[House Hearing, 116 Congress] [From the U.S. Government Publishing Office] EXAMINING DISCRIMINATION IN THE AUTOMOBILE LOAN AND INSURANCE INDUSTRIES ======================================================================= HEARING BEFORE THE SUBCOMMITTEE ON OVERSIGHT AND INVESTIGATIONS OF THE COMMITTEE ON FINANCIAL SERVICES U.S. HOUSE OF REPRESENTATIVES ONE HUNDRED SIXTEENTH CONGRESS FIRST SESSION ---------- MAY 1, 2019 ---------- Printed for the use of the Committee on Financial Services Serial No. 116-21 [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] EXAMINING DISCRIMINATION IN THE AUTOMOBILE LOAN AND INSURANCE INDUSTRIES EXAMINING DISCRIMINATION IN THE AUTOMOBILE LOAN AND INSURANCE INDUSTRIES ======================================================================= HEARING BEFORE THE SUBCOMMITTEE ON OVERSIGHT AND INVESTIGATIONS OF THE COMMITTEE ON FINANCIAL SERVICES U.S. HOUSE OF REPRESENTATIVES ONE HUNDRED SIXTEENTH CONGRESS FIRST SESSION __________ MAY 1, 2019 __________ Printed for the use of the Committee on Financial Services Serial No. 116-21 [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT] ______ U.S. GOVERNMENT PUBLISHING OFFICE 37-520 PDF WASHINGTON : 2020 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 PETER T. KING, New York 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 SEAN P. DUFFY, Wisconsin ED PERLMUTTER, Colorado STEVE STIVERS, Ohio JIM A. HIMES, Connecticut ANN WAGNER, Missouri BILL FOSTER, Illinois ANDY BARR, Kentucky JOYCE BEATTY, Ohio SCOTT TIPTON, Colorado DENNY HECK, Washington ROGER WILLIAMS, Texas JUAN VARGAS, California FRENCH HILL, Arkansas JOSH GOTTHEIMER, New Jersey TOM EMMER, Minnesota VICENTE GONZALEZ, Texas LEE M. ZELDIN, New York AL LAWSON, Florida BARRY LOUDERMILK, Georgia MICHAEL SAN NICOLAS, Guam ALEXANDER X. MOONEY, West Virginia RASHIDA TLAIB, Michigan WARREN DAVIDSON, Ohio KATIE PORTER, California TED BUDD, North Carolina CINDY AXNE, Iowa DAVID KUSTOFF, Tennessee SEAN CASTEN, Illinois TREY HOLLINGSWORTH, Indiana AYANNA PRESSLEY, Massachusetts ANTHONY GONZALEZ, Ohio BEN McADAMS, Utah JOHN ROSE, Tennessee ALEXANDRIA OCASIO-CORTEZ, New York BRYAN STEIL, Wisconsin JENNIFER WEXTON, Virginia LANCE GOODEN, Texas STEPHEN F. LYNCH, Massachusetts DENVER RIGGLEMAN, Virginia 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 Oversight and Investigations AL GREEN, Texas Chairman JOYCE BEATTY, Ohio ANDY BARR, Kentucky, Ranking STEPHEN F. LYNCH, Massachusetts Member NYDIA M. VELAZQUEZ, New York BILL POSEY, Florida ED PERLMUTTER, Colorado LEE M. ZELDIN, New York, Vice RASHIDA TLAIB, Michigan Ranking Member SEAN CASTEN, Illinois BARRY LOUDERMILK, Georgia MADELEINE DEAN, Pennsylvania WARREN DAVIDSON, Ohio SYLVIA GARCIA, Texas JOHN ROSE, Tennessee DEAN PHILLIPS, Minnesota BRYAN STEIL, Wisconsin C O N T E N T S ---------- Page Hearing held on: May 1, 2019.................................................. 1 Appendix: May 1, 2019.................................................. 41 WITNESSES Wednesday, May 1, 2019 Clarke, Kristen, President and Executive Director, Lawyers' Committee for Civil Rights Under Law........................... 9 Cross, Rachel J., Policy Analyst, Frontier Group................. 7 Lynch, James, Chief Actuary, and Senior Vice President, Research and Education, Insurance Information Institute................. 12 Rivera, Joshua, Data and Policy Advisor, Poverty Solutions at the University of Michigan......................................... 11 Van Alst, John W., Attorney, National Consumer Law Center (NCLC), and Director, Working Cars for Working Families, an NCLC Project........................................................ 5 APPENDIX Prepared statements: Clarke, Kristen,............................................ 42 Cross, Rachel J.............................................. 51 Lynch, James,................................................ 57 Rivera, Joshua............................................... 63 Van Alst, John,.............................................. 74 Additional Material Submitted for the Record Green, Hon. Al: Written statement of the National Association of Insurance Commissioners.............................................. 165 Written statement of the National Association of Mutual Insurance Companies........................................ 167 Written statement of UnidosUS................................ 173 Barr, Hon. Andy: Report entitled, ``Fair Lending: Implications for the Indirect Auto Finance Market,'' prepared by Charles River Associates for the American Financial Services Association, dated November 19, 2014.................................... 180 Report of the Insurance Research Council entitled, ``Auto Insurance Affordability: Cost Drivers in Michigan,'' dated April 2019................................................. 323 Joint insurance trades letter, dated April 29, 2019.......... 342 Written statement of the National Association of Mutual Insurance Companies........................................ 343 Written statement of the National Automobile Dealers Association................................................ 349 Beatty, Hon. Joyce: Written responses to questions for the record from Joshua Rivera..................................................... 359 Steil, Hon. Bryan: Written statement of the American Property Casualty Insurance Association................................................ 362 Zeldin, Hon. Lee: Written statement of the National Automobile Dealers Association................................................ 364 EXAMINING DISCRIMINATION IN THE AUTOMOBILE LOAN AND INSURANCE INDUSTRIES ---------- Wednesday, May 1, 2019 U.S. House of Representatives, Subcommittee on Oversight and Investigations, Committee on Financial Services, Washington, D.C. The subcommittee met, pursuant to notice, at 10:05 a.m., in room 2128, Rayburn House Office Building, Hon. Al Green [chairman of the subcommittee] presiding. Members present: Representatives Green, Beatty, Tlaib, Garcia of Texas; Barr, Posey, Zeldin, Loudermilk, Davidson, Rose, and Steil. Ex officio present: Representatives Waters and McHenry. Also present: Representative Budd. Chairman Green. Good morning, everyone. The Oversight and Investigations Subcommittee will come to order. The title of today's subcommittee hearing is, ``Examining Discrimination in the Automobile Loan and Insurance Industries.'' 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 the subcommittee may participate in today's hearing for the purposes of making an opening statement and questioning the witnesses. Mr. Budd appears to be here; I ask unanimous consent that he be allowed to participate. Without objection, it is so ordered. Today's hearing, as I indicated, is styled, ``Examining Discrimination in the Automobile Loan and Insurance Industries'', and it is long overdue. I now recognize myself for 5 minutes to give an opening statement. This hearing is long overdue. The empirical evidence has shown that non-white vehicle buyers who were more qualified than white buyers received costlier loans 62.5 percent of the time, according to a 2018 study by the National Fair Housing Alliance. African-American males pay more for dealer markups than white males, and these dealer markups, which are unrelated to credit risk, allow for predatory pricing and invidious discrimination. A 2015 Consumer Federation of America (CFA) study found that a driver living in a predominantly African-American neighborhood can expect to pay insurance premiums that are on average 70 percent more than similarly situated drivers in neighborhoods where African Americans are in the minority. A 2017 CFA study found that most auto insurers charge middle-aged women higher rates than men. In Houston Texas, women with perfect driving records pay on average $75 more than men with the same record, at the same address, with the same vehicle. At $1.26 trillion, auto loans are the third largest household debt category, after mortgage loans and student loans. What once was considered a luxury is now a necessity as evidenced by the fact that nearly every American household has at least one car. Let me summarize by saying this: I said that this hearing was long overdue and it is because the empirical evidence that I have cited--which is but a scintilla of what is available-- seems to overwhelmingly indicate that certain persons pay more than others for the same product, and these bits of evidence have been called to our attention by way of advocacy groups and persons who are intellectuals, scholarly persons, and they use a method called Testing. And using this Testing, this empirical evidence seems to be valid; it seems to indicate that there are problems that we have to address. The purpose of this hearing is not only to expose the empirical evidence but also to address or conclude in some way that there are means by which we can address the problems that will be called to our attention. I am grateful that we have witnesses here who can give us the additional information that we will need to draw our conclusions, but I also want to close with this: I think that we owe it to ourselves as a country to make sure every person is treated fairly when making an automobile purchase. We have the ability and the power and the authority in Congress to do this. The question is, do we have the will to make a difference for people who are in need of a car, who are probably not making as much money as Members of Congress but who dearly need the opportunity to have this necessity so that they can get to and from their jobs. And my hope is that we will be able to achieve that fairness through this hearing. With this said, I am going to reserve the balance of my time for Chairwoman Waters. I will now yield to the ranking member of the subcommittee, Mr. Barr. Mr. Barr. Thank you, Chairman Green. And I want to start by acknowledging our Subcommittee Chairman, Al Green, for convening this hearing. Thank you, Mr. Chairman, especially for all your efforts over the course of your congressional career and working to combat discrimination, race discrimination which is to be commended, and I am pleased to participate in this hearing. I welcome all of our witnesses; all of your work and efforts are a great value to our committee and we thank you for being here today. Obviously, race discrimination is abhorrent and it should not be tolerated in auto lending or the insurance industries or any industry. There is anecdotal evidence of auto lenders charging different rates based on a borrower's race so we are here to understand today how the industry has changed, how it prices risk without discriminating on the basis of race, and what needs to be done to make sure that everyone is treated fairly regardless of race. This committee has spent considerable time examining discrimination in auto lending and we take this issue very seriously. The committee has investigated and released several reports on this matter in the past and our work is ongoing. Studies purporting to show widespread discrimination in auto lending have routinely proven flawed, for example a 2013 report by the Consumer Financial Protection Bureau (CFPB) contended that dealer-assisted auto financing has a ``disparate impact on the price of credit for consumers in protected classes.'' As it turned out, the CFPB's methodology for counting African Americans was off by 41 percent, among many other problems. The committee's investigation of the matter found that the CFPB knew its methods were deeply flawed and prone to significant error, according to internal documents, but the CFPB released its study anyway. To be clear, racial discrimination in auto lending is illegal, as it should be. We must ensure that the industry does not move backwards and we need to base that work on accurate data and reliable findings. States continue to be the most effective regulators of the auto lending industry as has been the case for almost a century. This was certainly true in 2010 when the Dodd-Frank Act was enacted. Section 1029 of the Dodd- Frank Act specifically excluded auto dealers from the CFPB's jurisdiction; it was obvious then as it is now that States were best positioned to ensure that consumers are protected. We will also examine the auto insurance industry. The United States has the largest and most competitive market for auto insurance in the world. Like auto lending, the auto insurance industry is regulated at the State level. In fact, these are two of the most heavily regulated industries in the entire financial services portfolio. The price for auto insurance varies throughout the country depending on a litany of variables, and the competitive nature of this industry has led to more affordable prices for consumers. As with so many other things, the key to lower prices is free markets. We need look no further than Illinois to understand why; in Illinois, an inadvertent lapse in regulations allowed a flood of insurance companies to enter the market, that competition drove down prices for consumers, and the Illinois Legislature chose not to enact new rate regulations. It is not just competition that affects the price of insurance, State laws greatly affect prices as well. In 2019, Michigan was ranked as the State with the highest insurance premiums. The high prices in Michigan reflect decisions by the State Legislature and the insurance commissioner; Michigan's No-Fault Insurance Law provides for potentially unlimited lifetime medical assistance for people involved in accidents and this law leads to abuse of the legal system and fraud. Insurance experts have drawn a direct correlation between the No-Fault Insurance Laws and Michigan's high rates. Unlawful discrimination in setting insurance rates is inexcusable and it is already prohibited at the Federal and State level. In addition, industry standards forbid the use of certain factors such as race when determining rates. Those in the industry have worked to create a system where the insurance credit-ratings are purposely blind to the race of an applicant. I look forward to working with the chairman to ensure that discrimination does not occur in auto lending or insurance, but I also do not want to eliminate risk-based pricing in a way that could increase premiums for most drivers. Thank you. And I yield back. Chairman Green. Thank you. And the Chair will now yield 1 minute to the Chair of the full Financial Services Committee, the Honorable Maxine Waters. Chairwoman Waters. Thank you very much, Mr. Chairman. Buying a car is a significant purchase for many Americans and should be a fair and transparent transaction, free of discrimination. Unfortunately, this is not the case for persons of color. The National Fair Housing Alliance, in a test of auto lending discrimination, found that nearly two-thirds of minority loan applicants received higher-cost financing options from automobile dealers than less-qualified white applicants. Last Congress, Republicans impeded enforcement of fair lending laws, making discrimination potentially worse. They used the Congressional Review Act to rescind the Consumer Financial Protection Bureau's much-needed guidance to indirect auto lenders on how to comply with the Equal Credit Opportunity Act. Discrimination also