[House Hearing, 116 Congress]
[From the U.S. Government Publishing Office]
DRIVERS OF DISCRIMINATION: AN EXAMINATION
OF UNFAIR PREMIUMS, PRACTICES, AND POLICIES
IN THE AUTO INSURANCE INDUSTRY
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON HOUSING,
COMMUNITY DEVELOPMENT,
AND INSURANCE
OF THE
COMMITTEE ON FINANCIAL SERVICES
U.S. HOUSE OF REPRESENTATIVES
ONE HUNDRED SIXTEENTH CONGRESS
SECOND SESSION
__________
MARCH 4, 2020
__________
Printed for the use of the Committee on Financial Services
Serial No. 116-89
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
__________
U.S. GOVERNMENT PUBLISHING OFFICE
42-865 PDF WASHINGTON : 2021
--------------------------------------------------------------------------------------
HOUSE COMMITTEE ON FINANCIAL SERVICES
MAXINE WATERS, California, Chairwoman
CAROLYN B. MALONEY, New York PATRICK McHENRY, North Carolina,
NYDIA M. VELAZQUEZ, New York Ranking Member
BRAD SHERMAN, California ANN WAGNER, Missouri
GREGORY W. MEEKS, New York FRANK D. LUCAS, Oklahoma
WM. LACY CLAY, Missouri BILL POSEY, Florida
DAVID SCOTT, Georgia BLAINE LUETKEMEYER, Missouri
AL GREEN, Texas BILL HUIZENGA, Michigan
EMANUEL CLEAVER, Missouri STEVE STIVERS, Ohio
ED PERLMUTTER, Colorado ANDY BARR, Kentucky
JIM A. HIMES, Connecticut SCOTT TIPTON, Colorado
BILL FOSTER, Illinois ROGER WILLIAMS, Texas
JOYCE BEATTY, Ohio FRENCH HILL, Arkansas
DENNY HECK, Washington TOM EMMER, Minnesota
JUAN VARGAS, California LEE M. ZELDIN, New York
JOSH GOTTHEIMER, New Jersey BARRY LOUDERMILK, Georgia
VICENTE GONZALEZ, Texas ALEXANDER X. MOONEY, West Virginia
AL LAWSON, Florida WARREN DAVIDSON, Ohio
MICHAEL SAN NICOLAS, Guam TED BUDD, North Carolina
RASHIDA TLAIB, Michigan DAVID KUSTOFF, Tennessee
KATIE PORTER, California TREY HOLLINGSWORTH, Indiana
CINDY AXNE, Iowa ANTHONY GONZALEZ, Ohio
SEAN CASTEN, Illinois JOHN ROSE, Tennessee
AYANNA PRESSLEY, Massachusetts BRYAN STEIL, Wisconsin
BEN McADAMS, Utah LANCE GOODEN, Texas
ALEXANDRIA OCASIO-CORTEZ, New York DENVER RIGGLEMAN, Virginia
JENNIFER WEXTON, Virginia WILLIAM TIMMONS, South Carolina
STEPHEN F. LYNCH, Massachusetts VAN TAYLOR, Texas
TULSI GABBARD, Hawaii
ALMA ADAMS, North Carolina
MADELEINE DEAN, Pennsylvania
JESUS ``CHUY'' GARCIA, Illinois
SYLVIA GARCIA, Texas
DEAN PHILLIPS, Minnesota
Charla Ouertatani, Staff Director
Subcommittee on Housing, Community
Development, and Insurance
WM. LACY CLAY, Missouri, Chairman
NYDIA M. VELAZQUEZ, New York STEVE STIVERS, Ohio, Ranking
EMANUEL CLEAVER, Missouri Member
BRAD SHERMAN, California BILL POSEY, Florida
JOYCE BEATTY, Ohio BLAINE LUETKEMEYER, Missouri
AL GREEN, Texas BILL HUIZENGA, Michigan
VICENTE GONZALEZ, Texas SCOTT TIPTON, Colorado
CAROLYN B. MALONEY, New York LEE M. ZELDIN, New York
DENNY HECK, Washington DAVID KUSTOFF, Tennessee
JUAN VARGAS, California JOHN ROSE, Tennessee
AL LAWSON, Florida BRYAN STEIL, Wisconsin
RASHIDA TLAIB, Michigan LANCE GOODEN, Texas, Vice Ranking
CINDY AXNE, Iowa Member
C O N T E N T S
----------
Page
Hearing held on:
March 4, 2020................................................ 1
Appendix:
March 4, 2020................................................ 35
WITNESSES
Wednesday, March 4, 2020
Collins, Erin, Vice President, State Affairs, National
Association of Mutual Insurance Companies (NAMIC).............. 11
Dwyer, Elizabeth Kelleher, Superintendent of Insurance, the State
of Rhode Island, on behalf of the National Association of
Insurance Commissioners (NAIC)................................. 8
Heller, Douglas, Insurance Expert, Consumer Federation of America
(CFA).......................................................... 4
Larkin-Thorne, Sonja, Consumer Advocate and retired insurance
executive...................................................... 9
Poe, Eric S., Chief Operating Officer, CURE Auto Insurance....... 6
APPENDIX
Prepared statements:
Collins, Erin................................................ 36
Dwyer, Elizabeth Kelleher.................................... 46
Heller, Douglas.............................................. 51
Larkin-Thorne, Sonja......................................... 71
Poe, Eric S.................................................. 77
Additional Material Submitted for the Record
Clay, Hon. Wm. Lacy:
Written statement of the American Property Casualty Insurance
Association................................................ 99
March 2007 report of Commissioner Kevin M. McCarty, Florida
Office of Insurance Regulation, entitled, ``The Use of
Occupation and Education as Underwriting/Rating Factors for
Private Passenger Automobile Insurance..................... 111
Written statement of the Insurance Information Institute..... 138
Written statement of the U.S. Chamber of Commerce............ 141
Poe, Eric S.:
Supplemental materials supplied after the hearing............ 149
DRIVERS OF DISCRIMINATION:
AN EXAMINATION OF UNFAIR
PREMIUMS, PRACTICES, AND
POLICIES IN THE AUTO
INSURANCE INDUSTRY
----------
Wednesday, March 4, 2020
U.S. House of Representatives,
Subcommittee on Housing,
Community Development,
and Insurance,
Committee on Financial Services,
Washington, D.C.
The subcommittee met, pursuant to notice, at 2:04 p.m., in
room 2128, Rayburn House Office Building, Hon. William Lacy
Clay [chairman of the subcommittee] presiding.
Members present: Representatives Clay, Sherman, Beatty,
Maloney, Vargas, Lawson, Tlaib, Axne; Stivers, Posey,
Luetkemeyer, Tipton, Zelden, Steil, and Gooden
Ex officio present: Representative McHenry.
Also present: Representatives Davidson and Budd.
Chairman Clay. The Subcommittee on Housing, Community
Development, and Insurance will come to order. Good afternoon,
and welcome to our hearing.
Without objection, the Chair is authorized to declare a
recess of the subcommittee at any time. Also, without
objection, members of the full Financial Services Committee who
are not members of this subcommittee are authorized to
participate in today's hearing.
Today's hearing is entitled, ``Drivers of Discrimination:
An Examination of Unfair Premiums, Practices, and Policies in
the Auto Insurance Industry.''
I will now recognize myself for 4 minutes for an opening
statement.
Insurance marketing and advertising are ubiquitous.
Everyone is familiar with the Geico camel, the Allstate mayhem,
or NFL athletes selling auto insurance via colorful and often
memorable ads. What is also widespread is the use of credit-
based insurance scores in the insurance industry. Most
consumers are likely unaware that if they are behind on one of
their bills due to injury or illness, this delinquency might
negatively affect the premium that they pay for their auto
insurance.
The Federal Insurance Office (FIO) reported in 2017 that
approximately 18.6 million Americans live in ZIP Codes in which
auto insurance is deemed unaffordable. That report considered
the average annual written liability premium for the 28 percent
of American ZIP Codes that the Office deemed to be
traditionally underserved communities.
Auto insurance premiums, for example, are based in part on
driving-related factors such as an individual consumer's
driving record, but the use of such factors can raise questions
about fairness. For example, racial profiling and other
policing tactics that disproportionately target lower-income
and minority communities could artificially increase traffic
violations for certain consumers.
Think of Ferguson, Missouri. In other words, not only is
the police department discriminating against you, but then you
are paying more for your insurance, simply because you were
profiled.
While the two do not seem to be related, the consequences
of higher insurance premiums are quite profound, because then
you have less money to save in order to purchase a home, send
your children to college, or invest for your family's future.
In short, they serve to exacerbate the racial wealth gap.
And when auto insurance premiums utilize non-driving-
related factors such as a consumer's credit history, home
ownership status, professional occupation, and education or
attainment, communities of color tend to benefit less, based on
the available research. The use of these factors also raises
questions about fairness, because it does not have an obvious
correlation to losses covered in an auto insurance policy. And
because these factors can sometimes serve as proxies for
socioeconomic status and race, some consumer and civil rights
groups have pointed out the problems, going back many years,
and we will soon hear testimony about these practices.
Certainly, based on today's written testimony, the auto
insurance industry does not intend for this result, but
nonetheless it demands action on behalf of American consumers.
The draft bill by my colleague, Congresswoman Tlaib, asks the
Federal Insurance Office to study the possibility of disparate
impact in auto insurance pricing. I am hopeful that we will
learn from this conversation how we can ensure reasonable
pricing for all insurance consumers.
And I now recognize the ranking member of the subcommittee,
Mr. Stivers, for 5 minutes.
Mr. Stivers. Thank you, Chairman Clay, for convening this
important hearing. I can say with great confidence that
everyone in this room agrees that discrimination in all of its
forms is abhorrent. It is also illegal.
But tragically, our country has a lengthy history of
redlining and other forms of discrimination in financial
products. This history promoted landmark Federal legislation
prohibiting discrimination, including the Fair Housing Act and
the Equal Credit Opportunity Act. Because insurance products
are primarily regulated by the States, I am grateful that we
have somebody testifying on behalf of the National Association
of Insurance Commissioners who will be able to speak to the
history and what happened at the State level, as far as State
laws and enforcement mechanisms designed to root out and
eliminate discriminatory practices in the insurance industry.
So, Mr. Chairman, I think this hearing is justified, and we
should periodically examine what is going on with insurance
markets and ensure that our State-based regulatory framework is
working to protect policyholders and prevent discriminatory
practices.
Now, whether the legislation that has been posted for this
hearing is justified, is another matter. Our subcommittee has
considered several bills aimed at reducing the number of
factors that insurers can use to assess risk, most notably the
PAID Act, sponsored by Representative Bonnie Watson Coleman.
That bill would restrict insurers from considering 11 different
factors, while setting actuarily-sound rates. And I suspect
that the PAID Act divided committee Democrats. Therefore, we
are looking at a bill that our colleague from Michigan, Ms.
Tlaib, has authored, that would eliminate one factor, and that
is credit scores.
