2023 Student Loan Borrower Survey Supporting Statement B

Student Loan Borrower Survey Program

(OMB Control Number: 3170-XXXX

OMB Expiration Date: XX/XX/XXXX

Supporting Statement For

Student Loan Borrower Survey Program

(OMB CONTROL NUMBER: 3170-XXXX)

Part B

1. Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection method to be used. Data on the number of entities (e.g., establishments, State and local government units, households, or persons) in the universe covered by the collection and in the corresponding sample are to be provided in tabular form for the universe as a whole and for each of the strata in the proposed sample. Indicate expected response rates for the collection as a whole. If the collection has been conducted previously, include the actual response rate achieved during the last collection.

The Consumer Financial Protection Bureau (CFPB, Bureau) has acquired a nationally representative sample of de-identified consumer credit records (the “Consumer Credit Information Panel” or “CCIP”) from one of the three nationwide consumer reporting agencies. The basis for the sample is a 1-in-50 random sample of all credit records drawn from the consumer reporting agency’s archives. At the end of each month, the Bureau receives updated credit records for these sampled consumers (if available) and a 1-in-50 sample of credit records that were newly created since the previous quarter. This sampling process was designed to provide an ongoing panel of credit records that remains representative of the universe of credit records at each point in time. The contract with the consumer reporting agency also allows for CFPB testing to verify that the panel remains representative.

The de-identified credit records that the Bureau receives carefully exclude any direct identifying information to maintain the anonymity and protect the privacy of sampled customers. The records include information about the types of credit accounts, such as student loans and credit cards, which are included in each consumer’s credit record (though the identity of the creditor is excluded from the information the Bureau receives). The data also include information on non-credit-related debts that have been reported by third-party collection agencies, monetary-related public records (such as tax liens and bankruptcy filings), and details of any credit record inquiries made by lenders in response to an application for credit. The credit information in the CCIP is used to monitor conditions in consumer credit markets, to study consumer behavior regarding credit, to evaluate the effects of consumer regulations, or to address other issues in support of the Bureau’s research, monitoring, and supervisory missions.

The sample for this information collection will consist of one sample from the CCIP. The sample will consist of a new sample of consumers who have a student loan on their credit record in the Consumer Credit Information Panel. The sampling process may include sampling schemes other than simple random sampling, including stratification or oversampling populations of interest (such as consumers with a previously delinquent student loan and consumers of different age groups). The survey is anticipated to be sent to approximately 15,000 people. Based on historical response rates we conservatively expect approximately 3,000 total responses. Survey weights would reflect the sample design and would be adjusted to account for systematic differences in nonresponse along dimensions measured in the CCIP.

2. Describe the procedures for the collection of information.

As described above, we plan to survey a sample of individuals with credit records that are included in the CCIP. The persistent consumer record numbers for sampled credit records will be sent to the consumer reporting agency. The consumer reporting agency will identify the consumers associated with each sampled credit record and will mail the survey instrument which was designed by the Bureau staff. The consumer reporting agency’s subcontractor will receive the mail or online responses, remove any direct identifying personally identifiable information (PII) that may have been included in the consumer’s response, and send the de-identified data in electronic form to the CFPB. This process allows the Bureau to survey consumers without revealing direct identifying PII to the CFPB and has been successfully used for the National Survey of Mortgage Originations (NSMO), American Survey of Mortgage Borrowers (ASMB), the Survey of Consumer Views on Debt, and the Making Ends Meet survey.

The field period will be approximately eight weeks and include up to seven first-class mailings. We expect all sampled consumers will receive an initial mailing with a cover letter introducing the survey along with a paper questionnaire, postage-paid return envelope, and five-dollar cash incentive. We also expect the cover letter to include an offer of $20 cash to be received upon completion of the survey, and an additional $10 cash to be received if that completion occurs electronically. A reminder letter is sent one week later. In week five, consumers who have not yet completed the survey or opted out receive a second reminder letter with a replacement questionnaire, postage-paid return envelope, and a reminder of the offer of $20 to be received upon completion of the survey (along with the additional $10 to be received for an online completion). A final reminder letter is sent to remaining non-respondents in week seven. As noted below, we may adjust our incentive structure according to a review of best practices in the field and/or consultation with experts in survey methodology. The other three mailings consist of letters providing the respondents with their post-completion incentive payments in weeks 5, 9, and 14.

