SPST 0203 2022 Statement A Small Business Lending Survey FINAL

SUPPORTING STATEMENT

Small Business Lending Survey

3064-0203

The Federal Deposit Insurance Corporation (FDIC) is requesting approval from the Office of Management and Budget (OMB) for the reinstatement of a previously approved and subsequently discontinued information collection related to the FDIC Small Business Lending Survey (SBLS). The SBLS is a nationally representative survey of FDIC-insured depository institutions (commercial banks and savings institutions, henceforth referred to as “banks”) designed to produce a description of banks’ small business lending practices and volumes that is otherwise not available. An initial SBLS was collected in 2016 (SBLS 2016) and the proposed information collection is planned for 2022 (SBLS 2022).

The proposed collection will provide updated and more in-depth information on banks’ small business loan processes and underwriting, markets and competition, and volume of small business originations, compared to that collected in SBLS 2016. In addition, SBLS 2022 asks new questions related to bank small business lending on the use of financial technology, response to the coronavirus pandemic, and use of automation in their small business lending programs. The information included in SBLS 2022 is complementary to information currently captured in other regulatory reports and surveys.

The survey is sponsored by the FDIC and will be administered by the U.S. Census Bureau (Census).

A. JUSTIFICATION

1. Circumstances that make the collection necessary:

Small businesses are an important component of the U.S. economy. According to the Small Business Administration (SBA), small firms accounted for almost half of private-sector employment and over 65 percent of net new jobs between 2000 and 2019.¹ Many small businesses have little or no direct access to capital markets and are thus reliant on bank financing, both for operating expenses and for investment. For banks, small business lending is an important way that they help meet their communities’ needs, especially for the many banks that primarily focus on commercial rather than consumer lending.

Despite the importance of small businesses to the U.S. economy and the importance of bank lending to small businesses, there is little high quality data on small business lending by banks. In the Consolidated Report of Condition and Income (referred to henceforth as the Call Report), banks report their commercial lending in total and separately for three loan size categories ($100,000 or less, greater than $100,000 to $250,000, and greater than $250,000 to $1 million). Researchers and policy makers often use the total amount of outstanding commercial loans of $1 million or less from the Call Report as a loan-size proxy for bank small business lending due to its availability and a lack of alternatives. However, loans with origination amounts of more than $1 million are often also extended to small businesses. For example, the maximum loan amount for an SBA 7(a) loan is currently $5 million but the Call Report data do not separately break out loans with origination amounts of greater than $1 million. In addition, the Call Report data contain no information on borrower firm size.

The FDIC’s SBLS 2016 provided information to help quantify the extent of misstatement of small business lending when using the loan-size proxy. SBLS 2016 asked banks that had data systems that can capture borrower gross annual revenue (GAR) to split their fourth quarter 2015 reported Call Report C&I lending by borrower GAR. Using the survey responses of banks with $1 billion to $10 billion in assets, and assuming that businesses with a GAR of $10 million or less are small, the FDIC documented that less than half (48.0 percent) of C&I loan dollars outstanding to small businesses are from loans originated for $1 million or less.²

The Government Accountability Office (GAO) concluded in a 2018 report that using the Call Report proxy to measure small business lending limits “regulators’ and policymakers’ ability to assess actual changes in banks' small business lending, including any effect of regulation,” relying in part on the findings of SBLS 2016.³ As a result, the GAO recommended that the Board of Governors of the Federal Reserve System, the Office of the Comptroller of the Currency, and the FDIC reevaluate, and modify as needed, the requirements for the data banks report in the Call Report to better reflect lending to small businesses. In response, these agencies published a request for comment in 2019 seeking feedback on ways to improve the usefulness of the Call Report data on loans to small businesses.

The proposed SBLS 2022 seeks to collect additional information on how to improve Call Report data on small business lending. The proposed collection asks basic questions about banks’ ability to easily retrieve information on small business loans by business G AR, whether they use GAR or loan exposures to define small businesses, and if so, what the thresholds are. SBLS 2022 also asks similar questions about banks’ lending volumes for business purposes as in 2016, but by finer gradations for both business size and loan size and by both loans outstanding and loan originations. This combined information will help policy makers understand how best to collect accurate information on small business lending while exerting the least burden on banks.

The proposed SBLS 2022 will provide other important data to complement existing sources of data on small businesses lending. Similar to SBLS 2016, SBLS 2022 will document the current types of credit offered, information banks use to underwrite loans, the market area for small business loans, competition for small business lending, and the practices used to conduct small business lending. The information gleaned from repeating these topics will help policy makers understand the extent to which ongoing consolidation in the banking industry—in 2016 there were a little over 6,000 banks in the U.S. and currently there are less than 5,000—may have affected how banks conduct their small business lending and small businesses’ access to credit.

The proposed collection will also provide new information on banks’ current or planned use of financial technology, whether and how banks use automated lending, and the effects of the coronavirus pandemic with respect to their small business loan programs.

