1559-NEW_supporting_statement_A_20160208

DEPARTMENT OF THE TREASURY

WASHINGTON, D.C. 20220

Supporting Statement

Preliminary Evaluation of the Community Development Financial Institutions Fund

Bank Enterprise Award Program

A. Justification

1. Circumstances necessitating collection of information

Pursuant to the Riegle Community Development Banking and Financial Institutions Act of 1994, as amended (the Act, 12 USC 4701 et seq.), the Bank Enterprise Award Program (BEA Program) complements the community development activities of insured depository institutions by providing financial incentives to expand investments in CDFIs; and to increase lending, investment, and service activities within economically distressed communities where at least 30 percent of residents have incomes less than the national poverty level and where the unemployment rate is at least 1.5 times the national unemployment rate.

This data collection is the first step towards a larger research effort to conduct an evaluation of the BEA Program. This will entail: 1) a detailed analysis of program’s history and evolution; 2) an analysis of the differences in the distribution and socioeconomic characteristics of BEA qualified census tracts as compared to Community Reinvestment Act (CRA) qualified census tracts; 3) an analysis of the characteristics of applicants and awardees, and the types and levels of activities undertaken by applicants and awardees; 4) a comparison of BEA applicants with peer banks and the communities they served using summary bank data and Home Mortgage Disclosure Act data to analyze variations across BEA and CRA qualified areas; 5) the online survey of all BEA applicants for the 2012-2014 period, 6) semi-structured interviews with subsets of applicants; and 7) an integrated summary of these findings about the program and recommendations for further quantitative designs for evaluation of the program.

Treasury will use the information from this current data collection for three purposes:

1. To gather opinions from Bank Applicants’ BEA and CRA investment managers about the functioning of the BEA Program.

2. To gather opinions of all BEA program applicants for 2012, 2013 and 2014 rounds on the relationship between the BEA Program and the Community Reinvestment Act (CRA) regulations.

3. To chronicle the BEA Program as it is currently configured and to better understand how the program is perceived to operate as an incentive for program applicants and awardees to invest and maintain activities in BEA qualified census tracts.

Each purpose is discussed in detail below in relation to each of the two data collections: 1) an online survey, and 2) the semi-structured interviews with a subset of respondents. The information collected utilizing these methods will assist CDFI Fund in assessing the opinions of the BEA Program applicants on bank investment patterns in the both BEA and CRA qualified distressed communities. The online survey is the focus of this OMB clearance request.

Purpose 1: Gather opinions from Bank Applicants’ BEA and CRA investment managers about the functioning of the BEA Program

Addressed by: Online Survey

The CDFI Fund is requesting clearance to conduct a one-time online survey of the population of FDIC-insured financial institutions that applied for Bank Enterprise Awards during calendar years 2012, 2013 and 2014 (including award recipients and unsuccessful applicants). The first purpose of this survey is to gather opinions from Bank Applicants’ BEA and CRA investment managers about the functioning of the BEA Program. The BEA Program was created to: 1) provide both CDFI and non-CDFI banks an incentive for direct investment in statutorily defined distressed communities, and 2) provide non-CDFI banks an incentive to invest in CDFIs working in those statutorily defined distressed communities. Because bank investment decisions are driven by a myriad of factors, from macro-economic conditions to civic pride to profit motives, this survey is not intended to establish causation or measure the impact of the BEA Program. This study can, however, document through administrative data on investments in BEA qualified census tracts reported by awardees and applicants that did not receive an award (because the investment was made prior to the decision as to whether the applicant qualified for an award), and, through the survey, gather opinions from the banks applying for awards on how the prospect of the BEA award may have affected their investment decisions.

Purpose 2: Explore concerns raised by the GAO on whether the BEA Program is complementary or duplicative of the Community Reinvestment Act (CRA) regulations in banks’ investment decisions in distressed communities

Addressed by: Secondary Data Analysis

The second purpose of the study is to explore concerns about of the BEA Program raised by the GAO that the program is essentially duplicative of the Community Reinvestment Act in providing an incentive for banks to invest in distressed communities. The current study will gather opinions from bank investment managers on whether the CRA and BEA accomplish the same objectives within the same target populations.

