SPST_A_unbanked_bank_ svy_9_30_revised

SUPPORTING STATEMENT A

National Survey on Banks’ Efforts to Serve

The Unbanked and Underbanked

OMB No. 3064-0158

INTRODUCTION

The FDIC is requesting approval from the Office of Management and Budget (OMB) to revise its data collection on the efforts of FDIC-insured depository institutions (commercial banks and savings institutions, referred to henceforth as “banks”) to meet the financial services needs of individuals who do not have an account at a bank or credit union (the “unbanked”) and of individuals who have a deposit account but also rely on alternative, non-bank financial service providers (such as check cashing firms or payday lenders) for transaction or credit services (the “underbanked”). The information collection effort consists of two related survey instruments: one a survey of insured depository institution headquarters offices regarding corporate business strategies for serving the unbanked and underbanked, including bank product and services available to these households; and the other designed to gather information from branches of FDIC-insured financial institutions on specific methods used to reach the underserved and specific products and services offered at each location.

The 2011 Survey of Banks’ Efforts to Serve the Unbanked and Underbanked (2011 survey) gathers information about products and services that help bring unbanked and underbanked consumers into federally insured institutions. The 2011 survey is the second collection regarding this effort. The first data collection was approved by OMB in 2008 and conducted that same year.

Main Differences between the 2008 and 2011 Survey instruments

There are two main differences between the 2011 and 2008 surveys. First, the 2011 Bank Survey effort will survey both bank branches and bank headquarters; the 2008 survey included only bank headquarters. Surveying bank branches will help the FDIC gain a better understanding of the products and services offered at the branch level and will allow for a more meaningful geographic analysis of product and service offerings. Two very similar survey instruments have been submitted for OMB approval. The headquarter survey contains all branch survey questions, plus additional questions on strategies and obstacles that would typically be relevant to only bank headquarters.

Second, the 2011 survey is streamlined and much shorter than the 2008 survey. The 2008 had an estimated completion time of 290 minutes, while the 2011 survey instruments are estimated to take an average of 30 minutes. The FDIC hopes that a shorter survey instrument will increase the survey response rate and reduce burden on respondents.

The 2011 survey has been shortened through the following types of revisions:

1. Removing or consolidating questions by prioritizing topics of interest.

2. Reducing the level of detail asked in the question

3. Minimizing the use of open ended questions and structuring questions as multiple choice to reduce the burden on the respondent.

For questions that ask about fees, an additional instruction specifying the format in which the number should be entered has been added. For example, ‘enter dollar amount in format $XX.XX).

A more detailed comparison of the 2011 survey instrument (headquarter survey) relative to the 2008 survey is attached as Appendix A. The detailed comparison follows the structure of the 2008 survey.

A. JUSTIFICATION

1. Circumstances and Need

This information collection is mandated by section 7 of the Federal Deposit Insurance Reform Conforming Amendments Act of 2005 (“Reform Act”) (Pub. L. 109-173) and compliments the FDIC’s other initiatives in the area of economic inclusion. The Reform Act requires the FDIC to conduct ongoing surveys “on efforts by insured depository institutions to bring those individuals and families who have rarely, if ever, held a checking account, a savings account or other type of transaction or check cashing account at an insured depository institution into the conventional finance system.” The Reform Act further requires that the FDIC consider the following factors in conducting the survey: (1) the extent to which insured depository institutions promote financial education and financial literacy outreach; (2) the financial education efforts that appear to be the most effective in bringing unbanked individuals and families into the conventional finance system; (3) the efforts of insured institutions to convert unbanked money order, wire transfer, and international remittance customers into conventional account holders; (4) the cultural, language and identification issues as well as transactions costs that appear to most prevent unbanked individuals from establishing accounts; and (5) an estimate of the size and worth of the unbanked market in the United States.

