UPDATE 12.17.2020:
The Consumer Financial Protection Bureau this week finalized its Small Business Regulatory Enforcement Fairness Act (SBREFA) report, as part of a settlement agreement with CRC, NALCAB and two small business owners.
“The rulemaking is proceeding and we look forward to finally seeing a small business data collection rule,” said CRC Deputy Director Kevin Stein. “We’re also glad to be moving closer toward much-needed transparency around small business lending by race, ethnicity, gender, and neighborhood of small business borrowers in order to identify where discrimination and redlining are occurring, and to help policymakers determine whether further steps are needed to ensure equal access to fairly priced credit for all small businesses.”
CRC and other plaintiffs will meet with the CFPB to discuss when a Notice of Proposed Rulemaking will be released for public comment.
Original 12.14.2020:
In response to the Consumer Financial Protection Bureau’s small business data collection rule outline, the California Reinvestment Coalition on Monday, Dec. 14 issued the following comments to the bureau’s Office of Small Business Lending Markets.
The California Reinvestment Coalition (CRC), its members, and allies submit these comments in response to the Consumer Financial Protection Bureau (CFPB)’s release of its Outline of Proposals Under Consideration and Alternatives Considered for Section 1071 of the Dodd-Frank Act governing small business lending data collection and reporting.
The groups endorsing these comments represent Community Development Financial Institutions (CDFIs), Community Development Corporations (CDCs) small business technical assistance providers, legal service offices, civil rights groups, and other community-based organizations deeply concerned about the historic lack of access to responsible credit endured by businesses owned by women and people of color, by small businesses, and by businesses located in neighborhoods of color.
CRC builds an inclusive and fair economy that meets the needs of communities of color and low-income communities by ensuring that banks and other corporations invest and conduct business in our communities in a just and equitable manner. CRC and our member organizations have been fighting predatory financial practices, supporting small businesses, and fighting redlining for over 30 years. We have been advocating for small business data collection and transparency within the small business market for decades.
CRC and co-plaintiffs were pleased to have settled our case against the CFPB, resulting in a definitive timeline for CFPB to develop the 1071 rule. We appreciate that the CFPB is following through on its commitments and obligations, and has released the SBREFA Outline. While we find the CFPB’s efforts to be in good faith, we find substantial gaps and deficiencies suggested in the Outline.
We urge the CFPB to adhere to the following principles in further developing the 1071 Rule:
1. Focus on furthering the Rule’s stated purpose of furthering fair lending enforcement and helping to identify community development needs and opportunities.
2. Provide for broad coverage of lenders, applicants and transactions, including merchant cash advance products which expose the fact, and costs, of banks failure to lend to BIPOC and women-owned businesses.
3. Require detailed and disaggregated data collection for factors such as pricing (APR) credit score (FICO) and underwriting (Reasons for Denial), in order to allow analysis to hone in on whether and how discrimination in small business lending is occurring. Robust data collection will allow the public to have a much greater understanding of gaps in lending to all borrowers in the marketplace, and easily identify unmet borrowing needs.
4. Recognize that society’s interest in eradicating discrimination and closing the racial wealth gap far exceeds any negligible risk of privacy violations or regulatory burden on financial institutions that already collect much or all of the relevant data.
5. Ensure that all data is made publicly available, as only with such a high degree of transparency can the public have confidence that lenders will improve problematic and discriminatory practices that have thrived in the secrecy of darkness, that enforcement agents will bring cases against those that do discriminate to finally provide redress to victims of discrimination, and that policymakers can work to address any legitimate issues that prevent small, women- and minority-owned businesses from attaining responsibly priced credit.
Tragically, lending discrimination remains a fact of life, including for small business owners of color. There can be no doubt that lending discrimination has existed in the small business market for years. CRC member organizations have long reported on and lamented the failure of banks to lend to qualified small businesses, particularly those owned by women and people of color. CRC analysis of the limited data sets that have existed has found disparities in lending to minority-owned businesses. Just as with the subprime mortgage crisis, lightly regulated high-cost lenders have filled the space left by banks that fail to provide credit access, leaving small business owners of color and women-owned businesses burdened with Merchant Cash Advance and other harmful products that wreak havoc on business and personal finances.