2. Use of information collected:

The FDIC will use the information collected from the survey to produce a report of main findings, scheduled to be published in 2024, and will share the findings with the industry, policy makers, and academics. We plan to explore whether there are differences in bank practices for banks of different sizes and for banks with different business models, and to document changes over time in responses to questions that were retained from SBLS 2016. The FDIC may also use the collected data to produce more detailed research studies that focus on specific topics.

3. Consideration of the use of improved information technology:

The Census Bureau, with which the FDIC has contracted to collect the survey data, will collect this data using a web interface that will be user-tested to ensure comprehensibility and ease-of-use. In addition, the survey is organized into distinct sections that can be accessed independently and answered by different bank personnel, reducing the need to potentially coordinate among several respondents within a bank. To help ensure that the appropriate bank personnel respond to survey questions to which they have subject-matter expertise, we provide suggested bank personnel for each section.

4. Efforts to identify duplication:

There is no other survey or report that collects as comprehensive a set of information of small business lending by banks. There are other information collections which have questions that duplicate a small number of those from the proposed survey, but these collections are either not nationally representative or have a narrow focus that limits their value as a replacement for the proposed survey.

For example, the Board collects a report of small business lending in connection with its measurement of monetary aggregates, bank credit, and the flow of funds. The information on small business lending is reported only in aggregate and is intended to produce a picture of current credit conditions and identify changes that may indicate potential problems in this important economic sector. In contrast, the SBLS seeks to provide a more comprehensive view of how small business lending is conducted by banks and how differing approaches relate to the volume and measurement of small business loans.

In early 2020 FDIC staff approached staff from the Board of Governors of the Federal Reserve System, the Federal Reserve Bank of Kansas City, and the Federal Reserve Bank of New York who are currently or have previously worked on other surveys related to small business lending. The purpose of this outreach was to ensure that the questions in SBLS 2022 are complementary to, and not duplicative of, existing surveys. In each case, the staff of the other organizations confirmed that the information in the SBLS is unique and encouraged a second SBLS collection since the proposed collection helps make sense of the findings in their own contemporaneous surveys.

FDIC staff also discussed the proposed survey or solicited expert feedback from staff at other institutions who are not currently working on related surveys, but have an interest in the subject. In particular, FDIC staff are in regular engagement with staff from the Consumer Financial Protection Bureau (CFPB), the agency mandated under section 1071 of the Dodd-Frank Act to implement an annual data collection of small business lending. The information collected under the proposed survey is expected to inform the development of the CFPB’s efforts. The CFPB expects to produce a notice of proposed rulemaking in September 2021; there is currently no specific timeline for a final rule. We also engaged with staff from the Small Business Administration who expressed interest in the results from a second collection of the SBLS.

5. Methods used to minimize burden if the collection has a significant impact on a substantial number of small entities:

The FDIC has made a concerted effort to streamline the survey and reduce the burden associated with providing responses. This effort included conducting three rounds of cognitive testing with 46 banks of different sizes headquartered in 23 states and serving different types of market areas, with the aim of ensuring that the survey will capture useful information while minimizing response burden. In response to feedback from the cognitive testing, the FDIC has included in the survey only questions that rely on expert knowledge and do not require the gathering of data, or questions that require only data that can be provided from core data systems or from existing internal reports. Additionally, the FDIC has reduced the number of questions that will be answered by small banks, specifically those with less than $1 billion in assets.

5. Consequences to the Federal program if the collection were conducted less frequently:

This information collection is conducted periodically. The FDIC is proposing a second collection of the SBLS. The last collection occurred in 2016 and the current collection would take place in 2022. In the SBLS, questions may differ from collection-to-collection in order to give special attention to questions of most interest at the time.

5. Special circumstances necessitating collection inconsistent with 5 CFR Part 1320.5(d)(2):

None. The information is collected in a manner consistent with 5 CFR Part 1320.5(d)(2).

8. Efforts to consult with persons outside the agency:

A notice seeking public comment for a 60-day period on the reinstatement of this information collection was published in the Federal Register on April 21, 2021 (86 FR 20697). No comments were received.

9. Payment or gift to respondents:

No payments or gifts will be given to respondents.

9. Any assurance of confidentiality:

The Census Bureau will maintain the confidentiality of the data and information that are the subject of this study in accordance with applicable law, including but not limited to the Privacy Act. Although Title 13 U.S. Code Section 9 does not apply to the work, information, and data that are subject to this study, Census will, nonetheless, hold all data and information obtained in this survey up to the strict standards of Title 13 U.S. Code Section 9. The Census will take all necessary steps to protect the privacy of the respondents and commercial and proprietary information to the extent allowed by law.

9. Justification for questions of a sensitive nature:

The study does not request any information of a sensitive nature.

