Breed Discrimination in the Homeowners Insurance Industry

November 2020

Together with several other animal welfare groups, this white paper was created to address breed discrimination in the homeowners insurance industry. The collective goal is to end breed discrimination and urge the National Association of Insurance Commissioners to evaluate the use of breed discrimination lists. The paper is complete with testimonies and arguments, and divided into sections focusing on breed lists and discrimination, unjustified use of breed as actual risk, and recommendations. You can read the full paper below or download the PDF version.

Download the PDF

Executive Summary
I. Introduction
II. Breed Lists and Discrimination
   A. Uninformed Consumers
   B. People of Color
   C. Consumers of Moderate and Low Means
III. Unjustified Use of Breed as Actual Risk
   A. Lack of Reliable Data
      1. Dog Bite Studies
      2. Insurance Claims
   B. Rejection of Breed Discrimination 
      1. Governmental Opposition
      2. Non-governmental Opposition
IV. Recommendations
   A. Data Call 
      1. Basis for Creation of Breed Lists
      2. Use of Breed Lists
      3. Dog Bite Claims
   B. Interim Measures
      1. Moratorium
      2. Notice and Education
   C. Respond to Data Call

Executive Summary

This white paper discusses the discriminatory impact of the insurance industry’s use of dog breed lists to deny homeowner and renters insurance coverage and renewals, create policy exclusions, and place limitations on coverage. As the standard-setting organization for the insurance industry, the National Association of Insurance Commissioners (NAIC) is in the best position to investigate and respond to the concerns raised by use of these lists. This topic is particularly ripe for review in light of the increased adoption and fostering of pet dogs during the Covid-19 pandemic and because of the NAIC’s increasing sensitivity to discrimination by the insurance industry and its continuing concern for making insurance available and affordable for consumers.

Dog breed lists are a collection of breeds considered by an insurance company to be uninsurable based on behavior assumed to be intrinsic to a breed (e.g., biting behavior). Insurance companies create these lists without any reliable actuarial data showing a difference in behavior, yet they rely on the lists as if breed assumptions create a risk that is valid across the board. Consideration of a specific dog’s bite history or behavior is irrelevant if the dog’s breed is listed.

The use of breed lists has a detrimental impact on three groups—uninformed consumers, people of color, and consumers of low or moderate means. With regard to the first group, few homeowners consider the impact of a dog’s breed on their insurance coverage when choosing to bring one into the family, and many are surprised when they learn that they have little to no coverage because of their choice. Of those affected, few are aware of other options to obtain coverage. Consumers affected are left with two choices—either go underinsured or uninsured or end their relationship with their beloved pet.

Use of breed lists also raises concerns of fair and equitable treatment of people of color. Research suggests a correlation between breed type and race as it relates to perceptions of ownership, with a strong correlation between pit bulls and the African American community. Scholars opine that this is especially true as it relates to breed-specific laws, which may be a product of the fear associated with stereotypical images and activities of minority groups. This fear may also be associated with an insurance carrier’s choice of which breeds to include on its list.

Consumers of low or moderate means are also impacted because of the higher costs associated with obtaining coverage when they have been denied or their liability coverage restricted because of their pet’s breed. They are also placed at a disadvantage against preferred customers who might receive a waiver for their particular dog because of the business they bring to the carrier.

The impact of the insurance industry’s use of breed lists is particularly troubling because there is no reliable actuarial data that supports making a distinction between breeds, nor is there evidence that insurance claims for these breeds is financially significant for insurance carriers relative to other paid losses. Further, breed discrimination in general has been rejected by numerous governmental and non-governmental organizations, including the White House, the American Veterinary Medical Association, and national animal welfare groups, among others.

Because of the NAIC’s position and ability to respond to industry shortcomings, the NAIC should explore the discriminatory impacts of breed lists by issuing a data call seeking information relevant to determine whether reliance on breed lists is justified, take interim measures during the data call to restore affected consumers’ ability to acquire affordable insurance, and once data collection is complete, determine whether breed lists are a legitimate underwriting variable. 

I. Introduction

In 2020, Covid-19 and protests resulting from police violence against minorities changed the American political and social landscape. Both of these events require us to reflect on what role companies and industries play in preventing race-based discrimination and disparities. The NAIC acknowledged this need at their 2020 Summer Meeting when it considered the role of the
insurance industry. Indeed, NAIC President and Director Ray Farmer emphasized the importance of seizing this “historic opportunity as a regulator, a community, and as an important sector of the financial services industry to commit real, meaningful and lasting change.”¹

Making lasting change to address institutional biases in the insurance industry requires all existing practices be analyzed for intentional and unintentional discrimination, even those practices that might not have come under scrutiny in the past. Dog breed lists used by insurance companies to exclude or limit liability coverage in homeowners and renters insurance policies² comprise such a practice. Scrutiny is critically important now as the substantial increase in adoption and fostering of dogs due to the coronavirus pandemic³ has placed many insurance consumers at risk of losing their coverage.

The vast majority of insurance carriers⁴ use dog breed as a factor in determining whether to issue a homeowners insurance policy or to issue a policy with exclusions or monetary limitations. Dog breed lists used by insurance companies are comprised of large dogs that, because of their breed, are considered—without proof—to be more aggressive than other dogs. These lists usually include the following breeds: pit bull,⁵ mastiff, German shepherd, Akita, Staffordshire terrier, Chow Chow, Alaskan malamute, Doberman pinscher, Great Dane, Siberian husky, Rottweiler, Presa Canario, and wolf hybrids.⁶

Use of dog breed lists discriminates against three groups of homeowners insurance consumers: the uninformed, people of color, and people of moderate or low means. Even though there is no scientific evidence or actuarially supported data to prove certain dog breeds are inherently more inclined to bite than other breeds, many insurance companies deem owners of these breeds ineligible for homeowner policies or endorse their policies with flat exclusions or dollar limitations. The result is an undesirable choice faced by homeowners: remain uninsured or underinsured or give up what many consider a key member of the family.

Every American homeowner, regardless of race or means, has an interest in protecting the home they worked so hard to acquire from damages claims by third parties, and the NAIC’s mission includes protecting the public interest and facilitating the fair and equitable treatment of insurance consumers. Breed discrimination is an issue that impacts insurance consumers across the country. As a countrywide standard-setting association, the NAIC is in the best position to recommend fair and uniform solutions and should do so. 

II. Breed Lists and Discrimination

Use of breed lists can result in discrimination against three groups of insurance consumers: consumers who are unaware if any alternative exists, people of color, and people of moderate or low means.

A. Uninformed Consumers

The majority of insurance consumers are not well-versed in insurance. According to a survey by Insurance.com, 48% of the homeowners surveyed did not understand homeowners liability coverage and 24% did not even know the amount of liability coverage they carried and/or had never read their own policies.⁷  And when it came to educating themselves prior to acquiring their own homeowners insurance, one-third of the surveyed consumers had made no effort to compare rates to ensure they received the best value.⁸ If one were to ask consumers whether they know specifics of their coverage for incidents involving dogs or even asked about that coverage prior to purchasing, the percentages are likely greater. With over 63 million households in the United States owning dogs,⁹ that is a lot of uninformed consumers.

This lack of knowledge can impact consumers in a variety of ways, leaving consumers with little or no insurance coverage if they actually were to have a dog bite claim. First, insurance coverage plays little role in a homeowner’s decision to acquire a dog, regardless whether the dog was acquired after deliberate and careful thought or as a result of an unexpected opportunity or necessity.¹⁰ Further, breed itself is not always a factor in a homeowner’s decision to acquire a pet dog. Although breed may be a factor in a deliberate choice of a particular dog,¹¹ choice of breed may play a lesser role in other circumstances,¹² thereby making liability coverage even less of a factor in decision making.

Second, a consumer who is refused homeowners insurance or has limitations placed on the policy based on a dog’s breed is often unaware of other options available to obtain coverage. For example, a consumer who is denied insurance online, especially a consumer who has not made an effort to compare other companies or policies, may not know that they should research other companies, contact an agent for guidance, or even seek information from the state insurance commissioner.

Third, a consumer might not even know that coverage is lacking. For example, according to the Insurance.com study, 52% of the homeowners surveyed who acquired a dog after their policy began did not inform their insurer of their canine addition to the household.¹³ Depending on the insurer, if an incident were to occur involving the dog thereafter, the homeowner might find the dog’s breed a bar to coverage. Further, if a homeowner has purchased both a homeowners policy and an umbrella policy, the homeowner might think the umbrella policy will cover any shortage under a homeowners policy. However, if the umbrella policy only pays shortfalls after payment under a homeowners policy, the umbrella policy might not cover liability for incidents by certain dog breeds if those breeds were excluded under the homeowners policy.

Thus, breed exclusions and limitations from homeowners insurance take advantage of the uninformed consumer, leaving homeowners open to devastating liability should an excluded incident occur. In light of the NAIC’s mission to facilitate fair and equitable treatment of insurance consumers, the NAIC should take considered review of the impact of insurance breed lists.