A second non-survey component of the current study will analyze the distressed communities using administrative data from the BEA applications linked to secondary data from the Bureau of the Census and other sources such as Home Mortgage Disclosure Act (HMDA) data to examine the variations across in BEA and CRA qualified census tracts receiving investments. It will document the economic characteristics of BEA-eligible census tracts and compare those characteristics with the same indicators for CRA-eligible census tracts to show the extent, if any, to which the two programs serve the same or different target populations. As such, this research design does not seek to establish that the BEA program induced the variations in investment patterns. Because this is secondary data analysis, and no information will be collected from the public, OMB clearance is not required for this phase of the study.

Purposes 1 and 2

Addressed by: Semi-structured Interviews

Semi-structured interviews will be conducted twice, once prior to the online survey, and again, with different respondents, after data collection for the online survey is completed. The pre-survey triangulation interviews (20 respondents in total) and post –survey complementary interviews (20 respondents total). The interviews conducted prior to the online survey will be the triangulation interviews, which will target four (4) distinct groups with no more than 2 to 8 respondents per group. The complementary interviews will be conducted after online survey data collection with 2 to 8 respondents in four (4) distinct categories and include four different questionnaires to complement the results from both the triangulation interviews and the online survey. The four distinct groups targeted for the interviews are; 1) Indirect Activity stakeholder institutions received a BEA Program Award and indirectly invest in distressed communities through CDFI-related activities; 2) Direct Activity stakeholder institutions directly invested in distressed communities by providing both Distressed Community Financing Activities and Service Activities; 3) Non-awardee stakeholder institutions that applied for the BEA Program Award in 2012, 2013 or 2014 and did not receive an award; and 4) CDFI Partner stakeholder organizations which are indirect beneficiaries of the BEA Program award through a partnership with an institution which received a BEA Program award.

The purpose of the complementary interviews is to collect contextual data to enrich interpretation of both the online survey results (Purpose 1) and the patterns of investment in BEA and CRA qualified distressed communities (Purpose 2). Specifically, the pre-survey triangulation interviews were designed to gather in-depth information on BEA Program award investments, and their investment patterns in distressed communities. Once the online survey data collection is completed, the post-survey complementary interviews will be recalibrated to provide further clarifying, contextual information on the survey responses.

Purpose 3: Assess Current Configuration of the BEA Program and Explore Future Evaluation Designs

Addressed by: Online Survey, Secondary Data Analysis, and In-Depth Interviews

The third purpose of this current data collection is to chronicle the BEA Program as it is currently configured and provide Treasury an analysis of the above findings to explore and inform future evaluation designs for the BEA Program. Treasury is interested in exploring the feasibility of comparing the investment patterns of banks that applied for and/or received BEA Program Awards with a constructed peer group of similarly sized and situated banks that did not apply for and/or receive BEA Program Awards. Summary FDIC branch bank data coupled with HMDA data can provide a preliminary, limited peer group analysis for the current study. However, more robust peer group analysis depends on getting approval for detailed transactional data from the Federal Financial Institutions Examination Council. Treasury has a pending request for such data with the three federal financial institutions regulatory agencies.

2. Method of collection and use of data

The CDFI Fund will collect data one time. Respondents will complete and submit the online survey utilizing a customized web-based instrument. The CDFI Fund will use the data collected from this survey to gather the opinions of the CRA and BEA Program investment managers on bank investment decision in BEA and CRA qualified distressed communities. The analysis of the BEA related investments in the distressed communities will use data from the BEA applications linked to secondary data from the Bureau of the Census and other sources to examine economic indicators in census tracts receiving investment. The CDFI Fund will share the findings with the Office of Management and Budget (OMB), the Government Accountability Office (GAO), Congress, industry stakeholders, senior management, and the general public. The research will also help to inform future research and evaluation designs of the program.