But despite today's hearing being focused on a narrower
bill, the underlying set of questions remains. First, should
policyholders' premiums reflect their underlying risks? Second,
does credit reporting information help predict the risk of an
individual so they can set actuarily-sound rates for that
individual? Lastly, do high insurance costs in Michigan, with
its unique laws relating to personal injury protections,
justify the Federal Government upending a State-based
regulatory system that has successfully protected policyholders
for decades?
In assessing these questions, it seems to me that the
underlying premise of this bill is flawed. Removing credit
scores in assessment of risk will not resolve Michigan's self-
imposed affordability problems.
But we will need to take a step forward, and I am excited
to have this hearing to look into what is going on. I think we
all want to do what we can to make sure that the system is
properly underwriting the risk of individuals. And I am glad to
hear that Michigan is actually taking up its problem with
personal injury protection with State-based legislation to fix
that problem, because as a neighbor to the south, I can tell
you that insurance rates in Ohio are much, much more affordable
than they are in Michigan, and our demographics are very, very
similar. So, I am glad that Michigan is taking up their own
problem.
Mr. Chairman, I do appreciate you calling this hearing, and
I really look forward to hearing from the witnesses.
I yield back, Mr. Chairman.
Chairman Clay. The gentleman from Ohio yields back. I now
recognize the gentlewoman from New York, Mrs. Maloney, who is
also the Chair of the House Committee on Oversight and Reform,
for 1 minute.
Mrs. Maloney. I thank the chairman for calling this very
important hearing on the impact of discriminatory practices in
the auto industry, especially on lower-income and minority
communities. This cannot be discounted. It is very serious.
There have been studies which found that on average, a good
driver in a predominantly African-American community will pay
considerably more for State-mandated auto insurance coverage
than a similarly-situated driver in a predominantly white
community.
This is serious discrimination. There are often subjective
calculations, and discriminatory practices can rob hard-working
Americans of the economic mobility that they desperately need.
I believe this hearing is incredibly important. There are
factors that insurance companies use to calculate premiums,
like insurance scores based on credit scoring, and the use of
non-driving factors has been proven to disproportionately
impact low-income and minority communities.
This is a serious challenge. I am pleased that the chairman
is working on this, and I am here to work with you on
legislation and other actions you think may be necessary. I
yield back. Thank you.
Chairman Clay. The gentlewoman from New York yields back.
Today, we welcome the testimony of: Douglas Heller,
Insurance Expert, Consumer Federation of America; Elizabeth
Kelleher Dwyer, Superintendent of Insurance, the State of Rhode
Island, on behalf of the National Association of Insurance
Commissioners; Eric Poe, Chief Operating Officer, CURE Auto
Insurance; Sonja Larkin-Thorne, Consumer Advocate--it says in
parentheses, ``retired,'' but I don't believe that.
[laughter]
And finally, we have Erin Collins, Vice President, State
Affairs, National Association of Mutual Insurance Companies.
The witnesses are reminded that your oral testimony will be
limited to 5 minutes. And without objection, your written
statements will be made a part of the record.
You will each be recognized for 5 minutes to give an oral
presentation of your written testimony.
We will start with Mr. Heller for 5 minutes.
STATEMENT OF DOUGLAS HELLER, INSURANCE EXPERT, CONSUMER
FEDERATION OF AMERICA (CFA)
Mr. Heller. Chairman Clay, Ranking Member Stivers, and
members of the subcommittee, thank you for the opportunity to
present testimony today.
I am Douglas Heller, an insurance expert for the Consumer
Federation of America, and for reference, I am also an
appointee to the Federal Advisory Committee on Insurance, and
the California Automobile Assigned Risk Plan.
Since the purpose of auto insurance is mandated by law in
almost every State, we believe government has a special
obligation to ensure that insurance is available, affordable,
and priced fairly in the marketplace. However, in most States,
the market for auto insurance is not fair for low- and
moderate-income consumers and in communities of color, even if
the drivers have a perfect driving record.
The data I will share today comes from recent testing that
CFA conducted that reviewed online premium quotes from one of
the nation's largest auto insurers. Each quote represents the
cost of a minimum coverage policy for a 35-year-old driver with
a clean driving record, and each is for a driver who lives at
the same address, drives the same car, and has the same daily
commute.
The first test was for a man who is an investment banker
with a graduate degree. He owns his home and he has been
insured for the past 3 years. The insurance company quoted him
a premium of $718 for 6 months.
I then tested the same driver, except instead of a male
driver, I asked for the price for a female driver. The basic
liability policy increased for her to $813, a 13 percent gender
tax on women. If instead of having a master's degree, she only
has a high-school diploma, her premium jumped another 5 percent
to $852. And when I changed her to a renter instead of a
homeowner, still at the same address, her premium increased
another $84, to $936 for 6 months.
And then finally, if instead of an investment banker, she
is a supermarket cashier, she will now pay $1,079 for the same
coverage that the highly-educated male investment banker got
for $718, even though they both have been accident- and ticket-
free for all 20 years they have been driving.
All told, in order to comply with the law each year, this
working-class woman must come up with $722 more than the
socioeconomically advantaged man.
Now, while no insurance companies formally base rates on
income, many use these various proxies for income as tools to
slice and dice and price consumers in a way that consistently
leaves lower-income drivers with good records subsidizing much
less expensive premiums for financially secure customers.
These premiums I quoted didn't include the effects of two
other non-driving factors, credit history and current insurance
status, that can account for hundreds and even thousands of
dollars in surcharges. In all but three States, auto insurers
use credit-based scores to determine premiums, and the
following Consumer Reports data from Florida corresponds to
what we see in most States.
In Florida, a safe driver with excellent credit paid $1,409
per year, on average, for insurance. However, if the same
driver merely had a good but not excellent credit rating, the
premium jumped by $312. And the same driver with poor credit
paid $3,826 for the same coverage that the excellent credit
driver got for $1,409. That is a 172 percent rate hike for poor
credit, and it means that two-thirds of the total insurance
premium is attributable to their credit history, even when
their driving history was impeccable.
But nothing highlights the problem quite like one finding
from Consumer Reports: In most States, the impact of having
poor credit is greater than the impact of a drunk driving
conviction. In Florida, a safe driver with a poor credit score
paid $1,552 more for auto insurance--that is 68 percent more--
than a convicted drunk driver who happens to have excellent
credit.
The last driver of discrimination that I will talk about is
the surcharge for a break in coverage. This break could mark a
period in which the customer was driving uninsured, or maybe
didn't have a working vehicle, or it could even be that they
did not carry coverage while they were deployed overseas.
Returning to the supermarket cashier, we said that instead
of having insurance for the past 3 years, she is not currently
insured because she has not needed coverage, and her premium
jumped 41 percent. When we tested a member of the National
Guard, who selected the pull-down menu option on the insurance
company's website that said, ``I am not insured because I was
deployed,'' he was charged $213 more on a 6-month policy. That
is a 25 percent penalty for his service. And in our research,
we have found evidence of this patriot penalty in at least 21
States.
Mr. Chairman, we are very grateful that you have begun
assessing the unfair situation that the poor face when buying
State-required auto insurance due to their socioeconomic
status, and as you expose this new form of redlining, it will
become more and more clear that action is required to rectify
these discriminatory practices. And we particularly appreciate
the work of Representatives Tlaib and Watson Coleman for their
leadership on this issue.
Thank you.
[The prepared statement of Mr. Heller can be found on page
51 of the appendix.]
Chairman Clay. Thank you. Mr. Poe, you are now recognized
for 5 minutes.
STATEMENT OF ERIC S. POE, CHIEF OPERATING OFFICER, CURE AUTO
INSURANCE
Mr. Poe. Thank you, Chairman Clay and subcommittee members,
for having me here. My name is Eric Poe. I am the chief
operating officer for CURE Auto Insurance, a direct writer,
not-for-profit auto insurer in the States of New Jersey and
Pennsylvania that does not employ the use of credit scores,
education, or occupation, but will be forced to adopt credit in
the next year due to competitive reasons.
As the original insurance whistleblower that has crusaded
against this practice for the last 15 years, testifying across
this country, and particularly 12 years ago, I was before this
Financial Services Committee on this issue, I have not received
warm welcomes from my industry at trade luncheons, but I have
won over my conscience on how we do business.
In my limited time here today, I want to quickly address
two questions that are fundamental to this subcommittee: first,
why does the industry use socioeconomic rating factors; and
second, how do they actually get this in front of, and passed
by, regulators throughout the State? And then in summary, I'll
talk about what the unintended consequences are in this country
when we use these socioeconomic rating factors, and what would
happen if we actually support the ban in the PAID Act.
Starting with why, there are two subcategories of why the
industry does this. First, the oldest reason in America, for
profit and greed. The reality is that higher-income drivers
produce the highest profits for our industry. Therefore, any
proxy that has anything to do with income will produce the same
results. If you run a business and you want to make a profit,
of course you are going to adopt practices that give you the
biggest profits.
Second, data mining. Unbeknownst to most people in the
room, if you simply get a quote for car insurance on Geico and
you never even buy a policy through Geico, in their terms and
conditions, you agreed to allow them to run your credit report,
how often you actually make payments on your lease payments, et
cetera, and they actually can share it with every marketing
partner they have.
So moving on to the how, and to get to the how, before I do
that, I want to share with you the results of the largest study
ever done regarding occupations. In this country, in the year
2000, there was a million-policy study of occupational groups
outside of students. The three highest propensity of getting in
a car accident in this country were doctors, attorneys, and
architects. And that flies in light of the notion that these
are valid predictors of risk.
So, ask yourself, how do we get these factors used in every
State that we write this business, in terms of the insurance
industry, and that is because they have done a successful job
of redefining the word, ``risk.'' If you look at every single
State in this country, you are not allowed to adopt factors
that don't show a valid predictor of risk.
What the industry has done successfully over the last 30
years is redefine what does that risk mean, by using a term
called, ``loss ratio.'' Loss ratio is simply a measurement of
profitability. For the industry, if I say that I have a loss
ratio of 90 percent, that means I actually make 10 cents on
every dollar that I collect.
So what they have done is, they have taken factors and they
have provided studies that showed that these factors correlate
to loss ratios. Those loss ratios are simply saying that they
are more profitable. So yes, any variable that ties to income
is going to show loss ratios that correlate to those factors,
but they are not dealing with the risk of somebody driving
unsafely on the road.
In terms of the impact that this has on the country, we can
look at the State of New Jersey. In the year 2003--there
wasn't, for 30 years, a mandated insurance in the State of New
Jersey. There wasn't a single car insurance company that was
allowed to use credit scores, education, or occupation at all.
So, we can look at that State as an isolated State.
Since 2004, since they let credit scores, education, and
occupation in the State of New Jersey, bringing in Geico and
Progressive, two things have happened. Geico is now the largest
writer of car insurance in New Jersey, collecting $1.7 billion
annually, in just New Jersey alone, and Progressive is not too
far behind.
But more importantly to this subcommittee is that the
uninsured motorist rate in the State of New Jersey has gone up
90 percent in 15 years. So if you want to know what the impact
is, we could look at New Jersey for what that does country-
wide.