exists in the auto insurance industry. For example, the Consumer Federation of America found that auto insurers charged women and persons living in predominantly African-American communities disproportionately higher premiums. These types of practices warrant congressional scrutiny, analysis, and ultimately legislation. With that, Mr. Chairman, I yield back the balance of my time. Chairman Green. The gentlelady yields back. The Chair now recognizes the ranking member of the Full Committee, the Honorable Mr. McHenry. Mr. McHenry. Thank you, Chairman Green. And thank you for your leadership, especially on this topic. And thank you to Ranking Member Barr, as well, for his leadership. So let me be clear, there is no place for discrimination, period. The idea that lenders or insurers would charge a different rate based on skin color, ethnicity, or gender is contrary to everything that we stand for as a society. To have this discussion, we must have the most accurate statistics available. As Congressman Barr outlined, the CFPB used an algorithm that attempted to identify the race of borrowers based simply on their name and address. I don't have to look any further than my own community to see the failure of that practice, seeing as there are two families named McHenry in my hometown, and our race and ethnicity is different. So a name does not do justice nor does a zip code simply do justice to an understanding of race. As it turns out, the CFPB later acknowledged that algorithm had a 20 percent error rate. And then an independent analysis of it said that it could be as high as 40 percent. Both the lending and insurance industries, which are the focus of today's hearing, are regulated at the State level. So I wanted to understand if there any limitations on the States taking appropriate action to make sure that this does not happen. So with that, thank you, Mr. Green. Chairman Green. The gentleman yields back. We now welcome our witnesses: Mr. John Van Alst, an attorney at the National Consumer Law Center, and the director of Working Cars for Working Families, an NCLC project; Ms. Rachel J. Cross, a policy analyst at Frontier Group; Ms. Kristen Clarke, president and executive director of Lawyers' Committee for Civil Rights Under Law; and Mr. Joshua Rivera-- Ms. Tlaib, a member of this committee, is from Michigan, so she will say more about Mr. Rivera. Ms. Tlaib. Oh, are you recognizing me, Mr. Chairman? Chairman Green. Yes. Ms. Tlaib. Thank you so much. I just want to thank so much Mr. Rivera, who co-authored, ``Auto Insurance and Economic Mobility in Michigan,'' but I also want my colleagues to know our emphasis and advocacy around non-driving factors, which go beyond just race-based kind of discrimination. I am so thankful for the leadership of Ranking Member Barr, Chairman Green, and our incredible Chairwoman Maxine Waters for making this a very important critical issue to talk about the fact that we are increasingly using non-driving factors, people who become widowed, retired, and so forth are being considered as factors against car rates. Thank you so much, Mr. Chairman. Chairman Green. Thank you for your remarks. Moving on, the final witness will be Mr. James Lynch, the senior vice president of research and education at the Insurance Information Institute. I would like to welcome and thank you all for being here. The witnesses will each be recognized for 5 minutes to give an oral presentation of their testimony. And without objection, the witnesses' written statements will be made a part of the record. Once the witnesses have finished presenting their testimony, each member of the subcommittee will have 5 minutes within which to ask questions. On your table, and this is directed to the witnesses, you will see three lights: the green light means that you may go; the yellow light is the indicator that you have 1 minute left and you are running out of time; and the red light of course means that you are out of time. The microphones are very sensitive so please make sure you speak directly into them. With that, Mr. Van Alst, you are now recognized for 5 minutes. STATEMENT OF JOHN W. VAN ALST, ATTORNEY, NATIONAL CONSUMER LAW CENTER (NCLC), AND DIRECTOR, WORKING CARS FOR WORKING FAMILIES, AN NCLC PROJECT Mr. Van Alst. Chairman Green, Ranking Member Barr, and distinguished members of the subcommittee, thank you for inviting me here today to discuss discrimination in cars. I am an attorney with the National Consumer Law Center where I work with Legal Services attorneys, government attorneys, and private attorneys across the country, all of whom help low-income families with issues with cars and car finance. I also direct NCLC's Working Cars for Working Families project which works to ensure that families get a fair deal when buying and financing a car, and that the lack of a car does not stand in the way of a family's ability to become economically successful. A car can provide physical mobility but also economic mobility, allowing families to get to work, live in more affordable housing, and take advantage of educational opportunities. While cars can be a tool to help escape poverty, they are very expensive. In 2018, the average used car price exceeded $20,000, and for a consumer with subprime credit, monthly payments were over $400 and the average interest rate was over 16 percent. But for some consumers, buying, financing, and using a car can be even more expensive because of their race and ethnicity. Some are charged hundreds and even thousands of dollars more in interest rate markups. These are discretionary increases that the dealer makes to the interest rates which have already been objectively set based on the consumer's credit risk. In 12 cases that the NCLC co-counseled between 1998 and 2007, we coded millions of transactions for race based upon driver's license data. Within each credit tier, dealers marked up African Americans' interest rates almost twice as often as whites. And when African Americans rates were marked up, the markups were on average almost double those charged to whites. And the disparities don't stop at financing. Some consumers are more likely to be pressured to buy add-on products such as service contracts, and GAP. They are then charged more for these same products they are pressured to buy. In 2017, we examined a data set of millions of add-on transactions and found that Hispanic car-buyers were charged higher markups for add-ons at both the State and dealer level. These disparities make cars more expensive for some races and ethnic groups. They unnecessarily increase the cost of a car and increase the chance of default. In some cases, these higher costs keep families from getting a car at all. This contributes to disparities we see across the country in terms of access to cars. For families at or below the Federal Poverty Guidelines, 13 percent of white households lack access to a car compared to 31 percent of African-American households and 20 percent of Hispanic households. Many of the disparities we will discuss here today are only possible because the market for cars and especially car- financing used cars is opaque and inconsistent. Dealers have tremendous discretion to charge consumers different prices and the price setting takes place in a back room where dealers have to decide quickly how far they can push the consumer, and no one knows what anyone else pays for the car, the financing or the add-ons. A more consistent and transparent marketplace would not only benefit consumers of color, but everyone including dealers and finance entities that want to compete fairly and openly on price and quality on a level playing field. Towards this goal, we must ban dealer interest-rate markups, amend the Equal Credit Opportunity Act to enable data collection, prohibit discrimination in the pricing of goods and services in addition to discrimination in financing, increase enforcement of the Equal Credit Opportunity Act; and increase enforcement of general protections against general abuses in the sale or financing of cars. Given the evidence we have seen in discrimination in financing and other parts of the sale transaction, it is likely that many other abuses from yo-yo sales, to failure to pay off existing liens, are more likely to affect people of color. I commend the subcommittee for holding today's hearing on such an important topic and we stand ready to work with this subcommittee and other interested parties in bringing consistency, transparency, and fairness to the auto market. Thank you. [The prepared statement of Mr. Van Alst can be found on page 74 of the appendix.] Chairman Green. Thank you. We will next hear from Rachel J. Cross. You have 5 minutes. STATEMENT OF RACHEL J. CROSS, POLICY ANALYST, FRONTIER GROUP Ms. Cross. Thank you for the opportunity to appear before you today to discuss the state of U.S. auto lending. I am a policy analyst with Frontier Group and I am also testifying on behalf of the U.S. PIRG, the Public Interest Research Group with whom we co-authored our recent report, ``Driving into Debt,'' which examines how auto lending has changed since the Great Recession and what this means for consumers. In much of the country, owning a car is a virtual necessity. It is how many of us get to work, to school, to the grocery store and doctor; a car, in short, is the price of admission to living a full and productive life, but owning a car is also expensive and has driven millions of households to take on debt. Right now, Americans owe more for our cars than we ever have before; outstanding auto debt is over $1.2 trillion, and since the end of 2009 that amount has increased by over 50 percent. But it is not just that the overall auto debt has reached historic levels; the number of Americans who owe for their cars is also the highest in U.S. history and consumers are at risk, delinquencies are rising, the percentage of auto debt that is seriously delinquent, meaning 90 days late at least or more, is the highest it has been since 2012 and it is still rising. More than 7 million Americans have missed at least 3 monthly car payments. These numbers are concerning on their own, but what makes them deeply troubling is that they are happening in a strong U.S. economy, something important has been happening in the auto credit market. The lending practices since the recession that have boosted auto sales have come at a cost, the increased risk of financial instability for millions of American households. There are a few key steps that have gotten us to the point that we are at now. In the aftermath of the 2008 financial crash, investors and lenders alike noticed that auto debt performed relatively well compared to other securities during the recession and this sparked more interest in bringing more borrowers into the auto credit market and lenders of all types did so. First, they loosened standards for prospective borrowers. We found in our report that auto debt has risen across all income levels but it has risen the fastest among those with the lowest incomes. Since 2009, borrowing by residents of low- income neighborhoods has increased nearly twice as quickly as borrowing by residents of the highest-income neighborhoods. Lending in the sub-prime market followed a similar trajectory. Lenders have used other tools to bring borrowers in the marketplace, including the lengthening of loan terms. Extending a loan term brings down that monthly payment, which is an important measure for determining affordability particularly for lower-income consumers, but it also means that the consumer will pay more over the life of the loan due to interest payments, and will spend more time underwater or owing more for their car, and that is even worse. Consumers with a 6-year loan are twice as likely to default as others with a 5-year loan, and those borrowers are also more likely to have a poor credit history, lower income, and to pay higher interest rates than other borrowers. In 2017, 42 percent of all loans generated had terms of 6 years or more. This period also saw the rise of more outright abusive and predatory tactics in one specific part of the auto credit market: dealer financing. A direct loan that a consumer gets directly from a financial institution like a bank or credit union is, generally speaking, a safer bet for consumers. Indirect lending, however, is when the consumer finances through a dealership, where the dealer is a creditor, assigns the loan to another financial institution, and often has the consumer sign a retail installment sales contract. Dealer-arranged financing has weaker regulations and oversight than direct loans, enabling a number of abusive and predatory tactics. One area of abuse is excessive interest rates. When a dealer sells its financed contract to another lender, they are able to mark up that interest rate and pocket the difference as profit. Having consumers sign a retail installment sales contract also allows the dealer to charge an interest rate that can sometimes exceed State usury limits. In one example, a package of securities that Santander Bank was selling to investors was found to have 57 percent of all loans generated in the State of New York to be carrying interest rates that are so high, they would have been illegal if they had been from direct loans between a consumer and a bank. But because they were dealer- financed and indirect loans, it was legal. There has also been evidence of lenders failing to verify the income of borrowers, and even car salesmen inflating the consumer's income to ensure that they qualify for financing even if they ultimately cannot afford it. In key respects, auto lending in the last decade has been a Groundhog-Day repeat of the exact same practices that brought us to the 2008 market crash. Dealer-arranged financing has also enabled discriminatory pricing. Since its creation, the Consumer Financial Protection Bureau has investigated a number of large captive finance groups that provide indirect financing for charging borrowers of color higher interest rates than similarly situated white borrowers. These include some of the largest indirect lending firms in the nation like Toyota Motor Credit. These policies led to many African-American borrowers paying $200 more on average for financing. The CFPB investigations have repeatedly found that lenders giving dealers the ability and incentive to mark up interest rates enables this kind of discrimination. However, since the congressional repeal of the CFPB's Indirect Auto Guidance and changes in leadership, we have seen a lack of will to continue protecting consumers as they deserve. These are only a few examples of the way that dealer financing threatens the financial well-being of Americans. The entire list of threats that consumers, in particular the most vulnerable amongst us face, is appalling, and that so little action has been taken to stop these predatory behaviors is even more appalling. Thank you. [The prepared statement of Ms. Cross can be found on page 51 of the appendix.] Chairman Green. We will now hear from Kristen Clarke. You are recognized for 5 minutes. STATEMENT OF KRISTEN CLARKE, PRESIDENT AND EXECUTIVE DIRECTOR, LAWYERS' COMMITTEE FOR CIVIL RIGHTS UNDER LAW Ms. Clarke. Chairman