12. Estimate of hour burden including annualized hourly costs:

Estimated Annual Burden Hours:

Summary of Estimated Annual Burden (OMB No. 3064-0203)
IC Description	Type of Burden (Obligation to Respond)	Frequency of Response	Number of Respondents	Number of Responses / Respondent	Hours per Response	Annual Burden (Hours)
Respondents with Gross Annual Revenue Less than $1 Billion	Reporting (Voluntary)	One time	1,152	1	3.5	4,032
Respondents with Gross Annual Revenue $1 Billion or More	Reporting (Voluntary)	One time	848	1	6.5	5,512
Total Estimated Annual Burden Hours:						9,544

Source: FDIC.

Estimated Annual Cost of Burden Hours:

Estimated Category of Personnel Responsible for Complying with the PRA Burden	Estimated hourly Compensation	Estimated weights
		Banks with less than $1 billion in assets	Banks with $1 billion or more in assets
Executives and Managers4	$123.30	90%	50%
IT Specialists5	$92.30	10 %	50%
Weighted hourly compensation		$120.20	$107.8
Estimated hours		4,032	5,512
Estimated cost of burden hours		$484,646	$594,194
Total estimated annual cost of burden hours		$1,078,840

Source: Bureau of Labor Statistics: “National Industry-Specific Occupational Employment and Wage Estimates: Industry: Credit Intermediation and Related Activities (5221 And 5223 only)” (May 2019), Employer Cost of Employee Compensation (December 2020), Consumer Price Index (December 2020).

Note: The 75th percentile wage information reported by the BLS in the Specific Occupational Employment and Wage Estimates does not include health benefits and other non-monetary benefits. According to the December 2020 Employer Cost of Employee Compensation data compensation rates for health and other benefits are 34 percent of total compensation. Additionally, the wage has been adjusted for inflation according BLS data on the Consumer Price Index for Urban Consumers (CPI-U) so that it is contemporaneous with the non-wage compensation statistic. The inflation rate was 1.71 percent between May 2019 and December 2020.

Total Estimated Annual Labor Cost: $1,078,840

13. Estimate of start-up cost to respondents:

There are no anticipated capital, start-up, or operating costs.

14. Estimates of annualized cost to the federal government

The estimated annual cost to the FDIC is approximately $346,038.

14. Analysis of change in burden:

The estimated annual burden has increased by 2,044 hours from 7,500 hours in the SBLS 2016 to an estimated 9,544 hours for the proposed SBLS 2022. The increase is due to a redistribution of the expected type of respondent with a shift of 348 respondents from the “Respondents with Gross Annual Revenue Less than $1 Billion” category to the “Respondents with Gross Annual Revenue of $1 Billion or More” category, which has a higher estimated time per response. In addition, the FDIC has increased its estimated time per response for both respondent categories as shown below:

	SBLS 2016			SBLS 2022
IC Description	No. of Respondents	Time per Response (Hours)	Estimated Annual Burden	No. of Respondents	Time per Response (Hours)	Estimated Annual Burden
Respondents with Gross Annual Revenue Less than $1 Billion	1,500	3	4,500	1,152	3.5	4,032
Respondents with Gross Annual Revenue $1 Billion or More	500	6	3,000	848	6.5	5,512
	2,000		7,500	2,000		9,544

14. Information regarding collections whose results are planned to be published for statistical use:

The FDIC will use the information collected from the survey to produce a report of main findings, scheduled to be published in 2024, and will share the findings with the industry, policy makers, and academics. We plan to explore whether there are differences in bank practices for banks of different sizes and for banks with different business models, and to document changes over time in responses to questions that were retained from SBLS 2016.

17. Exceptions to expiration date display

None.

18. Exceptions to certification

None.

ATTACHMENTS

A1. FDIC Small Business Lending Survey

A2. Advance Letter (SBLS-A)

1 SBA, “Frequently Asked Questions About Small Businesses,” https://cdn.advocacy.sba.gov/wp-content/uploads/2020/11/05122043/Small-Business-FAQ-2020.pdf, accessed June 9, 2021.

2 For more findings from SBLS 2016 on the measurement of small business lending using the Call Report proxy, see

FDIC, “Small Business Lending Survey,” 2018, https://www.fdic.gov/bank/historical/sbls/index.html, and Jacob Goldston and Yan Y. Lee, “Measurement of Small Business Lending Using Call Reports: Further Insights From the Small Business Lending Survey,” (Staff Study, no. 2020-04, Federal Deposit Insurance Corporation, July 2020), https://www.fdic.gov/analysis/cfr/staff-studies/2020-04.pdf.

3 GAO, “Community Banks: Effect of Regulations on Small Business Lending and Institutions Appears Modest, but Lending Data Could Be Improved,” https://www.gao.gov/assets/700/694255.pdf, accessed July 16, 2021.

4 Occupation (SOC Code): Management Occupations (110000).

5 Occupation (SOC Code): Computer and Mathematical Occupations (150000).