Now, what would the impact be in terms of banning this
country-wide? Simple. Don't let my industry convince you that
this is going to make rates go up for everybody. It doesn't
make any sense. People, we are not talking about banning the
use of seat belts and headlights and blinkers. We are talking
about, there is going to be an aggregate loss to the system
every year, in every State in which insurance is written. All
we are talking about is how we are going to fairly distribute
what those rates are going to be for the people in this
country.
There are 30 million Americans who are currently driving
without insurance because they can't afford it, and it is the
obligation of this country to do something about it.
Thank you very much.
[The prepared statement of Mr. Poe can be found on page 77
of the appendix.]
Chairman Clay. Superintendent Dwyer, you are now recognized
for 5 minutes.
STATEMENT OF ELIZABETH KELLEHER DWYER, SUPERINTENDENT OF
INSURANCE, THE STATE OF RHODE ISLAND, ON BEHALF OF THE NATIONAL
ASSOCIATION OF INSURANCE COMMISSIONERS (NAIC)
Ms. Dwyer. Thank you, Chairman Clay, Ranking Member
Stivers, and members of the subcommittee. Thank you for the
invitation to testify today.
We appreciate the subcommittee's efforts to explore
discrimination in automobile insurance. State insurance
regulators are committed to ensuring that customers are treated
fairly and have access to affordable auto insurance products.
Protecting policyholders and addressing any unfair treatment of
consumers is crucial to the work we do and the bedrock
principles of State insurance regulation.
Auto insurance underwriting and rate-setting are important
for both insurer solvency and consumer protection, and
insurance regulators have robust authorities to address
discriminatory practices in these areas. Most States review
rate filings to ensure that insurance companies are using rate
factors that correlate with risk, a loss or expenses, and are
based upon supportable actuarial evidence. The more
underwriting factors that are used that correlate to risk of
loss, the more accurate the risk assessment and the rate.
This regulatory oversight helps to ensure policyholders are
charged a reasonable rate, and an insurer is taking the
appropriate amount of risk to maintain solvency, and therefore
be able to pay policyholders.
However, our regulatory framework recognizes that certain
risk classifications, even when accurately correlated with
risk, may be inconsistent with other public policies and also
provide regulators with the authority to ensure that rates are
not excessive, inadequate, or unfairly discriminatory. States
have significant authority under their own fair trade practices
statues, which prohibit insurers from refusing to insure or
refusing to continue to insure or limiting the amount of
coverage available to an individual because of race, sex,
marital status, religion, or national origin.
Through these authorities, insurance regulators strive to
ensure that consumers are treated fairly and to maintain the
critical balance between insurance solvency and product
availability and affordability.
To help us implement our statutory authorities, insurance
regulators have tools designed to identify problematic
activity, the rate approval processes, and market conduct
regulation and examination. State regulators also recently
enhanced their toolkit by collecting detailed auto insurance
data. Regulators can select a region, city, or ZIP Code, and
look at the coverage type, premium, and severity of losses in
that area, and compare it to demographic data in their States.
If States see anomalies or reasons for concern, they can follow
up and look at the individual company. The NAIC is also
completing a public report, which will show maps at ZIP Code-
level for every State under the same metrics.
While insurance regulators have broad authorities to
address unfair treatment of consumers, we acknowledge that
insurance rating and underwriting have become increasingly
sophisticated, with the advent of complex algorithms and the
use of big data and artificial intelligence. Technological
developments have the potential to improve how an insurer does
business and can benefit policyholders. However, we recognize
the complexity of these processes and the importance of
ensuring that they comply with State insurance laws designed to
protect consumers from illegal practices.
To that end, insurance regulators, through the NAIC, are
exploring insurers' use of big data for claims, marketing,
underwriting, and pricing, and developing resources to help
with evaluation of complex models, including developing best
practices to serve as a guide for review of predictive models.
In conclusion, we share your goal of preventing unfair
treatment of insurance consumers and remain committed to
addressing this important issue through our supervision of the
insurance sector. States have the enforcement authority, data,
expertise, and understanding of local market dynamics critical
to auto insurance, and are in the best position to determine
regulatory and legislative approaches to address discriminatory
practices.
For these reasons, the NAIC opposes Federal legislation
that would preempt State insurance regulatory authority in
these crucial areas and would limit a State's flexibility to
regulate such practices in a manner it deems appropriate for
the auto insurance market.
Thank you for the opportunity to testify, and I would be
pleased to answer any questions.
[The prepared statement of Superintendent Dwyer can be
found on page 46 of the appendix.]
Chairman Clay. We will now go to Ms. Larkin-Thorne for 5
minutes. You may proceed.
STATEMENT OF SONJA LARKIN-THORNE, CONSUMER ADVOCATE AND RETIRED
INSURANCE EXECUTIVE
Ms. Larkin-Thorne. Thank you, Chairman Clay, Ranking Member
Stivers, and members of the subcommittee. Thank you for the
opportunity to testify this afternoon on pending legislative
proposals regarding the use of credit scoring in the
underwriting of personal automobile insurance, and for allowing
me to share my insights and actual experience.
My name is Sonja Larkin-Thorne. I have over 40 years of
experience in the insurance industry and I currently Chair the
Consumer Data Subcommittee of the Connecticut Department of
Insurance Advisory Council on Technology. My testimony will not
attempt to repeat what you already know or have heard about the
use of insurance credit scores in personal auto insurance
underwriting or the many studies on this topic over the last
20-plus years.
My testimony focuses on the evolution of underwriting of
personal automobile insurance from traditional underwriting to
the use of credit scores, and now, to the use of big data. I
hope my testimony will also highlight how the collection and
analysis of these new data sources reaches every aspect of a
policyholder's life, but does so with little or no regulatory
oversight, and notably less oversight than is applied to credit
scores.
I am an insurance underwriter by training. I spent the
first 10 years of my insurance career working in or managing
underwriting departments. When I began my career, the manual
process of collecting and analyzing underwriting data
represented a significant expense that necessarily contributed
to the cost of policyholder premiums.
More than 40 years ago, insurers began to analyze and
utilize data sources that could reduce underwriting costs while
continuing to accurately reflect the likelihood and expense of
a policyholder filing an auto insurance claim. I recall
training my underwriting team on how to read a credit report,
while they also reviewed traditional motor vehicle records and
the policyholder's paper application for automobile insurance.
We had no desktop computers or electronic submission of
applications back then. Each piece of data was assigned a code
for input into the company computers. Little did we know, that
was the beginning of big data.
There is no escaping that data is the foundation of the
insurance industry and that the insurance industry has always
collected data and made long-term predications regarding
pricing, loss trends, and profitability. However, just like the
days of traditional underwriting gave way to underwriting that
included the use of insurance credit scores, the heydays of
insurance credit scores are on their way out.
The continued difficulty of explaining the correlations
between how one pays their bills and the likelihood and expense
of a policyholder filing an auto insurance claim has caused the
largest insurers to move beyond insurance credit scores and to
look at and use other data to enhance their underwriting and
pricing.
For example, California--and I grew up in California and
lived in California for most of my life--one of the largest
personal automobile insurance markets in the country, does not
allow the use of insurance credit scores, yet the State remains
one of the most competitive personal auto insurance markets in
the country. This means insurance companies in California and
elsewhere have figured out how to price personal auto insurance
without using credit scores.
The use of credit scores in personal automobile insurance
underwriting increasingly is being enhanced or replaced with
incredible amounts and new types of alternative, unregulated
personal individual data. Commonly referred to as, ``big
data,'' these extremely large datasets can be analyzed to
reveal patterns, trends, and associations related to human
behavior, interactions, shopping habits, driving patterns, and
demographics like race, occupation, education, marital status,
voting history, et cetera.
It is this big data, often produced by unregulated
algorithms that insurance companies use in insurance department
rate filings, to which I wish to draw the subcommittee's
attention. Any legislative or regulatory solution that seeks to
achieve the goals behind the legislation before the
subcommittee today needs to look further than the somewhat
outdated and narrow issue of insurance credit scores.
To this end, I would like to draw the subcommittee's
attention to the current work of the National Association of
Insurance Commissioners. The NAIC recognizes and has drawn
appropriate attention to the use of credit scores and to the
use of big data. Currently, there are 3 working groups looking
at this issue, and each working group incorporates the nation's
56 insurance regulators.
My recommendations regarding the legislative proposals
pending before the subcommittee are informed by my firm belief
that insurance consumer protection is best served at the State
level. Absolutely, there is a role for this subcommittee and
for the Federal Insurance Office to work with our State
insurance regulators to make sure that credit scores, big data,
and any other datasets are used to fairly and responsibly set
personal automobile insurance rates. I caution the subcommittee
against imposing a Federal solution to address an issue for
which the State insurance regulatory system is best designed.
[The prepared statement of Ms. Larkin-Thorne can be found
on page 71 of the appendix.]
Chairman Clay. And Ms. Collins is now recognized for 5
minutes.
STATEMENT OF ERIN COLLINS, VICE PRESIDENT, STATE AFFAIRS,
NATIONAL ASSOCIATION OF MUTUAL INSURANCE COMPANIES (NAMIC)
Ms. Collins. Good afternoon, Chairman Clay, Ranking Member
Stivers, and members of the subcommittee. Thank you for the
opportunity to testify here today. My name is Erin Collins, and
I am the Vice President of State Affairs for the National
Association of Mutual Insurance Companies.
NAMIC membership includes more than 1,400 regional and
local mutual insurance companies on Main Streets across
America, as well as many of the country's largest national
insurers, who collectively write about half of the nation's
auto insurance business.
NAMIC is here today to speak in strong opposition to
legislation designed to institute broad prohibitions on the use
of legitimate and predictive underwriting tools in auto
insurance. It is our view that Federal legislation in this area
is unwarranted and unnecessary, and it has been demonstrated
time and time again that such underwriting restrictions harm
policyholders by driving up insurance costs across-the-board.
I want to begin with an overview of insurance underwriting
and risk-based pricing. The goal of insurance underwriting is
to correlate the prices for insurance policies as closely as
possible to the likely cost of claims generated by those
policies. The more accurately a company targets the actual
costs, they better they are able to serve their policyholders,
which, in turn, enables them to offer more products to more
individuals. Simply put, accurate underwriting enables more
coverage for more consumers.
Utilizing predictive information is unavoidably central to
this process. Insurance differs from most other products
because the actual cost of providing insurance is unknown at
the time the product is offered, and the customary laws of
supply and demand do not apply.
Looking back at historic losses helps to forecast future
losses, but prior claims alone do not provide enough
information to serve as an adequate predictor of future risk.
Therefore, to more accurately make these predictions, various
factors are used to analyze the risk of future losses from any
given policy.
In today's auto insurance market, a multitude of risk-
predicting factors may be considered, including things like
driving history, multi-policy discounts, vehicle information,
vehicle safety equipment, age, credit history, miles driven, et
cetera. How or even whether particular factors are used differs
by insurer.
However, today's actuarial science indicates that the most
accurate risk assessment is achieved through a combination of
risk factors. It is important to note that these factors are
actuarially-based tools that are correlative and do not, and
generally cannot demonstrate causational relationships. A
causational relationship would indicate that a given factor
results in a loss. Even a fact pattern of a dozen prior
speeding tickets and prior accidents does not result in a
future loss.