3. Use of Information Technology

The survey will be administered using a customized online survey instrument as the primary mode of data collection. Two weeks after sending an email containing a link to the online survey instrument, follow-up telephone calls will be attempted with non-respondents offering them the option of completing the survey by telephone. The contractor will use the online form to capture telephone survey responses.

4. Efforts to identify duplication

The BEA Program Evaluation does not duplicate information collected from other sources.

5. Impact on small entities

This collection of information is not expected to have significant impact on small entities.

6. Consequences of less frequent collection and obstacles to burden reduction

The BEA Program survey is a one-time data collection and data are not available from any other source. This information is critical to ongoing program assessment, including identifying appropriate output and outcome measures, and refining existing reporting requirements.

7. Circumstances requiring special information collection

The proposed data collection activities are consistent with the guidelines set forth in 5 CFR 1320.5 (d)(2). There are no special circumstances that require deviation from these guidelines.

8. Solicitation of comments on information collection

Comments on the Evaluation of the Community Development Financial Institutions Fund Bank Enterprise Award Program were solicited in the Federal Register on March 18, 2015 (Document 2015-06233). The CDFI Fund received two responses. The CDFI Fund received comments from the Community Development Bankers Association and one individual. Please see Appendix A for the comments that were received and an explanation as to how they were addressed in the application.

9. Provision of payment to respondents

No payments or gifts will be made to respondents.

10. Assurance of confidentiality

With respect to confidentiality of information supplied as part of the Evaluation of the Community Development Financial Institutions Fund Bank Enterprise Award Program, the CDFI Fund is subject to the protections afforded by the Freedom of Information Act (Freedom of Information Act (FOIA), 5 U.S.C. § 552) exemptions regarding confidential business information, the release of which is likely to cause substantial competitive harm, as well as personal privacy protections under FOIA. Additional protections may be applicable under the Privacy Act (Privacy Act of 1974, 5 U.S.C. § 552a). The data protocols will ensure participant anonymity and, therefore, confidentiality. Via the survey instrument, cover letter and email, potential respondents will be assured that the information provided will not be associated with their names, organizations, businesses or communities—i.e., that their responses will be combined with other responses and reported in aggregate form only. Further, procedures to assure confidentiality and data security will be in place for all data to be collected through the online survey.

11. Justification of sensitive questions

No questions of a sensitive nature are asked in the online survey. However some respondents might consider some information about their businesses or organizations to be proprietary. This will be addressed by implementation of the confidentiality procedures described above.

12. Estimate of the hour burden of information collection

The hour estimate is based on the 156 BEA Program applicants (including program awardees and unsuccessful applicants) that applied for an award in during calendar years 2012, 2013 or 2014, and the estimated number of hours the online survey should require. The estimated burden hours for completing the online survey (65.4) is guided by the results of a cognitive survey pretest that indicated an average of 35 minutes to complete the instrument, and a projected 70 percent response rate yielding 109 respondents. .

13. Estimate of total annual cost burden to respondents

It is not expected that this information collection will have a cost burden to respondents other than the hour burden described in item number 12. The hourly cost burden is estimated at $85.77, yielding a cost per response of $51.46 and a total annual cost burden of $5,609.36. No purchases of equipment or services will need to be made by respondents for the sole purpose of completing the online survey.

14. Estimate of annualized cost to the Government

The cost to the Government is the CDFI Fund staff and contractor time associated with background research, requirements gathering, research design, instrument design and pretest, systems development, data collection activities, analysis, and reporting.

15. Any program changes or adjustments

This submission is a new request for approval. Therefore there is no change in burden.

16. Plans for information tabulation and publication

The contractor will analyze the survey data and generate descriptive statistics related to the BEA Program. The survey data will be coded and analyzed to show response patterns and correlations among the responses. The coded survey data will be linked with institutional and service area data to show correlations between institutional or service area attributes and survey responses. The survey data will also be linked with institutional data from the applications and reports filed as part of the BEA Program. Throughout the analysis, the team will generate charts and graphs by institution size, which will serve as the framework for a final summary report.

17. Reasons for not displaying expiration date of OMB approval

The CDFI Fund will display the expiration date of the OMB approval on the online survey.