B. People of Color

Breed lists can also have an impact on people of color, which also calls into question whether insurance consumers are treated fairly and equitably.

The Fair Housing Act (FHA) prohibits discrimination “against any person in the terms, conditions, or privileges of sale or rental of a dwelling, or in the provision of services or facilities in connection therewith, because of race, color, religion, sex, familial status, or national origin.”¹⁴ The regulations of the U.S. Department of Housing and Urban Development (HUD) provide that such discrimination includes “[r]efusing to provide . . . hazard insurance for dwellings or providing such services or insurance differently” based on those factors.¹⁵ Although insurance is mainly state regulated, courts have held that the FHA applies to property and hazard insurance as well as homeowners insurance.¹⁶

Discrimination in the insurance industry in general has a long history and includes race-based discrimination in the issuance of life, auto, and homeowners insurance policies. Indeed, the NAIC itself recognized the industry’s blighted past at its 2020 Summer National Meeting, with experts discussing significant instances of discrimination.¹⁷ In the life insurance context, for example, Black policyholders received policies valued at 2/3 that of White policyholders, yet Black policyholders were charged 30% to 40% more in premiums.¹⁸ Discrimination was also a result of underwriting guidelines for auto and home insurance. Auto insurance guidelines in Texas, for example, required that insurance applicants have been previously insured, which detrimentally impacted minority groups who were uninsured prior to the insurance mandate enacted in 1991.¹⁹ Underwriting guidelines for homeowners insurance that based rates on the age and value of the home also resulted in unfair racial discrimination.²⁰ And redlining, of course, which based risk solely on the location of the neighborhood, resulted in widespread racial
discrimination in the auto and homeowners insurance industries.²¹

The insurance industry has also faced extensive criticism and legal challenges based on its discrimination in issuing homeowners insurance policies in the past. For example, the industry was subject to a congressional hearing in 1997 examining redlining and credit screening practices. Allegations against the industry included the continuing use of home age and value in underwriting guidelines, costly yet inferior policies issued to consumers in primarily minority communities (including limitation of policyholders to recovery of market value rather than replacement costs), and even marketing disparity in terms of the lack of insurance agents and advertising in minority areas, making the acquisition of insurance by minority consumers in these areas much more difficult than in non-minority areas.²² And in Texas Department of Housing and Community Affairs v. Inclusive Communities Project, Inc., ²³ the United States Supreme Court held that disparate-impact liability applies to the FHA, because it “plays a role in uncovering discriminatory intent: It permits plaintiffs to counteract unconscious prejudices and disguised animus that escape easy classification as disparate treatment. In this way disparate impact liability may prevent segregated housing patterns that might otherwise result from covert and illicit stereotyping.”²⁴ This edict allowing imposition of disparate impact liability was then placed in HUD regulations.²⁵

Homeowner insurance exclusions based on breed bans raise a similar specter of discrimination. Although insurance companies may argue that the discrimination is based on risk (i.e., the risk of an insurance claim based on an incident involving a dog), several scholars have argued that focus on the breed of dog can result in unfavorable discrimination based on race. Ann Linder, a Research Fellow with the Animal Law & Policy Program at Harvard Law School, in analyzing impacts of breed discriminatory legislation on race found that there is a correlation between breed type and race as it relates to the perception of owners. In her study, she presented six pictures of different dog breeds and asked the study participants who they thought would own each type of dog based on gender, race and ethnicity, and age.²⁶ Linder determined that “unlike the other five breeds that were tested, pit bulls were perceived as most commonly belonging to people of color-- specifically, young, Black males,” and this perception was consistent among study participants, regardless of the participant’s race or gender.²⁷ Linder concluded that, although “racial connotations of each breed vary substantially,” the findings from her study “are consistent with [the] idea that pit bulls are being targeted, in part, due to racial bias and variables beyond the risk posed by the dogs themselves.”²⁸

Other scholars have also noted a link between race and breed type. In the context of breed specific laws, one scholar notes that such laws are:

the product of animus. Certainly there is animus against particular dog breeds endemic in the laws, but more importantly, . . . these laws are the product of animus against the owners  of such breeds, or at least the stereotypical owners. The war on pit bulls, in particular, is characterized by elements of moral panic against the persons who have so long been associated with these dogs—“gang members,” “drug dealers,” and “urban youth,” groups that all point back to fear of racial and ethnic minorities.²⁹

Another scholar also notes the connection between pit bulls and race. She notes that the 1980s saw a rising connection between pit bulls and gang violence. She states: “[T]he dogs themselves began to be figured as carriers of the contagion of criminality. This image was, it seems, propped up at least in part by the association of ‘pit bulls’ with the hip-hop music scene, itself a strongly racialized genre in the United States.”³⁰

Because breed exclusions in homeowners insurance may result in a discriminatory impact on racial minorities and, as discussed further below, there is no rational justification for continuing such exclusions, the risk of a dog incident under a proposed homeowners policy should be based on a fact-based analysis of an individual dog’s true risk, taking into account both the dog’s behavior and the owner’s ability to maintain safe interactions between the dog and the community. Indeed, even in the FHA context, HUD guidance recognizes the need for a fact based inquiry into the risk posed by an assistance animal rather than a reliance on breed:

A housing provider may . . . refuse a reasonable accommodation for an assistance animal if the specific animal poses a direct threat that cannot be eliminated or reduced to an acceptable level through actions the individual takes to maintain or control the animal (e.g., keeping the animal in a secure enclosure). . . . Pet rules do not apply to service animals and support animals. Thus, housing providers may not limit the breed or size of a dog used as a service animal or support animal just because of the size or breed but can, as noted, limit based on specific issues with the animal’s conduct because it poses a direct threat or a fundamental alteration. ³¹

Following this type of standard will avoid even the perception of a racial bias because the focus is based on actual conduct of a specific dog rather than its breed.

C. Consumers of Low or Moderate Means

Breed lists also limit the ability of consumers of low and moderate means from obtaining sufficient homeowners coverage when they have pet dogs included on the lists. Many homeowners whose dogs are excluded by their preferred insurance carrier are forced to either purchase coverage through another carrier at a higher cost—even if there is no evidence of an increased risk from that particular dog—forego coverage because they are unable to afford the cost, or part with their canine family member to eliminate potential liability.

Moderate and low means consumers are also at a disadvantage because they have fewer bargaining chips than those with more assets. Although insurance carriers allege that dogs on the breed list are an undesirable risk, many carriers will overlook that risk by waiving underwriting guidelines if the insurance consumer is a preferable or profitable customer. For example, if a consumer has a primary residence, a vacation home, and several vehicles to insure, a carrier may choose to waive or remove restrictive endorsements based on breed rather than risk losing the customer. Similar accommodations might be made for consumers who have both personal and commercial accounts based on concern that loss of the personal account might lead to loss of the more valuable commercial account.

Further, even where a carrier attempts to accommodate a homeowner by imposing conditions on ownership of the dog for purposes of coverage, moderate and low means consumers are hindered by their lack of disposable income. For example, a carrier might agree to write a homeowners policy covering a pet dog whose breed is included on the carrier’s breed list if the homeowner reduces the risk of liability, such as by erecting a six-foot high fence with a locking gate around the property. However, if the homeowner has no ability to pay for the accommodation, the homeowner is left in the same position the homeowner would have been had no accommodation been made at all.

Given the sensitivity that the NAIC has to meeting the needs of consumers in a fair and equitable manner, any justification for underwriting guidelines that discriminate against consumers— especially against the groups of consumers discussed here—should be viewed with skepticism. Insurance carriers that use breed lists base their usage on the assumption that these breeds create a greater risk of liability. However, as discussed below, no evidence exists showing a greater risk, so the given justification fails.

III. Unjustified Use of Breed as an Absolute Risk

The insurance industry’s use of dog breed as an indicator of risk is unjustified for two reasons. First, there is a lack of reliable data showing that the listed breeds are more aggressive or bite more often than non-listed breeds. Second, discrimination against breeds based on assumptions regarding risk has been rejected by a variety of governmental and non-governmental organizations as well as state and local governments.

A. Lack of Reliable Data

The justification that the insurance industry uses for breed discrimination is twofold: (1) research by the Center for Disease Control (CDC) regarding risk of bites from certain breeds; and (2) the number of claims paid by the industry for dog incidents.³² The industry’s reliance on these arguments is unjustified, however, because there is a lack of reliable data showing increased risk by breed and because dog incident claims based on breed are small in comparison to other property and liability claims faced by the industry.

