A Biased, Broken System: Examining Proposals to Overhaul Credit Reporting to Achieve Equity

Chairwoman Waters, Ranking Member McHenry, and distinguished members of the Committee on Financial Services, my name is Dan Quan, and I am an adjunct scholar at the Cato Institute’s Center for Monetary and Financial Alternatives. I am also a co-founder and General Partner of Nevcaut Ventures, a venture capital fund that invests in financial technology (FinTech) startups. In addition, I provide advisory services through my own firm, Banks Street Advisory, and as a Senior Advisor with McKinsey. Previously, I was Senior Advisor to the Director at the Consumer Financial Protection Bureau (CFPB) and led its FinTech initiative, Project Catalyst.

I am honored to take part in today’s hearing entitled, “A Biased, Broken System: Examining Proposals to Overhaul Credit Reporting to Achieve Equity.” My testimony will focus on the public credit bureau proposal that was first developed by the think tank Dēmos¹ and later endorsed by then-presidential candidate Joe Biden.²

At $14.6 trillion,³ the consumer credit market in the United States is the largest and most competitive in the world. This vibrant credit market depends on a functioning credit reporting industry that is mostly made up of three national credit reporting agencies, Equifax, Experian, and TransUnion. TransUnion alone maintains 3.1 billion tradelines, processes more than 2 billion credit updates monthly, and matches these to more than 236 million active credit consumer profiles in the United States.⁴

The credit reporting agencies have been the subject of serious criticisms, most of which concern three areas:

Accuracy. A number of studies and surveys have indicated that credit reports have accuracy issues. The most comprehensive and authoritative study conducted by the Federal Trade Commission (FTC) shows that almost one in four consumers have an error on their credit reports that could affect their credit score, and 5% of consumers had errors that could result in less favorable credit terms.⁵

Security. American consumers can not be assured that their sensitive personal information is safe and secure. The Equifax data breach in 2017 that exposed the personal information of 147 million American consumers is one of the largest in recent history. Still, credit bureaus are not subject to any federal supervision on data security.

Equity. Forty-five million Americans are excluded from the mainstream credit system according to a CFPB study published in 2015.⁶ Black, Hispanic, and low-income consumers are disapproportionately credit invisible or unscorable.⁷

In response to these concerns, the Biden campaign, drawing on a proposal by Dēmos, a think tank based in New York City, endorsed the idea of a public credit bureau to supplement or even replace the private credit reporting agencies. The Dēmos proposal calls for:

1. Establishing a public credit bureau within the CFPB.
2. A 7‑year transition period within which credit data will be furnished to the public bureau. During this time, private credit bureaus are required to provide their (historical) credit data to the public credit bureau. At the end of the transition period, lenders are only allowed to use data from the public credit bureau. The private credit bureaus known today will essentially cease to exist.⁸
3. Credit information can only be used for lending. The use of credit information for employment, housing, and insurance will be prohibited.⁹
4. The public credit bureau will develop new algorithms that minimize disparate racial impact by including non-traditional data sources such as rental history and utility bills.

The Dēmos proposal’s policy goals are to “improve equity, transparency, accuracy, accountability, appropriateness, security and public awareness of credit information.”¹⁰

These are certainly laudable goals. However, I believe that they can best be achieved by retaining and improving upon the existing, private sector credit reporting industry than by establishing a public credit bureau.

First, improving equity in credit access is best achieved by private sector innovations. The proposed bureau is supposed to achieve more equitable credit access by taking advantage of “new algorithms” that will consider new data sources such as rental history and bill payment. However, it is doubtful that by doing so the bureau will improve upon what the existing system is capable of achieving. This is so in part because the algorithms in question are not so new.

The so-called “new algorithms” have been developed by for-profit companies to tackle the credit access challenge without any government interventions.¹¹ With the advancement of open banking and the underlying financial infrastructure built by data aggregators, consumers now have the ability to permit third-parties to access their financial data housed within financial institutions. Riding this wave of innovation, Experian, FICO, and Finicity partnered to create Experian Boost and Ultra FICO. Both programs operate on an opt-in basis and are entirely free of charge. They allow consumers to use their on-time bill payment history and streaming subscriptions to increase their FICO scores. With the consumer’s explicit consent, the payment data flows directly from the consumer’s checking account to Experian and is incoporated in the “new algorithm” developed by FICO. According to Experian, users can boost their credit score by an average of 12 points, and to date, credit scores have been collectively increased by more than 69 million points.¹²

The FICO score is important, but it is not the only metric lenders use to assess credit risk. A growing number of lenders, most of them venture-backed, see the promise of using on-time payment history data in credit decisioning and have developed cash flow based credit underwriting models (i.e., another “new algorithm”) to lend to consumes who lack a credit history. In fact, such “new algorithms” received the attention of 5 federal banking regulators, which issued a joint statement affirming the benefits and promise of the use of cash flow data.¹³

Many technology-driven companies are already innovating to serve the need of the 45 million consumers who have been traditionally left behind. The public credit bureau will not add any value to the emerging market-based solutions. Nor is there any good reason to suppose that the bureau would have the know-how and experience to develop better solutions.