2. Use of Information Collected

The FDIC will use the information to produce a study on banks’ efforts to serve the unbanked and underbanked populations in the U.S. The study results will be included in biannual reports to Congress, as required by the Reform Act, and are intended to inform policy makers and guide future policy decisions with respect to the FDIC’s economic inclusion initiatives and efforts of banks to serve the unbanked and underbanked.

To permit broad representation of banks’ efforts to serve the unbanked and underbanked, the FDIC plans to gather information from approximately 480 headquarters offices and 1,300 branches of various size classes and geographic locations.

Participation in the information collection effort is voluntary for all banks. No information about any specific bank or bank customers collected in the survey questionnaire will be disclosed to the public.

3. Use of Technology to Reduce Burden

To minimize burden on respondents, the surveys will be web-based, with an estimated completion time of 30 minutes.

4. Efforts to Identify Duplication

The Survey of Bank’s Efforts to Serve the Unbanked and Underbanked is the first large-scale, national survey to collect information about banks’ efforts to serve the unbanked and underbanked populations. Moreover, the survey effort will reduce the amount of potentially duplicative information by utilizing bank asset data from call reports submitted by banks for December 31, 2010. In addition, information as to whether branch locations are in low- and moderate-income areas will be obtained from the address information gathered in the FDIC’s Summary of Deposits survey from each June 30^th.

5. Minimizing the Burden on Small Banks

To minimize burden on all respondent banks, the time estimated to complete the survey has been reduced significantly. The last survey required an average of 290 minutes to complete. For the current survey, the average completion time is expected to be 30 minutes. The survey effort will be web-based and will provide respondents with the option of saving their answers at any point, exiting the survey, and completing it at a later time.

5. Consequences of Less Frequent Collection

This information collection will be conducted once every two years to comply with the legislative mandate.

7. Special Circumstances

There are no special circumstances.

8. Consultation with Persons Outside the FDIC

The FDIC published a “first” Federal Register notice on December 23, 2010 (75 FR 80811. No comments were received in response.

Pursuant to the PRA, the FDIC published a second Federal Register notice on March 17, 2011 advising the public that the collection has been submitted to OMB for review.

One comment letter from the Center for Financial Services Innovation (CFSI) was received in response to the second Federal Register notice. The commenter was generally supportive of the survey and noted that the information collected will be very useful in raising awareness about the types of products that are relevant to the unbanked and underbanked populations. The commenter letter’s made a number of specific suggestions; these, along with , and the FDIC responses, are summarized below.

• Add metrics for understanding usage of financial products and services.

Although acknowledging the survey’s effectiveness in inquiring about banks’ efforts to serve unbanked and underbanked populations, the commenter suggested that the survey gather information about actual customer usage of products to better assess their impact on the underbanked population. As directed by Section 7 of the Federal Deposit Insurance Reform Conforming Amendments Act of 2005, this survey is designed to measure banks’ efforts to bring individuals and families into the conventional finance system. The survey questions were written in accordance with this mandate, to gather information specifically on the products and services that banks offer to unbanked and underbanked consumers, and the other activities they undertake to reach these populations. Adding detailed account-level questions and other questions regarding customer usage of specific products would lengthen and complicate the survey, and could reduce response rates.

In January 2009 and June 2011, the FDIC sponsored the National Survey of Unbanked and Underbanked Households, a supplement to the Current Population Survey. That survey collects data on consumer usage of bank and non-bank financial products.

• Guidance on usage questions.

The commenter specifically suggested that the FDIC ask respondents to report the percent of their checking, savings, and small-dollar loan customers according to census tracks or zip codes. However, i Insured financial institutions typically do not maintain this level of data on product usage by census track or zip code. Therefore, requesting such information, therefore, would likely be more burdensome than is desirable and counterproductive to efforts to shorten the survey response time. Nevertheless, the survey will allow the FDIC to provide some degree of geographic analysis based on branch location, and whether the branch is in an LMI or non-LMI census tract.

• Add ranges in answer choices.