1. Dog Bite Studies

With regard to breed aggression, the insurance industry relies on CDC research on fatal and nonfatal dog bites.³³ Reliance on this work is misplaced, though, because not only is it outdated,³⁴ but its reliability has since been undermined by the CDC itself based on data gathering limitations, including the inability to accurately identify breeds and to collect accurate data on dog bites.³⁵ Other studies purporting to show a distinction based on breed share similar limitations³⁶ as well as being challenged as anecdotal rather than data-driven.³⁷ Further, additional studies have conflicting views of the impact of breed on dog bites. Those studies indicate that other factors play a more significant role in aggression than breed, such as whether the dog is male and intact, chained, mishandled or a victim of prior abuse or neglect.³⁸ Even other studies suggest that small dogs not on the breed list have comparable bite incidents as larger listed dogs, and therefore, breed should not be used as an indication of a propensity for aggression.³⁹

In addition, whether a breed is placed on an insurance list may be swayed by sensationalized media reports of dog attacks (which served as the springboard for enactment of state and local breed-specific legislation in the 1980s), ⁴⁰ and those eye-catching reports may focus more on the breed of the dog rather than on the actual facts leading to the attacks. For example, in 2001, a well-publicized attack in California involved two Presa Canario dogs that mauled a woman to death.⁴¹ The trial for the mauling revealed that the dogs had previously tried to attack people and other dogs several times, the dogs were not owned by the defendants but were merely being cared for while the owner served prison time,⁴² and one defendant knew that the other defendant, who was with the dogs at the time of the attack, could not control the dogs.⁴³ And when one digs further into the facts, it is revealed that the dogs in the attack had been bred to be “vicious guard dogs.”⁴⁴ Each of these factors impacted the fatal attack far more than the mere breed of the dogs. Yet the Presa Canario suddenly appeared on insurance breed lists, even though the breed was not one of the breeds noted by the CDC in its original study.⁴⁵

Because of the detrimental impact on homeowners by the use of breed lists, reliance on dog bite studies using inaccurate data on the actual breeds being charged with a higher risk of biting is unwarranted and unfair. A variety of factors influence whether a dog will bite, and a young, intact, unrestrained and provoked male dog of any breed is more likely to be involved in a biting incident than, for example, an old, deaf, partially blind and arthritic Rottweiler with no history of biting.⁴⁶ The key is that some dogs do present a higher risk of a dog-related claim, but the risk is based on the dog’s individual behavior and surrounding circumstances, not on the breed. Therefore, the insurance industry can protect itself by evaluating the risk on a case-by-case basis rather than roping in pet dogs based on breed even if individually they present a lower level of risk.

2. Insurance Claims

In October 2020, the Insurance Information Institute (III) issued estimates on the number of dog bite claims nationwide. According to III calculations, the number of dog “bite” claims increased from 16,919 claims in 2003 to 17,802 claims in 2019, while the average cost per claim more than doubled from $19,162 to $44,760 for the same time period. The total for dog bite claims paid for 2019 was $796.8 million.⁴⁷ The increased costs of claims, however, are due to increased medical costs and an upward trend in the size of settlements, judgments and jury awards.⁴⁸ Further, although the claims were labeled as dog “bites” in III’s estimates, the numbers actually included both bites and other dog-related injuries, such as injuries due to children, cyclists or the elderly being knocked over by dogs.⁴⁹

However, despite the millions paid for dog incidents, that cost is only a drop in the bucket when contrasted with the amount paid by the insurance industry for other types of claims. Homeowners insurance losses for property damage made up over 98% of losses in 2018, and total liability losses—where dog incidents fall—were less than 2% of the amount.⁵⁰ Professor Larry Cunningham, using data from 2002, explains how the insurance statistics are misleading:

The III states, “[d]og bites now account for almost one quarter of all homeowner’s insurance liability claims costing $345.5 million.” Some perspective is in order. For every $100 in premiums, insurers spend $77 paying claims. Of that $77, the overwhelming majority ($72, or 93.5%) is spent on paying property damage claims. Liability claims only amount to $5, or 6.5%, of total claims. Even then, dog bites only constitute a percentage of that figure. Put into perspective, the money paid out in dog bite claims is negligible when compared to the overall amount of money paid out for other types of claims. Damage due to lightning, fire, and mold all individually account for more claims payouts than all liability claims combined.⁵¹

Thus, in the broader context of all claims paid under homeowners insurance policies, claims based on dog incidents are only a small portion. Making wholesale exclusions of dogs based on breed makes little difference to the overall losses faced by the insurance industry. And liability for dog bite claims can still be mitigated by exclusions and limitations based on actual risks presented by an insured’s dog.

In addition, insufficient actuarial data exists as to the impact of claims caused by listed versus unlisted dogs. If listed breeds do not actually increase claims higher than those for unlisted breeds, than the breed list has no rational basis. 

B. Rejection of Breed Discrimination

A number of governmental and non-governmental organizations have rejected the concept of breed discrimination, including the CDC.⁵² This rejection is most clearly seen in the context of breed-specific legislation (BSL). A short review of BSL and its rejection is instructive for breed based discrimination in insurance.

BSL is state or local legislation enacted largely in the 1980s in response to media reports of dog attacks. The intent of BSL is to ban or place other restrictions⁵³ on specific breeds in an attempt to prevent dog bite injuries. One of the main objections to BSL is the same as for insurance breed lists—i.e., that no reliable evidence exists showing that dogs of specified breeds bite more frequently than others.⁵⁴ Further, BSL is difficult to enforce,⁵⁵ and when it is enforced, application of BSL requirements is inconsistent. For example, a number of studies show that visual identification of a dog’s breed based on physical features is often inaccurate,⁵⁶ and in a study of shelter workers, some dogs breeds’ are deliberately mislabeled to increase the possibility of adoption.⁵⁷ The application of BSL to mixes of dogs on the list makes inaccurate identification even more probable.⁵⁸ In addition, BSL results in unintended consequences, including dog abandonment or relinquishment to a shelter resulting in increased levels of euthanasia, behavioral issues and health impacts because of the dog’s isolation from people,⁵⁹ and the limitation of jurisdictional movement by owners of regulated dogs.⁶⁰ These consequences could also result with insurance breed lists if homeowners are forced to part with their dogs.

1. Governmental Opposition

A number of governmental organizations have rejected breed discrimination for the reasons stated above. For example, as noted before, the CDC questions the efficacy of its own studies and argues that breed alone is an insufficient basis for determining the dangerousness of a dog. As it states in its report:

Although fatal attacks on humans appear to be a breed-specific problem (pit bull type dogs and Rottweilers), other breeds may bite and cause fatalities at higher rates. Because of difficulties inherent in determining a dog’s breed with certainty, enforcement of breed-specific ordinances raises constitutional and practical issues. Fatal attacks represent a small proportion of dog bite injuries to humans and, therefore, should not be the primary factor driving public policy concerning dangerous dogs. Many practical alternatives to breed-specific ordinances exist and hold promise for prevention of dog bites.⁶¹ 

Indeed, the Obama Administration agreed with this view when it responded to an online petition seeking federal prohibition of BSL. The White House stated: “We don’t support breed specific legislation . . . . As an alternative to breed-specific policies, the CDC recommends a community-based approach to prevent dog bites. And ultimately, we think that’s a much more promising way to build stronger communities of pets and pet owners.”62

The United States Postal Service (USPS), whose letter carriers are one of the most frequent dog bite victims,⁶³ partners with State Farm and the American Veterinary Medical Association and focuses on dog bite prevention rather than breed discrimination.⁶⁴ For example, each year during National Dog Bite Awareness Week, USPS provides information to homeowners on how they can protect their carriers from dog attacks. USPS also offers alternative options for mail delivery when a specific dog’s behavior presents a problem and provides carriers with handheld scanners to note the presence of a dog at an individual address.⁶⁵ This year the USPS has also begun a program in several states whereby carriers will place stickers on mailboxes indicating homes with dogs.⁶⁶

In addition, in a growing trend,⁶⁷ at least 22 states ban the use of BSL in some fashion and focus instead on the behavior of individual dogs through dangerous dog and other breed-neutral laws.⁶⁸ For example, many of these states allow local regulation of dogs at large and dangerous or vicious dogs, but are prohibited by state law from using breed as a basis for that regulation.⁶⁹ Others take an approach focusing on a dog’s behavior rather than the dog’s breed. The State of Washington, for example, mandates that BSL restrictions cannot apply to dogs that pass an “American kennel club canine good citizen test or a reasonably equivalent canine behavioral test.”⁷⁰

Importantly, some states also specifically prohibit breed discrimination in the context of insurance. Pennsylvania, for example, provides that “No liability policy or surety bond issued . . . may prohibit coverage from any specific breed of dog.”⁷¹ Similarly, in Michigan, the Department of Insurance and Financial Services has issued a bulletin interpreting the state’s insurance law. In regard to underwriting, the bulletin states:

The Essential Insurance Act, specifically MCL 500.2103(2), permits insurers to deny, cancel, or non-renew coverage to a “person who insures or seeks to insure a dwelling that has physical conditions that clearly present an extreme likelihood of a significant loss under a home insurance policy.” Pet ownership, by itself, does not cause an otherwise eligible person to become ineligible for homeowners insurance because a pet does not “clearly present an extreme likelihood of significant loss.” The Essential Insurance Act does not allow companies to deny, cancel or non-renew coverage based on the insured’s possession of a particular animal.⁷²

The bulletin, however, does not prohibit an insurance company from taking into account the true risk presented by a dog. According to the bulletin in the context of rating, “Michigan law does not prohibit the imposition of surcharges based on dog breeds if the surcharge is actuarially supported.”⁷³ Thus, if an insurance company can show statistical data supporting an increased risk based solely on dog breed, then a surcharge would be allowed; however, if the company does not have that support, making surcharge distinctions based on breed is not allowed.