The U.S. Small Business Administration Paycheck Protection Program (PPP), part of the nation’s COVID response, reflected these disparities and reinforced redlining dynamics by funneling small business relief primarily through the very banking institutions that have not served the small businesses meant to benefit from the 1071 Rule. CRC surveys of bank COVID responses found that nearly all bank respondents acknowledged prioritizing existing clients over businesses that did not have a prior banking relationship, especially during the first round of PPP funding, and the limited data available showed that only 2% of PPP loans were made in support of Black-owned businesses and only 6% of the recipients were Latinx-owned businesses. And the data reported do not reveal information about applications taken and denied.
The Center for Responsible Lending identified a number of structural barriers built into the PPP that disadvantaged small and self-employed businesses, especially those owned by people of color. And, disturbingly, a recent study of the PPP conducted by the National Community Reinvestment Coalition found that Black and White matched-pair testers experienced different levels of encouragement to apply for loans, different products offered and different information provided by bank representatives.
Honor the statutory purpose of the Rule: Further fair lending and identification of community need
The statutory purposes of Section 1071 must be honored and vigorously pursued. In developing the rule, the CFPB must hew closely and always return to the Congressionally stated purpose of the rule. According to the statute, “the purpose of this section is to facilitate enforcement of fair lending laws and enable communities, governmental entities, and creditors to identify business and community development needs and opportunities of women-owned, minority-owned, and small businesses.” Both aspects of this stated purpose – to further fair lending enforcement, and to help the public identify community development needs – argue in favor of detailed data and full public disclosure.
Provide broad coverage of lenders, applicants, and transactions
To be effective, the rule must provide for broad coverage of lenders, applicants, and transactions.
What is a “small-business applicant?” We support the proposed second alternative, a size standard of a maximum of 500 employees for manufacturing and wholesale industries, and a maximum of $8 million in gross annual revenue for all other industries. We do so because this will capture the greatest number of small business credit applications. Note that it is critically important that data fields be created to allow for analysis looking at smaller buckets of small businesses based on gross revenue, number of employees, etc., which will be discussed further, below.
Who is covered? The definition of “women-owned, minority-owned and small business” should provide for the broadest possible coverage. While we appreciate that the vast majority of such businesses would be covered by a definition that focused on small businesses, including small business owned by women and people of color, we recommend that the definition be broadened to also include businesses owned by women and people of color that may not be deemed “small businesses.” It might even be easier administratively for small business credit providers to collect data for all applicants as opposed to developing systems for screening out the minority of applicants that are not covered.
“Minority-owned” and disaggregate race and ethnicity data. We support the proposal to consider whether 50% or more of the ownership interest or profit/loss resides with a member of a minority group. But we strongly disagree with the proposal to NOT use the disaggregated race and ethnicity categories that are part of the Home Mortgage Disclosure Act (HMDA) data collection regime. CRC members fought for the inclusion of disaggregated race categories in HMDA and think such data are equally relevant and important in the small business lending context. Detailed race and ethnicity data will further the purpose of the rule to further fair lending and help policymakers address community needs.
We strongly object to the suggestion that the race or ethnicity of an applicant for credit should only be determined by self-reporting from the applicant. HMDA has successfully managed to require lenders to make determinations based on visual observations and encouraged them to do so by considering surnames when borrowers choose not to self-identify. We think this approach should work as well in the small business context and should be embraced as a way to enhance data collection and fair lending enforcement.
Which financial institutions are covered? We support the Bureau’s inclination to cover all categories of institutions, including: banks, credit unions, non-banks, commercial finance providers, nonprofits, and government lenders. We think the only size or volume-based threshold that should apply is in proposed Option One, an exemption for financial institutions that make fewer than 25 loans in either of the last 2 years. A similar threshold has been established in the 2015 HMDA rule. Increasing the volume threshold may disproportionately harm rural communities that may find that small lenders makes up a significant portion of the local lending market. An asset threshold would not add value and appears unworkable for non-bank lenders. If multiple lenders are involved in making the credit available, we support the position of NCRC that the data collection resides with the company most closely interacting with loan applicants.