The commenter also suggested that, throughout the survey, answer choices be expanded to include a range of possible responses, rather than a choice of one response or another, especially questions related to the characteristics of small-dollar loan products. Many questions throughout the survey, especially those asking about fees, allow respondents to write in a specific numerical value. Except for the amount of the loan, the questions about the characteristics of small, unsecured personal loans are Yes/No questions that will identify loan products that conform to the FDIC’s Small-Dollar Loan Template. (See http://www.fdic.gov/smalldollarloans/)

• Expand on priorities and obstacles in offering products and services to the underbanked.

For purposes of gaining a deeper understanding of the specific ways in which banks aim to serve the underbanked population, the commenter recommended adding a question about strategic priorities for serving the underbanked. The survey contains questions about retail strategies that banks use to make doing business with the bank more welcoming or convenient, and asks respondents to specify the effectiveness of each strategy in developing relationships with unbanked and underbanked consumers. These questions will provide detailed information on the ways in which banks aim to serve unbanked and underbanked individuals.

The survey also contains a question about the obstacles that banks face in serving unbanked and underbanked consumers. Based on the comment letter’s suggestion, the FDIC added an answer choice of “marketing products effectively” in order to understand whether banks find it difficult to market their products and services to underserved populations.

• Add questions about mobile banking and mobile payments.

Finally, the commenter suggested that the FDIC add specific survey questions to elicit information on mobile financial services offered by banks. At this time, mobile financial services have not been widely deployed, and are not generally targeted to underserved populations, so the FDIC does not feel it would be useful to add questions about these topics to the current survey. The FDIC will continue to monitor developments in these areas, and may consider adding questions about mobile financial services in future surveys.

The FDIC initiated a pilot test of the survey to obtain qualitative feedback from nine pilot institutions. Based on the feedback received, some technical edits were made to the surveys to increase clarity and level of understanding of instructions, definitions, and questions.

9. Payment or Gift to Respondents

No gifts or payments will be given to respondents.

10. Confidentiality

No issue of confidentiality is anticipated because the FDIC will not receive any information that could potentially identify a specific respondent bank. Contact information is gathered only to facilitate data collection efforts. The FDIC’s contractor will remove all individual respondent and bank identifying information from the dataset delivered to the FDIC. Upon satisfactory completion of the contract, the FDIC’s contractor will destroy all data collected as part of this survey effort. The FDIC or other bank regulators will not be able to link survey responses to individual institutions. The FDIC will release a report containing only aggregate results. Survey data will be released only to the extent permitted by the Freedom of Information Act (5 U.S.C. 552).

11. Information of a Sensitive Nature

The study will not request any personally identifiable information on individual bank customers, such as name, address, or account number.

12. Estimate of Annual Burden

The FDIC anticipates that the survey will take 30 minutes on average. Therefore, the burden estimates are as follows:

Headquarters Survey: 469 respondents x 0.5 hours = 234.5 hours.

Branch Survey: 1300 respondents x 0.5 hours = 650 hours.

Total burden: 234.5 hours + 650 hours = 884.5 hours

13. Capital, Start-up, and Operating Costs

None.

14. Estimates of Annualized Cost to the Federal Government

The estimated cost to the FDIC is $500,000.

15. Reason for Change in Burden

The burden decrease of 3396 hours is the result of a redesign of the survey to decrease the estimated completion time by an average of 260 minutes.

16. Publication

It is the intent of the FDIC to publish only general findings of the study, in terms of descriptive statistics presented in tables. In addition to total universe results, headquarters level data will also be presented by asset size category (tiers), while branch level data will be presented by region, tier and LMI/non-LMI classification. Examples of the types of tables that will appear in our report are given below.