Vermont, Connecticut, and Massachusetts take a similar approach through desk drawer rules. Vermont’s Department of Financial Regulation indicates in Objection Letters that the Department does “not allow the blanket exclusion of entire canine breeds.”⁷⁴ Similarly, Connecticut provides in its Objection Letters: “Exclusions for breed of dog exclusions are not acceptable in CT. The Department has allowed companies to file an endorsement that excludes coverage for a specifically named dog, but not breeds. The endorsement must be made optional. It may not be used as a mandatory endorsement.”⁷⁵ In addition, Massachusetts’ Base Checklist for Property and Casualty Insurance indicates as an additional casualty provision: “Dog exclusion language must specify the breeds deemed aggressive by the company and provide support for any excluded individual dog not within an aggressive breed. (NB: All excluded breeds/dogs must have a prior history of biting.)”⁷⁶

Maryland, although allowing use of breed lists, prohibits an insurer from canceling or refusing to underwrite or renew an insurance “risk or class of risk” unless the standards “are reasonably related to the insurer’s economic and business purposes.”⁷⁷ The Maryland Insurance Administration has interpreted this to mean that underwriting decisions must

be made solely on the basis of a reasonable application to relevant facts of underwriting principles, standards and rules that can be demonstrated objectively to measure the probability of a direct and substantial adverse effect upon losses or expenses of the insurer in light of the proposed rating plan or plans of the insurer then in effect.⁷⁸

The Administration further explains that the insurer has the burden to justify any refusal to underwrite with evidence, which might include state or national data, showing the relationship between the exclusion and the insurer’s economic and business purposes.⁷⁹ Therefore, to use a breed list, an insurer in Maryland must have a justifiable—perhaps even a statistically justifiable—basis for doing so.

A number of states have also introduced bills to ban breed discrimination in insurance policies.
Massachusetts recently considered House Bill 1038, which provides:

An insurance company offering homeowners insurance coverage or renters insurance coverage that issues a policy or contract insuring against liability for injury to a person or injury to or destruction of property arising out of the ownership or lease of residential property shall not refuse to issue or renew, cancel or charge or impose an increased premium or rate of such a policy or contract based in whole or in part upon the harboring of a specific breed of dog upon the property.⁸⁰

In addition, in 2018 Massachusetts passed a law requiring insurance carriers providing homeowners insurance coverage to provide data on dog-related incidents for a three-year period beginning January 1, 2019.⁸¹ Information to be provided for each incident includes, among other things, the breed or predominant breed of dog, training, past behavior, location of incident, provocation, and municipal designation as a dangerous dog.⁸²

Illinois introduced a bill that would allow insurance companies to cancel or impose increased premiums for dogs declared vicious under state law, but it would prohibit the same based on a dog’s breed alone.⁸³ New York also introduced a bill that would “prohibit[] insurers from canceling, refusing to issue or renew, or charging higher premiums for homeowners' insurance based on the breed of dog owned.”⁸⁴ Because the timing of the consideration of these bills coincided with the coronavirus pandemic, action on them has been delayed.

2. Nongovernmental Opposition

A number of high-profile non-governmental organizations also oppose breed discrimination, including veterinary organizations, lawyers, animal welfare organizations, and dog groups and encourage the public to challenge the use of breed discrimination.⁸⁵ A few of these organizations are discussed below.

Veterinarians
The American Veterinary Medical Association (AVMA) is very clear in its opposition to breed discrimination in the context of both BSL and homeowners insurance. With regard to BSL, the AVMA explains: 

Any dog can bite, regardless of its breed. It is the dog’s individual history, behavior, general size, number of dogs involved, and the vulnerability of the person bitten that determines the likelihood of biting and whether a dog will cause a serious bite injury. Breed-specific bans are a simplistic answer to a far more complex social problem, and they have the potential to divert attention and resources from more effective approaches.⁸⁶

The AVMA also notes the problems involved with BSL, including the difficulty of identifying dog breeds, the lack of data showing risk distinctions between breeds, euthanasia of dogs that do not present a high risk of harm, and the general ineffectiveness of the regulation.⁸⁷ The group further opposes BSL because it discourages responsible pet ownership of other breeds because BSL “give[s] a community a false sense of security” and ignores the value of the listed breeds as police, military, rescue and service dogs.⁸⁸

The AVMA asserts similar arguments as a basis for opposing breed discrimination in the insurance context. The AVMA also warns veterinarians about providing behavioral evaluations to insurers because temperament tests have limitations. As explained by Dr. Bonnie Beaver, an AVMA Executive Board member and certified veterinary behaviorist, “there are many situations in which a dog may behave aggressively, and temperament tests can't rule out the possibility of aggression.”⁸⁹ Instead, the AVMA encourages preventing dog bites through responsible pet ownership. To that end, the AVMA created a Task Force on Canine Aggression and Human Canine Interactions, which recommended a multipronged and multidisciplinary approach for communities that focuses on cooperation, investigation, data collection, education, and media control.⁹⁰ AVMA also recommends to people who have been impacted by insurance breed lists that they should work toward enacting state legislation that would prohibit breed discrimination by insurers.⁹¹

The American Veterinary Society of Animal Behavior (AVSAB) opposes BSL for the same reasons as the AVMA—misidentification, ineffectiveness, and false security—and reduction of dog bites requires preventive measures rather than breed discrimination.⁹² In particular, AVSAB notes that the actual breed of a dog may not be known by the owner or even trained professionals because visual identification is unreliable, especially in the case of mixed breed dogs, whose DNA might indicate a certain breed makeup even though the dogs bear no actual resemblance to that breed.⁹³

In addition to challenging the scientific evidence purporting to support BSL, AVSAB points to other studies indicating aggression problems with non-listed breeds, showing that “[b]reed alone is not predictive of the risk of aggressive behavior” and that dog and owner behavior must be viewed on a case-by-case basis.⁹⁴ Therefore, rather than back BSL, the organization supports the use of dangerous dog laws, which focus on individuals and not breeds.⁹⁵

A number of state and local veterinary medical associations also oppose breed discrimination.⁹⁶

Lawyers
Like the veterinary organizations and for the same reasons, the American Bar Association (ABA) has taken a strong position against breed discrimination. In 2012, the ABA approved a resolution that “urges all state, territorial, and local legislative bodies and governmental agencies to adopt comprehensive breed-neutral dangerous dog/reckless owner laws that ensure due process protections for owners, encourage responsible pet ownership and focus on the behavior of both dog owners and dogs, and to repeal any breed discriminatory or breed specific provisions.”⁹⁷

Animal Welfare Organizations
The American Society for the Prevention of Cruelty to Animals (ASPCA) also condemns breed discrimination as an inappropriate measure for controlling the risk of harm from dogs. The core of the ASPCA’s argument relates to the balance between controlling the risk and the right to an individual’s choice of breed. In the ASPCA’s view, BSL is ineffective and discriminates against responsible dog owners and better results will ensue from enforcement of existing leash and dangerous dog laws. Further, the ASPCA castigates breed discrimination by the insurance
industry because use of breed lists “force[s] responsible home owners/pet guardians to choose between a beloved family pet and insurance for their home—a choice no one should be compelled to make.”⁹⁸

Best Friends Animal Society and the Humane Society of the United States (HSUS) share similar views. Best Friends asserts that even if dogs look a certain way, they do not act the same way, and it also challenges breed discrimination because it is not only ineffective but also costly.
To buttress its support, Best Friends provides the public with a calculator to determine the fiscal impact of breed discrimination on their communities.⁹⁹ HSUS additionally stresses the strain on shelters as dogs are forced out of their homes, limiting space for other dogs.¹⁰⁰

The margin-left: 30px; (NACA), a professional group for animal control officers, also takes an opposing position on breed discrimination and recommends a four pronged approach to dog bite prevention derived from a recommendation by the National Canine Research Council:

1. Focus on the behavior of the dog and the behavior of the owner.
2. Improved husbandry practices, better understanding of canine behavior, education of parents regarding safety around dogs.
3. Consistent enforcement of dangerous dog/reckless owner ordinances in communities.
4. Effective laws that hold owners responsible for custody and control of dogs regardless of breed or type.¹⁰¹

