ABA - American Bankers Association

02/27/2024 | Press release | Distributed by Public on 02/26/2024 21:47

The true cost of too much data

A 2024 ABA survey shows that the CFPB's 1071 final rule will significantly increase the cost of small business credit.

By Dan Brown and Kathleen RyanABA Data Bank

S
ection 1071 of the Dodd-Frank Act amended the Equal Credit Opportunity Act (ECOA) to require financial institutions to collect and report to the Consumer Financial Protection Bureau (CFPB) 13 data points regarding applications for credit by women-owned, minority-owned, and small businesses. Section 1071 also authorizes the CFPB to require the collection of additional data, but only if such data "would aid in fulfilling the purposes" of Section 1071: "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." On March 30, 2023, the CFPB published a final rule to implement section 1071 (final rule). The final rule expanded on the 13 statutorily mandated data points and requires financial institutions to collect and report 81 data fields.

CFPB cost estimates

Section 1022(b)(2)(A) of the Dodd-Frank Act requires the CFPB, when promulgating a rule, to consider the potential benefits and costs to consumers and "covered persons" (in this case, financial institutions that lend to small businesses), including the potential reduction of access by consumers to consumer financial products or services resulting from such rule. Because Section 1071 relates to business credit, rather than consumer credit, the CFPB considered the potential benefits and costs to small businesses, and the potential reduction in credit access for small businesses.

To comply with the statute, the CFPB sought to estimate the cost of implementing the final rule ("one-time cost") and the annual ongoing cost of compliance with the final rule ("ongoing cost"). The CFPB estimated one-time costs based on a 2020 survey it conducted before issuing a notice of proposed rulemaking. That survey instructed respondents to estimate costs assuming the CFPB would require institutions to collect only the 13 statutorily mandated data points. 105 financial institutions responded to the survey, with 42 banks and credit unions responding.

The CFPB did not conduct a survey to estimate the annual ongoing cost of the final rule. Instead, the CFPB relied on its estimates for the number of hours required to report data under the 2015 Home Mortgage Disclosure Act (HMDA) final rule. The CFPB "adjusted" these estimates to the number of data points required under the 1071 final rule. In addition, the cost estimates for the 2015 HMDA final rule were not based on a survey of covered financial institutions. Instead, the CFPB identified the tasks involved in collecting and reporting HMDA data by interviewing 20 financial institutions, 9 vendors, and 15 government agencies. Based on these interviews, the CFPB estimated the hours needed for each task and multiplied the hours by the average salary of compliance officers.

The CFPB's reliance on HMDA cost estimates to generate estimates of the ongoing cost was inappropriate. HMDA requires the collection and reporting of data from applications for home mortgage credit, which for many lenders involves a single line of business, a single loan origination system, and a uniform application form used by nearly all lenders. In addition, the majority of mortgage lenders have long used loan origination systems from which the HMDA data can be extracted and reported to the CFPB. In contrast, small business lending typically involves a more complex and diverse set of loan products offered by multiple lines of business, each of which may use a unique loan origination system or no loan origination system at all. And there is no widely used standard application form for small business lending, unlike mortgage lending.

ABA survey

Since publication of the final rule in March 2023, small business lenders have been assessing the obligations it would impose, if and when it goes into effect, and developing associated budgets. During the week of Feb. 5, 2024, the American Bankers Association conducted an anonymous survey of member banks on costs to comply with the final rule; our survey asked about both one-time and ongoing costs associated with their compliance imposed by the final rule. The survey data show that the final rule will significantly increase the cost of small business lending, which would not only disproportionately harm smaller banks but also smaller businesses-as a previous ABA staff analysis predicted.

ABA analyzed survey results from 297 banks. Figure 1 shows the number of respondents by different asset groups. As the graph shows, banks with $1-10 billion in assets accounted for more than 120 respondents, and there were more than 80 responses from banks with less than $500 million in assets.

[Link]

Figure 1: Survey respondents by asset group (click to enlarge)

Respondents allocate a significant percentage of their total lending to small businesses. Figure 2 is a histogram outlining the distribution of the percent of small business lending as a share of total lending. More than 30 percent of respondents (90 of the 297) reported that small business lending makes up a major portion of their loan portfolios. This not only highlights the negative implications of the 1071 final rule for many banks, but it also reinforces that a sizable share of the nation's banks are overwhelmingly focused on serving America's small businesses.

[Link]

Figure 2: Survey respondent small business lending concentration (click to enlarge)

The survey responses that banks provided to ABA reflect bankers' actual experience with project management planning, initial implementation of the final rule, and discussions and estimates from IT, operations, compliance and legal staff and third-party vendors.

Summary of findings

ABA's survey shows that the CFPB's analysis grossly underestimated the costs associated with implementing and complying annually with the final rule. For example, while the Bureau said it would only cost entities between $44,800 to $77,800 to build the systems and implement processes necessary for compliance with the rule, respondents to the ABA survey said one-time costs would range between $112,685 and $7,474,186.

Annual, ongoing compliance cost estimates also showed a significant difference, with the Bureau estimating that those costs would range between $8,349 to $278,618 based on the complexity of the institution, while the ABA survey found that those costs would be between $71,944 and $2,010,125.