What forms of credit should be covered, and why Merchant Cash Advance should be covered? We believe strongly that all forms of credit should be covered. Certainly, bank transactions should be covered so enforcement agencies, policymakers and the public can see the extent to which banks are failing to serve businesses owned by people of color and women. Additionally, Merchant Cash Advance, factoring, and leases should be added to activities the CFPB proposes to cover. The Bureau undermines its proposal to define covered lenders broadly by then exempting a significant number of transactions that define important and highly relevant categories of lenders.
This is perhaps one of the most glaring gap in the Outline – the suggestion to create exemptions for consumer-designated credit, leases, factoring, trade credit, and merchant cash advances. Any rule that allowed for such a gaping hole in coverage would substantially and perhaps fatally undermine the statutory purpose of the Rule. How can enforcement agencies, policymakers, and the public know whether small business lending discrimination is occurring if we do not seek data on higher costs products that CRC believes are targeted to protected classes under the Equal Credit Opportunity Act?
In its comments on the CFPB’s Outline, Opportunity Fund notes compellingly that, “According to a white paper, ‘Key dimensions of the small business lending landscape,’ published by the Bureau in 2017, the estimated number of accounts for factoring products was estimated to be at eight million, MCAs with one million, and equipment financing at nearly nine million. These three products accounted for over 18 million accounts, 2.5 times the estimated 7 million term loan accounts. While the dollar amount of factoring and MCA products is smaller than term loans, quantifying factoring and MCAs by the number of accounts illustrates that these types of credit are widespread. Additionally, leasing products make up 13% of the small business financing market share in dollar terms, further indicating that those products should also be included.” The National Community Reinvestment Coalition estimates based on its analysis of MCA company filings, that the number of MCAs has increased significantly since the CFPB white paper.
Further, according to LendingClub, “Federal Reserve research shows that minority-owned businesses are twice as affected by “potentially higher-cost and less transparent credit products”—a term the Federal Reserve uses to refer specifically to merchant cash advance and factoring financing.”
A September 2017 survey of CRC member organizations working with small businesses found that 87% of respondents reported that small businesses were receiving MCAs, but 63% reported that this was not helpful to the small businesses. CRC’s CDFI members and allies report the frustration and the wasted resources they expend to refinance small businesses out of Merchant Cash Advance and similar products.
Through one of our members, CRC spoke to a firefighter in Chino who opened a successful home care business, only to suffer cash flow problems when he had a heart attack that forced him to reduce his work hours. He turned to Merchant Cash Advance, but one MCA turned into 15 MCAs, and the more MCAs he had and the higher the charges, the harder it was for him to find someone to help him get out of the cycle of debt. Sadly, this story is all too typical. When CRC, Woodstock Institute, Main Street Alliance, and National People’s Action organized a convening of small business owners to talk to an interagency panel in San Diego a few years ago, in which the CFPB participated, 11 of 12 participating small business owners had taken out at least one MCA and regretted it.
The state of California has defined Merchant Cash Advance and required MCA, factoring, and leasing transactions to be subject to pricing disclosures. And the Department of Financial Protection and Innovation has recently taken action against MCA lenders, treating such transactions as loans, particularly where future receivables are at issue and where the small business borrower bears the risk of repayment, as is the case with most MCAs.
To be sure, MCAs and other higher cost products exist and thrive because banks and other more conventional lenders have failed to adequately serve women-owned, minority-owned, and small businesses. They target these businesses because these businesses have been redlined and shut out of mainstream banking. But if we do not require MCA reporting, we will be letting banks off the hook by failing to reveal the extent to which businesses of color and women-owned businesses are having to turn elsewhere for financing, and at what cost. Given their significant role in the marketplace and with businesses meant to benefit from the rule, as well as how their success highlights the fair lending failures of mainstream financial institutions and the harm imposed on women and minority-owned businesses, MCAs must be covered by 1071 data collection requirements.
What constitutes an “application?” This is an important issue, for if the definition of applicant is too narrow, the data will ignore discrimination that occurs frequently when small business owners make credit inquiries. For example, businesses owned by people of color and women might be urged to develop and submit business plans instead of being offered the loan applications they seek. The work of NCRC, referenced above, shows that discrimination is occurring in the pre-application phase. We cannot allow this data collection effort to unduly narrow the funnel so that interactions between financial institutions and small business owners seeking credit remain hidden from potential scrutiny. We support the Bureau’s inclination to consider both oral or written requests to be “applications,” but argue against any proposed exemptions for: (2) re-evaluation, extension, and renewal requests.