Dog-Specific Groups
The American Kennel Club (AKC) opposes breed discrimination in terms of BSL and by insurance companies for the same reasons as veterinarians and animal welfare organizations. With regard to BSL, the AKC’s objection focuses on the vagueness of breed descriptions, society’s changing perception of which dogs are considered dangerous, and the lack of evidence showing a distinction in aggression between breeds.¹⁰² It also criticizes the inappropriate focus of breed discrimination on the breed rather than on the dog owner responsible for a dog’s behavior.¹⁰³ In addition, according to the AKC, since any breed can be trained to be aggressive, imposition of breed bans will merely place the aggressiveness training on other breeds.¹⁰⁴ Instead, the AKC encourages the use of dangerous dog laws that focus on the individual dog and not the breed.¹⁰⁵

As far as homeowners insurance goes, the AKC supports consideration of the actual risk of a specific dog and not merely its breed. According to its policy statement:

The American Kennel Club believes that insurance companies should determine coverage of a dog-owning household based on the dog’s deeds, not the dog’s breed. If a dog is a well-behaved member of the household and the community, there is no reason to deny or cancel coverage. In fact, insurance companies should consider a dog an asset, a natural alarm system whose bark my deter intruders and prevent potential theft.¹⁰⁶

Like others, the Association of Professional Dog Trainers opposes breed discrimination, asserting in its position statement: “Canine temperaments are widely varied, and behavior cannot be predicted by physical features such as head shape, coat length, muscle to bone ratio, etc. The only predictor of behavior is behavior.”¹⁰⁷ And the National Canine Research Council considers BSL to be an ineffective policy and too difficult to enforce. As noted earlier, the group recommends a multi-factor alternative to breed discrimination and rejects the notion that certain breeds are more likely to bite than others.¹⁰⁸

IV. Recommendations

Absent regulation, insurers are authorized to define their risk tolerance by selecting or avoiding certain risks. However, as discussed above, underwriting guidelines that allow denial or exclusions from coverage for a homeowner’s dog merely because the dog’s breed is included on the company’s unsubstantiated breed list is unfair and conflicts with the NAIC’s concern with the availability and affordability of insurance, which was expressed repeatedly during the 2020 Summer Meeting.¹⁰⁹

Arbitrary breed lists targeting pet dogs are inconsistent with this concern because denial or limitation of a policy based on breed means that an affected homeowner will have no, fewer or more expensive options. Because of the importance of available and affordable insurance, a risk that will burden only some consumers—especially consumers who may already suffer from lack of information and means—should be justified based on reliable data. If breed lists are not a justified risk, then insurers should not limit access to insurance for owners of pets on those lists.

The NAIC is in the best position to seek this justification and, indeed, must do so. The NAIC’s Unfair Trade Practices Act considers certain activities to be unfair trade practices if an insurer has committed the activity “with such frequency to indicate a general business practice to engage in that type of conduct.”¹¹⁰ Unfair trade practices includes “unfair discrimination,” which itself includes:

Making or permitting any unfair discrimination between individuals or risks of the same class and of essentially the same hazard by refusing to insure, refusing to renew, canceling or limiting the amount of insurance coverage on a property or casualty risk solely because of the geographic location of the risk, unless such action is the result of the application of sound underwriting and actuarial principles related to actual or reasonably anticipated loss experience.¹¹¹

Use of breed discrimination lists to deny or limit homeowners insurance is a general business practice that can have an impact on geographic location, especially as it relates to minority communities. Therefore, the NAIC should determine whether the use of breed lists is supported by actuarial data or related to actual or reasonably anticipated losses. To that end, recommendations are provided below to address this issue.

A. Data Call

One of the main objections to breed lists is the lack of actuarial data to support discrimination between breeds. To address this issue, the NAIC should issue a data call to collect state-specific and national data on the risks associated with listed breeds. In this way, the NAIC and the insurance industry can make fact-based decisions on the validity of breed lists. 

1. Basis for Creation of Breed Lists

To understand an insurance company’s own justification for a breed list, it is important to have
information on the reasons the insurance company adopted the list in the first place. Relevant
data to this issue include:

• Identity of each underwriting carrier that uses a breed list, dog breeds included on that specific carrier’s list, and the year the list was created;
• Reasons for the carrier’s creation of the list;
• Underlying data used to identify which dogs should be placed on the list;
• Frequency of review of the list to determine whether changes would be appropriate (e.g., adding to or taking breeds off the list); and
• Actual changes to the breed list by each carrier, with dates, identifying what changes were made and why. 

2. Use of Breed Lists

Further data vital to understanding the discriminatory impact of breed lists on consumers include how insurance companies actually use the list when deciding whether to issue or renew a policy, create an exclusion, or impose a liability limitation. Relevant data for this issue include, for each carrier:

• Guidance provided to ensure consistency in application of the breed list in underwriting guidelines;
• Procedures for determining and/or confirming breed identity;
• Instructions and/or training provided to adjusters in identifying breeds on the list;
• Number of policies and the geographic locations of homeowners that were denied, cancelled, nonrenewed, given an exclusion and/or imposing a liability limitation based on the breed list; and
• Procedures for informing consumers if there are alternative insurance options for homeowners with dogs on a breed list.

3. Dog Bite Claims

To determine whether the breed lists have actually reduced the number and cost of dog bite claims to any significant degree, the NAIC should call for statistics on dog bite claims before and after creation of the dog breed list. Information on the carrier’s classification of claims would also be relevant. Collected data should include:

• Number and value of claims in years prior to use of a breed list and years after;
• Identification of claims by breed (to determine impact of non-listed breeds vs. listed breeds)¹¹² and circumstances surrounding bite (e.g., unprovoked vs. provoked);
• Explanation of whether bite claims are tracked and coded and whether each claim is documented as just a medical payment claim, just a liability claim, or both;
• Explanation of what incidents are classified as bite claims (e.g., a dog bite versus a dog that knocks over a person causing injury); and
• Number of claims settled and reasons for settlement.

Massachusetts’ Dog Claim Reporting Form, included here as Appendix 1, sets out other claim specific data that could be collected for study.

B. Interim Measures

Because of the implications that a breed list presents to uninformed consumers, people of color, and consumers of low and moderate means, the NAIC should consider breed lists to be presumptively unfair and discriminatory until proven otherwise. Therefore, during the data call, the NAIC should take measures to protect these consumers. Measures that the NAIC can take include issuing the following:

1. Moratorium

During the period of the data call, it would be appropriate for the NAIC to request a moratorium on the use of breed lists by insurance carriers until data has been collected, analyzed, and determined to provide a justification for breed lists.

2. Notice and Education

If a moratorium is not requested or complied with, the NAIC could require carriers to notify consumers who are denied coverage or have their policies subject to exclusion or limitations based on owning a specific breed of dog and provide materials educating the consumer on alternative options. Such notifications might include:

• Basis for denial of coverage, exclusion or limitation if the denial, exclusion or limitation was based on ownership of a specific breed of dog;
• Full disclosure of exclusions or limitations placed prominently on the Declarations Page if the exclusions or limitations are based on ownership of a specific breed of dog; and
• Disclosure of any potential gap in coverage between a consumer’s homeowners policy and umbrella policy due to a dog on the breed list.

Information that should be provided consumers who are denied coverage or are subject to an exclusion or limitation based on dog breed includes:

• Information regarding alternative options, if any, for coverage of homeowners with dogs on the breed list;
• For carriers who sell online, a prompt after a denial, exclusion or limitation that recommends that the consumer contact an agent who can help if alternative coverage is available;
• Information on coverage buy back options; and
• Information on coverage by the FAIR plan or non-admitted carriers.

C. Respond to Data Call

Once data from the data call has been collected, the NAIC should validate “the legitimacy of all underwriting variables.” To that end, the Society of Actuaries may be a key component in the review, as the Society has committed itself to “provide trusted and objective actuarial research, analysis and insight on important societal issues.”¹¹³

When review of the data is complete, the NAIC should make a determination on whether use of breed lists is a justifiable basis for denying coverage or renewal or creating an exclusion or limitation on a homeowners policy. If the NAIC finds there is no justification, the NAIC should take measures to end the use of discriminatory breed lists.