Any assertion that factors should be held to a causational
standard fundamentally misunderstands the methodology of
actuarial science and would exclude practically every exercise
to try to assess risk.
The US auto insurance market is one of the most competitive
of any industry in our economy, and competition ensures that
insurers have every incentive to accurately and appropriately
match the rate to the actual risk that is posed by a driver.
Those companies that predict claim costs better than their
competitors are typically more successful.
Ultimately, we believe that competitive markets are the
most effective way to ensure lower prices, widespread
availability of insurance products, superior service, and
product innovation. The State-based system of insurance
regulation has allowed that competitive market in the U.S. to
flourish. This robust regulatory system is designed with
consumer protection foremost in mind, and regulates virtually
every aspect of the insurance business. Under State law,
insurers are prohibited from setting rates that are excessive,
inadequate, or unfairly discriminatory against any individual.
Stringent anti-discrimination prohibitions are also in place in
the States, and rating factors must be actuarily sound in order
to be used by law. There is no data to indicate that these
provision or their enforcement have been inadequate or lacking.
As I close, I would reiterate that banning the use of
underwriting factors would simply disrupt and substantially
weaken auto insurance markets across the country, undermine the
State-based system of insurance regulation that serves the U.S.
well, and ultimately harm the very consumers such action
purports to help. Therefore, NAMIC respectfully encourages the
subcommittee to reject the legislative proposals being
discussed here today.
Again, thank you for the opportunity to speak here today,
and I look forward to answering any questions you may have.
[The prepared statement of Ms. Collins can be found on page
36 of the appendix.]
Chairman Clay. I want to thank all of the witnesses for
your testimony today, and now we will begin the 5-minute
questioning period. I will start off by recognizing myself for
5 minutes.
Mr. Poe, you have testified previously that your company
has chosen not to use certain non-driving-related factors such
as credit history, despite the competitive advantage of doing
so, because of concerns about the fairness of using such
factors. Can you talk about which factors your company
deliberately uses, chooses not to use, and how you believe that
these company practices make auto insurance premiums more
equitable and affordable for consumers?
Mr. Poe. Yes. Thank you, Chairman. We don't use any non-
driving factors that are what we think are obvious corollaries
to socioeconomic proxies. So we don't use--right now, we
examine all filings by other carriers, and I can tell you, as a
matter of fact, that my industry uses full-time employment as a
decision of whether or not you are eligible for the preferred
rates of a car insurer. So if you simply lose your job, your
rates will be more, and you will not be eligible for a
preferred company.
What my industry does is they tend to actually file two
separate companies with the same trademark name. For example,
Geico has three trademark named companies in almost every State
they do business: Geico; Geico Indemnity; and Geico Casualty.
But you only get told to go to Geico.com and fill out a quote
because you might save 15 percent.
The reality is that most people who go to Geico and put in
their information have no idea that on the basis of your
educational attainment and your job, whether or you are in a
high-paying job, you will be rejected from the preferred
companies of Geico, and only be eligible for a higher rate for
car insurance and not be notified, because there is no Federal
requirement and there is no State requirement to say that you
were rejected on the basis of your educational attainment and
your occupation.
There are other statuses like home ownership status. There
is a reason why all these advertisements talk about bundle,
bundle, bundle. Does a poor person have a house? A poor person
barely has a window or a pot to pee in, to throw it out of.
So, mandating insurance becomes a public policy
requirement. In terms of this committee, there is a
requirement, in our opinion, there is an obligation for
legislators to make sure that if you are going to mandate car
insurance, you can't just let poor people pay more than rich
people. That is the bottom line.
Chairman Clay. In other words, you are telling me that
there is an extra tax for being poor.
Mr. Poe. Exactly.
Chairman Clay. It is imposed on these people by the
industry.
Mr. Poe. Yes. And unlike the traditional methods of
predatory lending, where you can actually draw a conclusion of
a credit risk because somebody doesn't earn as much money and
they are more likely going to default on their loan because
they don't have assets to collateralize, here we are talking
about mandating car insurance, and I can be a safe driver my
entire life. But simply because I am poor, I am going to pay 40
percent more for car insurance, for no reason whatsoever except
for that I drive less profits than the wealthier people in this
country.
Chairman Clay. Thank you for your response. This next
question is for Mr. Heller and Ms. Larkin-Thorne. The insurance
industry has stated that it is regulated at the State level,
unlike most other large industries. Mr. Heller and Ms. Larkin-
Thorne, if the insurance companies don't change the way they
treat certain discrete and insular minorities with regard to
premium pricing, are the individual States capable of mandating
change? We will start with you.
Ms. Larkin-Thorne. Thank you, Mr. Chairman. Yes, they are.
California is a prime example of a State that took control of
insurance rating factors. The people, the voters of California
changed that law. And I say that any State voters who want to
build the type of coalition that was built in California, that
took on insurance companies spending millions of dollars
against their initiative, they can win.
I believe in the power of the vote, I believe in the power
of the people, and I think if they want to change, they can.
And every Member who thinks that there is a problem in their
State should be talking to their insurance commissioners. They
should be talking to their legislatures. And I think that they
can facilitate the type of change they want. You look at
California. You look at Hawaii. They did it.
Chairman Clay. Mr. Heller, are the States capable of
mandating change?
Mr. Heller. Mr. Chairman, unfortunately, I think many
States have fallen down and failed on this issue. I agree with
Ms. Larkin-Thorne. I am a Californian and I work very closely
with the California regulators. The people of California did
speak. Unfortunately, most States don't have a ballot
initiative the way California did to enact that law and
overcome the resistance.
Unfortunately, regulators around the country have fallen
flat on this issue, and we need the support from the Federal
Government to say you need to at least have a bare minimum,
because if you are going to make people buy insurance, we are
going to make sure that you have a floor of protections for
consumers, so low-income consumers and communities of color
have access to products in their State.
Chairman Clay. Thank you for your response. My time has
expired. The gentleman from Ohio, Mr. Stivers, the ranking
member of the subcommittee, is now recognized for 5 minutes.
Mr. Stivers. Thank you. Thanks, Mr. Chairman, for holding
this hearing. And I thank all of you for your testimony, and I
would like to start with Ms. Dwyer.
Can State insurance regulators be trusted to prevent
discrimination in the auto insurance markets?
Ms. Dwyer. Yes, we can.
Mr. Stivers. What mechanisms do they currently employ to
prevent this?
Ms. Dwyer. We currently, in every State except Illinois,
review all rate filings from all auto insurance companies. We
look at both the form and the rates being charged. We look at
their algorithms, and we test for factors such as suspect
classifications. We would never allow anyone to rate on race,
national origin, or religion, and that is pretty much standard
in all 50 States.
Mr. Stivers. Great. A question for Mr. Heller, given that
you said that our State insurance regulators have fallen down
on the job. They have been on that job for 75 years, since D-
Day. Would you prefer, then, a Federal regulator, for instance?
Mr. Heller. No.
Mr. Stivers. So, you would prefer--
Mr. Heller. Thank you, Mr. Stivers. I actually--
Mr. Stivers. --Federal laws to override McCarran-Ferguson?
Mr. Heller. I think that it might be good to get rid of the
antitrust exemption that the insurance industry is given
through McCarran-Ferguson. That is true.
But Ranking Member Stivers, I actually believe in State
regulation. I think it is the way to go, because I do think
that at the local level, we can do a better job. But what I
have seen in too many States, is--
Mr. Stivers. I only have 5 minutes, so be quick.
Mr. Heller. --a failure to protect the most vulnerable
consumers, and I think efforts here in Congress to support a
basic floor of rules and protections would be valuable.
Mr. Stivers. Okay. For Ms. Dwyer and Ms. Collins, does
anything in this bill by the Congresswoman from Michigan, Ms.
Tlaib, change the insurance rates in Michigan suddenly, if this
bill were to pass? Would it change insurance rates in Michigan
for automobile drivers, any of them?
Ms. Collins. Thank you. I don't think it would
demonstrably--
Mr. Stivers. The studies--would that change insurance rates
for anybody?
Ms. Collins. It would not demonstrably improve insurance
rates, no. I am sensitive to the concerns of the Michigan
insurance market. I have spent a large part of my career
working on those issues, and they are not problematic because
of underwriting factors. They are built upon the problems of
the no-fault system with unlimited lifetime medical benefits.
Mr. Stivers. Ms. Dwyer, are you familiar with what is going
on in Michigan with regard to personal injury protection and
what the Michigan legislature and the governor are doing to
actually try to fix that, to make insurance more affordable in
Michigan?
Ms. Dwyer. Yes, and I am absolutely not an expert, but it
is my understanding that there have been changes that will go
into effect in July that should be beneficial.
As far as your original question, what it would do,
removing one factor would reallocate, so some people would get
increases, and some people would get decreases. It would be
disruptive to the market, but it wouldn't appreciably lower
what everyone is paying.
Mr. Stivers. And Ms. Collins, let's say for a second we
were to only go to driving factors. Are you familiar with the
study from Stanford University that indicates that minorities
are much more likely to be pulled over and ticketed than white
drivers? And if Congress would decide to preempt State laws and
only move to driving factors, excluding all non-driving
factors, what would that do to minority policyholders?
Ms. Collins. I am not specifically familiar with that
study. Generally speaking, in terms of driving records, I do
know that they are notoriously inaccurate, and I think that we
have shown that the majority of drivers in America are
benefitted from a full picture of risk, by as many factors as
can be contemplated.
Mr. Stivers. Is it possible that if we move to only driving
factors, potential institutional racism that happens in some
places where minority drivers are more likely to be pulled over
and ticketed would actually result in higher insurance rates
for minority drivers?
Ms. Collins. Based on your description of that study, yes,
that does seem logical.
Mr. Stivers. That is a study from Stanford University, so I
will summarize it for you.
And this question is for Superintendent Dwyer. Do you think
the bills attached to this hearing raise any concerns about
broad Federal overreach into our State-based system?
Ms. Dwyer. Yes, I do. The other factor that has to be taken
into account is financial solvency. So the more rating factors
that are used, the better predictor of risk, the more solvent
the company is. And why is that important? Because they pay
consumers. The worst thing out there would be paying premiums
and not having your claim paid.
Mr. Stivers. Great. Quickly, just by a show of hands, is
there anyone on the panel who does not support the State-based
regulation for insurance under McCarran-Ferguson?
I will note the absence of a raised hand, which means
everyone on the panel does support McCarran-Ferguson and the
State-based policy. Is that correct? Can you show me by a nod
of the heads or something? Great.
Mr. Chairman, my time has almost expired. I would like to
submit for the record letters from the Insurance Information
Institute, the American Property Casualty Insurance
Association, and the U.S. Chamber of Commerce, that all have
some concerns about the draft legislation we are talking about
today, and I would appreciate them being entered into the
record.
I yield back the balance of my time. Thank you.
Chairman Clay. The gentleman yields back, and without
objection, the documents are submitted for the record.
And I have to note that my friend from Ohio has adopted Mr.
