June 28
th
, 2024
Associate Commissioner Philip Barlow
District of Columbia Department of Insurance, Securities, and Banking
1050 First St NE #801
Washington, DC 20002
202-442-7823
Re: Comments on Report on Market Conduct Examination—Evaluating Unintentional Bias in
Private Passenger Automobile Insurance
The Consumer Federation of America (CFA) submits these comments on the Department of
Insurance, Securities, and Banking’s (DISB) draft Report on Market Conduct Examination—
Evaluating Unintentional Bias in Private Passenger Automobile Insurance. CFA is an association
of over 250 consumer organizations that works to advance consumer interests through
research, advocacy, and education.
We thank DISB for investigating unintentional bias in auto insurance and appreciate this
report’s important research identifying a “racial premium gap,” where Black and Latino
customers pay substantially more for auto insurance than white customers. While this report
marks a critical first step toward ending biased pricing, it also recognizes that there is more to
uncover and understand with respect to the premium gap. In particular, this report does not
examine a variety of socioeconomic rating factors and other factors that may contribute to the
disparate impact seen in this market.
Accordingly, CFA urges DISB to follow up this research with an investigation of factors that drive
the racial premium gap. This investigation and analysis should examine insurers’ use of
socioeconomic rating factors, including: 1) consumer credit information, 2) education level, 3)
job or occupation, 4) homeownership status, 5) prior carrier and coverage limits, and 6) lapses
in insurance coverage. Additionally, while this initial report examined the role of gender and
marital status, we believe it would be valuable to include those characteristics in future
research.
The District of Columbia requires drivers to purchase and maintain auto insurance coverage,
and District statutes also require that auto insurance rates not be excessive, inadequate, or
unfairly discriminatory. There is, therefore, a special obligation on the part of the District and
DISB to monitor and regulate the auto insurance market in order to root out and prevent bias
and unfair discrimination. DISB’s draft report found clear evidence of unintentional bias in auto
insurance. The report reviewed recent auto insurance applications from consumers who live in
Washington, D.C. and used the Bayesian-improved first name surname geocoding (BIFSG)
methodology to infer the race of these policyholders. The report concluded that “inferred Black
drivers pay 1.46 times as much as inferred white drivers, whereas inferred Hispanic drivers pay
1.20 times as much, and inferred Asian drivers pay 1.02 times as much. In dollar terms, that
translates into the average annual premium of $705 for white drivers, $1,031 for Black drivers,
$849 for Hispanic drivers, and $722 for Asian drivers.”
1
This means Black drivers pay on average
$326 more for insurance than white drivers, which the report refers to as the “Black/white
premium gap.”
Previous reports on auto insurance discrimination have uncovered similar results, where safe
drivers in predominantly Black or Latino neighborhoods are charged higher auto insurance rates
than safe drivers in predominantly white neighborhoods. For example, in 2017 ProPublica
found that the overwhelming majority of auto insurers were charging residents of these
neighborhoods over 10% more in premiums, that many insurers were charging those residents
30% higher premiums, and that many of these pricing disparities did not appear to be justified
by risk.
2
In 2015, a Consumer Federation of America report found that good drivers living in
predominantly Black communities paid on average 70% more for auto insurance compared to
similarly situated drivers in predominantly white communities.
3
Insurance companies use many different rating factors to evaluate risk and charge consumers
premiums, and as the DISB report notes, “companies vary in the way factors are used in
developing rates.” However, insurers’ use of socioeconomic factors is widespread and harmful
to large numbers of consumers—and we believe it is a key driver of the racial premium gap and
unintentional bias in auto insurance. Below are several examples of how these factors impact
consumers’ premiums.
In July 2023, CFA released a report on consumer credit information and auto insurance costs
across the United States of America. The report analyzed premium data from fall 2020 and
found that consumers pay dramatically higher premiums based on their credit information,
even if all their other qualities remain equal. Our analysis found that American consumers with
perfect driving records and excellent credit paid an average annual auto insurance premium of
$470 for auto insurance. If those same consumers had fair credit, their average premium
increased to $701, even if their driving records were perfect. Good drivers with poor credit
faced even higher premiums, averaging $1,012 for basic coverage across the country—or 115%
more compared to drivers with excellent credit.
4
1
“Report on Market Conduct Examination: Evaluating Unintentional Bias in Private Passenger Auto Insurance.”
Department of Insurance, Securities, and Banking. Available at
https://disb.dc.gov/sites/default/files/dc/sites/disb/page_content/attachments/Unintentional%20Bias%20report
%20-%20v.2%20draft.pdf.
