A new study looking at potential unintended bias in auto insurance finds that while black and Latino drivers pay more for coverage than other drivers, they also have more claims and driving violations.
video
ملاحظة:
إذا لم تشتغل الحلقة إختر سيرفر 1 أو 2
The study authors agree that higher losses and poorer driving records help explain the higher insurance costs for black and Latino drivers, and they acknowledge that they can’t answer with confidence without further research into why these drivers have more claims and driving errors.
But the authors also point out that other characteristics related to classification or policies, certain deductibles, financial status, infrastructure, law enforcement, and other factors may contribute to higher claims and premiums for blacks and Latinos. They argue that these other factors are worth studying in the future.
The report, “Assessing Unintentional Bias in Private Passenger Auto Insurance,” published by the District of Columbia’s Department of Insurance, Securities, and Banking, analyzed data from single-vehicle/single-driver private passenger auto insurance companies in the District of Columbia—a group that represents 64 percent of all auto policies in the district. Researchers also interviewed industry representatives, consumers, and subject matter experts.
“This investigation is an essential step toward understanding and addressing financial barriers that may unfairly impact certain racial groups,” said CIMA Commissioner Karima Woods. “Our goal is to ensure that D.C. residents are treated fairly, regardless of race, and that insurance premiums reflect actual driving risks.”
The insurance industry was quick to claim that the study confirms that auto insurance pricing is not unfairly discriminatory and is not based on race.
Premium Gap
According to the analysis, black drivers pay an average of 1.46 times more than white drivers, which equates to an annual premium gap of $326. Hispanic drivers pay 1.20 times more than white drivers, and Asian Pacific Islander drivers pay the same as white drivers.
In dollar terms, this translates to an average annual premium of $705 for white drivers, $1,031 for black drivers, $849 for Hispanic drivers, and $722 for non-resident U.S. drivers.
DISB sought to identify factors that might explain average premium gaps, specifically the average premium gap between blacks and whites. DISB looked at age, policy type, driving record, claims history, gender, and other factors to determine whether and how these factors affect insurance premiums.
Analysts looked at cumulative paid losses by race and found that black drivers, as a group, accounted for more claims on average than white, Hispanic or African American drivers. Black drivers’ average losses were 2.38 times those of white drivers. Thus, according to the study, black drivers are more expensive as a group than other groups.
Driving record, which the study says is a “practical and reasonable measure” because it is logically linked to risk, explains some of the gap in average insurance premiums between blacks and whites. Black drivers are more likely to be involved in accidents caused by driver error, and the gap is even larger among those with a drunk driving record.
But even after taking all these factors into account, there was still a $271 gap in average insurance premiums between black and white drivers.
Expected and unexpected results
A number of findings from the District of Columbia Department of Insurance, Securities and Banking’s Unintentional Bias Assessment Report were as expected: Except for younger drivers, younger drivers are less expensive; previous accidents on driving records are associated with higher premiums; premiums do not vary much by gender; and insurance costs are lower for older vehicles.
Other findings weren’t necessarily expected: The premium gap narrows as people and vehicles age; there are proportionally more older black drivers; black and Hispanic drivers have more minor accidents or accidents at their own fault; gender splits are similar regardless of race, though Hispanic drivers are more male; vehicle ages are similar for blacks and whites, slightly older for Hispanics, and newer for API.
Another interesting finding: Drivers who use agents pay less on average, and black drivers are less likely to use agents.
Correlation With Race
The authors hypothesize that since insurance companies’ price plans do not vary by applicants’ race, if outcomes differ on average by race, the rating characteristics must also differ. “In other words, there must be some factor and/or step in the rating plans that are related to race in a way that produces different average premiums by race for metropolitan drivers,” they write.
“What explains the difference in average insurance premiums by race? There is no evidence that black and Latino drivers are more likely to take risks while driving or have more accidents. However, there are a variety of factors that may explain this disparity in insurance premiums, including credit-based insurance scores and homeownership discounts. Additionally, some insurance companies set auto insurance premium rates using predictive models. Since premiums are determined mechanically, there may be some factors and/or steps in those models that correlate with race in a way that produces different average premiums by race for metropolitan drivers,” the study says.
Examples of such factors that insurance companies use but do not analyze include credit-based insurance scores; discounts on homeownership, education level, or multiline insurance; payment methods; and telecommunications.
The report describes these factors as “important discriminatory factors in shaping insurance premiums,” while questioning the use of such factors where “there are no rational explanations for why they predict accident risk; or where there is evidence that they are related to race in ways that would widen the gap in average insurance premiums between blacks and whites.”
For example, multi-line discounts reward customers who have more assets to insure. Full-payment discounts also benefit those who have the ability to pay up front. This could mean that white drivers might be required to pay an average of $705 up front, while black drivers might be required to pay an average of $1,031 up front.
Why More Claims?
The authors call for additional research into the types and causes of claims by black and Latino drivers to see if infrastructure or other changes might help reduce the claims disparity. Additionally, they call for investigation into why black and Latino drivers are more likely to commit driving violations, and whether this might be due to differences in enforcement rather than differences in driving practices.
Analysts also point out that some of the factors that influence costs also favor wealthy policyholders. For example, wealthy drivers are better able to pay some insured losses outright without filing claims, are more likely to have secure off-street parking, and are more likely to live in neighborhoods with better road conditions.
A previous study of market behaviour by the Motor Insurance Authority, which looked at specific insurers’ models used in motor insurance, concluded that the models “operated as intended and did not include any factors that would directly lead to bias in the rating process.” However, that review was unable to determine whether unintended bias existed.
The authors of this new report on unintended bias say additional study is needed into the types and causes of the much higher losses for black and Hispanic insureds to see if infrastructure or other changes might help reduce the claims disparity.
The Insurance Authority said it is exploring additional studies and hopes to develop a “balancing test” to assess whether certain factors should continue to be allowed in determining insurance premiums.
Insurer Reaction
Insurance groups responded to the study by claiming it proves that auto insurance rates are based on risk and loss criteria, not race. They also questioned how DISB identifies race in its analysis.
“It is important to understand that insurance companies do not collect race information or use race information in any way, including to set premiums,” the American Property & Casualty Insurance Association (APCI) said in a statement. “While DISB created and used racial categories in its reporting, insurance companies set rates based on risk, not race. DISB’s analysis found that as a group paid higher premiums, they also had more accidents and more claims, generated more losses, and therefore paid more premiums. Higher losses and more claims translate into higher premiums.”
The National Association of Mutual Insurers said insurers do not use a policyholder’s race as a factor in providing or pricing coverage and questioned the study’s methodology. “Even with flawed methodology, the study simply does not support the notion that groups in protected classes are subject to unfair discrimination,” the NAMI said.
Because the data it used from insurance companies did not include any information on race, DISB inferred race using a geocoding methodology that leverages U.S. Census data and names and addresses, a methodology that regulators including the Consumer Financial Protection Bureau use when conducting analyses.