Green of Texas's tactic as far as a show of hands.
Mr. Stivers. It is a good tactic.
Chairman Clay. I now recognize the gentlewoman from
Michigan, Ms. Tlaib, for 5 minutes.
Ms. Tlaib. Hello, everyone. Thank you so much for being
here. This is one of the most critical issues right now in the
State of Michigan that is keeping people in poverty, because of
these high rates.
We can talk a lot about some of the other broken system,
but one question I have, to follow up on my colleague's
question, is how many of you on the panel support proxies that
discriminate?
Okay. Ms. Collins, I have some questions for you. What is
your marital status?
Ms. Collins. Thank you, Congresswoman. I am married.
Ms. Tlaib. Yes. Credit score?
Ms. Collins. I do have a credit score.
Ms. Tlaib. Do you know what it is?
Ms. Collins. Not offhand, no, I don't.
Ms. Tlaib. How about your education level?
Ms. Collins. I have a master's degree.
Ms. Tlaib. How about your ZIP Code, if you would like to
provide that.
Ms. Collins. I have a ZIP Code. I prefer not to share it.
Ms. Tlaib. Sure. These are the questions that are asked of
my residents, which have nothing to do with whether or not they
are a good driver. It has nothing to do with whether or not
they are susceptible to accidents. These are proxies to
discriminate, and it is very frustrating for my residents, from
their credit score, to their education level, to even be asked
that, or them turning from married to widow, that they see a
300 percent hike in their auto insurance rates.
And so one question to Mr. Heller, one of the studies that
Poverty Solutions at the University of Michigan did was they
looked at someone with a decent credit score, and someone with
a better credit score, and the person who had a lower credit
score was paying 300 percent more than the other person, but
the other person, the one with the better credit score, had a
DUI, driving under the influence, violation.
What would justify--and again, this is a study by the
University of Michigan showing a 300 percent hike, again, of
the person who is the safer driver versus the one who would
drink and drive?
Mr. Heller. Thank you, Representative Tlaib. Unfortunately,
that is not just a problem in Michigan. That is what we see in
most States. Consumer Reports, a couple of years ago, reported
that in most States the drunk driver with excellent credit pays
less for insurance, on average, for the same product than the
poor credit, excellent driver. And what the insurance industry
sees in the poor credit, excellent driver, is somebody who,
with other factors like their job title or their education, is
likely to buy a basic limits auto insurance policy, and that is
just not that attractive. There is not a big lifetime value to
selling sur to somebody who is only buying the minimum limits.
But that person with the drunk driving violation and the
excellent credit score, and perhaps also the investment banker
with a college degree, might buy home insurance as well, and
maybe life insurance, an umbrella policy. Maybe they will
insure their boat. The lifetime value is what the insurance
companies are looking for. And that may be fine for other
products, but in virtually every State in the country, auto
insurance is required by law, and that is why we believe we
have to have some public policy standards that say some factors
are not acceptable because they are keeping people from getting
into the market and leaving them uninsured.
Ms. Tlaib. And one of the things I want to refer to you,
Ms. Collins, and I would like the chairman to submit for the
record, are ProPublica and Consumer Reports reports which found
that members of the National Association of Mutual Insurance
Companies were charging higher rates in ZIP Codes where most
residents are minorities than ZIP Codes that were predominantly
white neighborhoods with similar accident calls.
Chairman Clay. Without objection, it is so ordered.
Ms. Tlaib. Thank you. Ms. Collins, justifying racial
disparities by pointing to differences in risk is an argument
that falls apart when we really investigate the data. Why would
the association be against a study, just to investigate the
data regarding disparate impact of auto insurance rates? Why
don't you want to know if it impacts your consumers, the ones
who are paying into the system, to make sure they are not being
discriminated against?
Ms. Collins. NAMIC is supportive of objective studies that
look at all underwriting and rate impactors--
Ms. Tlaib. So you are okay with the Federal Government--
Ms. Collins. We are okay with the FTC study. There are
numerous studies that look at these factors, in Arkansas, for
example.
Ms. Tlaib. Are you okay with the Federal Insurance Office
investigating? Why wouldn't they be objective?
Ms. Collins. If you're speaking to the measure in this
committee, please note that I said, ``objective study.'' I
believe that studies should not be built with conclusions in
mind, and the language of that particular bill seems not to
meet that test. So, we would oppose that measure.
Ms. Tlaib. One of the things in some of the comments by
some of your members has been that low-income Americans spend
more money on cigarettes and alcoholic beverages. Do you
remember that study and those comments coming from your
association?
Ms. Collins. I am assuming you are referring to the letter
you referenced in the last hearing.
Ms. Tlaib. That's right.
Ms. Collins. I am glad you brought that up, because there
was an article that grossly mischaracterized a letter that we
wrote directly in response to the--
Ms. Tlaib. Well, then why not support--
Ms. Collins. --FIO study's specific questions.
Ms. Tlaib. But why not, Ms. Collins, support actually
investigating whether or not there is disparate impact in your
industry?
Ms. Collins. We were responsive to the FIO's specific
questions.
Ms. Tlaib. Why not support the Federal Insurance Office to
provide this data and information, so we make sure that our
folks are not being discriminated against.
Chairman Clay. The gentlewoman from Michigan's time has
expired.
Ms. Collins. Thank you, Mr. Chairman. I would be happy to
schedule some time to sit down with the Congresswoman and go
through those issues at her leisure.
Chairman Clay. Okay. I now recognize the gentleman from
Florida, Mr. Posey, for 5 minutes.
Mr. Posey. Thank you, Mr. Chairman. It has been my
observation over the years that the government closest to the
people works best. One of the biggest problems I observed in
the Florida legislature was that members of the legislature who
had not served in local government thought that when they got
elected to the legislature, that meant they were the boss of
local government. And, at the same time, they hated the Federal
mandates that would get passed down from time to time.
Hypocritically, I might add, one day, actually at 10:00 in
the morning, they passed a joint resolution to Congress and
said, ``Do not send us one more of your unfunded mandates,''
period, end of subject, exclamation point. At 2:00, 4 hours
later, they passed an unfunded mandate on local governments.
And then, one of the problems I see here in Washington is
that Members who haven't served in local governments or State
governments seem to think that suddenly they are the boss of
the State governments and the local governments, which makes it
even worse.
And on the regulatory side, we have talented leadership in
our States and their insurance commissioners, and historically,
States regulated insurance and helped tailor insurance products
to the needs of the people in their States. Today, we have
legislation before us that would intervene or interfere in the
State-led insurance sector by dictating coverage and premium-
making practices. This would prohibit the use of credit scores,
credit reports, or ZIP Codes, as mentioned earlier, as if
somebody in Manhattan should pay the same price for liability
insurance as somebody in rural Wyoming. I don't know how you
can think that, but obviously, there shouldn't be any
discrimination whatsoever in the insurance industry or any
other industries.
And to follow up with Ms. Collins, what is the evidence
regarding discrimination in coverage decisions in pricing of
auto insurance?
Ms. Collins. Thank you, Congressman. There is no evidence
that I am aware of that there is any unfair discrimination in
underwriting or rating in auto insurance. In fact, the State-
based system is quite adept at regulating the market.
Superintendent Dwyer and her colleagues are robust regulators
who ensure that there are no unfair discriminatory practices in
the auto insurance market, or any insurance market, for that
matter, and you will find no greater advocate of the State-
based system than NAMIC.
Mr. Posey. Is there a case you are aware of where auto
insurers discriminated based on race or gender?
Ms. Collins. No. I am not aware of, again, any unfair
discrimination in auto insurance.
Mr. Posey. How would the industry assess this bill to
prohibit the use of a credit report, ZIP Code, credit score,
other consumer data or information?
Ms. Collins. I think credit-based insurance scores have
been the subject of quite a lot of study over the years, and
they have a variety of authors--States, localities, departments
of insurance--and they all reached the same conclusion, which
is that credit-based insurance scores are highly predictive of
risk.
And I should note that credit-based insurance scores are
not the same thing as a credit report. Where a credit report
refers to delinquency, a credit-based insurance score aims to
predict an insurance loss. And what the studies have shown is
that it is incredibly predictive in doing that. So, to remove
that asset and tool in matching rate to risk would demonstrably
harm the insurance market, and ultimately, harm consumers.
Mr. Posey. Many comments have been written on the power of
market competition to drive out discrimination. It makes sense
that an auto insurance company would find little profit in
deciding to charge a higher price to persons of color or women,
or not to sell them insurance if a competitor would provide the
coverage or could bid the business in a way with a lower rate.
Can you please share with me what we know about the power
of competition in discouraging discrimination, let's just say
in auto insurance, for example?
Ms. Collins. Sure. The State-based system in the United
States has fostered the most competitive and powerful insurance
market in the world, and that only helps consumers. It provides
more coverage opportunities, more products, more innovation,
and that only creates more space and more entities to compete
for those policyholders' business. And to that end, it would
ultimately be a factor that would eliminate that kind of bias
rather than increase it.
Mr. Posey. Thank you. I have read that regulators in
California are thinking of prohibiting discounts to certain
classes of drivers because they believe such discounts are not
based on the likelihood of lower claims, and that such
discounts result in discrimination. While this does not sound
like a good policy, it is also a powerful example that States
already have the responsibility to regulate their own insurance
markets, making the kind of Federal legislation we are
examining here inappropriate in preempting the strong State-
based system of insurance regulation.
Can you please comment on this, and whether California's
plan to prohibit certain discounts is a good way to fight
discrimination?
Ms. Collins. I don't believe that an initiative like that
would at all address the concerns that it purports to protect.
In fact, I think it would hurt the consumers ultimately,
because it would restrict further the ability of insurance
companies to accurately rate risk and offer their customers
discounts that they find valuable.
Chairman Clay. The gentleman's time--
Mr. Posey. Thank you, Mr. Chairman, for the time.
Chairman Clay. --has expired. The gentleman from Florida,
Mr. Lawson, is recognized for 5 minutes.
Mr. Lawson. Thank you, Mr. Chairman, and witnesses, welcome
to the committee. This question is more of an observation, and
I think anyone can answer it. The National Highway Traffic
Safety Administration (NHTSA) confirmed that male drivers cause
6.1 million accidents annually, while there are only 4.4
million crashes per year with women at fault. Yet, women are
more likely to pay higher rates than their male counterparts,
according to the research just completed by the CFA.
The CFA found that Progressive charged 40-year-old female
drivers living in Tampa 32 percent more for the same coverage
than males. The insurance company's use of gender as a rating
factor does not seem to reveal much in the way of a consistent
risk assessment.
Do you think regulators should reconsider allowing
companies to continue using it at all? I will start with you,
Mr. Heller.
Mr. Heller. Thank you, sir. The use of gender in auto
insurance is something that surprises people, particularly
women who think that they are going to get a better deal. You
cited some statistics. And while it is true that in some
companies, young men still pay more than young women, not only
has the Consumer Federation of America found, but also several
insurance industry research studies and the California
Department of Insurance have found that women pay more for auto
insurance now, and it is just striking to people. It is also
unfair. That is why about seven States have prohibited the use
of gender, California most recently.