Table 1. Banks' priority on providing financial services to unbanked and underbanked populations (HQ) (Percent of Column Total)

Table 2. Regulatory concerns as obstacles to providing financial services by asset size tier (HQ) (Percent of Row Total)

Table 3. Average minimum account balance to avoid low-balance fees for entry-level checking account (Branch) (Percent of Column Total)

Table 4. Average fees for auxiliary products offered to customers of most basic, entry-level accounts or non-customers by asset size tier (Branch) (Percents or Dollars)

Table entries will consist of valid statistical estimates of universe percents and means. These entries will be weighted estimates, derived as described in Section 2.2 of Part B of our supporting statement for PRA submission. Although specific estimates of standard errors will not be presented in the tables, a description will be included regarding the general magnitudes of the standard errors to provide the reader with an understanding of the precision of the estimates. Also, no differences in estimates between universe subgroups will be discussed in the report, unless the differences are significant at the 10% level of significance (and this criterion will be noted in the report).

Because the new survey will differ substantially from the original survey, there will only be a small number of estimates that are comparable between the two surveys. Furthermore, these estimates will only be for the HQ survey since the original survey did not include a branch survey component. Any difference between comparable estimates (whether percents or means) for the two HQ surveys will only be noted in the report if the difference is significant at the 10% level of significance. Based on the target precision levels for the two HQ surveys (to estimate a 50% characteristic to within ± 5% with 90% confidence), sample percents that are in the proximity of 50% will have to differ by at least 7 percentage points to indicate that, with 90% confidence, the two universe percents are different (since in such instances the 90% confidence interval for the difference in the universe percents will not include zero).

17. Exceptions to Expiration Date Display

None.

18. Exceptions to Certification

None.

Appendix A

Comparison of the 2011 survey instrument (headquarter survey)

relative to the 2008 survey

a. Instructions

In addition to a public burden statement, the 2011 headquarters survey as modified includes a notice about how the responses and bank identifying information will be handled.

The 2011 headquarters survey also includes questions requesting the bank’s consent to allow branches to participate in the branch survey (2011.S12 and 2011.Q44-45). The consent question also appears at start of the survey and also the end for respondents who didn’t answer or refused earlier. A follow up question about the reason for refusing is also included at the end of the survey. This last question should inform our efforts to seek bank participation (2011.Q45).

The 2011 survey includes screening questions to ensure that respondents are qualified to answer the survey (2011.S13), confirm whether a bank is a unit bank (i.e., banks with only a main office with no branches) (2011.S16), and confirm that the responding branch is a full-service branch (only full-service branches are part of the branch universe).

The 2011 survey also includes two questions that describe the type of customer the institution serves and whether it is an internet bank. Question S14 differentiates institutions that mostly target business customers from institutions that focus on retail consumers. Question S15 identifies internet banks. Both business and internet flags will be considered during the analysis of the data and might help explain differences in results.

b. Retail Bank Information

This information was excluded from the 2011 survey to reduce the burden on respondents.

c. Education and Outreach

• In the 2011 survey, this section revolves around identifying the type of financial education (Q35) or outreach (Q37) activities conducted by the bank and their effectiveness (Q36, Q38). These cover content that were included in a number of 2008 questions (2008.QIIB3,QIIC,QIID,QIIG,QIIH)

• From the 2008 survey experience we learned that banks do not typically distinguish unbanked from underbanked consumers in their education and outreach efforts. Therefore, the 2011 survey does not make this distinction when requesting information on efforts that target these populations.

• Questions regarding targeted marketing (2008.QIIF) and expansion of services to unbanked/underbanked (2008.QIIi) are included in the 2011 survey in the section on ‘Bank efforts to identify Unbanked/Underbanked population (2011.Q1,Q2,Q3A)’. The 2011 survey added a question about the most effective marketing channels used (2011.Q4). No questions are asked on the specific type of population targeted (e.g. working poor, urban, African Americans, etc.)(2008.QIIF2).

• Questions regarding challenges related to providing financial products and services to unbanked/underbanked (2008.QIIL, QIIM) are included in the 2011 survey in a section focused on challenges of serving un/underbanked (2011Q39-Q41). The list of possible challenges is more extensive than in 2008. For instance, it includes 'Marketing products effectively to the unbanked/underbanked' (Q39I) as a possible challenge, which was added in response to a public comment letter.