Green, Ranking Member Barr, and members of the subcommittee, my name is Kristen Clarke and I am president and executive director of the Lawyers' Committee for Civil Rights Under Law. Thank you for the opportunity to testify today about the discrimination in the automobile loan industry. The Lawyers' Committee is a national non-partisan civil rights organization created at the request of President John F. Kennedy to activate the private bar in the fight against discrimination, and for over 55 years we have been on the frontlines of our nation's fight for justice in the areas of economic justice, voting rights, fair housing, criminal justice, education, hate crimes, and more. We know that eliminating lending discrimination, root and branch, particularly across the auto industry must stand as a core civil rights priority today. Discriminatory auto lending poses grave financial danger to African-American, Latino, and other vulnerable consumers; increasingly, lenders are taking advantage of a highly unregulated and unmonitored market to promote discriminatory and predatory practices. We thank this committee for sounding an alarm about this and undertaking careful fact-finding to identify a remedial response. All Americans need reliable transportation to access jobs and schools, as has been noted, and we know that communities of color that are deprived of economic investment often lack effective public transportation, those who can't rely on public transportation often need an affordable loan to finance the purchase of a car and to help them build good credit. As the third largest source of outstanding household debt after mortgages and student loans, car loans and their impact on communities of color warrant greater scrutiny. There is a stark racial wealth divide in our country today. A typical white family has $140,000 in wealth, while a typical African-American family has just over $3,400. Abusive, predatory automobile lending and lack of access to equitable financial services contribute to and exacerbate this racial wealth gap. Widespread racial discrimination in the market is not new and has been well documented over the past 30 years. African Americans and other consumers of color face discrimination in various ways when purchasing and financing a car. They are often charged higher prices, higher interest rates, and more expensive insurance rates on the basis of their race or ethnicity, and they are also targeted for predatory sales and repossession schemes. Racial bias also seeps into the industry by way of discretionary dealer markups and often indirect auto lenders fail to have controls in place to prevent discretionary markup- pricing disparities resulting from car dealers' racial bias, resulting in people of color being burdened with more expensive loans than white consumers. While much lending discrimination can be identified through statistics, there is also evidence of intentional discrimination throughout the industry. In 2014 the Justice Department secured settlements with Auto Fare and Southeastern Auto Corp in North Carolina for engaging in reverse redlining. The dealerships targeted black consumers with unfair and predatory credit practices and the dealership operator expressed his view that African-American customers have fewer credit options, making them more likely to accept predatory contracts. A recent 2018 study by the National Fair Housing Alliance used matched-pair testing in Virginia that showed that almost 63 percent of the time, borrowers of color received costlier loans even though they were more qualified than their white counterparts. Under this Administration, the Justice Department has retreated from Fair Lending enforcement against dealers, not surprisingly after the rollback of the critical 2013 CFPB Guidance we saw several lenders who had previously implemented flat-fee models, revert back to discretionary dealer-rate programs. What must be done? We ask Congress to enact robust legislation to increase protections for borrowers of color by prohibiting or significantly limiting discretionary dealer markups. We need more oversight over the Justice Department and the CFPB to understand why their enforcement activity has come to a grinding halt. And finally we need more transparency and data collection to help us get to better understand that the extent of discrimination across the auto lending industry. Thank you. [The prepared statement of Ms. Clarke can be found on page 42 of the appendix.] Chairman Green. Thank you. Mr. Joshua Rivera, you are now recognized for 5 minutes. STATEMENT OF JOSHUA RIVERA, DATA AND POLICY ADVISOR, POVERTY SOLUTIONS AT THE UNIVERSITY OF MICHIGAN Mr. Rivera. Chairman Green, Ranking Member Barr, and distinguished members of the subcommittee, thank you for inviting me to testify today. I appreciate the opportunity to discuss auto insurance and affordability and the disproportionate impact of certain rate-setting practices on low-income and minority drivers. Our team at Poverty Solutions first got the idea to research auto insurance directly from Detroit residents; to our surprise, in conversation after conversation with local stakeholders, the issue of auto insurance just kept coming up as a major poverty issue. When we started to look at the data what we found stunned us. Michigan has the most expensive automobile insurance in the U.S., with an estimated annual premium in 2018 of $2,610, almost double the national average. And with an average annual premium of $5,414, Detroiters fac the most expensive car insurance rates in the country. The sticker shock prompted us to research how the State got here in the first place. The resulting policy brief, ``Auto Insurance and Economic Mobility in Michigan: A Cycle of Poverty,'' documents the problems and offers potential solutions for auto insurance reform in Michigan. I have included a copy of the policy brief with my written testimony for the record. Michigan is not alone in facing rising auto insurance rates. In 2018, 184 million U.S. drivers, or nearly 4 out of every 5 drivers, faced a rate increase. In addition to Michigan, Louisiana, Rhode Island, and Florida also faced annual average premiums that are above $2,000. The U.S. Treasury Department's Federal Insurance Office deems auto insurance unaffordable in areas where premiums exceed 2 percent of a zip code's median household-income. Using recent data, we found that auto insurance rates represent more than 2 percent of median household-income in 97 percent of Michigan's zip codes, yet the burden is substantially greater for lower-income and minority communities. In places like Flint, rates eat up between 8 and 24 percent of residents' pre-tax income and we also found that rural communities were especially hit hard by high rates; this matters because these communities deal with particularly poor roads and they have to drive longer distances for employment, which makes affordable transportation matter just that much more. These costs make it harder for people to move up the economic ladder especially for low-income families, locked out of the auto insurance market by a lack of affordable coverage options. Take the case of a family where the household head is above the age of 65 and relies on Social Security as a primary source of income: in 2018, the maximum monthly Social Security benefit was $2,788 a month, and it would take nearly 2 months' worth of Social Security benefits to cover the annual cost of insurance in Detroit. Why are rates so high? There are many factors at play, from broader economic trends to rising health care costs and differences in State regulatory practices, yet one reason why rates vary so considerably between drivers is that insurance companies use non-driving characteristics such as gender, zip code, and credit scores to set premiums for customers. It is reasonable to ask whether the use of non-driving factors in setting premiums is unfair and whether there is potential for discrimination in rate setting. Of these factors, credit scores are by far the biggest cost drivers for consumers, with rates more than doubling for those of poor versus excellent credit, thus a single mother in Detroit with a perfect driving record but bad credit could be charged one of the highest auto insurance premiums in the entire country. What can be done? While numerous policy--to reduce auto rates, are typically from the purview of States, Congress has expressed interest in curtailing rate-setting practices at the Federal level. In our report we called on Michigan to curb the use of non-driving factors in setting rates following the model of several other States that already do this. Prohibiting the use of all non-driving factors in rate setting may not be feasible as insurance companies must be able to develop actuarially sound models, however, other States have struck compromises on this. California, for example, has established reasonable rate-setting guidelines and prohibits the use of credit scores. Low-income drivers in cities pay nearly 60 percent more for auto insurance than high-income drivers. In Los Angeles, the gap is only 9 percent, likely due to regulations placed on insurers. And in Hawaii, where the use of credit scores has been banned since 1987, the commissioner of insurance testified before Congress in 2007 that despite the ban, markets remained competitive and healthy. In closing, with sensible reforms to our auto insurance policies, we can lower transportation costs and dramatically improve economic opportunity for families. Thank you, again, for inviting me to share my research findings with you. I would be happy to answer any questions. [The prepared statement of Mr. Rivera can be found on page 63 of the appendix.] Chairman Green. Thank you. Mr. James Lynch, you are now recognized for 5 minutes. STATEMENT OF JAMES LYNCH, CHIEF ACTUARY, AND SENIOR VICE PRESIDENT, RESEARCH AND EDUCATION, INSURANCE INFORMATION INSTITUTE Mr. Lynch. Thank you. I would like to take a moment to thank you, Representative Green, and the entire committee for giving me the opportunity to speak today. My name is James Lynch. I am chief actuary and senior vice president of research and education at the Insurance Information Institute in New York. Founded in 1960, we are the trusted source of unique data-driven insights to inform and empower consumers. Our members include 8 of the 10 largest personal auto insurance writers in the United States. We provide objective, fact-based information about insurance, information that is rooted in economic and actuarial soundness. I am a fellow of the Casualty Actuarial Society, the leading property casualty actuarial organization in the world, and I serve on the Society's board of directors. I have more than a quarter century of experience in property casualty insurance and reinsurance, and have held senior actuarial positions at QBE The Americas, and White Mountains Reinsurance of America. And today I would like to discuss how companies set rates for automobile insurance. Because of court cases and Federal legislation that stretch back decades, insurance companies are primarily regulated at the State level. Every insurance company must satisfy the laws and regulations of every State it operates in plus the District of Columbia, so most large insurers have 51 sets of laws to follow and 51 sets of regulators to satisfy. Every State regulates what insurers can charge for personal auto insurance. State laws ensure that rates aren't too high because no State wants its consumers overcharged, or too low because if rates are too low, an insurance company might lack the funds to pay all the claims that it has said it will pay, and are lower for drivers who are less likely to be in a crash and higher for drivers who are more likely to be in a crash. The company can only offer a discount if it can show that the customer is less likely than the average customer to suffer an insured loss; it can only surcharge if it can show the customer is more likely than the average customer to suffer a loss. Insurers can't change rates daily or weekly, the way a grocery store can change the price of a gallon of milk. They must notify the State, usually beforehand, what they intend to charge. In some States, the department of insurance must explicitly approve changes in advance. The result is that auto insurance is not priced according to the law of supply and demand; it is a cost-plus product. Insurers estimate what they will pay out in claims then add in expenses and a reasonable profit. In addition to State regulators, auto insurers operate in an extremely competitive environment and an important part of that competition is to develop sophisticated plans that properly assess each customer's likelihood of being in a crash. Insurers use teams of actuaries to figure out how to set rates. They look for characteristics that successfully predict the accident rate, the most famous perhaps is driving record; drivers who have avoided accidents for several years are less likely to be in an accident in the future but driving record is not the only factor. The strongest by most accounts is location, which tells a lot about the number of vehicles per square mile, and just like with bumper cars, the more cars there are in an area, the more likely they are to crash into each other. There are certain things that is important to know about these rating variables. First, they work, they are effective at gauging the likelihood that a customer will be in an accident. Second, they are selected after rigorous actuarial analysis; every rating variable has been proven effective through analysis of actual data. Third, they are filed in advance with State regulators along with statistical proof of their effectiveness; in some States they must be approved in advance and they can't be changed without going through the same regulatory process. Any Federal effort to oversee rating variables will overlap rigorous efforts that States already undertake. Fourth, companies constantly review how effective these factors are. If they don't work in the world, they are adjusted or abandoned. Fifth, the factors can change over time and actuaries adjust those factors as a result, for example gender, is a well-known commonly used variable and part of the reason it has been effective is that men drive more miles than women but that is changing. From 1963 to 2013, the number of miles the average man drove increased by about a third but the number of miles the average woman drove increased nearly 90 percent. The predictive power of gender as a rating variable has changed because the more miles you drive, the more likely you are to be in a crash and women are approaching men and sometimes exceeding men in that respect. Sixth, insurers are constantly looking for new variables and when they find one, the new one can change how much emphasis is placed on the old ones. And last, but certainly not least, the setting of private passenger auto insurance rates is a color-blind process. Insurers do not gather information based on race or income nor do they discriminate against anyone on the basis of race or income and they do not adjust the rates based upon any proxy for race or income. Thank you for