Second, transparency is a real issue that requires a different kind of policy intervention. The Dēmos proposal argues that existing credit algorithms are opaque and lack useful information for consumers to improve credit. This is a fair criticism, but creating a public credit bureau that provides free credit reports and scores will not solve the problem.

Today consumers who have an internet connection or a smart phone can obtain free credit reports and credit scores relatively easily. However, most of them (myself included) still have a hard time understanding why their credit applications are rejected by lenders.

Ultimately it is lenders (not credit bureaus) that are responsible for the issues in transparency. Lenders are required under the Equal Credit Opportunity Act (ECOA) and the Fair Credit Reporting Act (FCRA) to issue adverse action notices to consumers whose credit applications are declined. Those notices must explain why adverse actions were taken. The intention of such notices is to provide transparency to the credit underwriting process so that consumers may improve their odds of getting approved in the future. Unfortunately, almost every lender simply complies for the sake of complying by issuing adverse action notices that contain cryptic reason codes that most consumers do not understand. This is a long-standing issue. The CFPB hosted an adverse action notice tech sprint last October¹⁴ in an attempt to use its persuasion power to nudge the industry to come up with more consumer-friendly solutions, and this is where the focus should be.

Third, the public credit bureau will not necessarily improve data accuracy. The Dēmos proposal recommends that data furnishers be held accountable for providing accurate information to the public credit bureau. At the same time, the public credit bureau will use “the most robust methods available to ensure that credit information is accurate and to avoid mixed files (cases where one person’s credit accounts are mixed into someone else’s file).”¹⁵

If consumers are to achieve financial freedom, their credit files must contain accurate information. But steps have already been taken toward this end. Improving credit report accuracy has been one of the CFPB’s top policy priorities.¹⁶ Since 2012, it has been supervising the credit reporting bureaus for accuracy¹⁷ while holding data furnishers accountable for their failure to supply accurate information to them.¹⁸ Although we do not yet know just how successful its efforts have been—the last FTC accuracy study was conducted a decade ago—it behooves policymakers to request a new study to determine what those efforts have accomplished before considering any radical policy reform.

Furthermore, assuming that consumers’ credit reports still contain many errors today, it isn’t clear how a public credit bureau would succeed in reducing the same errors that affect its own reporting. For the 7‑year transition period it would inherit the same errors in files transferred to it from the private credit bureaus. And thereafter it would continue to receive files from the same error-prone sources the private bureaus rely upon.

Fourth, accountability can be achieved without establishing the public credit bureau. The Dēmos proposal correctly states that consumers have a right to dispute errors in their credit files. Although there are deep disagreements about the dispute process,¹⁹ both H. R. 3621 and H. R. 5332 contain proposals to improve it. The merits of those proposals can be debated, but fundamentally, it is possible to improve the dispute process in the existing system. 

Fifth, appropriateness can be achieved without the public credit bureau. The Dēmos proposal would allow credit information to be used in lending only.²⁰ However, the FTC in 2007 conducted a study that shows credit-based insurance scores are “effective predictors of risk under automobile policies.”²¹ There can be a healthy policy debate (and the commission of new studies if needed) about whether credit reports can be used for employment, insurance, or other purposes. Congress and states can pass laws to expand or limit the use of credit reports and the private sector will simply comply.

Sixth, it is not clear why proponents of a public credit bureau think it would be more secure than the private sector. While the Equifax data breach is surely damaging, one should not forget the consequences from the 2015 data breach at the Office of Personnel Management that exposed sensitive information of more than 22 million federal employees.²² There is no safe haven in today’s networked world. What is urgently needed is enhanced data security supervision rather than a public credit bureau.

Seventh, substantial public awareness and credit education has already been effectively achieved without a public credit bureau. While there is surely room for improvement, public awareness of credit reports has significantly increased as consumers can obtain free credit reports and scores via many channels.²³ Today, consumers can receive free updated credit reports and scores as frequently as every week. This is a significant improvement from the early days when consumers had to proactively request free credit reports via the arcane and clunky annu​al​cred​itre​port​.com every 12 months.

I have so far demonstrated that all the policy goals in the public credit bureau proposal can be as effectively achieved via innovation and by reforming the existing credit reporting industry. There are also significant negative consequences of establishing a public credit bureau that make the proposal undesirable.