In addition to identifying the specific challenges, the 2011 survey also intends to quantify the extent to which these challenges are an obstacle in serving unbanked/underbanked (major obstacle, minor obstacle, etc,). The 2011 survey also drills down on the types of competition and regulatory challenges faced by financial institutions (2011Q40-41).

• The 2011 survey does not have specific questions on financial education materials that were included in 2008 (2008. QIIB, QIIB1, QIIB2) since it was unclear whether banks were reporting standard product brochures in this category. Instead, specific financial education materials can be captured as the “other” option of 2011.Q35-Q36.

• In order to have a more concise survey, no information is requested regarding how long the bank has performed the activities( 2008.QIIC1), about the specific topics addressed in financial education efforts (2008.QIIC2), or about research performed by the bank on unbanked/underbanked population (2008.QIIJ).

• Open ended question QII3 of the 2008 Survey (2008.QIIB3) was replaced with 2011Q32.

• No data is gathered on payroll cards (2008. QIIE)

Retail Branch Information

• The 2011 and 2008 surveys both gather information on the type of retail strategies used by banks to attract underserved consumers (2011.Q5-Q6). Specifically, 2011.Q5 consolidates and simplifies QIIIA-C of the 2008 survey. The retail strategy option of ‘Automated telephone banking (IVR)’ was added as a result of the survey cognitive testing.

• The 2011 survey asks an additional question regarding effectiveness of these retail strategies (2011.Q6).

• No information is requested on specific extended hours (2008.QIIIA) or specific foreign languages used in/at the branch (2008.QIIIB).

• Questions regarding the services offered by the bank (e.g., money order, check cashing, etc.)(2008.QIIID) are included in the ‘Auxiliary Service’ section of the 2011 survey (2011.Q30)

Services Provided to Non-Customers

• The 2008 questions about services offered by the banks (e.g., check cashing, money orders, remittances, etc.) (2008.QIVA, QIVD) are simplified in 2011.Q30. In the 2011 survey, Q30 explicitly asks if these services are offered to customers of the bank’s most basic or entry-level account and to non customers. Results of the pilot indicated that this clarification was needed given that product fees might vary by account type.

• In the 2011 survey check cashing questions (2011.Q30A) focus on payroll checks, as opposed to a variety of types of checks.

• Questions regarding consumer identification to cash a check (2008.IVC) were also simplified to focus only on non-US passport and other foreign ID (2011.Q32A-B).

• Question 2008.IVB,IVC1, IVC2, IVE were removed since they asked for more in depth information about products and services included in the survey. 2008QIVE can be indirectly captured in the 2011 survey through questions about challenges in serving unbanked (2011.Q39, Q41)

Account Opening & On boarding Process for New Customers

• In terms of ID requirements for opening an account, the 2011 survey (2011.Q25) focuses only on whether a bank accepts non-traditional forms of ID or tax ID (e.g., matricula consular, foreign passport, and ITIN) since other standard IDs will generally be accepted by banks. Thus, some of the options included in 2008 QVA are not included in the current survey. 2008 survey question QVB is also not included to keep the section short.

• Account screening questions are similar to the 2008 survey (compare to 2011.Q26-29). The current survey asks about the process of rejecting an application (2011.Q26 compared to 2008.QVE), and the most common reasons for rejecting an account (2011.Q28-29 compared to 2008.QVG).

• Questions about specific screening tools (2008.QVC-QVD, QVF) are not included since 2011.Q8-9 capture the extent to which these screening tools preclude account opening.

• Similar to the 2008 survey (2008.QVH), the 2011 survey includes a question on the availability of a stepping stone account if the application for a conventional account is denied (2011.Q27). In order to keep the survey short, no information is requested on the specific characteristics of this account (2008.QVH1). However, some banks may include the characteristics of this stepping stone account in the checking account section (2011. Q8-17).