your time. And I would be happy to respond to any questions you may have. [The prepared statement of Mr. Lynch can be found on page 57 of the appendix.] Chairman Green. Thank you. And I thank all of the witnesses at this time. Let me yield myself 5 minutes. And I would like to start with something very basic. Do you believe that testing is a means by which you can determine whether discrimination exists in auto lending? If so, would you just kindly extend a hand into the air, I would like to make a record? I take it, Mr. Rivera, you do not believe that testing is a means by which we can determine whether discrimination exists? Mr. Rivera. The question in auto lending is a little bit outside of my purview so I will defer to the experts on auto lending. Chairman Green. You will defer. Mr. Lynch. I am in the same boat. Chairman Green. You will defer. Mr. Lynch. Auto lending is a little bit outside of my world. Chairman Green. Well, persons who claim expertise--Ms. Clarke, Ms. Cross, and Mr. Van Alst--indicate that testing is a good tool. With this tool, have we been able to validate the claims that this invidious discrimination exists as it relates to people of color and women? Let me start with Mr. Van Alst. Mr. Van Alst. Testing that has happened more recently, looking at auto finance, has certainly been consistent with the findings of other analysis. I am more familiar with the use of large data sets to identify discrimination that has taken place and the report that you referred to earlier from the National Fair Housing Alliance, found similar results using testing. I will say in the auto arena, testing can be and is a good tool, but it is difficult, and in fact the National Fair Housing Alliance did have difficulty in getting straight information from dealers in order to make testing effective. Chairman Green. Have you found that it is the markup that becomes the problem, the discretionary mark-up? Mr. Van Alst. I found that everything that is discretionary at the dealer level becomes a problem. We had certainly seen through litigation that these dealer interest-rate markups which are in addition to the interest rates already set based on risk pricing led to discrimination, and I want to say that dealers are understandably defensive when you point this out, and the fact that we see evidence of discrimination doesn't mean that dealers are necessarily overtly wanting to discriminate against races. I am sure, when you talk to dealers, they don't want to do that but they have to quickly size up someone and decide how far they can push them and what we see from the data is that what winds up happening is they use race to decide how far they can push someone in terms-- Chairman Green. Let me move to, Ms. Clarke, quickly. Ms. Clarke, give me your thoughts on the dealer markups, and we are talking about in addition to the pricing that has already taken place in terms of the lending? Ms. Clarke. So, matched-pair testing in the fair housing context has been a bedrock tool for identifying when landlords discriminate on the basis of race, ethnicity, or national origin and it is absolutely a powerful tool that can be used to identify when we are seeing potential discrimination rear its head in the auto lending industry as well. The National Fair Housing Alliance study that was released in 2018 did a very good job of deploying matched-pair testers out in Virginia and identified disparate treatment on the basis of race. Those black consumers who were exploring buying a car and accessing a loan were absolutely treated differently, they were given different information about the cost of loans, different information about the cost of the vehicle, different information about the markup, and we would not have been able to capture those differences in treatment unless we deployed in a very carefully controlled setting trained-- Chairman Green. Ms. Clarke, I am going to-- Ms. Clarke. --testers. Chairman Green. --have to ask you to wrap that one up and let me ask you another question that relates to the methodology used by the CFPB with surnames and geography to perform a study. Can you give us your opinion about that, please, in terms of it being valid or not? Ms. Clarke. In my view, these kinds of studies and tests are one factor among many that we should look at in identifying discrimination. We do these kinds of analyses in the voting rights context, in the Title VII employment discrimination context, and once you have conducted an analysis that suggests you have identified discrimination, that is when you move further, you depose witnesses, you subpoena documents and get other evidence that might ultimately reveal and help you conclude whether or not you have a lender engaged in unlawful discrimination, so it is one important factor. Chairman Green. Has any court invalidated the study that the CFPB performed? Ms. Clarke. I am not aware of any court that has invalidated this study and I appreciate some of the questions and concerns that have been raised but again this is just one factor that should be further corroborated by testimony and other documents; further analysis of data that may reveal unlawful discrimination. Chairman Green. Thank you. I now recognize Ranking Member Barr for 5 minutes. Mr. Barr. Thank you, Mr. Chairman. And let me start with just making an observation, and maybe it is just me, or maybe it is just central Kentucky in my congressional district, and I certainly defer to the expertise of the panel who have studied this issue with great intensity, in terms of what you have seen, in terms of discrimination, racial discrimination in auto lending but in my experience, and again maybe this is just in my congressional district, I don't know a single racist auto dealer, I don't. I have never ever observed in my interaction with any auto dealer, a desire to not sell a car. And in arranging financing, my experience in actually listening to my constituents of all races, is that auto dealers want to arrange financing and get a car moved off their lot and in the possession of their customers, regardless of race. And what I would say also about the indirect lending model is that contrary to what we have heard today in a lot of the testimony, dealer-arranged financing helps, it doesn't hurt, consumers and customers and it does so because the dealer has the ability to have the scale and the purchasing power to actually interact with banks in a way that the consumer cannot. The dealer has the ability to bring volume to a bank and arrange financing in a way that a cherry-picked customer cannot and so what I would say is remember dealer-arranged financing is a service, it is a service provided by the dealer that the customer may choose to use or choose not to use and no one is forced to use a dealer to arrange for financing, that is something that we haven't heard in the testimony here today. A car buyer opts, chooses to use dealer-assisted financing, he or she as a result receives a significantly more convenient buying experience and can achieve a lower interest rate. Now, last Congress when we rolled back the CFPB's Indirect Auto Financing Guidance, in no way did we amend the fair credit Laws or hinder their enforcement. In fact, the National Automobile Dealers Association encourages dealers to adopt a robust DOJ-based Fair Credit Compliance Program and under this program dealers are required to be able to demonstrate through documentation any deviation in discount rates between individuals which must be based on legitimate business reasons. I just wanted to make that statement and I invite the panel if they disagree with that, to talk about that in your testimony later. But at this point let me move on to insurance and I want to first recognize my colleague from Michigan, the gentlelady from Michigan, for justifiably expressing concern about high rates in Michigan. And Mr. Rivera, to that point and that issue in preparation for this hearing, I reviewed not only your testimony but also your report and op-ed on Michigan auto insurance rates that you released in March. After reading that report, I thought you made a very compelling case for the Michigan State Legislature to reconsider some of its policies, but nowhere in your written testimony did you mention Michigan's No-Fault Auto Insurance Mandate that is paired with the only unlimited personal injury protection (PIP) requirement in the country. Your report states that this No-Fault PIP requirement currently accounts for 42 percent of the average premiums paid by the Michigan driver. On top of this, you state that Michigan does not impose a medical fee schedule resulting in the average auto accident claim in Michigan totaling over $75,000, which is more than 5 times the average of the next closest State. Why is that information excluded from your testimony and don't you think that may be why Michigan premiums are so high? Mr. Rivera. Well, thank you for the excellent question. And thank you for reading the brief. As you know, we attached the brief to the written testimony and for the purposes of the written piece, really wanted to focus on what is really one of two issues: one is, why are rates so high just generally, and that is around the medical cost State practices that we all know so well; but then there is a second question, which is, why do rates differ between people? And that is where we really thought we had to incorporate this discussion around non- driving factors for the simple fact that you could get charged more if you lost your spouse, you could get charged more if you live in a rural area, and for factors like that we wanted to have a focus for the purposes of making sure that we also answered the prompt of the title of the testimony. Mr. Barr. Mr. Lynch, in my remaining time, why do insurance companies use credit scores? Mr. Lynch. Because they are very effective. They work and they are very good at predicting how likely a person is to be in an accident in the coming 6 months or a year. Mr. Barr. Is it ever used as a single variable to determine a rate or is it part of a larger algorithm? Mr. Lynch. No. It is always part of a larger algorithm. Mr. Barr. And can you explain how consumers' prices would be effective, if insurers were not allowed to use that? Mr. Lynch. Well, then, what would happen is that for some people and by some measures most people their rates would go up and for other people their rates would come down and it would come to an average. In the State of Michigan, the average rate would be twice the average in the typical State but what would happen is that your drivers who present less risk would be in essence overcharged and people who are at greater risk of being in an accident would be undercharged. Chairman Green. The gentleman's time has expired. The Chair now yields to the Chair of the Full Committee, Chairwoman Waters. Chairwoman Waters. Thank you very much, Mr. Chairman. Since 1991, academic studies have demonstrated that when auto dealers have the discretion to mark up the interest rate on auto loans, non-white borrowers are disproportionately harmed. Nearly 2 decades ago, the National Consumer Law Center brought several successful class actions that revealed racial disparities in dealer markups of interest rates. I would like to ask, I think it is Mr. Van Alst, to describe this academic research and how prevalent such discriminatory markups happen; are you familiar with the research that was done? Mr. Van Alst. Yes. We were co-counsel in 12 cases that were litigated between 1998 and 2007. We analyzed millions of transactions. We coded these transactions for race based upon driver's license data. At the time we were involved in these cases, 14 States collected race information for driver's licenses and we were able to therefore code these transactions by race. This is something that can't be done today. I know we have had lots of discussion about data and lots of folks have indicated an interest to have good and accurate data, working on these problems but currently the Equal Credit Opportunity Act prohibits the collection of race data in these transactions and so we can't replicate today the same analysis that was done at the time this litigation took place. Chairwoman Waters. What year was this done? Mr. Van Alst. This was between 1998 and 2007 and so that is not only us but enforcement entities and States as well are both prohibited from collecting this data and prohibited from using that data because they don't have it, towards enforcement and this is a great hurdle whether we talk about State enforcement in this area or Federal enforcement, without the data, without knowing what is happening, we are forced to rely on--as has been pointed out--measures that are perhaps slightly less accurate than if we had good data about the race of people involved in these transactions. Chairwoman Waters. Do you have any studies or information, not polling data but stories that have been told by individuals about what happened to them when they went to purchase a car? Mr. Van Alst. I have reams of stories about abuses that people suffer when they go to purchase a car both from the time when I worked at Legal Services and since then when I have dealt with attorneys across the country, however very few consumers know that they have been discriminated against in terms of financing. This is not something that you can see, you can't tell that you were marked up 3 percent, your financing was, and the next person wasn't marked up at all, consumers have no way of knowing that this is happening to them, so I hear stories about all sorts of other things but consumers can never know that this is being done to them. Chairwoman Waters. But there are many stories in the African-American community about how they were treated when they went to buy a car and now, some African-American consumers will ask someone to go with them because of the way they have been basically abused, they have been bullied, they have been lied to, and so these stories are legendary about what happens to African Americans and women when they are trying to purchase a car. And I know that the members of this committee, every member of this committee has heard these stories about what happens when women and people of color and African Americans go to purchase a car. So I wanted to find out a little bit about that study and perhaps what we need to do to keep proving, to keep giving the information because it is not that this is not happening, it is because the auto dealers have a lot of influence in this Congress, and Members who believe it is their responsibility to protect the automobile dealers and not the consumers and so this is a very difficult issue. You saw what happened before when it came before this committee and you saw how they rolled out the lobbyists, how they put the money into the campaigns, and you saw the results coming out and so this is difficult work and we have to keep doing it, we have to keep finding ways to get more information and not just rely even on the stories that we hear oftentimes but the fact that this has to change. I yield back the balance of my time. Chairman Green. The gentlelady yields back. The Chair now recognizes the gentleman from Florida, Mr. Posey, for 