Multiple requests are multiple. If a small business submits multiple requests for credit at one time, perhaps inquiring about several products at once, these should be considered several applications for credit for 1071 reporting purposes. So, too, should the amount of credit requested by the applicant be collected and reported.
Collect disaggregated and detailed data to permit robust analysis and ferret out discrimination
While providing for broad coverage, the rule must also collect disaggregated and detailed data to further rigorous analysis that will help determine whether discrimination is occurring.
For example, while we support a broad definition of small business so that many credit transactions are covered, we are particularly concerned about the smallest of businesses that may need smaller credit amounts and that are the bedrock of local communities in light of their local hiring, provision of local-serving products and services, and general support for their local neighborhoods. In fact, 95% of businesses, 97% of minority-owned businesses, and 98% of women-owned businesses have less than $1
million in revenue, and need financing under $100,000. That is why the 1071 data must allow for comparisons of the experiences of businesses with lower revenue with those of higher revenue, between self-employed businesses and those with a few or many employees, etc. To that end, we strongly support the CFPB including additional discretionary data points, as is within the Bureau’s authority.We look forward to seeing detailed data relating to mandatory reporting requirements concerning loan product type (including credit card lending and lines of credit), type of guarantee (which should include any additional or personal collateral required), loan term, and loan purpose.
APR pricing data is a must. How can we have a data collection regime designed to further fair lending enforcement, if we ignore and screen out information about whether high-cost lenders are targeting business owners of color or women-owned businesses, or if lenders are charging more to their female borrowers or customers of color as compared to white or male borrowers?
The state of California has instituted a policy of requiring pricing data disclosures for commercial financing though our first in the nation truth in lending law for small businesses. To implement this policy, the state Department of Financial Protection and Innovation is proposing Annual Percentage Rate (APR) data to be disclosed to small businesses and is determining how best Merchant Cash Advance and similar credit providers can calculate the APRs on their loans. New York’s Small Business Truth in Lending Act, recently passed by the legislature, requires APR disclosure by statue.
We support the recommendations and positions of the Responsible Business Lending Coalition in regard to how APR calculations can be standardized and straightforward for financing providers not already required to calculates an APR under California and New York law. APR is the only true, tested, transparent, and comparable tool for quantifying price information, and 1071 data collection must include such data. To do otherwise would be akin to earlier iterations of HMDA data which only allowed analysis as to whether a loan was made, not whether it was high cost or conventional. Lenders flooding neighborhoods of color with high-cost loans would be seen as well serving otherwise underserved markets. But not every loan is a good loan. Pricing data, all in APR more specifically, is key to achieving 1071’s statutory purpose.
Other discretionary data fields should be included. We support the inclusion of additional suggested data fields, such as time in business, six-digit NAICS Codes, and number of employees. These data points may go to a credit provider underwriting decisions and must be accounted for so credit providers cannot, as HMDA reporters have done for years, hide behind data NOT collected as the reason and justification for their lending disparities.
Including fields for Reasons for Denial could help explain declinations, and lead to needed policy change. We strongly recommend that 1071 data include all HMDA Actions Taken, and Reasons for Denial. When the data become public, we will no doubt see disparities. And no doubt credit providers will say that the disparities can be explained by underwriting decisions that 1071 data do not account for. But capturing Reasons for Denial can help policymakers and the public determine if there are legitimate reasons that small businesses are not deemed qualified for certain forms of credit, and enable us to work towards solutions. Lenders should be required to report the reasons for denial (deficiencies in: business credit score, personal credit score, debt to income ratio, collateral, size of loan, years in business, inadequate loan documentation, debt service coverage ratio, etc.), as well as provide supporting data to justify that reason.
Data should be collected upfront. The Bureau wonders if data should be collected upfront, or whenever the lender determines. We think it is important to collect data upfront. Much discrimination in the market likely occurs in the early stages of application. Lenders should obtain all information immediately so they can report on the action taken on all loan applicants.