Appendix

¹ Race & Insurance, NAIC Summer National Meeting 2020, trans. 17:37 – 17:52 (Aug. 17, 2020), youtube.com/watch?v=H6NwNJ0AmtE&t=1398s (hereinafter Race & Insurance Transcript).
² Reference to “homeowners policy” in this paper refers to both homeowners and renters policies.
³ See Kim Kavin, Dog Adoptions and Sales Soar During the Pandemic, WASH. POST, Aug. 12, 2020, https://www.washingtonpost.com/nation/2020/08/12/adoptions-dogs-coronavirus/.
⁴ Carriers such as Allstate, GEICO, Nationwide, and Travelers discriminate against certain breeds of dogs. State Farm appears to be the only large insurance carrier that does not use a breed list or discriminate based on the breed of a homeowner’s pet dog. Instead, State Farm looks to the particular dog’s bite history to determine whether to exclude the dog from coverage. See Mark Fitzpatrick, Restricted Dog Breeds for Homeowners Insurance,
ValuePenguin (Aug. 19, 2020), https://www.valuepenguin.com/homeowners-insurance-restricted-dog-breeds.
⁵ Importantly, the generic term “pit bull” does not refer to a recognized breed. Although the United Kennel Club recognizes the specific “American Pit Bull Terrier,” the American Kennel Club does not and instead recognizes the
American Staffordshire terrier, Staffordshire bull terrier, and bull terrier. Compare United Kennel Club, American Pit Bull Terrier (last accessed Nov. 16, 2020), https://www.ukcdogs.com/american-pit-bull-terrier, with Am. Kennel
Club, American Staffordshire Terrier (last accessed Nov. 16, 2020), https://www.akc.org/dog-breeds/american-staffordshire-terrier/.
⁶ The breeds listed here are compiled from a review of filings of the fifty largest homeowners insurance companies. See Amy Danise, The Prohibited Lists: Home Insurance Dog Breed Restrictions, Everquote (Apr. 1, 2019),
https://www.everquote.com/blog/home-insurance/homeowners-insurance-dog-breed-restrictions/.
⁷ Les Masterson, Homeowners Insurance Survey 2018: Almost Half Don’t Understand Liability Home Insurance, A Quarter Have No Idea How Much They Have, Insurance.com (July 5, 2018),
https://www.insurance.com/coverage/homeowners-insurance-survey-2018.
⁸ Id.
⁹ Ins. Information Inst., Spotlight On: Dog Bite Liability (Oct. 16, 2020), https://www.iii.org/article/spotlight-on-dog-bite-liability (hereinafter Spotlight on Liability) (citing American Pet Products Association’s 2019-2020 National Pet Owners Survey). The same APPA survey indicated that there are 89.7 million dogs owned in the United States. Id.
¹⁰ For example, a person might spontaneously decide to adopt a stray dog or might be obligated to take a dog because a family member or friend has become ill or died.
¹¹ Katrina E. Holland, Acquiring a Pet Dog: A Review of Factors Affecting the Decision-Making of Prospective Dog Owners, 9 ANIMALS 124 (2019) (noting that factors influencing a person’s decision to acquire a dog include
household composition, dwelling type, socioeconomic status, ethnicity, prior dog ownership, dog appearance and age, breed popularity, and pre-acquisition behavior).
¹² See id.
¹³ Masterson, supra note 7.
¹⁴ 42 U.S.C. § 3604.
¹⁵ 24 C.F.R. § 100.70(d)(4). HUD’s regulation was upheld in National Fair Housing Alliance v. Prudential Ins. Co., 208 F. Supp. 2d 46, 57 (D.D.C. 2002) (concluding after extensive analysis that “HUD's interpretation of section
3604 as set forth in its regulations is a permissible construction of the FHA, and, accordingly, that the scope of section 3604 extends to the provision of homeowners insurance”).
¹⁶ See, e.g., United Farm Bureau Mut. Ins. Co. v. Metropolitan Human Relations Comm’n, 24 F.3d 1008, 1015 – 16 (7th Cir. 1994); Wai v. Allstate Ins. Co., 75 F. Supp. 2d 1, 6 (D.D.C. 1999); Lindsey v. Allstate Ins. Co, 34 F. Supp.
2d 636, 641 – 43 (W.D. Tenn. 1999); Dunn v. Midwestern Indem. Mid-Am. Fire & Cas., 472 F. Supp. 1106, 1109, 1112 (S.D. Ohio 1979).
¹⁷ See Race & Insurance Transcript, supra note 1.
¹⁸ Id. at 24:46 (statement of Dr. Robert Klein, Robert W. Klein & Associates).
¹⁹ Id. at 81:25 (statement of Birny Birnbaum, Executive Director, Center for Economic Justice).
²⁰ Id. at 82:01 – 82:35 (statement of Birny Birnbaum, Executive Director, Center for Economic Justice).
²¹ Id. at 34:34 – 35:17 (statement of Dr. Robert Klein, Robert W. Klein & Associates).
²² Homeowners Insurance Discrimination, Hearing before the Committee on Banking, Housing and Urban Affairs, U.S. Senate, 103d Cong., 2d sess., 131 – 32 (May 11, 1994) (Statement of the NAACP Legal Defense and
Educational Fund, Inc. (May 18, 1994)); see also Douglas A. Wissoker, Wendy Zimmermann, George Galster, Testing for Discrimination in Home Insurance, URBAN INSTITUTE ELEVATE THE DEBATE (Dec. 1997),
https://www.urban.org/sites/default/files/publication/67051/307555-Testing-for-Discrimination-in-Home-Insurance.pdf (finding discrimination in the provision of housing insurance in the context of Hispanic consumers in Phoenix, Arizona and Black consumers in three boroughs of New York City).
²³ 576 U.S. 519 (2015). 
²⁴ Id. at 40. 
²⁵ 24 C.F.R. pt. 100. HUD’s disparate impact regulations were amended in September 2020. The amendments and preamble to the final rule are available at https://www.hud.gov/sites/dfiles/ENF/documents/6111-F-03%20Disparate%20Impact%20Final%20Rule%209-3-20%20FOR%20POSTING.pdf. 
²⁶ Ann Linder, The Black Man’s Dog: The Social Context of Breed Specific Legislation, 25 ANIMAL L. REV. 51, 59 –60 (2018).
²⁷ Id. at 60 & n.55. 
²⁸ Id. at 64. 
²⁹ Ann L. Schiavone, Real Bite: Legal Realism and Meaningful Rational Basis in Dog Law and Beyond, 25 WM. & MARY BILL OF RIGHTS J. 65, 111 – 12 (2016).
³⁰ Erin Tarver, The Dangerous Individual(’s) Dog: Race, Criminality and the ‘Pit Bull,’ 55 CULTURE, THEORY & CRITIQUE 273, 281 (2013) (citations omitted); see also id. at 273 (“The concomitant revulsion toward both
dogfighting and ‘pit bulls’ suggests an expression of fear of a perceived threat to normative whiteness, insofar as these ‘dangerous’ dogs are figured as carriers of the contagion of racial abnormality.”).
³¹ HUD Guidance, Assessing a Person’s Request to Have an Animal as a Reasonable Accommodation Under the Fair Housing Act 14, Jan. 28, 2020, https://www.hud.gov/sites/dfiles/PA/documents/HUDAsstAnimalNC1-28-2020.pdf?utm_medium=email&utm_source=govdelivery.
³² See Larry Cunningham, The Case Against Dog Breed Discrimination by Homeowners’ Insurance Companies, 11
CONN. INS. L.J. 1, 14 – 15 (2004-2005). 
³³ Id. at 19 – 20. 
³⁴ The CDC studied a 20-year period from 1979 to 1998 for fatal dog bites, id. at 17 – 18, and for non-fatal dog bites, years that fell in the 1990s and 2003. Id. at 20 – 24. For a detailed discussion of the CDC studies, including their limitations, see id. at 17 – 24. 
³⁵ For example, the fatal dog bite research relied in part on the accuracy of news reports culled from a computer search, which in turn relied on the ability of the researchers to identify search terms sufficient to find all news reports of fatal dog bites. Id. at 18. As explained by Dr. Julie Gilchrist, CDC epidemiologist: “‘No centralized reporting system for dog bites exists, and incidents are typically relayed to a number of entities, such as the police, veterinarians, animal control, and emergency rooms, making meaningful analysis nearly impossible. Moreover, a pet dog that bites an owner or family member might go unreported if the injury isn't serious.’” R. Scott Nolan, The Dangerous Dog Debate, JAVMA NEWS, Nov. 1, 2017, https://www.avma.org/javma-news/2017-11-15/dangerous-dog-debate (quoting Dr. Gilchrist). Because of this unreliability, the CDC stopped gathering breed data on dog bite fatalities in 1998, the last year included in its studies. Id. 