18. Explanation of exceptions to certification statement

Not applicable.

Collections of Information Employing Statistical Methods

See Supporting Statement Part B.

Date of Comment	Author Name	Author Position	Organization	File Name	Category	Comment	CDFI Fund Response
5/18/2015	Jeannine Jacokes	Chief Executive Officer	Community Development Bankers Association	20150518 – Jeanine Jacokes - Community Development Bankers Association	Utility	Concerned that “while such a survey will yield some broad data, the draft survey posted for public comment may be too general to capture how the program influences behavior.”	The data collection is part of a larger research effort to conduct an Initial BEA Program Evaluation, which involves: a detailed analysis of program’s history and evolution; a systematic analysis of the difference in the distribution and socioeconomic differences of BEA qualified census tracts as compared to Community Reinvestment Act (CRA) qualified census tracts; an analysis of the characteristics of applicants and awardees, and the types and levels of activities undertaken by applicants and awardees; a comparison of BEA applicants with peer banks and the communities they served including variations across BEA and CRA qualified areas; the online survey of all BEA applicants for the 2012-2014 period, and in-depth, semi-structured interviews with subsets of applicants; and an integrated summary of these findings about program and recommendations for research designs for further quantitative evaluation of the program. These are not being conducted under this collection. The semi-structured interviews by design will require nine or fewer respondents for each data collection instrument. Treasury will use a multi-strand design for the qualitative interviews, with half of the semi-structured interviews conducted prior to the survey and the second half conducted following the survey. Four unique instruments have been designed for Stage 1 of the interviews based on the stakeholder category. The number of interviews for each category ranges from two to eight. Following the survey, interview protocols will be designed to incorporate key findings from the survey, as well as any findings from the initial interviews. Again, there will be no instances where more than nine respondents will receive the same instrument, with interviews for each group ranging from three to seven. The attached document outlines the qualitative protocol design.
5/18/2015	Jeannine Jacokes	Chief Executive Officer	Community Development Bankers Association	20150518 – Jeanine Jacokes - Community Development Bankers Association	Burden	Commented that “some of the draft survey questions appear to be duplicative (e.g. Question #2 and #7 under both the CDFI Support Activities and Distressed Communities Activities).”	The CDFI Fund and the contract team have reviewed these questions and other questions and have concluded that there is no redundancy and that these questions are logically distinct and clear. Question 2 addresses targeting areas of greater economic distress, whereas question 7 pertains to increases in investment. Moreover, the contractor has conducted cognitive pretesting with 5 applicants, which validates that these questions are clear and logically distinct.
5/18/2015	Jeannine Jacokes	Chief Executive Officer	Community Development Bankers Association	20150518 – Jeanine Jacokes - Community Development Bankers Association	Clarity	Stated that “the draft survey places too heavy an emphasis on evaluating the influence of the Community Reinvestment Act (CRA) versus the BEA Program. In reality, CRA and BEA work in a highly complementary fashion, rather than compete.”	The survey seeks to address issues raised by the GAO in its report on the BEA program (GAO-06-824), which asserted that “… CRA provides banks with a regulatory incentive to undertake [BEA] award-eligible activities.”
5/18/2015	Joyce Dillard	Individual	NA	20150518 – Joyce Dillard	Other	Stated that “Census tracts are not being reflected with income defined representing current residents. We ask that census tracts be updated with the most current AMERICAN COMMUNITY SURVEY (ACS) data.”	The comment is not responsive to the survey questions but rather directed at the way the BEA program measures distress. Further, the comment fails to appreciate that while the five-year ACS data is updated annually, a completely new five-year estimate will only be available in 2016. Moreover, annual updates would introduce a great deal of complexity and cost in administering the program. Furthermore, analysis of the socioeconomic characteristics of BEA distressed communities indicates that the high poverty and unemployment rates persist over time. Finally, the commenter should note that the CDFI Fund recently updated BEA distress criteria with median family income and these data are available online (see: http://www.cdfifund.gov/what_we_do/acs/BEA-Eligibility.asp)

5