5 minutes. Mr. Posey. Thank you, Mr. Chairman, and Ranking Member Barr, for holding this hearing today. There is no place for discrimination against any American citizens. Allowing discrimination to persist would dishonor those who fought so bravely to defend our nation and currently are online to do so, we all abhor discrimination clearly. For today's hearing I think we ought to follow the principle of evidence-based decision-making. Evidence-based decision-making encourages us to look beyond first impressions, not for the purpose of dismissing concerns but to reach a deeper understanding on why something is the way that it is and that is what we must do to solve problems like we have before us today. That is the spirit I believe we ought to have in approaching today's hearing; we must collaborate to see the full story with respect to auto insurance and markups of auto insurance interest rates. Often, there is more going on than first impressions might tell us. Perhaps a solution is something different than calling the usual cast of regulators and sending them forth. We would all find it reasonable that a business seeking profits probably shouldn't care about ethnicity, age, or gender of their customers. Clearly, a business shouldn't make any more money by merely charging more for certain groups and market economy with competition we have in this country, ought to result in a person taking his business elsewhere and for businesses competing for this business to offer him a better deal than someone who discriminates. Now, that is how we expect the markets to work and our competition usually drives down prices for everyone. For the panelists, if the differences we see in the auto insurance rates in the interest markups are solely due to ethnic, age, or gender bias, why doesn't the marketing competition drive these practices out of existence? And we usually start from left to right, so let us start from right and go back left today. Mr. Lynch. That is your right? Mr. Posey. Yes. Mr. Lynch. Okay. So insurance companies--a lot of the discussion today has involved the use of the word ``discretion,'' and once insurance companies have made their determinations based upon the risks that they see and the data that they observe year in, and year out, once those rates are set for automobile insurance, the world of discretion ends; there is no discretion left, the rate is what the rate is and it is based upon the characteristics that any customer presents and those and all of those have been linked to higher or lower propensity to be in an accident and that is simply how insurance is priced and how insurance is done and once that is in place, in auto insurance there is no more discretion. Mr. Posey. Okay. Thank you. Mr. Rivera? Mr. Rivera. Hello, thank you for the question. Some of the factors used in auto insurance have been found to not be correlated with risk. There are certainly non-driving factors you want to consider; they need to be included for insurance companies to have sound rates. A good example of this is gender, the way in which it is treated has been found in research to vary from company to company in which sometimes you get a discount, and other times you get charged a little more. It also varies by State and it is this inconsistent application that could be potentially unfair for the consumer going from one State or another to have their gender matter differently, as one potential way in which risk might not be fair. Another is the use of things like credit scores. I agree with my colleague that credit scores are not correlated with race but we have to ask the question, why people of color are disproportionately represented in the ranks of Americans with poor or no credit and it is not because they lack financial skills, it is because of systemic, sometimes implicit discrimination that happens elsewhere in the market that ends up showing at things like credit scores. So I would say well, I think there are compromise factors looking State to State they have looked at it, there are certainly some factors that deserve a little more scrutiny for thinking about what their basis really is, and I think for insurance companies themselves, that transparency is not always available to consumers. It is often very hard to get information about what factors they are using and one of the biggest factors that can weaken a good free market is imperfect information. Mr. Posey. Thank you. Ms. Clarke? Ms. Clarke. The foreclosure crisis and resulting recession resulted from us relying on the market frankly and from insufficient oversight and regulation on banks that were peddling sub-prime and predatory mortgage products to consumers, many of them people of color. Likewise, in the auto industry, we don't want to return to that era. We need strong Federal oversight and regulation to ensure that consumers are being treated fairly and equitably and most importantly to ensure that practices like markups are not rearing their ugly head. Mr. Posey. Thank you. Ms. Cross? Ms. Cross. One reason perhaps that the market hasn't corrected this so to speak is because a lot of times borrowers with poor credit histories or in areas of other disadvantaged situations know that they have no other option and that puts them in a very vulnerable state when they go to buy a car and to get financing, to be kind of at the will of creditors who know that they are able to take advantage of that situation. Mr. Posey. Okay. Mr. Van Alst. Another reason that competition hasn't cured these problems is that this is not a transparent and clear marketplace; you have to actually spend hours at the dealership to even find out what terms you will eventually be offered, typically you have to have already gone through the sales process back at the F&I Office before you can even find out what sort of finance terms are going to be offered, so it is much more difficult to do comparison shopping unless you are prepared to spend days and days to try to go from one dealer to another. Mr. Posey. Thank you, Mr. Chairman. Chairman Green. The Chair was quite liberal with the time; the gentleman yields back. The gentlewoman from Michigan, Ms. Tlaib, is recognized for 5 minutes. Ms. Tlaib. Thank you, Mr. Chairman. Thank you so much, Mr. Rivera, for being here. Would it surprise you to know that a driver with a DUI and a good credit score will pay less for auto insurance than a driver with no DUI and a good driving history who has a decent credit score, not so great? Mr. Rivera. When I found out I was shocked, yes. Ms. Tlaib. Tell us a little bit more about that? I want you to tell people here, the non-driving factor, the use of a credit score, a person who committed DUIs who has a really bad driving history compared to somebody who doesn't have a great credit score but for the fact that--and let me tell you I have seen people's credit scores impacted by the fact they have become a widow, that they are retired, I know a woman who worked at Beaumont Hospital for 25 years, had been driving for 55 years, and when she retired, her car insurance went up $350, and when she called, they said it was because her credit score was impacted, however she is driving less, she is not getting any tickets, so how is this legal? How is this possible? Mr. Rivera. Yes. So some of the research has said that one of the reasons why credit scores are a factor is it predicts the likelihood of a claim but we really need to unpack that. Why do folks with low credit scores need to make a claim? And the answer to that is, think about the situation that gets you to there in the first place; for many Americans, their credit score is influenced by factors outside of their control, a family emergency, loss of a spouse, getting behind because of the high cost of college or education, and for those-- Ms. Tlaib. But Mr. Rivera, they are saying if somebody has a low credit score, the likelihood of them committing a fraudulent claim is higher, so they are saying because they are poor, because they have a low credit score, the likelihood of them committing a crime is higher, isn't that discrimination? Mr. Rivera. Yes, I believe so. It disproportionately impacts low-income, working-class families. Ms. Tlaib. That is right. So Mr. Lynch, there are two drivers, true fact, who can talk to each other from across the street in their driveways, they are the same age, they have the same driving history, the same car, the same credit score, what I can't figure out is why a driver who lives in Detroit on the side of Mack Avenue gets quoted double the price for basic minimum coverage than a driver who lives right across the street on Mack Avenue on the Grosse Pointe Side? Mr. Lynch, can you explain why it is determined by the auto insurance industry that that side of the street pays $3,000 more per year than the driver in the Grosse Pointe area? Mr. Lynch. That is because of the way that the territories have been drawn and those-- Ms. Tlaib. But you are basing it on zip codes, right, not driving history and driving record, say it, you are using-- Mr. Lynch. It is being-- Ms. Tlaib. --zip codes. Mr. Lynch. --based upon the territory which is the zip, the-- Ms. Tlaib. Zip codes. Mr. Lynch. --codes are usually used to determine what the territories are, those territory factors, the boundaries of them, are being constantly reevaluated by insurers and in fact one of the things that insurance companies are trying to get a better handle on, is something called, I think it is called geospatial coding which is the use of the exact longitude and latitude to set rates so that some of those-- Ms. Tlaib. Mr. Lynch-- Mr. Lynch. --discrepancies-- Ms. Tlaib. We call that redlining. It is a discriminatory practice to base it not solely on people's driving history. And the non-driving factors are hurting families. And I really do have a question for Mr. Rivera because I have limited--Mr. Rivera, what research did you find in the difference between the rates between zip codes? Because I know according to the Consumer Federation of America, a zip code in Detroit with an 8 percent white population adjacent to zip codes with 85 percent white population pays 65 percent more in auto insurance, is that correct? Mr. Rivera. That is correct. Ms. Tlaib. For instance, even though we are talking about this being a white-black issue, and we have talked about this, I don't see this as a white-black issue anymore, I am talking to people and it is true, African Americans in my district pay a lot more but what I am showing now is that 97 percent of Michigan's zip codes pay unaffordable car insurance rates. This has nothing to do with no-fault. These are non-driving factors. They are using gender and Mr. Rivera, when you talked about gender, I was a little taken aback, you are saying that someone, a woman and a man, you are using that as a factor in calculating rates, not where a person works, and you asked us where we work and you calculate the length of the driving between your home and your workplace, so why isn't that used? Why are gender and people's marital status and their credit score being used to calculate rates? Mr. Rivera. Well, from my perspective, they shouldn't be. They are being used because insurance companies think that those folks will file more claims and they want to charge them more and the unfortunate effect of this is that the people who need the cheapest auto insurance the most, are the ones who get charged more and those in relatively wealthy areas, relatively wider areas who have relative privileges, get charged less, despite the fact that they could afford it. Ms. Tlaib. And lastly, Mr. Chairman, if I may, I did pass out and ranked so you all can see this is not just a Michigan issue, but car insurance rates are high risk especially with a lot of our residents. The use of non-driving factors is really putting more people into poverty and it is something that our families are facing every single day. And so, I did pass out the ranking from Florida to Kentucky to New York and Ohio. So I hope my colleagues consider this as they look at the bill that I have right now, that prohibits the use of credit scoring in calculating car insurance rates. Thank you, Mr. Chairman. Chairman Green. Thank you. The gentlelady's time has expired. The Chair now recognizes the gentleman from Georgia, Mr. Loudermilk, for 5 minutes. Mr. Loudermilk. Thank you, Mr. Chairman. I thank everyone on the panel for being here today. This is an issue we need to address but it is an interesting issue, in that most if not all of our States by law require drivers to have some form of auto insurance, and as someone who has recently been involved in a very bad accident, was hit by someone who had no insurance, it is obvious to me why we do require people to have insurance. On the other hand, insurance is not a one-size-fits-all thing, and what I found through my experience in life is when you have a one-size-fits-all, it is only one very narrow segment of a consumer base that it fits; everyone else is then harmed in some way or the other because it doesn't fit where they are and insurance is one of those areas that I see that with one-size- fits-all, is generally going to raise the price for everyone and when that price goes up, those who come from low-income families, as I grew up, are harmed the most, and so that is my concern of where we are going in all of this and my serious concern over the proposed bill. Mr. Lynch, I understand we do use credit ratings which, from the information that I have derived, is a fairly accurate way of--not the sole way but a way of helping to determine a risk factor because insurance is all about risk. I remember growing up that if you were a male, you were going to pay more in car insurance up until you were 25-years-old; I don't know if that is still the case. I remember when I turned 25, that was a magic age of, now I get cheaper insurance, and it happened, even though my wife, she was already you know, she was a less risk her car, she paid a little less on it because I guess statistics showed that men had a little heavier foot than women did. The point I have is--my understanding of the rates in insurance are based on multiple risk factors because you do not have a crystal ball, you cannot sit down and predict if someone is going to have an accident at some time and is it going to be their fault so you have to have tools and I know it has been touched on already, but how does the credit rating apply in helping to make those determinations of risk as you also use other tools, how do they work together? Mr. Lynch. It is a type of analysis that actually has become a lot more sophisticated in recent years. One of the things that actuaries have gotten really good at is understanding how those individual rating factors interact and overlap or miss each other and so they adjust rates for that, so for example when you see the kind of a one-way analysis that you have seen with gender rating in particular, where you take six factors and hold them constant and then only change gender, when you go from community to community, that does give you a window but you can't project forward and extrapolate that experience to all women who are