³⁶ Common limitations of studies include the shortcomings of data collection (e.g., reliance on media reports, limited emergency room data, or unvalidated self-reporting), exclusion of data from consideration, misidentification of the dog breed, and lack of data indicating the prevalence of a certain breed within the location being studied. See Cunningham, supra note 32, at 20 – 27. 
³⁷ Id. at 23 – 27. 
³⁸ Grant Barnhard, Breed-Specific Legislation: The Pitfalls of Pit Bull Prohibitions, 4 MID-ATLANTIC J. L. & PUB. POL’Y 59 (2018) (citing B. Beaver et al., A Community Approach to Dog Bite Prevention: American Veterinary Medical Association Task Force on Canine Aggression and Human-Canine Interactions, 218 J. AM. VETERINARY MED. ASS’N 1732, 1733 (2001), https://www.avma.org/sites/default/files/2020-03/dogbite.pdf; Malcolm Gladwell, Troublemakers: What Pit Bulls Can Teach Us About Profiling, NEW YORKER, Feb. 6, 2006, https://www.newyorker.com/magazine/2006/02/06/troublemakers-malcolm-gladwell; Gary Patronek et al., Co-Occurrence of Potentially Preventable Factors in 256 Dog Bite-Related Fatalities in the United States (2000-2009), 243 J. AM. VETERINARY MED. ASS’N 1726, 1726 – 36 (2013)
³⁹ Id. at 85 – 86 (discussing Deborah Duffy et al., Breed Differences in Canine Aggression, 114 APPLIED ANIMAL BEHAVIOR SCI. 441, 445 (2008), http://140.122.143.143/yuyinghs/yuyinghsu/papers/DuffyHsuSerpell2008.pdf).
The referenced study indicated that, although Rottweilers and Doberman pinschers showed a higher degree of aggression to strangers, the same was true of Yorkshire terriers and poodles. Id. at 85. Akitas, boxers, Australian
cattle dogs, German shepherds and pit bulls showed a higher level of dog-directed aggression, and Basset hounds, beagles and cocker spaniels showed the highest aggression to owners. Id. And the smaller dogs in the study—
Dachshunds, Chihuahuas, and Jack Russell terriers—showed consistently high aggression levels to all targets. Id. Similarly, dog bite statistics from the City of Denver for 2019 indicated that Labrador retrievers, Chihuahuas, border collies, and boxers ranked in the top ten dogs for bites by breed, with Labrador retrievers ranking at the top. See Erin Powell, Everything You Wanted to Know About Dog Bites in Denver, 9 News, Feb. 18, 2020,
https://www.9news.com/article/news/local/next/dog-bites-denver-2019/73-81bbc2d0-2c10-4600-bc03-5b46391f2daa.
⁴⁰ See Cunningham, supra note 32, at 6 – 7. 
⁴¹ Bob Egelko, Appeal for Murder Rap in Dog-Maul Case / Attorney General Says Judge Had No Right to Let Knoller Off So Easy, S.F. CHRON., Apr. 12, 2003, at A15, https://www.sfgate.com/bayarea/article/Appeal-formurder-rap-in-dog-maul-case-Attorney-2622631.php; Jane Meredith Adams, Officials Working to Build Case in Fatal Dog Attack, CHI. TRIB., Mar. 3, 2001, https://www.chicagotribune.com/news/ct-xpm-2001-03-03-0103030092-story.html. 
⁴² The defendants were lawyers who, three days after the dog attack, formally adopted the inmate—a client—as their son. Egelko, supra note 41, at A15. 
⁴³ Id. at A15. 
⁴⁴ Adams, supra note 41. 
⁴⁵ See Jeffrey Sacks, Leslie Sinclair, Julie Gilchrist, Gail Golab, & Randall Lockwood, Breeds of Dogs Involved in Fatal Human Attacks in the United States between 1979 and 1998, 217 J. AM. VETERINARY MED. ASS’N 836, 837 –38 (2000) (hereinafter CDC Study) (listing breeds in fatal dog attacks between the designated period). 
⁴⁶ Cf. Cunningham, supra note 32, at 12 (describing a couple who lost their homeowners insurance because they owned a Rottweiler who was “deaf, partially blind, and ha[d] arthritis”). 
⁴⁷ Spotlight on Liability, supra note 9. 
⁴⁸ Id. 
⁴⁹ Id. 
⁵⁰ Insurance Information Institute, Facts + Statistics: Homeowners and Renters Insurance, Oct. 27, 2020, https://www.iii.org/fact-statistic/facts-statistics-homeowners-and-renters-insurance.
⁵¹ Cunningham, supra note 32, at 15 (quoting Insurance Information Institute, Dog Bite Liability, at http://www.iii.org/media/hottopics/insurance/dogbite/).
⁵² See Patronek, supra note 38, at 1726, 1726 (2013) (concluding that most dog-bite related fatalities “were characterized by coincident, preventable factors; breed was not one of these” and that the study “supported previous
recommendations for multifactorial approaches, instead of single-factor solutions such as breed-specific legislation, for dog bite prevention”). 
⁵³ Other restrictions may include exclusion from the locality, increased insurance requirements for ownership, mandatory spay/neuter requirements, and classification as a “dangerous dog” and subject to dangerous dog
requirements relating to secured enclosures, signage, and leashing and muzzling when outside its enclosure. See Cunningham, supra note 32, at 8.
⁵⁴ Id. at 17 – 27. Professor Cunningham, in a review of scientific studies on dog bites, states: Numerous scientific studies have attempted to identify the number of annual dog bites, the dogs most likely to bite, the people most likely to be bitten, and the circumstances under which bites are most likely to occur. Such studies have not reached a uniform consensus and have left us with more questions than answers. Even the studies that have attempted to report on breeds’ proclivity to bite have cautioned that their research is incomplete and should not be used to justify breed discrimination by legislatures or insurers. Id. at 17. 
⁵⁵ Am. Veterinary Med. Ass’n, Why Breed-Specific Legislation Is Not the Answer (last accessed Oct. 12, 2020), https://www.avma.org/resources/pet-owners/why-breed-specific-legislation-not-answer. 
⁵⁶ K.R. Olson, J.K. Levy, B. Norby, M.M. Crandall, J.E. Broadhurst, S. Jacks, R.C. Barton, M.S. Zimmerman, Inconsistent Identification of Pit Bull-type Dogs by Shelter Staff, 206 VETERINARY J. 197 – 202 (2015) (noting that “[o]ne in five dogs genetically identified with pit bull heritage breeds were missed by all shelter staff” and that “[o]ne in three dogs lacking DNA for pit bull heritage breeds were labeled pit bull-type dogs by at least one staff member”).
⁵⁷ Christy Hoffman, Natalie Harrison, London Wolff & Carri Westgarth, Is That Dog a Pit Bull? A Cross-Country Comparison of Perceptions of Shelter Workers Regarding Breed Identification, 17 J. APPLIED ANIMAL WELFARE SCI. 322-39 (2014), https://www.tandfonline.com/doi/full/10.1080/10888705.2014.895904. 
⁵⁸ See Cunningham, supra note 32, at 32 – 34. 
⁵⁹ See id. at 42. 
⁶⁰ Am. Bar Ass’n, Resolution 100, at 8 (Aug. 2012), ohioanimaladvocates.org/wp-content/uploads/2019/06/ABA-Resolution-Against-BSL-201208.pdf. 
⁶¹ CDC Study, supra note 45, at 836.
⁶² Ban and Outlaw Breed Specific Legislation (BSL) in the United States of America on a Federal Level!, WE THE PEOPLE (Dec. 9, 2012), https://petitions.obamawhitehouse.archives.gov/petition/ban-and-outlaw-breed-specific-legislation-bsl-united-states-america-federal-level-0/. 
⁶³ Children, the elderly, and postal carriers are the most frequent victims of dog bites, with children being the most common victims and victims most likely to suffer severe injury. U.S. Postal Serv., National Dog Bite Prevention
Week, POSTAL BULLETIN 22490, at 3 (Mar. 29, 2018), https://about.usps.com/postal-bulletin/2018/pb22490/pb22490.pdf. 
⁶⁴ Indeed, USPS Manager of Safety Linda DeCarlo spoke out against BSL at a press conference on dog safety, noting that “[a]ny breed can bite.” Madison Fantozzi, Postal Service Sinks Its Teeth into Dog-Breed Legislation,
SCRIPPS HOWARD FOUND. Wire (May 15, 2014), shfwire.com/postal-service-sinks-teeth-into-dog-breed-legislation/.