The role of the firewall. While we appreciate the idea behind ensuring that underwriters have no access to demographic data for loan applicants in order to prevent overt discrimination, we wonder if such a firewall has ever truly existed. We are hopeful that robust 1071 reporting by lenders and oversight by CFPB will significantly advance fair lending enforcement, regardless of the strength of the firewall. Consistent with the idea of and opportunities presented by Special Purpose Credit Programs, we wonder if lender/underwriter interest in the demographic information of loan applicants will more likely be geared towards ensuring equal access and not discrimination. In other words, it may be that lenders interested in the demographic data of loan applicants are so interested because they want to ensure they are providing equal access to credit, not because they intend to deny loans to protected classes of borrowers.
In any balancing test, recognize the strong interest in detailed, broad, and public data collection over dubious concerns around regulatory burden and privacy
Society’s interest in eradicating discrimination and closing the racial wealth gap far exceeds any negligible risk of privacy violations.
By not disclosing any personally identifiable information, privacy risks are reduced. For years, banks have disingenuously advocated for the privacy interests of their borrowers, even though we are not aware of any privacy breach that resulted from HMDA data (and even though banks routinely fight virtually every other effort to protect bank customer privacy). HMDA has also witnessed the usage of modified approaches that might put exact data points into broader buckets that still allow for helpful data analysis.
The privacy interests of the lenders should not be considered at all, and do not appear to have been a concern for Congress. HMDA has worked for years without any data breaches to compromise individual privacy interests or impose unfair or undeserved harm on reporting lenders. As with HMDA, certain data points can be disclosed to the public in buckets or ranges.
We believe that most or all of the data that might be collected under 1071 is already collected. Lenders will necessarily collect much of this data in order to inform underwriting decisions. Additionally, bank regulators, the SBA, and the CDFI fund routinely require participants in their programs to collect data that may then get reported. Recent disclosures reveal the extent to which Economic Injury Disaster Loan (EIDL) and PPP lenders were required to collect data on their loan applicants, where such data have been made publicly available and where the sky has not fallen. This is not new, this is just standardizing and enhancing what already exists in incomplete and fragmented form.
The Bureau wonders if reputational risk, the need to respond to allegations of discrimination with further analysis, the costs of compliance, and a possible push towards standard underwriting will lead some financial institutions to stop offering some products. We note that none of these concerns are raised by the statutory purposes behind the Rule. Full transparency should protect responsible lenders from unfair
scrutiny and enforcement and set them apart from those less scrupulous. We also acknowledge the work that the CFPB did to estimate compliance costs, suggesting that costs appear reasonable and that costs passed on to small business customers will be minimal ($12 to $40). Experience with HMDA suggests a data collection requirement does not lead to lenders leaving the market. And a lack of detailed 1071 data will only allow predators and discriminating institutions to continue to unfairly extract profits while imposing harm on small businesses, female owners and BIPOC communities.
Make all of the data readily available to the public.
Data should be made available quarterly to the public, as has been the case with HMDA, despite recent rollbacks of this practice. Access to the data is important, and we want that access to be timely, for all stakeholders, so that the public is not relying on old data that may not reflect newer products, policies, and practices of lenders.
We also believe that all credit providers subject to 1071 should be required to provide their data to anyone within 30 days of a request to do so. Such has been the case for HMDA, and this has worked out well. CRC has requested HMDA data from HMDA reporters several times over the years. We see no reason why this should not be an option with small business data. If the CFPB will relieve lenders of any obligation to respond to individual requesters, as it seems inclined to do, we suggest this would be a stronger argument for quarterly public reporting of 1071 data. We support the Bureau in providing guidance to lenders before, during, after implementation.
The target date for full implementation and public data disclosure should be no later than January 1, 2023. We have waited for the data for over 10 years since the passage of the Dodd Frank Act, but really well-
beyond that as CRC members have for decades recognized small business data collection as perhaps the most impactful tool for increasing access to responsible credit for small businesses and, in particular, those owned by women and people of color.
We thank you for the opportunity to comment, and for your consideration of our views.
Very Truly Yours,
California Reinvestment Coalition
(Other Groups)
Download a copy of the letter
Sign-on to fight small business lending discrimination