drivers. Mr. Loudermilk. Right. Mr. Lynch. You can't do that because you haven't done anything to take into account that interaction. I can also state that all of these variables, all of them are constantly being reassessed by insurance companies. Gender was one that specifically insurance companies are currently looking very closely at because we are starting to understand the nature of gender identification in a very different way that we did even 10 years ago. And insurance companies have their eyes open, and they see those things as they are happening. And so they are working hard to make sure that their rating plans take that into account. And then as women have been driving more in recent years, that relationship between men and women, which at one time was thought to be exclusively because men have a heavier foot than women, it actually is being shown to be at least in some of the cases it appears to be just the number of miles being driven. Mr. Loudermilk. Okay. Mr. Lynch. The number of miles you are on the road tells you how likely you are to be in an accident, which is kind of a-- Mr. Loudermilk. Don't most States who have the predominant regulatory impact on insurance require that you only use risk- based factors in determining-- Mr. Lynch. Yes. Mr. Loudermilk. Okay. So quickly, Mr. Rivera, you revealed that there may be some discrimination in the actual credit reporting which is not directly effective on the insurance agency, shouldn't we focus on that, not eliminating the insurance companies from using something that is a tool, address the discrimination on the credit report side not eliminate their ability to use a tool? Mr. Rivera. I think it has to be examined because just because something is a risk factor, it doesn't mean that the reason for why it creates a risk factor is justifiable. In the case of credit scores, it might be correlated with risk, but you want to unpack it and I think as you know, I have kind of talked to today, some of the reasons for why it can become a risk factor are actually reasons that we may not want to penalize when it comes to pricing, there are certain consumers. Mr. Loudermilk. Thank you. I yield back. Chairman Green. The time has expired. The Chair now recognizes the gentlewoman from Ohio, Mrs. Beatty, who is also the Chair of our Subcommittee on Diversity and Inclusion. She is recognized for 5 minutes. Mrs. Beatty. Thank you very much, Mr. Chairman. And thank you to all of our witnesses here today. I am not sure where I should start with this, after reading everyone's testimony. Certainly, I associate myself with much of everything that Congresswoman Tlaib has said. You probably can tell by our tone that it is a subject that we want to weigh in on very much because when you talk about discrimination in any field, when you talk about disparities that we should be able to change, when you cite information in testimony and reports that oftentimes don't match up or you highlight things that don't make sense, it puts me on pause of, where do I start? So Mr. Rivera, you are up first. In your report you cited the findings by the Federal Insurance Office that nearly 19 million people live in zip codes where auto insurance is unaffordable, 19 million people out of 327 million people, meaning about 6 percent of the American people live in zip codes with unaffordable insurance. Now, you also stated, if we compare that with Michigan where you state 97 percent of the people live in zip codes with unaffordable insurance, now, you know, I am from Ohio, so I am not picking on you because it is Michigan, I am just giving you back what you gave to us, this to me seems like a Michigan problem. Why should we use Michigan, with its deeply flawed system, as evidence that the Federal Government needs to usurp State laws and change how insurance rates are calculated, explain that to me? Mr. Rivera. Of course. There are two questions there that I will think through. The first is that the question of jurisdiction is a question that I will leave open to Congress and it is not the subject of my expertise. The second is, I can provide recommendations for what could be done at any level of government. And I think Michigan is the canary in the coal mine; I don't think it is alone. You see rising insurance rates throughout the country and you see working-class drivers struggling to pay rates. And a part of that factor is the use of non-driving factors. So as Congress considers the policy levers in front of it, we have attempted to provide the best evidence for the case for or against, looking at those factors. Mrs. Beatty. Okay. Mr. Van Alst, let me just say, any time we are using zip codes, that poses a problem to me. People should not be judged, in my opinion, by their zip codes, their addresses, and where they live, but there was a bill that came up in our 114th Congress where we dealt with this, and it was a system that was used, the Bayesian Improved Surname Geocoding (BISG) methodology, and it used race and zip codes, so let me ask you, between 2003 and 2007, the National Consumer Law Center settled for more than $100 million on related class- action suits so I am going to ask you, did you use the BISG Methodology in that suit and if not, what did you use? And part of the reason, it was interesting as I was reviewing some of the testimonies here, I plugged in my surname and everything they asked me for, and it came up that there was an 80.24 percent chance that I was white. Now, clearly, saying there was an 80 percent chance that I was white by that zip code, the system is flawed when you do that, and conversely for a whole lot of people it would come up and actually maybe they are poor and maybe that they do live in an area and it will work against them and so I am curious, what system do you use? Because I am going to plug my name and address in and see what happens. Mr. Van Alst. In our litigation, what we actually did was pull race data that folks reported themselves to their State DMV or RMV when they got their driver's license. At the time we were doing our litigation, there were 14 States in the U.S. that collected race data, and because people move around, we were able to use that data to get statistically significant results in a number of States beyond those 14. That is no longer the case, those States no longer collect race data. There are issues with using surname and geocoding as a proxy for race. I think the CFPB and others certainly agree that there are issues with that unfortunately because the Equal Credit Opportunity-- Mrs. Beatty. I am going to have to stop because my time is running out, and I just want to quickly say, it works two ways because it could not only hurt African Americans but also benefit folks who might not be a minority, who live in a minority neighborhood. Thank you, Mr. Chairman. And I know my time is up. Chairman Green. The gentlelady's time has expired. The gentleman from Ohio, Mr. Davidson, is recognized for 5 minutes. Mr. Davidson. Thank you, Mr. Chairman. And I thank our witnesses. And following my colleague from Ohio, I have a similar observation about the difference between Ohio and Michigan, and we are not talking football but with insurance. If you look at the consequences. Ohio is less than half the cost for comparable coverage, and we are here talking about laws and public policy because laws and public policy have consequences and you can see clear differences in the policy requirements to write a policy in Michigan versus to write a policy in Ohio and that is embedded in the price. I have appreciated the education in preparation for this hearing and in the dialogue here about other factors that are going into the pricing and I guess some differences of opinion about how those factors are affecting the price. But Mr. Rivera, at the core, when you write a policy in Detroit, and you write a policy in Cleveland, Ohio, they are demographically very similar, so isn't the core price difference between the two States the content of the policy? Mr. Rivera. Well, certainly State-by-State trends impact the cost of auto insurance but for drivers in both Toledo and Detroit, their credit score or their gender can also impact their price, which is something that they share in common. Mr. Davidson. Okay. So there are factors that can influence it. But Mr. Lynch, as you look at these generic things like sex, gender, zip code, things like that that are not personalized, when you look at the rate differences on insurance, there are some people who are advocating for, everybody gets a policy, you get a policy and you get a policy and you get a policy, and it doesn't really matter what your driving records are like and we just want to average-price it, so everybody shares the risk which produces more risk in the pool and no one could be denied the ability to buy insurance or even have to pay a higher price. Why would we do that? Why wouldn't we say, you are a safe driver so you will pay less than somebody who is not a safe driver? If someone has 5 speeding tickets and a DUI, isn't it rational that this person would pay more then somebody who has never had a moving violation? Mr. Lynch. Well, driving record is used to evaluate whether a person is a good risk. It is one variable, but it is not the best predictor of risk, and that is a simple fact. I did a simple one-off analysis about the percentage of cars that were in an accident and the percentage of people who get a moving violation in a year. And I figured out that about 75 percent of Americans probably got a discount for good driving because they haven't had either of those happen in the last 3 years. So that would be 3 out of 4 people in this room getting that discount and there is a lot more to whether you are going to be likely to be in an accident than that. I have talked to some people, and that is actually kind of reflective of what the industry really is. The strongest predictor, the absolute strongest predictor is the number of cars in a territory, because the more cars there are per square mile, the more likely they are to bump into each other. And I live in New Jersey where we have about the highest rates in the nation and we have the highest vehicle density in the nation and that is no coincidence. Mr. Davidson. Right. So when you do that, even though you have all these other factors for the individual, you inherently have to come back to something like zip code, something like who drives through this area here and what kind of vehicle, the price for insurance for certain types of vehicles, that influences what kind of car you drive, that also affects your rate, correct? Mr. Lynch. Depending on the package of coverages of auto insurance, that can make a significant difference. Obviously, it costs more to have first-party coverage on, say, a $50,000 car than it does on a car that you got secondhand for $1,000 or $2,000 or $4,000. Mr. Davidson. Correct. And then my understanding is when you look at data analytics and of the panelists it seems that your organization would be in the best position to offer perspective on the data that is collected. Data is increasingly monetized. And I assume many of your companies are purchasing and acquiring data that allows for highly personalized pricing based off of a whole variety of data analytics. Can you comment on that please? Mr. Lynch. Well, yes. Data is critical and data is very expensive. It is very expensive to prepare the dataset, it is very expensive to preserve it. We did some work in the State of Florida and just cleaned up a small database of only about 300,000 records which is quite small and only about 10 variables and it cost us $15,000 for about 3-months' worth of work at the Triple-I. It is very expensive to maintain a dataset, and companies don't do that unless they can get actionable and important information from it. And that is why we developed the sets of rating factors that we developed; if it wasn't necessary to do that, in all likelihood it wouldn't happen. Mr. Davidson. Thank you. My time has expired, and I yield back. Chairman Green. The gentleman's time has expired. We will now recognize the gentlewoman from Texas, Ms. Garcia, for 5 minutes. Ms. Garcia of Texas. Thank you, Mr. Chairman. And thank you for putting this hearing together. I think it is a very important topic and I know that as a former Legal Aid lawyer in the consumer section, I certainly understand a lot of the issues that face particularly a lot of working-class people in my district, and it is sad to note that these issues were there many years ago and they are still here today. So thank you to all the panel for being here because this is something that directly impacts so many of us in and around Texas. But Mr. Chairman, I know that my colleague from Michigan has followed this issue because she is after all from the place where the cars are built, so I yield the rest of my time to my colleague from Michigan, Ms. Tlaib. Ms. Tlaib. Thank you so much. So Mr. Lynch, why does marital status have anything to do with somebody's driving history or driving record? Mr. Lynch. It is one of many factors that insurance companies have looked at, in that case, over the course of decades and determined that in certain conditions and in combination with other factors, it can make a difference as to how likely a person is to be in an accident. Ms. Tlaib. So Mr. Rivera, one of the things that I talked about is obviously the use of credit scores and the fact that I think a lot of my colleagues on both sides are agreeing that there is some flawed system to credit scoring, and I appreciate them saying that. I will tell you a story about somebody in my district who, at a very young age, went and served our country, and came back, and of course didn't build up their credit, through no fault of their own, and is trying to build up their credit as they come back after serving, after driving Humvees in Iraq. And they come here and they try to get car insurance and their credit score is being used as a factor--not where they live, the distance between their job and their workplace or any other driving history because they have been driving since they were 16. Again, I really want to unpack this idea around the use of non-driving factors because I do think even if we fix whatever other issues are going on, there are layers of issues when it comes to insurance coverage on home and auto but this use of non-driving factors, to me, is a really dangerous loophole in that we on the Federal level have a responsibility to push back against especially corporations using these factors because they are a factor to get around obviously the accountability of making sure it is fair and just for everyone. And as a woman, as soon as I am divorced, my credit score drops, even though I am still making the same amount of money-- it just drops. The use of credit scores to me is extremely dangerous. And I want Mr. Rivera to really dive in and talk about this unpacking, around the use of credit scoring. Mr. Rivera. Yes. Well, in many instances it can be egregious, particularly when you think about, a credit score is not just a reflection of your financial acumen, it is a question of what opportunity you have in the market. And in many