pre-1980s CGL policies being implicated, the litigation is bound to be complex and fact intensive. With this in mind, and the thousands of claims on the line and hundreds of millions of dollars at stake, the court cannot afford to perform the analytical gymnastics that the “unfortunate event” test requires when attempting to define “occurrence.” Complicating an already overly complicated question by having judges establish arbitrary lines regarding when an event and result are close enough in time and space to be considered one occurrence only serves to increase judicial uncertainty and incoherence.
Instead, New York should utilize Minnesota’s “actual-injury” test when determining occurrences where the policy language is unclear, because it provides a coherent way to navigate occurrence-counting in not just sexual abuse cases, but in all insurance contexts. It combines and refines the “cause” and “effects” test to create an easy method with which to determine occurrences, one that rejects the arbitrary lines that judges must draw to determine whether there has been a “single causal continuum” in the “unfortunate event” test.