2
“Minority Neighborhoods Pay Higher Car Insurance Premiums Than White Areas With the Same Risk.” By Julie
Angwin, Jeff Larson, Lauren Kirchner, and Surya Mattu. ProPublica and Consumer Reports. April 5, 2017. Available
at https://www.propublica.org/article/minority-neighborhoods-higher-car-insurance-premiums-white-areas-same-
risk.
3
“High Price of Mandatory Auto Insurance In Predominantly African American Communities.” By Tom Feltner and
Douglas Heller. The Consumer Federation of America. November 2015. Available at https://consumerfed.org/wp-
content/uploads/2015/11/151118_insuranceinpredominantlyafricanamericancommunities_CFA.pdf.
4
“The One Hundred Percent Penalty: How Auto Insurers’ Use of Credit Information Increases Penalties for Safe
Drivers and Perpetuates Racial Inequality.” By Douglas Heller and Michael DeLong. The Consumer Federation of
The report found similar conditions in the District of Columbia. Table 1 below shows that
consumers paid substantially higher average premiums if they had fair and poor credit, even if
their driving records were clean with no tickets, crashes, or claims filed.
Table 1: Average District of Columbia Auto Insurance Premiums By Credit Information
Consumer Credit
Information
Consumers With
Excellent Credit
Consumers With Fair
Credit
Consumers With
Poor Credit
Average Annual Auto
Insurance Premium
$557
$854
$1,306
Percentage Increase
Based on Premium
Information
0%
53%
134%
Insurers’ employment of credit information has a large impact on premiums, and in many cases
it has an even greater impact than driving records. A 2015 analysis by Consumer Reports
concluded that in the District of Columbia, drivers with a clean driving record but poor credit
paid $2,957 in premiums, while drivers with excellent credit but a driving while intoxicated
conviction paid $2,215—substantially less than drivers with poor credit.
5
Other factors have similar effects on consumers’ premiums. Studies have shown that
consumers are charged higher premiums if they have only graduated from high school instead
of college or if they have a master’s degree, if they work in lower-paying jobs as opposed to
higher-paying jobs, and based on their ZIP code or neighborhood. Some insurers charge
consumers higher rates if they rent their homes or if they are single instead of married, or
based on their gender.
Auto insurers’ use of the aforementioned socioeconomic factors disproportionately hurts
people of color—especially Black and Latino consumers. Table 2 shows how these factors affect
Black and Latino consumers, meaning they pay considerably higher premiums.
Table 2: Insurance Rating Factors By Demographics
Insurance Rating Factors
(Premium Discount or Penalty)
White^
Black
Hispanic or
Latino
Discounts
Percentage of Population
(except average credit score)
Homeowner
69.6%
41.4%
47.4%
Married
51.1%
29.3%
42.8%
America. July 31
st
, 2023. Available at https://consumerfed.org/wp-content/uploads/2023/07/Official-CFA-Credit-
Score_2023-FINAL-REPORT.pdf.
5
“The Secret Score Behind Your Rates.” Consumer Reports. July 30, 2015. Available at
https://www.consumerreports.org/cro/car-insurance/credit-scores-affect-auto-insurance-rates/index.htm.
Bachelor’s degree or higher
33.9%
22%
17%
Employed in management, business, arts,
and sciences
40.4%
29.9%
22.5%
Penalties
Home Renter
30.4%
58.6%
52.6%
Widowed, Divorced, Separated, Never
Married
48.9%
70.7%
57.2%
High school graduate or less
36.8%
45.3%
58.5%
Employed in service occupations
16.0%
24.5%
24.8%
Unemployed
4.2%
8.7%
6.3%
Credit score lower than 620*
(Significant Penalty)
5.4%
21.3%
11.2%
Average Credit Score*
(Lower Score = Higher Premium)
734
677
701
DISB’s report also examined cumulative paid losses by race, and found that as a group, Black
drivers represented more claims on average compared to white, Hispanic, or Asian drivers, with
their average losses being 2.38 times the losses of white drivers. DISB should conduct research
on the reasons for these losses and determine the likely causes; one possible reason for these
increased losses includes poorer and less frequently maintained infrastructure in Black
neighborhoods.
Evidence from this market conduct examination report and numerous other sources
demonstrates there are considerable unfair inequalities in the District’s auto insurance market,
as with other markets throughout the nation. We are heartened by DISB’s report and urge DISB
to build on this research through further analysis of the socioeconomic factors that cause the
racial premium gap in auto insurance.
CFA looks forward to working with DISB in this area. Please contact us at
[email protected] with any questions.
Sincerely,
Douglas Heller
Director of Insurance
Consumer Federation of America
Michael DeLong
Research and Advocacy Associate
Consumer Federation of America