Intrusion of Privacy. The public credit bureau may pose a significant threat to consumers’ privacy. From the moment consumers make their first credit transaction, the federal government would monitor every credit account applied for, every account opened and closed, every payment made (on time or late), every lender from which money was borrowed, where the consumer has lived, and even who they have worked for. Putting so much sensitive personal information in the hands of the government, especially when consumers have no way of opting out, should be a concern big enough to override any potential merits of the public credit bureau.

Huge Taxpayer Cost and Waste. The cost of setting up and running the public credit bureau needs to be seriously considered. For FY 2020, the CFPB had a workforce of 1,503 employees²⁴ and an operating budget of $580 million.²⁵ TransUnion, the smallest of the three national credit bureaus, employed 8,200 people worldwide and had an annual operating cost of $2.1 billion in 2020.²⁶ Of course those are not apples-to-apples comparisons, but they give us a sense of how much the federal government needs to be expanded to run the public credit bureau.

Anti-Competition and Stifling Innovation. The public credit bureau will crowd out private investment in the credit reporting industry. Funded by tax revenue, the public credit bureau will ultimately monopolize credit reporting. If members of this Committee are already concerned about the lack of competition among the three national credit bureaus today, the situation will be far worse once the public credit bureau becomes the sole place for lenders to turn to. The budding and promising innovation in expanding credit access will cease to exist.

The credit reporting industry has a number of health issues that merit appropriate legislative and regulatory interventions. Here is some food for thought.

Refreshing the Accuracy Study. There has not been a rigorous accuracy study since the FTC conducted one 10 years ago. Many changes have taken place in the past 10 years, including the increased regulatory scrutiny on accuracy and the substantial growth in consumer awareness thanks to the proliferation of free credit reports and scores.²⁷ The CFPB has indicated that it is exploring a new accuracy study,²⁸ and it has garnered support from both the industry and consumer adcovates. However, it has yet to conduct the study. Members of this Committee should insist on the completion of this study.

Supervising Data Security. The Equifax breach took place 5 years after the CFPB began to examine credit bureaus. However, the CFPB was never given the authority to oversee data security and as a result it was not able to assess the adequacy of Equifax’s information security.

There needs to be federal supervision of data security. When it comes to data security, credit bureaus should be held to the same standard as financial institutions. Congress needs to act quickly before the next breach hits.

Reforming the Dispute Process. It is vitally important that consumers have accurate information in their credit files. Inaccurate information and derogatory information are not the same, even though the inaccurate information that consumers try to get rid of during the dispute process is almost always derogatory. Derogatory information, if accurate, is equally valuable in helping lenders assess creditworthiness. There are already a number of proposals and I think they all merit serious considerations. I would like to offer two high-level suggestions. First, the dispute process should be made easy for consumers. A frictionless dispute process can yield high success rates and reduction in errors. Innovation can play an important role here. Second, consumers should be able to directly dispute with data furnishers and have the updates reflected in all bureau reports. Errors in furnishing and reporting lags are often times the source of inaccuracy in credit files.

Protecting Consumers’s Data Rights Under DFA Section 1033. Congress can pass legislation to improve accuracy, transparency, and security in credit reporting, but it would not be able to effectively bring credit invisible consumers into the credit system. Competition and innovation remain the best means to achieving equity in credit access. Much of the encouraging innovation in improving credit access discussed earlier in my testimony depends on the reliable access to consumers’ financial data housed within and controlled by financial institutions.

Section 1033 of the Dodd Frank Act requires financial institutions to allow third parties that are permissioned by consumers to access their financial records. This is called open banking or open finance. Consumers’ data is no longer compiled, furnished, and sold without their own consent. Instead, the consumer decides who can access their own financial records, for what purposes, and for how long. While much progress has been made, “many large financial institutions, which hold the majority of consumers’ data, have too often resisted the intent of Section 1033 by creating obstacles to easy data access.”²⁹

It has been more than 10 years since the Dodd Frank Act was enacted and 5 years since the CFPB first studied consumers’ data access rights. Now is the time for the CFPB to fulfill its mandate and write a pro-consumer, pro-competition, and pro-innovation rule. The rule should root out existing anticompetitive practices by requiring complete access to consumers’ financial data and explicit and transparent consumer consent and control.

The public credit bureau solution would be overkill for the long standing issues with the credit reporting industry. The solution is not only likely to fail to achieve the goals it sets out to reach, but it will also create a massive bureaucracy that wastes public funds, intrudes on privacy, and stifles innovation.

There is no need to tear down the entire system. With appropriate legislative and regulatory interventions and a conducive environment to foster innovation, Congress, regulators, consumer groups, and the private sector can work together to reduce racial disparities in credit, improve accuracy, and protect consumers’ sensitive data.

Thank you for the opportunity to provide this information, and I welcome any questions that you may have.