And I think that the point perhaps is that we need to
protect people who have to buy this product, and if the States
aren't taking on that role, then it is important for the
Federal Government to push them in that direction. That is why
this hearing is so important, if only to raise the attention
and make the State regulators pay attention. We appreciate it,
as we are working in the States for this protection, because
women should not pay more for auto insurance, and it is
happening across the country.
Mr. Lawson. Mr. Poe, do you want to comment on that?
Mr. Poe. Yes, just that we don't use--
Mr. Lawson. I can hardly hear you.
Mr. Poe. Sorry. Just that we have never actually charged
females more than males, so that data only suggests that once
again, males probably make a higher income. Higher incomes
generate better profitability, and that is probably what
justifies their algorithms, not based on our non-driving
factors, traditionally. So there is never a female who actually
pays a higher rate with our carrier.
Mr. Lawson. Does anyone else want to comment on that?
Ms. Dwyer. Thank you. When we look at an algorithm, when we
look at a rate filing, we make sure that every factor being
used is supported actuarily. So some of the information that
you are producing should be taken into account by the actuary,
but we do not allow them to simply use a factor that has no
actuarial basis.
Different companies are also going to use different
factors. So as an individual, if you receive a poor rate from
one company, you can shop that rate and get different amounts,
because they are looking at their overall risk. That
competition benefits the individual customer.
Mr. Lawson. Ms. Larkin-Thorne?
Ms. Larkin-Thorne. As a divorced mom who raised a child,
when I went from married to single or divorced, my insurance
rate did not change. It was the same. And so I look at it, and
I have friends who have had the same circumstances, and I have
not heard that complaint. And I talk to a lot of people about
their insurance.
Now I live in Connecticut and it may be the rates in our
State. But I was in California at the time I was divorced and
my rate absolutely did not change at all.
Mr. Lawson. Ms. Collins?
Ms. Collins. Thank you, Congressman. I would echo the
comments from Superintendent Dwyer that it is important to
recognize that each factor must be actuarily sound in order to
be utilized. I would also note that as I stated in the
beginning, we find that actuarial science most often dictates,
and the companies now employ, that it is a combination of all
factors. So there is not an instance where there is an
individual factor that dictates the rate or underwriting of an
individual.
Mr. Lawson. So basically what you are saying is--I know my
time is running out--in Florida, for example, in an area like
Tampa, females 40 years of age are paying a 30 percent higher
rate than the males in Florida. Mr. Heller, I don't understand
how that happens.
Mr. Heller. It does happen, and it happens because the
insurance companies are pricing the way they want for their
profitability. And unfortunately, regulators are not diving in
as deep as they can. With respect to Superintendent Dwyer, I
see this around the country, and I will note that the big irony
here is that some companies do charge women less than men. And
if the actuarial science was accurate, there is no reason that
one company would see women as more risky and another company
would see men as more risky. They are doing this for their own
marketing purposes, but consumers have to buy the product,
irrespective of their gender.
Mr. Lawson. Thank you. My time has expired, and I yield
back, Mr. Chairman.
Chairman Clay. The gentleman's time has expired. The
gentleman from Colorado, Mr. Tipton, is recognized for 5
minutes.
Mr. Tipton. Thank you, Mr. Chairman, and I thank the panel
for taking the time to be here.
I wanted to be able to get a little bit of clarity on some
of the comments that I have heard in regards to a ZIP Code
having an impact on assessing driving risk. Ms. Collins, does
the ZIP Code impact what kind of rate you are going to be
paying?
Ms. Collins. Thank you, Congressman. Territorial rating,
especially in dealing with auto insurance and other areas of
insurance, is important in that it helps to identify areas of
risk. For example, a risk associated with auto insurance is
density. The number of drivers on the road, the population on
the road, all factor into the risk of severity of accidents and
frequency of accidents. So, a car located in a rural plains
State is not going to have the same risk profile as a vehicle
in a downtown metropolitan area. So yes, to answer your
question, it is an important factor.
Mr. Tipton. I live in a rural area of Colorado, so I am
probably going to have a lower rate than somebody driving in
Washington, D.C.?
Ms. Collins. Yes. I think to the extent that the factor is
utilized on territory, yes, that situation, that factor would
show a lower risk.
Mr. Tipton. I did want to follow up with you a little bit
as well in regards to the credit scoring. You noted it is not
the credit score but being able to use it as a predictive
model. Can you expand on that a little bit?
Ms. Collins. Sure. A credit-based insurance score is a
score that uses pieces of a credit report to develop an overall
insurance score that relates, and directly relates to the risk
of insurance loss. And the studies that have been done on
credit-based insurance scores overwhelmingly show that it is
not just predictive, it is overwhelmingly predictive, and is
one of the most predictive factors of insurance loss that
exists.
Mr. Tipton. So the credit information, as you describe it,
applied correctly, can actually benefit the consumers by
lowering some rates?
Ms. Collins. Yes. Actually, the Arkansas study on credit-
based insurance scores showed that the vast majority of drivers
benefitted from the use of credit-based insurance scores.
Mr. Tipton. Thank you. And you also spoke to the study bill
that is being discussed here to a degree today. But studying
some of the underwriting factors nationally, this would make it
public information, is that accurate?
Ms. Collins. That is my understanding, yes.
Mr. Tipton. Okay. That would be an unprecedented step for
the Federal Government, wouldn't it?
Ms. Collins. To my knowledge, yes.
Mr. Tipton. So following that line of thought, what would
be the impact of disclosing this proprietary underwriting
information beyond the insurance market as a whole?
Ms. Collins. In my opinion, it would create, at minimum, a
dampening effect, but realistically, potentially the end of
innovation in auto insurance. It would create a situation in
which insurers are not able to develop new ways of assessing
risk or ways to develop value propositions to their customers
or to their potential customers. So, I think it would
demonstrably harm the market.
Mr. Tipton. So just saying, why have a problem with a study
bill, effectively, this study bill would have a dampening
effect and would probably impact nationwide some of the
innovation that is needed in the insurance market?
Ms. Collins. I think it would be wildly detrimental to
innovation, yes.
Mr. Tipton. Okay. We do have a well-established Federal
statutory precedent in this country in regards to regulation
matters relating to insurance belonging to the individual
States. Our States have had the authority, as the
superintendent had noted, to be able to develop the insurance
market within their borders. We just passed, in the Views and
Estimates, out of the hearing this last week, a proposal
regarding some of the State impacts that we are going to be
seeing, at least on the Republican side, to be able to have it
at the State level.
So, Ms. Collins, would the study bill before the committee
today jeopardize the State-based approach to regulation?
Ms. Collins. Yes, I do believe it would. It certainly does
not take into account the 10,000 regulators across the United
States that have created and maintained such a powerful
insurance market in the States, and I think it would
demonstrably undermine the State-based system and how well they
regulate the market now.
Mr. Tipton. Thanks. And finally, Superintendent Dwyer,
almost every State in the country has adopted a law requiring
insurance premiums not to be excessive, inadequate, or unfairly
discriminatory. Could the public disclosure of proprietary
underwriting standards undermine the foundation of fairness
that individual States have implemented?
Ms. Dwyer. Yes, it could. Normally, those types of factors
are not public.
Mr. Tipton. Great. Thank you. My time has expired, Mr.
Chairman.
Chairman Clay. The gentleman's time has expired. I now
recognize the gentleman from Missouri, Mr. Luetkemeyer, for 5
minutes.
Mr. Luetkemeyer. Thank you, Mr. Chairman. And I welcome the
witnesses. I guess I want to start with, what is the definition
of insurance? I think it is the company accepting the risk to
pay the losses that are incurred as a result of an automobile
accident. Is that a fair explanation of what insurance is, and
the risk that you are taking?
Ms. Collins, you made the comment that the companies' rate
for risk, and I think two or three of you already said that the
rate factors are actuarily sound, so they can't go out, and
there are laws in place that prohibit discrimination for
certain things. So after that, we get to the point where, okay,
how do we charge for the risk?
In flood insurance, we have one rate. If you have a $50,000
house, you pay $750. If you have a $250,000 house, you pay
$750. Mr. Heller, do you think that is fair?
Mr. Heller. The one rate for flood insurance, I don't think
that is exactly how it works, because it is relative to where
you are vis-a-vis the flood map.
Mr. Luetkemeyer. It is pretty well one rate. We do not
discriminate between the size of the risk. Do you think having
one rate across-the-board is fair? That is my question.
Mr. Heller. I think that one rate would be an acceptable
way to go if we are trying to create the biggest pool to bring
people in. I think it is better to--
Mr. Luetkemeyer. If you have a $250,000 house, you would
rather pay less for your insurance than me, who would have to
subsidize your rate. Is that what you are saying?
Mr. Heller. No. I guess I am not following your question,
because there are different rates depending on what--
Mr. Luetkemeyer. Okay. Let's throw this out, since you are
not getting it right. Okay. If you want to have one rate for
everything--
Mr. Heller. Oh, I see.
Mr. Luetkemeyer. Okay. So, we are taking car insurance and
everybody pays the same rate, period, across-the-board. Is that
fair?
Mr. Heller. If that would be the case, no. It would not be
fair--
Mr. Luetkemeyer. In other words, if my son--he is a lot
older than 19, but let's say he is 19 and he has two DWIs and
two wrecks on his record, and he buys a new Corvette. Do you
want to pay the same rate as he does?
Mr. Heller. No. I believe that his risk would be higher--
Mr. Luetkemeyer. Okay. So we agree that there are rating
factors that should be an important part of this discussion.
Mr. Heller. That is absolutely right.
Mr. Luetkemeyer. Okay. So where do we go with the rating
factors? We have laws that say certain ones are legal and
certain ones are not.
Mr. Poe is concerned about some of the rating factors
because his company is getting beat by Geico and Progressive
because they decided to find different rating factors that they
could use to be able to frame their rates differently.
So, in a free market society, if I like Snickers more than
anybody else, and insurance companies decide they want to go
out here and figure out if people who eat Snickers candy bars
are more or less likely to have a wreck, do you think that is a
good rating factor, if they want to take the risk on whether
that is right or not?
Mr. Heller. Sir, the government--
Mr. Luetkemeyer. Answer the question. Yes or no?
Mr. Heller. They don't require people to buy Snickers. They
do require people to buy auto insurance. It is a different
market.
Mr. Luetkemeyer. Mr. Heller, they require them only to buy
liability and uninsured motorist, which is for somebody else,
not for you, okay.
Mr. Heller. And yet they discriminate based on their--
Mr. Luetkemeyer. Do you want to be hit by an uninsured
motorist and suffer the loss for lots and lots of health
issues, and not have insurance to pay for that, not have
somebody reimburse you for your car that has been damaged, if
you have a brand-new Cadillac? Is that fair? Yes or no?
Mr. Heller. That is not fair, and that is why we want to
get more people--
Mr. Luetkemeyer. Okay. This is why--
Mr. Heller. --into the market so they are not uninsured.
Mr. Luetkemeyer. --the States have mandated that you have
at least liability insurance. Now, if you have full coverage,
that is required by your lienholder, who, guess what, they want
to be able to pay for that car if it has a wreck, right? So,
let's don't go there now and say you are required to do this.