Deposit Products

• In the 2011 survey, deposit products are divided into two sections: checking account and savings account. Basic fees and features of these accounts are captured in a more comprehensive and systematic way.

The 2008 survey asked only for the minimum balance requirement of a basic checking account (2088.QVIA). The 2011 survey asks for minimum balance requirements, maintenance fees, and the minimum amount required to open an account. (2011. Q8-Q10).

• For maintenance fees associated with checking and savings accounts. (2011.Q9A, Q9C, Q20, and Q21) the 2011 survey asks whether a maintenance fee is assessed and, if so, how often the maintenance fee is charged (2011.Q9A_1, Q9C_1, Q20_1, Q21_1). This change was made as a result of the pilot, which found that some institutions charge maintenance fees on a quarterly basis.

• Regarding payment features of the checking account, similar to the 2008 survey(2008.QVIB), the 2011 survey asks whether check writing is part of the basic account features (2011.Q12), and whether there is a fee involved (2011 Q13). Questions regarding other payment features and/or the cost associated these features have been removed (2008.Q.VIB).

• In addition to collecting non-sufficient funds fee data (2008.QVI.C1 compared to 2011.Q15), the 2011 survey also asks more generally about the overdraft fee associated with the most basic account (2011.Q14A) and whether the overdraft program is automated or not (2011.Q14B). By splitting the question into two, the FDIC will be able to collect both adhoc and automated overdraft fees.

• In addition, the 2011 survey asks about the availability of line of credit or linked account options (2011.Q16). This last item was covered by Q.VI.C2 in the 2008, but less explicitly.

• Questions regarding low balance alerts and threshold NSF items (2008QVI.C3) are not included to reduce survey length.

• A question regarding whether the most basic entry level checking account is card based is added to the 2011 survey (2011.Q11). Electronic accounts are typically less costly for banks to offer and can be a way to offer low-cost accounts to underserved consumers.

• A question on whether partnerships are used to promote checking account products was added (2011.Q7B).

Savings Accounts

• Similar to the checking account section of the 2011 survey, the savings account section has additional questions to obtain more compressive information regarding basic fees and requirements for savings account (2011.Q18-21).

• The types of savings accounts available to customers (2008.VIIA) are included in 2011.Q24, but no information about the interest rate is collected to reduce the burden on responding banks.

• Questions 22-23 of the 2011 survey regarding an automatic saving option and the associated fee are new.

• The 2008 question regarding partnering with organizations to promote savings (2008QVIIB) is captured by 2011.Q7B.

Payment Products

• Questions regarding availability of funds from a deposited check (2008.VIIIA) are part of the checking account section of the 2011 survey (2011.Q17).

• Advance loan questions are not included (2008.VIIIB) since they were often confused with overdraft programs.

Credit Products

• Similar to the 2008 survey, the 2011 survey asks about the extent to which banks offer small dollar loans. The way the 2008 small-dollar loan questions were asked (2008.QIXB), did not allow us to identify which specific loan features differ from the FDIC Small-Dollar Loan Template.

• Specifically, the 2011 asks about the amount of the smallest unsecured personal loan (2011.Q33). If the minimum amount is below $5,000 or if the bank has no minimum established, the survey asks which of the loan characteristics are consistent with the FDIC Small-Dollar Loan Template (2011.Q34): repayment period, APR, and underwriting process. This structure will allow the FDIC to identify which features are consistent with the template and which ones are not.

• Unlike the 2008 survey, the 2011 survey does not ask about eligibility criteria for small dollar loans or to provide specific fees and terms of the loan (2008.IXA1. IXB1). Also, other unsecured credit products (2008 QIXA), tax refund anticipation loans (2008.QIXC), and credit cards (2008.QIXD) are excluded from the survey instrument to keep the survey as short as possible.

Lastly, the 2011 survey has only two open-ended questions. Question 42 asks respondents to describe the type of program they think would be effective in bringing unbanked/underbanked consumers to the financial services. Question 45 asks about the reason for not participating in the branch survey.

- 14 -