hearings that I have been to before this very committee, there is rampant evidence of discrimination and a whole host of markets that lead one to have a bad credit score. I will give you one example of something that is just down the block from me, there is something called L.A. Insurance. They used to sell 7-day insurance, it was a way of charging you an extremely expensive package just so you can meet your obligations to this State and then when the State took action against 7-day insurance, they started selling 6-month insurance. And if you called them, they said hey-- Ms. Tlaib. Thirteen days, too. Mr. Rivera. Thirteen days. And if you called them and you got on the phone with them, like I did, you would ask, hey, do I really need to have this for 6 months? And they kind of say, no, cancel it after this, once you send us the check. So in many ways, minority communities, low-income communities, just get up-charged every step of the way before they even get to the auto insurance policy and so it is only salt in the wound when the credit score is used again to determine their rate. Ms. Tlaib. So Mr. Lynch by and large, the insurance industry has never admitted that communities of color actually pay more for their car insurance, in fact they often become downright defensive about it. According to the balance in 2014, the National Association of Mutual Insurance Companies sent a letter--and I have it here and I would like to submit it for the record. The Federal Insurance Office insinuated that people of color can afford to pay more for car insurance because--wait for it--they spend money on their pets, toys, alcohol, tobacco, and recording equipment. Is that true? Mr. Lynch. What are you asking? Ms. Tlaib. Do you know about this letter? Mr. Lynch. I had nothing to do with composing that letter. But that letter does exist. And the minute I saw it, I thought, what could you possibly be thinking, that is not the-- Ms. Tlaib. But doesn't that-- Mr. Lynch. --way-- Ms. Tlaib. --already ensure-- Mr. Lynch. --the world operates. Ms. Tlaib. And I am sorry, Mr. Chairman, doesn't that already ensure they are using discriminatory--they are thinking of my people of color, in America, in this way, that they would, to use that as a basis. Mr. Lynch. I get where you are coming from, I really do because when I saw the hole it put in your stomach, it put it in mine as well. But I can tell you that the insurance industry doesn't operate that way and that was just an unfortunate and gross exception to the way that the industry normally operates. And you know what? We had nothing to do with that statement but I personally and sincerely apologize that any person here was subjected to that kind of thinking. Ms. Tlaib. Well, Mr. Lynch, I-- Chairman Green. The gentlewoman's-- Ms. Tlaib. --appreciate that you acknowledge that. Chairman Green. --time has expired. But please-- Ms. Tlaib. I do ask, Mr. Chairman, unanimous consent to insert for the record the letter-- Chairman Green. It will be admitted, as well as your other document styled, ``Auto Insurance Rate Oversight and Reform Subcommittee Members State by State.'' Do you desire to have that in the record? Ms. Tlaib. Thank you, yes. Chairman Green. Without objection, it is so ordered. The Chair will now recognize the gentleman from Tennessee, Mr. Rose, for 5 minutes. Mr. Rose. Thank you, Chairman Green, for calling this hearing. And I appreciate the opportunity to hear from the witnesses today. I understand my colleagues' concerns about the high prices in Michigan and concern that those prices may be impacted by certain demographics, by non-driving factors like income level, but Cleveland and Detroit are fairly similar in terms of demographics and income level yet Cleveland has dramatically lower insurance premium rates. According to Mr. Rivera's report, in Cleveland the average annual insurance premium is $1,277, while in Detroit the average premium is $5,414. Mr. Lynch, can the conditions that you have described in Michigan such as the no-fault insurance law partly explain the difference in insurance rates between Cleveland and Detroit? Mr. Lynch. That is an enormous part of the difference. No- fault automobile insurance--when actuaries look at all the different States, they look at the State of Michigan completely separate from the rest of the country because the laws there are drawn differently for auto insurance than they are anywhere else, and that is just a statement of a projective fact, it is not a value judgment. It is the most generous system in terms of claim payments. For a no-fault auto insurance claim, the limit on payments is infinity, it can be $5 million, it can be $10 million, it can be $20 million. And there are claims that come through of that size. In other States where no-fault exists, and it doesn't exist in most States, but where no-fault does exist, typically the limit is somewhere between $10,000 and $15,000, so when that $10 million claim comes through, the auto insurer pays the $10,000 or whatever their limit is and then it moves over into the health insurance system. But in Michigan, especially if you are dealing with Medicare or Medicaid, those types of insurance have any other insurance pay first, so if a 65-year-old was involved in a terrible accident in Michigan, they are going to get coverage first from the no-fault policy and it is going to pay and pay and pay forever and then they will never touch the Medicare. What that means is that insurance is a cost-plus product. Insurers estimate what the cost of claims is going to be and then they add in provisions to cover agents' commissions and taxes and licenses and fees and a reasonable provision for profit and when you do that, when you have an infinite limit, you know you are going to have a higher rate than you have in a State like Ohio where I don't believe it is a no-fault State. And I believe on liability coverages, the minimum limit is around $35,000 to $60,000. So the difference between a $60,000 claim and a $10 million claim is going to have a big influence on how much money every person in the State pays regardless of how that rate is arrived at. Mr. Rose. Thank you. Before coming to Congress, I was a private businessman, and also served briefly as Tennessee's Commissioner of Agriculture. I tend to think that government and regulation work best the closer they are to the parties being governed and regulated. The insurance industry is one of the few success stories in that it hasn't been swallowed whole by Washington. We have been regulating insurance at the State level for over 150 years, and it is a system that works for consumers in terms of accessibility and affordability. Members from both parties seem to agree on this. Even when crafting the Dodd- Frank Act and creating the CFPB, Democrats on this very committee stipulated that the CFPB was not an insurance regulator, they didn't want to supersede State-based insurance laws and commissioners. Like them, I am opposed to any potential overreach by Congress that diminishes a State's right to regulate insurance. I believe H.R. 1756 represents such an encroachment by Washington. Of all 50 States, only 3 have decided to prohibit the use of credit-scoring data in auto insurance underwriting: California; Massachusetts; and Hawaii. Now, I might disagree with their decisions because I think that data helps better price the product for consumers but that is the decision those State regulators have made and legislatures and government legislators and governors have made and I respect it. Tennessee and Michigan or any of the other 45 States could make the same decision, but it is their decision not ours, and I hope it remains that way. Thank you. And I yield back. Chairman Green. The gentleman's time has expired. The Chair now recognizes the gentleman from New York, Mr. Zeldin, for 5 minutes. Mr. Zeldin. Thank you, Mr. Chairman. I represent a district on Long Island in the State of New York. It is very difficult there, as it is in many other parts of the country, to get around without a car, whether it be to try to get to school or to try to get to work. Access to affordable financing and insurance is what helps them have that opportunity to have a car that can get them to work or to school. This isn't a wonky policy debate that we are having here; it really impacts each and every one of their lives very personally. In real life, absent a government takeover of the lending market, you always have divergent auto lending rates for different borrowers based on creditworthiness, the term of the loan, the price of the car, and the difference between State laws and regulations. Now more than ever, consumers can shop around, compare rates, and get the best deal. It is so critical and it has been said over the course of the hearing today that any discrimination related to race, gender, or ethnicity is entirely unacceptable. Extorting honest lenders and using spurious statistics to justify those attacks that some of what we are getting into here during this this hearing could hurt the very people we are supposed to be helping so it is very important to have the best possible data when reaching our conclusions that we are talking about. Using questionable disparate-impacted data means no actual discrimination has to be proven or occur for what could be a politically motivated lawsuit, it could be a bureaucratic attack to shut down an honest lender or an auto dealer. Meanwhile, rates go up and access to credit goes down and consumers, especially low-income ones, will suffer. When it comes to flawed mandates on auto insurance, we have also had State policies that I briefly mentioned earlier, that have led to rates skyrocketing, so any attempt to do that at the Federal level could be very problematic. My question first, for Mr. Lynch is, we have been talking about the State of Michigan a lot today and that is good. The State of Michigan mandates unlimited personal injury protection (PIP) and no-fault auto insurance. There are people at home who don't live in the State of Michigan, and they are listening to today's hearing. So, what does that mean? Can you explain it to them? And do any other States in the country do that? Mr. Lynch. Yes. They are about--and unfortunately, I didn't prep myself on that exact question but there are about a dozen States that have some form of no-fault automobile insurance. The idea of no-fault is that it kind of usurps the idea of liability and claims. A typical claim--we always think, who is at fault in the accident, and whoeveris at fault pays in the accident, but in States like Michigan, the alternative for medical injuries is no-fault insurance so it doesn't matter if you are in an accident and you are injured in the accident, it doesn't matter if you are at fault or not at fault, the insurance responds and pays so that that is the difference between no-fault and a typical tort system, which I believe is the case in Ohio, which is the other State that seems to have come up in a number of comparisons. Mr. Zeldin. The average cost of an automobile accident insurance claim in Detroit in 2017 was $51,000. Is that normal? Mr. Lynch. I would guess it is normal for Detroit but it is not normal for the rest of the country. Mr. Zeldin. Can you explain, I guess tied to the first question, how that mandate impacts the rates? Mr. Lynch. Well, what happens is no-fault is not the cause of that high rate. The actual cause is the fact that the no- fault is unlimited, so when you say a $50,000 claim, that is obviously an agglomeration of some small claims and some medium-sized claims and some large claims, and in Michigan, because of the no-fault law, the large claims can get really, really large, in fact there is an entire insurance entity that is ultimately responsible for claims that exceed I believe $550,000, no-fault claims and that is above what virtually any personal automobile insurance policy in the United States is, except for in Michigan. Mr. Zeldin. Mr. Rivera, your report got into--and I guess this question is for you and maybe Mr. Lynch but we only have a few seconds--a whole lot of different reasons why the Michigan rate in the State is so high. Can you speak to what that would do for the rest of the country if the Michigan model was used elsewhere around the country as far as what, in your report causes the high rates? Mr. Rivera. I would hate to extrapolate given that there are vast differences State by State but I do believe that the research cited by Mr. Lynch on this question of personal injury protection claims is worth considering as was in the brief I think, they play a part in averages, they can explain differences between people, which I think is the focus. Chairman Green. The gentleman's time has expired. The Chair recognizes the gentleman from Wisconsin, Mr. Steil, for 5 minutes. Mr. Steil. Thank you very much. And Mr. Chairman, thank you for holding today's hearing on what is a really important topic. There is no place for discrimination in our system across the board and I think we are identifying that here today. I also want to talk about the high rates of insurance and where, as policy folks, we can dive in to try to drive the cost of insurance down for everyone so everyone could benefit. I think the State-based insurance regulations allow us to compare how different policies work and apply those lessons to other States. We have come to observe these policy decisions yield sometimes bad outcomes for customers. I am from the State of Wisconsin, which actually has the second lowest insurance rate in the nation, averaging $951. We share a border with Michigan so snow in Wisconsin, snow in Michigan, sunny day in Wisconsin, sunny day in Michigan. And you see probably about the most divergent of pricing, you see one of the highest at $2,600 a month--well over twice the cost for a car to be insured in Michigan, directly opposite the line of a car to be insured in Wisconsin. And so, Mr. Lynch, could you comment on what Wisconsin is doing right in keeping these costs down? Mr. Lynch. Well, I wouldn't characterize personally, and in my position with a non-partisan organization, I wouldn't characterize it as right or wrong but I would characterize it. I can speak to the differences-- Mr. Steil. What is causing that? Mr. Lynch. As I said, in Michigan you have no-fault automobile insurance with no limit to the policy. And if you have a $10 million claim, that is a dollar on every person in the State of Michigan, give or take. A comparable claim in Wisconsin would be, say, $60,000, which is--how many people are there in Wisconsin? Sixty thousand divided by however many people there are that is what, like a dime of policy and those things add up over time so I think that is probably the biggest driver. I can tell you that to my knowledge, credit scoring is regulated in 40 States, it is banned in 2 or 3 but credit scoring is not the reason that rates are higher in Michigan or lower in Wisconsin, it is-- Mr. Steil. It is interesting-- Mr. Lynch. And I want to say one other thing-- Mr. Steil. I am going to reclaim my time. I appreciate-- Mr. Lynch. Sure. Mr. Steil. But we are so limited. I appreciate that take. Mr. Chairman, I would like to