You are not required to have full coverage unless you want it,
on your own car, unless a lienholder is there.
Mr. Heller. And, sir, all of the rate--
Mr. Luetkemeyer. Is that right? Am I correct?
Mr. Heller. --quotes that I gave are for liability-only
coverage. We discriminate on liability for credit. You use your
credit score just for the liability policy.
Mr. Luetkemeyer. I am not there yet. I am talking about
different criteria, and we are going to get there. Just a
second.
Mr. Heller. Let's only talk about liability-only, the one
that you are required to buy. I think that is fair.
Mr. Luetkemeyer. We are talking about different rating
factors here. My comment and question to you is, if I, as a
business in the free market, want to decide if anybody who eats
Snickers bars gets a discount because I think they are going to
be more--because of the sugar high they are on, they will be
able to react more quickly. Therefore, I think they are going
to be a better risk. Is that okay for me to do that?
Mr. Heller. Not in the auto insurance market, because there
is no justification for that.
Mr. Luetkemeyer. If I can actuarily show that, if these
ladies over here said I could actuarily show that, is that
fair?
Mr. Heller. No, I don't think that would be fair.
Mr. Luetkemeyer. You don't believe in actuarily having
sound rates?
Mr. Heller. I think that what we need to do is we need to
focus our rates on a product that we require people to buy,
based on driving rate.
Mr. Luetkemeyer. No, you are missing the whole point. You
are going off on another tangent, sir.
Mr. Heller. Forgive me.
Mr. Luetkemeyer. Sir, I have the microphone. I'm sorry. It
is my question.
Ms. Collins, do you think I am going down the right road
here?
Ms. Collins. Yes, Congressman. I think it makes perfect
sense. If you believe that rates should match risk then it
makes no sense to--
Mr. Luetkemeyer. In a free-market society, is it okay for
the company who decides they want to slice the bread thinner,
if they come out with a different way to look at this that is
within the law, and they want to decide if I can eat a Snickers
bar and I am more astute behind the wheel, I can be a better
driver, is it okay for them to give me a discount, and let
somebody else have to pay more? Is that fair?
Ms. Collins. So long as the factors are actuarily sound
and--
Mr. Luetkemeyer. And you approve, and they can actually
show that I am a better driver because of that?
Ms. Collins. We would support that, yes.
Mr. Luetkemeyer. Thank you very much. I yield back.
Chairman Clay. The gentleman's time has expired. I now
recognize the gentleman from Ohio, Mr. Davidson, for 5 minutes.
Mr. Davidson. I thank the chairman. And I thank the
witnesses for coming, and I'm hoping to learn more for the good
of the American people and the preservation of a very sound
State-based automobile insurance market.
Here we are yet again, visiting an issue that this
committee held a hearing on last year, and multiple bills and
amendments have failed since then. When we discussed this topic
last year, Congresswoman Beatty from Ohio put it well. She
said, ``This seems like a Michigan problem to me. Why should we
use Michigan, with a deeply flawed system, as evidence that the
Federal Government needs to usurp State laws and change how
insurance rates are calculated?''
I agree with her. Auto insurance costs an Ohio driver, on
average, just $952 a year, compared to $2,484 a year in
Michigan. Better yet, auto insurance costs $1,277 a year, on
average, in Cleveland, whereas it is $5,414 a year in Detroit.
Now, these two cities have similar demographics, and these
figures aren't mined. They are figures that Josh Rivera, from
the University of Michigan, gathered and wrote about. I would
like to enter his report into the record.
Chairman Clay. Without objection, it is so ordered.
Mr. Davidson. I thank the chairman.
So the conclusion I am supposed to draw from the dialogue
here, or from some of my colleagues, is that to use credit
scoring data by auto insurers amounts to some sort of racist or
misogynistic conspiracy by the auto industry, in response to
higher auto insurance rates in Michigan. Why not Cleveland? Is
there no one with a low credit score in Cleveland? Just
Detroit? They have low credit scores in Detroit and nobody in
Cleveland has low credit scores? There are no racial minorities
in Cleveland? There are no men and women differences in
Cleveland versus Detroit? It makes no sense to disrupt the
system based off of these arguments.
I would like to flesh out Congresswoman Tlaib's argument.
We keep on discussing driving history only for pricing. There
are a lot of folks who believe that minorities are more likely
to be pulled over, ticketed, or arrested. Does that factor in?
There are a host of factors that could explain some of these
differences, and the question is, can insurance companies take
into account actuarily sound data in their pricing models?
So Ms. Collins, what is the right way to factor these
considerations in under actuarial standards and the bounds of
the law?
Ms. Collins. Thank you, Congressman. Insurance companies
look at a variety of factors to try to identify risk as closely
as they can with the cost that would be incurred, and the
ability for them to be able to do that is what keeps the market
functioning. And the issues associated in Michigan have really
very little to do with that process, as we noted earlier, and
the medical costs associated in that system.
So to directly answer your question, the impact of not
being able to match that rate to that risk would cause
uncertainty for insurance companies, and that would lead to a
direct rise in costs.
Mr. Davidson. Thank you for the explanation and further
clarification that, frankly, my colleague from Michigan is
trying to deal with State law by imposing a Federal standard
that could make Ohio as bad as Michigan, frankly. Instead,
perhaps we could consider how the State legislature of Michigan
could change the liabilities that insurers could price, that
individuals could have in the market in Michigan. So that would
be in the statehouse in Michigan, not here in Congress, in
Washington, D.C.
My colleague would argue that we should do away with other
factors and just use driving history, but if it is true, then
how do we price in factors like liability in States like State
law? We should be able to use sound practices and we should be
able to do it based on the law of the jurisdiction where the
insurance coverage is being priced.
It seems to me that what my colleagues are actually wanting
is a sane price regardless of risk. They want to socialize the
entire risk pool so that everyone is treated the same.
However, risk-based pricing is a very basic lending
principle, and publishing insurers' proprietary underwriting
data will crush competition within the auto insurance industry.
Only those who have not spent time in the business, and haven't
developed their own intellectual property, would suggest doing
away with it.
I yield back.
Chairman Clay. The gentleman yields back. I now recognize
the gentleman from North Carolina, Mr. Budd, for 5 minutes.
Mr. Budd. Thank you for the time, Mr. Chairman. Also, thank
you for the time in your district. It has a very nice barbecue
restaurant. So, I look forward to having you in North Carolina
so you can try some Davidson County, Lexington-style barbecue.
Chairman Clay. Thank you.
Mr. Budd. I also want to thank the witnesses for being here
today. I have spoken on one of these two bills that we are
considering, multiple times, and it is clear there isn't
support in this committee for Congress to pull credit scoring
data out of auto insurance underwriting. So as I mentioned in
the past, auto insurance costs about $85 a month, on average,
and that is in North Carolina. That is the fourth-lowest rate
in the country. As we say in my district, and probably
elsewhere, ``If it ain't broke, don't fix it.''
So that said, the FAIR Study Act is new, and I have done my
due diligence. While you call it a study bill, it is anything
but, and it is certainly not fair, so quite a misnomer here.
Compliance costs aside, the bill directs the Federal
Insurance Office to publish all of the proprietary information
that it collects from auto insurers on an annual basis. The
bill would destroy the foundation of competition in the auto
insurance market. Everybody who has seen a football game, or
just anything on TV, has seen 15 ads for auto insurance, and
they inherently know that it is a competitive market, with lots
of options for consumers. Luckily, there is no way to quantify
that.
The Department of Justice, which is responsible for
enforcing anti-trust laws, uses something called the
Herfindahl-Hirschman Index. I have a couple of business degrees
from pretty schools and I had never heard of that. But I have
seen it recently with some bank mergers, and now I see it
related to this.
So when the score gets to 2,500 or above, the index signals
market consolidation that is actually harmful to consumers, so
it is a helpful index. But the Herfindahl-Hirschman Index for
the property and casualty insurance industry--remember now,
2,500 is harmful--but for the auto insurance industry, it is
302, and that was in 2018. So, very low. And that is down from
346, also another low number, in 1998. So, it has gotten even
better in 20 years.
Ms. Collins, can you explain for the committee the actual
impact of the FAIR Study Act, if it were enacted? If the FIO
published proprietary underwriting information from every
participating auto insurer, what kind of pressure would it
create for consumers?
Ms. Collins. Thank you, Congressman. If proprietary
information were to be made public, I believe that it would be
hugely detrimental to consumers. As I noted at the outset,
consumers are benefitted by a highly competitive market, which
the auto insurance industry is, in that they compete against
each other to most accurately match rate to risk, and that
benefits consumers both in their choice in products and the
level of protection that they feel that they need, and serves
to keep prices as low as possible.
And so, removing the ability of insurance companies to
compete against each other to try to find the most effective
ways to match rate to risk would serve to take away those
benefits to the consumers.
Mr. Budd. Thank you for that. My friends across the aisle
claim that they want to stop too-big-to-fail financial
institutions from forming or market monopolies taking hold, yet
here is an example of a proposal that would kill competition in
one of the most competitive markets in the country. So,
Economics 101, the competition brings down prices for
consumers--we all know that--which is what I thought was the
original goal of these bills.
But the irony is that these bills would actually hurt
instead of help the policyholders. By eliminating and studying
ways to reduce a vast amount of factors used in underwriting,
things which would help these companies become more competitive
and better consumers, including making proprietary underwriting
information publicly available, what you are actually doing is
raising rates for consumers who currently shop for insurance,
again, in this very competitive market.
So, a question. If insurance would have to price risk based
on guesswork, it is an obvious conclusion that the least risky
drivers will have to subsidize the riskiest drivers. So, Ms.
Collins, how would you respond to this?
Ms. Collins. Yes, Congressman, we would agree. By making
insurance companies guess at what their exposure would be in
order to remain solvent, they would necessarily have to
increase rates just to meet their statutory obligation to their
policyholders. So, we totally agree with that sentiment.
Mr. Budd. Thereby hurting consumers.
Ms. Collins. Absolutely.
Mr. Budd. Thank you for your time.
Chairman Clay. The gentleman yields back. Now, we go to the
gentleman from Wisconsin, Mr. Steil, for 5 minutes.
Mr. Steil. Thank you very much, Mr. Chairman. Thank you for
holding today's hearing, and thank you to our witnesses for
being here.
Can I just start off by getting a show of hands if you
believe that non-driving-related factors are predictive of
risk?
[show of hands]
Mr. Steil. Three out of five, maybe.
Mr. Heller. There are probably some that have some value,
but not the ones we have spoken about today. I do not believe
they are predictive of risk.
Mr. Steil. Very good. I appreciate that insight. Ms.
Collins, I would like to ask you, if we look at drivers' actual
records, driving violations, amount of severity, and then you
start to look at if there are other factors that an auto
insurer might need to account for in the underwriting to cover
operating costs such as the regulatory cost of providing
insurance in a jurisdiction, the severity of auto accidents and
the injuries they cause, the prevalence of auto theft, the
likelihood of filing an auto claim, the probability of
policyholders defaulting on their premiums or any debt owed as
a result of an accident, or the rate of uninsured motorists
nevertheless driving in any given State, and since insurance
companies have a solvency obligation to operate in a safe and
sound manner, should companies be prohibited by Federal law
from including any of those pricing inputs or other pricing
inputs that you are looking at into their underwriting models?