The Bureau also failed to adequately assess the proportional impact on smaller banks. The costs associated with this regulation will not only hurt small business lenders, but it will also negatively affect the small businesses (not to mention their employees, customers and vendors) that rely on America's banks for capital access. In light of the results of this survey, the CFPB should not only reconsider the final rule, but also consider better ways to assess the true cost of regulations on affected entities in future rulemaking.

Analyzing the costs

[Link]

Figure 3: Comparison of one-time cost estimates (click to enlarge)

[Link]

Figure 4: Comparison of ongoing cost estimates (click to enlarge)

Survey respondents said complying with the final rule would result in one-time costs between $112,685 - $7,474,186 (Figure 3) and annual ongoing costs between $71,944 and $2,010,125 (Figure 4). Figures 5 and 6 show how each cost component contributes to the top-line costs. For one-time costs, updating computer systems is the most expensive component, which will cost respondents between $19,244 and $2,266,939 and makes up 34 percent of the total one-time cost of the regulation. For annual ongoing costs, transcribing data will be the most expensive factor, which will cost affected banks between $11,033 and $222,688 and makes up over 16 percent of total ongoing costs.

[Link]

Figure 5: One-time costs by subcomponent (click to enlarge)

[Link]

Figure 6: Ongoing costs by subcomponent (click to enlarge)

As expected, costs were proportionately higher for smaller banks. But for many cost components, the cost estimates for certain components were higher for banks with less than $500 million in assets (Figures 7 and 8). For example, the one-time cost for updating computer systems was significantly more expensive for banks with $500 million or less in assets ($74,236) than banks with between $500 million and $1 billion in assets ($19,244).

[Link]

Figure 7: One-time cost estimates by asset group (click to enlarge)

[Link]

Figure 8: Ongoing cost estimates by asset group (click to enlarge)

The main reason is that many of the smallest banks affected by the final rule are not subject to other federal data collection obligations under HMDA and the Community Reinvestment Act, and they will need to make significant loan origination system upgrades or replacements to comply. In addition to the steep learning curve for smaller banks, other reasons for this surprising difference include the fact that smaller banks lack the specialized expertise to make these changes. As a result, they must acquire more third-party assistance than larger banks, and they can expect to make a proportionately higher investment in technology than larger banks to handle Section 1071 compliance.

[Link]

Figure 9: One-time costs as a percentage of revenue (click to enlarge)

[Link]

Figure 10: Ongoing costs as a percentage of revenue (click to enlarge)

Figures 9 and 10 illustrate that, proportionately, the costs are significantly higher for the smallest size groups compared to the largest size group. As the figures illustrate, both one-time and ongoing costs would exceed 1 percent of total revenue for the smallest banks, which exceeds the significance threshold established by the Small Business Administration Office of Advocacy. If a rule is deemed to have a "significant" impact on small entities, the rule will make it more difficult for smaller entities to compete in a particular sector of the economy than large businesses.

Shifting to the overall costs to the market, the CFPB estimated those costs would result in a market impact on banks and credit unions of between $147,000,000 and $159,000,000. Using the average impact from the ABA survey, ABA estimates that the one-time costs are expected to have a $6.87 billion impact on the banking industry. While the CFPB's market estimate includes roughly 100 credit unions, the ABA's analysis only looks at covered banks.

A similar disparity is seen in the CFPB's estimate of the impact of the ongoing costs. The CFPB calculated the overall market impact from ongoing costs as between $297 million and $313 million for banks and credit unions together. The ABA's survey reflects an annual market impact of as much as $1.95 billion for banks alone.

Finally, the CFPB acknowledges that lenders will pass on costs to small businesses. The Bureau acknowledges that lenders will raise rates and fees for small business loans and other credit products. The fact that the costs will be passed on to small businesses does not absolve the CFPB of responsibility for the costs to lenders. Higher costs mean reduced access to credit for small businesses, especially the smallest. The result is contrary to what Congress intended when it enacted Section 1071. Yet, as a previous ABA analysis indicated, while the Bureau conceded that lenders will pass on costs to small businesses, the CFPB made no attempt to quantify the rule's likely impact on interest rates and fees. ABA's survey indicates that those impacts will be significant.

Conclusion

In light of the rule's impact on small businesses, the rule should be rescinded and the CFPB should, at a minimum, take account of the "real" cost and reduce the data points to the mandated 13. Additional avenues to reduce costs on vulnerable entities while still meeting statutory requirements include increasing the covered transaction threshold (from 100 to 500) or exempting loans to larger businesses from the reporting requirements. Legislative proposals to make these changes have been introduced both in the House and Senate.

As the survey results illustrate, the cost for banks to comply with the final 1071 rule will be far higher than the CFPB projected, and the burden will disproportionately fall on smaller lenders and small businesses. The CFPB failed to look at the proportional cost of the rule, and it also significantly underestimated the true cost of the rule on lenders of all sizes. The ABA survey reinforces what our members and other stakeholders have been saying for some time: The CFPB has misjudged the cost to comply with this rule, which will reduce the funds available for lending to the very small businesses Section 1071 intended to help. In addition, the significant compliance costs will disincentivize banks to lend to small businesses, which again seems at odds with Congress' objective. By degrading the economic viability of banks that specialize in lending to small businesses, the rule will have a significant adverse impact on the communities where these banks and small businesses are located.

Dan Brown is an economist and senior director on ABA's economic research team. Kathleen Ryan is SVP for fair and responsible banking at ABA.