insert in the record a letter from the American Property Casualty Insurance Association. Chairman Green. Without objection, it is so ordered. Mr. Steil. And what it does is, it identifies and highlights some of these policy areas that we look at to ultimately drive down the cost of insurance and make it affordable for folks living both in Wisconsin and across the United States. When we look at the importance of risk-based pricing and doing that--I want to just go across the panel, if I can and just ask, if you believe in risk-based pricing because it seems like on some of our discussion today maybe that is or maybe that is not the case and so just to go across. Mr. Van Alst, do you believe in risk-based pricing on car insurance? Mr. Van Alst. For car insurance, yes. Mr. Steil. Ms. Cross? Ms. Cross. I defer to my fellow panelists, as I do not focus on insurance. Mr. Steil. All right. Ms. Clarke? Ms. Clarke. Yes. But but the devil is in the details. And it---- Mr. Steil. Okay. But risk-based pricing is-- Ms. Clarke. Yes. Mr. Steil. Fair enough. Mr. Rivera? Mr. Rivera. Risk-based pricing that doesn't discriminate on the basis of gender and other factors that are not proxies for income. Mr. Steil. Understood, the risk-based pricing at its core is an important aspect. Mr. Lynch? Mr. Lynch. I would be consistent with the State laws in every State in the United States and say yes, risk-based pricing is an appropriate way to price automobile insurance. Mr. Steil. Thank you. Going back to you, Mr. Lynch, as we look at insurance-based pricing, do you believe that non-driving related factors are predictive of driving risk? Mr. Lynch. Well clearly, they are, that is why they are used. Mr. Steil. And can you maybe describe for us the process that insurance companies are using to identify and evaluate these risks? Mr. Lynch. Well, when you price insurance, you use a two- step process and the first step is that you estimate what the average rate is going to be for the average consumer in wherever your book of business is, let us say it is the State of Michigan. And then after that you go through a process and you identify through actuarial analysis of your data, supplemented by data from other companies sometimes that are outside of the insurance industry, and then that is how you determine based upon the risks that you find that correlates strongly with the likelihood of being in an accident and that is how you come up with rating factors to adjust your rate come to a final rate. Mr. Steil. Thank you very much. I yield back my time. Chairman Green. The gentleman's time has expired. The gentleman from North Carolina, Mr. Budd, is recognized for 5 minutes. Mr. Budd. Thank you, Mr. Chairman, for letting me sit in on this hearing. And I want to thank the witnesses as well. It is not often that we have hearings about the insurance industry in the U.S. Congress, and that is because we have a State-based system that has been working well for over 150 years, so I believe that North Carolina insurers should answer to Raleigh and not to Washington, D.C., so again these are rare. Mr. Lynch, I have reviewed the Insurance Research Council's April 2019 Report on Auto Insurance Affordability in Michigan and I would like to ask permission to enter this to the record, Mr. Chairman, if that is okay? Chairman Green. Without objection, it is so ordered. Mr. Budd. Thank you, Mr. Chairman. I want to make sure that I have the facts right. How many States, including Michigan, allow for the use of credit-scoring data in auto insurance underwriting, Mr. Lynch? Mr. Lynch. I believe that number is 47. Mr. Budd. Okay, 47. So it strikes me as fair to say that Michigan is not unique in allowing for the use of this predictive-scoring data, so how is Michigan unique? Mr. Lynch. As I said before, the no-fault automobile insurance system and the unlimited no-fault makes it different from every other State, the unlimited nature of it is itself unique. Mr. Budd. So this IRC Report states that the average cost of a claim for an auto accident in Detroit is $51,000 on average per accident and the average cost of auto insurance coverage in the area is $5,400 a year. Now, I have two teenage drivers in a home that I insure on one policy and it is less than that a year, in North Carolina, so how much of an outlier are these figures when compared with other States like North Carolina and what is to blame for the high cost of these figures? Mr. Lynch. Well, this is something that I think I have answered once or twice but it never hurts to try to make it a little clearer, when you have unlimited claims, when your claim can be any size especially in the United States where health insurance is so expensive, you can get claims where people suffer tremendous injuries, and in Michigan if it is in an automobile accident, not in any other kind of accident but if it is in an automobile accident, then the automobile insurance system as a whole has to bear that cost and that cost ends up being directly reflected in the rates the consumers pay. Mr. Budd. So do you agree that State-based insurance legislators and regulators in Michigan can best solve this problem? Mr. Lynch. I work for an organization that doesn't make public policy recommendations and lobby so I can't really answer that. I can say that I have testified in the State of Michigan as they explore ways to alleviate this issue and that at the State level, they are very aware of it. Mr. Budd. Understood, and thank you, Mr. Lynch. I agree, and I think we have identified the cost of sky- high auto insurance rates in Detroit and who can fix the problem. I see no reason for Congress to supersede, in my case North Carolina, in regulating my State's insurance market and I also see no reason why my constituents should be forced to subsidize Michigan or any other State for their seriously flawed State insurance laws. With that, I sincerely want to thank Chairman Green for letting me join today's discussion. And I yield back my time. Thank you. Chairman Green. The gentleman yields back. Without objection, the Chair now recognizes Ranking Member Barr for an additional 5 minutes. Mr. Barr. Thank you once again, Mr. Chairman, for holding this hearing. And once again, I applaud you for your work in Congress and before Congress in combating invidious discrimination. And as Members on both sides of the aisle have expressed today, race discrimination, any kind of discrimination has no place in our society, certainly not in auto lending, certainly not in insurance or any industry whatsoever. I think what a lot of Members have expressed today and some of the witnesses as well is that we all want to get at discrimination, stop it, prevent it, but we also want to make sure that we don't eliminate risk-based pricing so that every good driver in America, and most people are good drivers, are not punished because they are not credited for their positive driving records, and I think it gets to this idea of socializing risk versus pricing insurance based on risk. Let me ask Mr. Lynch one kind of final question to summarize what we have been discussing today and that is, would elimination of insurance credit-scores undermine the risk-based pricing model? Mr. Lynch. Well, to the degree that you restrict the ability to use proven predictors of accident rates, you are going to have a less robust model, it is going to get weaker and that would be regardless of the variable that you were deciding to restrict. Mr. Barr. And let us really get at the core issue here, the core issue is that some of the witnesses on the panel are making the argument, as I perceive it, that credit scores are a proxy for race. Do you agree with that or not? Mr. Lynch. Well, no. They are not. Mr. Barr. And tell me why you disagree with the other witnesses on that point? Mr. Lynch. Oh, well, when we look at insurance, we are told what is fair and you--what you have to do is create a fair rate and a fair rate is one, as I said in my remarks, that is neither too high nor too low nor overcharges or undercharges a risk that presents themselves with a certain known risk characteristics, and all of that of course is silent as it should be to issues of race and income, so insurance companies do follow those laws and so that is kind of where I come to my conclusion. Mr. Barr. If we eliminated credit scores completely from the underwriting of insurance, would an African American with a pristine driving record be subject to a higher premium? Mr. Lynch. I have no way to answer that question because the industry does not gather information based upon race so I can't say. Mr. Barr. Well, forget the race, would drivers in general with good driving records, would they be subjected to higher premiums as a result of elimination of the use of credit scoring in underwriting premiums? Mr. Lynch. Well, most drivers have good driving records and the people who have good credit scores and good driving records and are presenting less risk to the insurance company would end up overpaying. Mr. Barr. Okay. Without objection, Mr. Chairman, I would like to submit several items to the record. The first is the Charles River Report on Fair Lending. The second is a Report on Auto Insurance Affordability and Cost Drivers in Michigan. The third is a letter to you and to me from the National Association of Mutual Insurance Companies, the Independent Insurance Agents and Brokers, the National Association of Professional Insurance Agents, and the American Property Casualty Insurance Association. And the fourth is a statement by the National Association of Mutual Insurance Companies. Chairman Green. Without objection, it is so ordered. Mr. Barr. And with that, Mr. Chairman, I appreciate once again, your leadership in holding this hearing. And I yield back. Chairman Green. Thank you. The Chair now yields 5 minutes to himself. And I thank the Members, all of whom have said they are very much interested in making sure that we do not have invidious discrimination. Let us talk for just a moment about the indirect auto lenders and how they work with the dealers, wherein the dealers are acting as agents to a certain extent of the lenders and once a price has been set for the purchase of a car with the indirect auto lender, the dealer has the option to mark up the interest rate and to do so without any rules or regulations for the most part and actually the dealer does it at his own discretion. Do you agree that this is the case, Ms. Clarke? Ms. Clarke. That is indeed the case. It is a widespread practice. Chairman Green. I have to move to the next person. Ms. Cross, do you agree? Ms. Cross. Yes. I agree. Chairman Green. And Mr. Van Alst? Mr. Van Alst. Yes. Chairman Green. Okay. So we agree that the dealers can mark up within their discretion the final interest rate. Now, let me ask this of you, Ms. Clarke. Between 2013 and 2016, the CFPB brought enforcement actions against four indirect auto lenders for violating the Equal Credit Opportunity Act by authorizing and incentivizing discretionary dealer interest rates. These markups resulted in non-white borrowers paying higher interest rates than non-Hispanic white borrowers. These actions garnered $104 million in restitution for borrowers. Under Director Mulvaney and Director Kraninger, however, the CFPB has not taken any public Fair Lending enforcement action against an indirect auto lender or any lender at all. What are the consequences of the CFPB's lack of enforcement of Fair Lending Laws in the context of auto lending? Let us start with you, Ms. Clarke, please. What are the consequences of the lack of enforcement? Ms. Clarke. We are implicitly sending a green light to dealers and lenders across the country that this kind of discriminatory lending practice is okay. And by rolling back-- rescinding the 2013 CFPB Guidance we have seen lenders and dealers actually abandoned efforts and proactive steps that they were taking to reduce and eliminate the incentive for dealers to engage in those markups. There are a lot of lenders who are imposing a flat rate which completely removed the incentive for dealers to markup loan interest rates so we are seeing the resurgence of discrimination as we see the Federal Government retreating from this space. Chairman Green. What is a possible solution? And I will start with whomever would like to give me the first answer. Mr. Van Alst. I can think of two. One, is to go ahead and prohibit this discretionary increase in the interest rate. There are many other methods that can be used to compensate dealers for the time or effort they spend trying to arrange financing, such as a flat rate. Two, data collection, as we have discussed already today-- without accurate and good data, we can't know what is going on. And currently, the Equal Credit Opportunity Act prohibits collection of race data in these transactions unlike the mortgage market. So it is very important that we gain that data to know what is happening. Chairman Green. Would simply letting the buyer know that this markup exists be a beneficial disclosure? Mr. Van Alst. We have found that throughout the auto sale and finance process, disclosures can often be really ineffective. Dealers who do this on a regular basis are very good at getting consumers to sign things, disclosing things, while they are covering up with one hand or turning the consumers' attention elsewhere, simply disclosing this won't fix the problem and in fact as we pointed out already, consumers have to go through a long, arduous process to ever get to the point where they had have something like that disclosed and you can't do that. Chairman Green. Let me move on to Ms. Cross. Ms. Cross, do you want to add something? Ms. Cross. Yes. One of the big things that you see happen is a lot of predatory tactics that disclosure won't necessarily fix 100 percent. Yo-yo financing is an abusive tactic where a consumer is given the information about their loan and they think it is a final deal. And then later a lender can call back--days or weeks later, the dealership will call back and say, ``Oh I am sorry, something went wrong with the financing. And we need to renegotiate and charge you a higher interest rate.'' Disclosure is a good principle in general, but it won't necessarily cover all our bases in some of those predatory tactics we see. Chairman Green. And Ms. Clarke, what would you recommend? Ms. Clarke. We need the CFPB and the DOJ to enforce the law. We need data collection to bring transparency to the practices of dealers and lenders. And finally, we need Congress to not abdicate its responsibility to eliminate racial discrimination, root, and branch across our country. It is not enough for the States to do this work. They lack the unique expertise and resources that are embodied inside of our Federal Government agency. So we need strong Federal precedents, if we are going to ever combat the crisis that we are up against. Chairman Green. Well, let me thank all of the witnesses for appearing today, and the Members as well. 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. And without objection,the hearing is adjourned. [Whereupon,at 12:09 p.m.,the hearing was adjourned.] [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]