Ms. Collins. Thank you, Congressman. Absolutely not. They
should not be prevented from doing that. The objective of
underwriting, again, is to match most closely the premium and
the rate to the risk that is represented, and the more tools
that are actuarily sound and on point to get closer to that
ideal, the more successful that insurance company is going to
be in getting the right product to the right consumer's risk,
and be able to write broader risks for more consumers.
Mr. Steil. Thank you. And I think it is worth pointing out
the challenges of when we socialize risk. And so, all things
being equal, with the average consumer's rates increase,
companies can no longer use predictive and approved factors. In
particular, in this instance we are talking about credit
history.
Ms. Collins. Yes, Congressman. Removing the innovation and
the trend towards finding new ways to identify and measure risk
would take us backwards in the United States to rather a crude
version of underwriting, and it would harm consumers in that
the insurance company would have to guess, again, to use the
other Congressman's word, at what the risk might be, and
therefore the lower-risk drivers would necessarily be
subsidizing the higher-risk drivers in order to create that
money for the pool to ensure that the insurance company could
meet their obligation to their policyholders.
Mr. Steil. And as we talked today, and we heard some of my
colleagues discuss involving the Federal Government in what has
traditionally been a State-regulated agency, could you provide
a little color and describe the process by which companies
identify and evaluate risk factors? And in particular, can you
highlight the role that State regulators are playing in
permitting the use of specific rating factors?
Ms. Collins. Insurance companies use a multitude of risk
factors. Not all insurance companies use all factors. As we
have noted, the actuarial science shows that it is most
successful when a combination of factors is in play. And this
is important in the context of State regulation, because the
State regulators review how insurance companies combine those
factors, as well as the individual factors themselves, to
ensure that nothing is unfairly discriminatory and that rates
are not excessive or inadequate.
Mr. Steil. So State insurance regulators are analyzing what
constitutes unfair discrimination and what is prohibited at the
State law level?
Ms. Collins. Absolutely, and they do so robustly.
Mr. Steil. I appreciate your testimony here today, and I
yield back.
Chairman Clay. The gentleman yields back. At this time, the
gentleman from California, Mr. Sherman, who is also the Chair
of our Subcommittee on Investor Protection, Entrepreneurship,
and Capital Markets, is recognized for 5 minutes.
Mr. Sherman. Mr. Chairman, I have been absent from this
room because we had the briefing on the coronavirus by our Vice
President. I am now totally reassured, maybe, and look forward
to hearing from our witnesses.
When I was 16, I was told that 16-year-olds are bad
drivers. Ms. Collins, of course, 16-year-olds, when they first
get their license, don't have bad driving records. Do most
insurance companies charge you more when you are 16 than when
you are 26, and is there a real correlation between being a new
driver with a spotless record and not being a good risk for the
insurance company?
Ms. Collins. Thank you very much for the question. Yes,
there is actuarial science to show that less-experienced
drivers and younger drivers do pose a higher risk of loss, and
that has been studied a multitude of times. And there was a
NHTSA study that showed that there are additional factors
associated with age. For example, the incidence of texting
while driving tends to diminish as a driver becomes more
experienced and older as well.
Mr. Sherman. Is it a matter of driving experience? Some
people get their license when they are 16. Some may first get
their license when they are 26. Do most companies look at age
or number of years driving or number of miles driven? Kids at
my high school drove a whole lot of miles. At my wife's high
school, even if you had your license, you took the subway. Is
it inexperience in terms of years holding the license,
inexperience in numbers of miles driven, or just age?
Ms. Collins. I believe that insurance companies use a
multitude of those factors to form a full picture of risk, but
I would defer to Superintendent Dwyer to elaborate.
Mr. Sherman. And while we do that, I will go to the other
end of the age spectrum. When my mother was 89, she still had
her driver's license. I wouldn't let me my kids drive with her.
Ultimately, I persuaded her not to drive at all. She hadn't had
an accident in at least a decade because she only drove like 12
miles a month. Trust me. Those are miles you didn't want to be
in her car. So is it legitimate, Ms. Dwyer, for them to look at
either young or old age in determining the level of risk?
Ms. Dwyer. When insurance companies look at any of those
factors and file them with their rating plan, they have to
provide the actuarial basis. So, you can't just guess.
Mr. Sherman. Right. They have to show you statistics. What
level of correlation or regression analysis--because some
things correlate a little bit and some things are one to one.
Is there any level of certainty or correlation that is
necessary?
Ms. Dwyer. I have an actuary look at every rate filing for
automobile insurance filed in my State. I am not an actuary, so
I am not absolutely sure on that. But I make sure that my
actuary is satisfied with the level of correlation. So it has
to be something significant, not a very vague correlation. It
has to be something significant.
Mr. Sherman. Is there any State that prohibits looking at
age or experience rather than actual tickets and accidents of
that individual driver?
Ms. Dwyer. I don't believe so, but we could--
Mr. Sherman. So as far as you know, every State would allow
it? Mr. Heller?
Mr. Heller. Thank you, Mr. Sherman, and I should mention I
am also a constituent of yours, so I am happy to see you here.
I believe that there are--while States don't prohibit that,
there are several companies now, for example, who won't even
ask how many miles you are driving, so they are not looking at
some of the real, legitimate factors that do relate, that we
can show the data. Instead, they turn to these other factors
that we have talked about all day, like your credit score and
your job title.
Mr. Sherman. Well, I am with the California Auto Club, and
they hound me 3 or 4 times before I finally am able to remember
how many miles I have on each odometer on each coast, and then
they give me a discount if I get the document in on time.
Obviously, number of miles driven is important.
Superintendent Dwyer, are there any States that require
companies to look at such a relevant factor as how many miles
do you drive every year?
Ms. Dwyer. I don't believe so. However, we have that in
most filings. Disclosure by the consumer is not necessarily the
best way to determine that, from what I have heard from
companies. They are getting this information through big data
occasionally. But they are using it.
Mr. Sherman. No, trust me. The Chinese don't know how many
miles I drive each of my cars yet, as far as I know. But if we
just look at whether you have had an accident recently, we are
missing out on the fact that my high school friends had
terrible attitudes towards driving, my mother had some
problems, and I only drive a few miles here in Washington.
Mr. Heller. Mr. Sherman, if I could clarify, though?
Mr. Sherman. I will ask the chairman if I have time.
Mr. Heller. Oh, I am sorry. Mr. Chairman, if I could just
clarify one point that was made by the superintendent, very
briefly. Just that California does require that miles are used
in rating. That is all. California law does require that.
Mr. Sherman. That is a good thing.
Chairman Clay. The gentleman yields back. At this time, I
am going to take 2 minutes to close, and in that 2 minutes, I
am going to yield to my friend from Michigan, since this
subject has generated quite a bit of discussion. It is her
bill, so I yield to the gentlewoman from Michigan, Ms. Tlaib,
for 2 minutes for closing remarks.
Ms. Tlaib. Thank you, Mr. Chairman. Look, I understand that
there is a whole formula within the industry, but I feel like
the formula right now has led towards discriminatory practices,
because the data and information out there is very clear, and I
think it is very important for all of us. And I will work with
colleagues on the other side of the aisle to actually have the
Federal Insurance Office look at it, whether or not there is
disparate impact in this formula.
It is the Federal Government's job, and I am not trying to
change State law. I am trying to ensure that people on the
ground, people like my community, which is the third-poorest in
the country, are not being directly discriminated against,
based on these factors.
And it is sincere. It is the fact that I have seen data
that it is preventing them from owning their own homes, because
they are paying these high rates. But I also feel like it is
very much discriminatory and dehumanizing to be asked about
education level, to be asked about marital status, and to see
that somehow you are now going to be treated differently,
solely based on your credit score.
For me, having a driving-under-the-influence violation is a
huge risk, but that person is going to pay less than the person
who doesn't have a great credit score. Just the data in Florida
is unbelievable, to see that somebody with poor credit is
paying $3,826, and somebody with better credit is paying
$1,400. What is going on here?
These are proxies, and we are not going to allow them. Mr.
Heller, we have to prohibit the collection of racial data by
insurance, and we do that. But these are now being used as
proxies.
Prove me wrong that they are not, by allowing a disparate
impact study to actually happen. What are you afraid of it
showing? Prove us wrong by allowing the Federal Insurance
Office to investigate this and look at this.
Thank you so much, Mr. Chairman, for this hearing. I know
many of my residents at home really do appreciate this.
Chairman Clay. The gentlewoman yields back, and I now
recognize the ranking member of the subcommittee, Mr. Stivers.
Mr. Stivers. Thank you, and I will be fairly brief. I will
conclude as I started. Thank you to the chairman for holding
this hearing.
Everyone in this room is opposed to discrimination. There
was great testimony by Ms. Dwyer about what our insurance
commissioners at the State levels are doing to make sure we
stop discrimination based on factors that they shouldn't use to
discriminate. But we also need to figure out how to create an
insurance system that is priced fairly for everyone.
And generally, under McCarran-Ferguson, I trust our State-
based regulators to do that job. I think there have been some
really good studies that have been done, including the Stanford
University study that shows if you move away from factors and
only do driving factors, you actually make the situation worse
for African Americans and other folks who face some
discrimination in the way that they are pulled over and the way
they are ticketed.
So, there may be no perfect system. There may be no perfect
underwriting factors. And I am not saying that there isn't a
study we couldn't work on. But the study as it is worded in
this bill that we talked about today has a lot of conclusions
that aren't based on any facts. And I wouldn't want to support
a study that already has conclusions in it. That is not fair. A
study should be a study that looks at things fairly, and then
makes conclusions based on what is observed, not conclusions
based on when you authorize the study.
So while I would be willing to work on that with the
gentlelady from Michigan, the chairman or the gentlelady from
Michigan, I think we would have to do that in an appropriate
way that doesn't start with conclusions. It might end with
conclusions, but it shouldn't start with conclusions, and that
is my concern about this study that was proposed today.
I think there has been a lot of great testimony by all of
the witnesses. Again, I want to thank all of the witnesses for
your perspectives, and while I trust our State-based
regulators, I am not saying there is not a Federal role. I
would want to make sure if there was a Federal role, it was
done in a thoughtful way, without conclusions up front. I yield
back, Mr. Chairman.
Chairman Clay. The gentleman yields back, and I would like
to thank our witnesses for their testimony today.
The Chair notes that some Members may have additional
questions for this panel, which they may wish to submit in
writing. Without objection, the hearing record will remain open
for 5 legislative days for Members to submit written questions
to these witnesses and to place their responses in the record.
Also, without objection, Members will have 5 legislative days
to submit extraneous materials to the Chair for inclusion in
the record.
This hearing is now adjourned.
[Whereupon, at 3:45 p.m., the hearing was adjourned.]
A P P E N D I X
March 4, 2020
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]