⁶⁵ U.S. Postal Serv., Community Activities: National Dog Bite Awareness Week (last accessed Nov. 2, 2020), https://about.usps.com/what/corporate-social-responsibility/activities/dog-bite-awareness.htm. 
⁶⁶ Kaitlyn Mattson, Postal Service Implements Dog Bite Prevention Program (July 8, 2020), https://www.avma.org/javma-news/2020-08-01/postal-service-implements-dog-bite-prevention-program. 
⁶⁷ Nat’l Canine Res. Council, Breed-Specific Legislation Is on the Decline (last accessed Nov. 7, 2020), https://nationalcanineresearchcouncil.com/breed-specific-legislation-on-the-decline/
⁶⁸ ASPCA, What Is Breed-Specific Legislation (Apr. 1, 2020), https://www.aspca.org/animal-protection/public-policy/what-breed-specific-legislation. States that prohibit or limit BSL include Arizona, California (except for
breed-specific spay/neuter requirements), Colorado (with some home rule city exceptions), Connecticut, Delaware, Florida, Illinois, Maine, Massachusetts, Minnesota, Nevada, New Jersey, New York, Oklahoma, Pennsylvania,
Rhode Island, South Carolina, South Dakota, Texas, Utah, Virginia, and Washington. Id.; Best Friends Animal Soc’y, Anti-Breed-Specific Legislation by State (last accessed Nov. 6, 2020), https://resources.bestfriends.org/article/anti-breed-specific-legislation-state. 
⁶⁹ See, e.g., FLA. STAT. § 767.14; 510 ILL. COMP. STAT. 5/24; MINN. STAT. 347.51.; TEX. HEALTH & SAFETY CODE § 822.047.
⁷⁰ REV. CODE WASH. § 16.08.110(1)(a). 
⁷¹ 3 PA. CONS. STAT. § 459-507-A. 
⁷² DEP’T INS. & FIN. SERV., BULLETIN 2019-20-INS (Oct. 31, 2019), https://www.michigan.gov/documents/difs/Bulletin_2019-20-INS_670400_7.pdf (citations omitted). 
⁷³ Id. 
⁷⁴ See, e.g., Objection Letter to Farm Family Ins. Co. (Oct. 14, 2019), SERFF Tracking No. ANPC-132098513. 
⁷⁵ See, e.g., Objection Letter to American Reliable Ins. Co. (Apr. 17, 2020), SERFF Tracking No. PENN132312596.
⁷⁶ Mass. Div. of Ins., Base Checklist for Property and Casualty Insurance (Jan. 2020) (emphasis added)
⁷⁷ MD. ANN. CODE § 27-501(a)(2). 
⁷⁸ Homeowner’s/Renter’s Insurance—Dog Bite Liability Exclusion for Certain Breeds of Dogs, MD. INS. ADMIN. BULL. 15-25 (Sept. 28, 2015), https://insurance.maryland.gov/Insurer/Documents/bulletins/15-25-homeowners-renters-insurance-dog-bite-liability.pdf. 
⁷⁹ Id. 
⁸⁰ H.B. 1038, 191st Mass. Gen. Ct., 2d Sess. (2019-2020). 
⁸¹ An Act to Protect Animal Welfare and Safety in Cities and Towns, ch. 219, Mass. Acts of 2018, § 32. 
⁸² Mass. Dog Claim Reporting—Instructions and Template (2019), https://www.mass.gov/doc/statutory-dog-claim-reporting.
⁸³ S.B. 2462, 101st Ill. Gen. Assembly, Reg. Sess. (2019-2020). 
⁸⁴ S.B. 7924, N.Y. State Leg., Reg. Sess. (2020). Other states that have considered similar legislation in previous years include Connecticut, Maine, and Maryland. See H.B. 5443, Conn. Gen. Assembly (2016); L.D. 1192, 126th Maine Leg., 1st Sess. (2013); S.B. 647, Md. Gen. Assemb., Reg. Sess. (2019). L.D. 1192, 126th Maine Leg., 1st Sess. (2013).
⁸⁵ See, e.g., Humane Soc’y of the U.S., Repealing Breed-Specific Legislation: Moving Beyond Breed to Save Dogs
and Strengthen Communities (last accessed Nov. 8, 2020), https://humanepro.org/sites/default/files/documents/repealing-breed-specific-legislation.pdf. 
⁸⁶ Am. Veterinary Med. Ass’n, supra note 55. 
⁸⁷ Id. 
⁸⁸ Id. 
⁸⁹ Id. 
⁹⁰ Am. Veterinary Med. Ass’n, A Community Approach to Dog Bite Prevention, 218 J. AM. VETERINARY MED. ASS’N 1732 (June 1, 2001), https://www.avma.org/sites/default/files/2020-03/dogbite.pdf.
⁹¹ Id. 
⁹² Am. Veterinary Soc’y of Animal Behavior, Position Statement on Breed Specific Legislation (last accessed Nov. 6, 2020), https://avsab.org/wp-content/uploads/2018/03/Breed-Specific_Legislation-download-_8-18-14.pdf. 
⁹³ Id. 
⁹⁴ Id. The non-listed breeds included Jack Russell and West Highland White terriers, Labrador, golden and Chesapeake Bay retrievers, American cocker spaniels, and boxers. Id. 
⁹⁵ Id. 
⁹⁶ See, e.g., Cal. Veterinary Med. Ass’n, CVMA Policy on Breed-Specific Legislation (last accessed Nov. 7, 2020), https://cvma.net/resources/cvma-policies/canine-policies/cvma-policy-on-breed-specific-legislation/; Chi.
Veterinary Med. Ass’n, CVMA’s Position Statement on Breed Specific (Pit Bull) Legislation (last accessed Nov. 7, 2020), https://www.chicagovma.org/cvma-s-position-statement-on-breed-specific-pit-bull-legislation/; South
Florida Veterinary Medical Association Supports Repeal of Miami-Dade County Pit Bull Ban, Local10.com (last accessed Nov. 7, 2020), https://www.local10.com/news/2012/07/25/south-florida-veterinary-medical-association-supports-repeal-of-miami-dade-county-pit-bull-ban/. 
⁹⁷ Am. Bar Ass’n, Resolution 100 (Aug. 2012), ohioanimaladvocates.org/wp-content/uploads/2019/06/ABA-Resolution-Against-BSL-201208.pdf.
⁹⁸ ASPCA, Position Statement on Breed-Specific Bans (last accessed Nov. 7, 2020), https://www.aspca.org/about-us/aspca-policy-and-position-statements/position-statement-breed-specific-bans. 
⁹⁹ Best Friends Animal Soc’y, Breed-Specific Legislation Cost Calculator (last accessed Nov. 7, 2020), https://resources.bestfriends.org/article/breed-specific-legislation-bsl-cost-calculator. 
¹⁰⁰ Humane Soc’y of the U.S., Repealing Breed-Specific Legislation (last accessed Nov. 7, 2020), https://humanepro.org/page/repealing-breed-specific-legislation. 
¹⁰¹ Nat’l Animal Care & Control Ass’n, NACA Is Opposed to Breed Specific Legislation (last accessed Nov. 7, 2020), https://www.nacanet.org/wp-content/uploads/2020/10/BSL-Guideline-1.pdf. 
¹⁰² Am. Kennel Club, Why Breed Bans Affect You (Feb. 1, 2018), https://www.akc.org/expert-advice/news/breed-bans-affect/. 
¹⁰³ Id. The AKC states: “A study published by the Journal of Animal and Veterinary Advances found that owner behavior has a direct impact on dog aggression and personality. The study of approximately 50 purebred breeds
concluded that the time an owner spends caring for and training a dog is inversely correlated to the level of aggressive behavior the dog exhibits.” Id. (citing Factors Links to Dominance Aggression in Dogs, 8 J. ANIMAL &
VETERINARY ADVANCES 336 (2009)). 
¹⁰⁴ Id. The AKC agrees that BSL is costly, but ineffective, fills animal shelters, and results in increased euthanasia
rates.
¹⁰⁵ Am. Kennel Club, Canine Legislation Position Statement: “Dangerous Dog” Control Legislation (last accessed Nov. 7, 2020), http://images.akc.org/pdf/canine_legislation/position_statements/Dangerous_Dog_Control_Legislation.pdf.
¹⁰⁶ Am. Kennel Club, Canine Legislation Position Statement: Homeowners’ Insurance and Dangerous Dogs (last accessed Nov. 7, 2020), http://images.akc.org/pdf/canine_legislation/position_statements/Homeowners_Insurance_and_Dangerous_Dogs.pdf. 
¹⁰⁷ Ass’n of Prof. Dog Trainers, Breed Specific Legislation Position Statement (2001), https://apdt.com/wp-content/uploads/2017/01/PositionStatement.BreedSpecificLegislation_and_FAQs.pdf.
¹⁰⁸ Nat’l Canine Res. Coun., Ineffective Policies (April 11, 2016), https://www.nationalcanineresearchcouncil.com/public-policy/ineffective-policies. 
¹⁰⁹ See, e.g., Race & Insurance Transcript, supra note 1, at 39:34 – 41:45 (statement of Dr. Robert Klein, Robert W. Klein & Associates); id. at 42:17 – 42:44 (statement of Chlora Lindley-Myers, Director, Missouri Department of
Insurance). 
¹¹⁰ Unfair Trade Practices Act, NAIC Model Laws, Regulations, Guidelines and Other Resources, § 3.B (Jan. 2004). 
¹¹¹ Id. § 4.G(3) (emphasis added).
¹¹² Because small dogs are often picked up when handling, there is a higher likelihood for a bite to the face. Facial bites often result in multiple, expensive reconstructive surgeries, and from a claim perspective, facial scars have the highest value of any scar. 
¹¹³ Soc’y of Actuaries, 2017-2021 Strategic Plan 5 (Nov. 2016)