Note that the precise details of the incentive structure and mailing sequence may change as we approach the release of the survey and finalize elements of the study design.

3. Describe methods to maximize response rates and to deal with issues of non- response. The accuracy and reliability of information collected must be shown to be adequate for intended uses. For collections based on sampling a special justification must be provided for any collection that will not yield “reliable” data that can be generalized to the universe studied.

Obtaining sufficiently high response rates is a challenge for any survey. The Bureau intends to include a cash incentive with each survey to boost response rates. Additionally, the Bureau will incorporate lessons learned by the Bureau from a pilot conducted in 2017 for the Bureau's Making Ends Meet survey. The Bureau has also learned through the National Mortgage Database team, comprised of staff from the CFPB, Freddie Mac, and the Federal Housing Finance Agency, which administers the National Survey of Mortgage Originations and the American Survey of Mortgage Borrowers using a similar sampling methodology.

When the Bureau successfully piloted the survey in 2017, we tested whether a ten-dollar incentive would increase response rates to the survey enough to justify the additional cost.¹ We found that the ten-dollar incentive did not significantly increase response rates enough to justify the additional cost, and therefore, we plan to use a five-dollar incentive for the pre-completion incentive for our survey.

Meta-analyses of mail surveys find that incentives given initially with the questionnaire yield significantly higher response rates than do incentives contingent on return of the survey alone, or no incentives at all; furthermore, monetary incentives produce a stronger effect than non-monetary incentives. Additional post-completion incentives are also a recommended approach to increasing response rates without increasing costs. Many recurring federally funded surveys use monetary incentives, including the Survey of Consumer Finances, the Survey of Income and Program Participation, and the National Survey of Drug Use and Health, and self-administered surveys such as the Survey of Doctorate Recipients, the National Survey of Recent College Graduates, and the National Survey of Mortgage Borrowers. Incentives have consistently been found to improve response rates across a variety of survey topics and modes. Incentives have been found to be cost-effective in different modes and often reduce the effort required to contact and interview sample persons or the number of follow-up mailings.

More recently, we conducted additional primary research on improving survey response rates with different incentive designs. The incentive structure outlined in item (2) above proved most cost-effective and showed a 20 percent increase in response rates over pre-completion incentives alone, with a large majority of responses coming electronically. We welcome public comment on additional ways to improve the cost-effectiveness or design of our incentives and may change our incentive design after consultation with other survey experts, review of current best practices in survey administration, and direct experimentation.

In addition, the Bureau believes that response rates can be sufficiently maximized through careful design of survey instrument and clear communication to potential respondents about the survey’s purpose, use, and confidentiality protections. Finally, because we will have the de-identified credit records of both respondents and non-respondents to the survey, we anticipate using this information to model survey nonresponse and to adjust sample weights to reduce the likelihood that the results of any analysis are not biased by correlation between nonresponse and observable credit characteristics.

The extensive information from the de-identified credit records for both respondents and non-respondents will provide a strong basis for investigating potential nonresponse bias relative to the CCIP. The data will, for example, permit us to examine differential rates of nonresponse correlated with credit score, dollar amounts of various types of credit lines, and demographics. Based on this analysis, the Bureau will also construct survey weights so that the survey will be representative of American consumers in the target population for a given survey. In the previous surveys the CFPB has done using credit-record data as a sampling frame, the Bureau found that consumers who are older and have higher credit scores are slightly more likely to reply. Observable credit characteristics only weakly predict who will respond. The nonresponse bias analyses will be of benefit to other federal agencies using samples based on commercially available administrative data.

For survey respondents, the Bureau will additionally be able to compare the self-reported demographic information in the survey (for example, education, age, income, and marital status) to the auxiliary demographic information included in the credit-record database. This comparison may shed light on the reliability of such auxiliary data and, thus, provide information that may be valuable to government researchers and others that rely on such data when direct measures of these characteristics are unavailable. In other surveys the Bureau has done using credit-record data as a sampling frame, the Bureau found that the auxiliary data are predictive of survey respondents’ self-reported demographic information, but there are significant differences for some consumers.

4. Describe any tests of procedures or methods to be undertaken.

The CFPB plans to use pretesting and cognitive interviewing to test research instruments on a small scale prior to their use in full-scale surveys. This can include a debrief with respondents after the pilot to ensure that they understood the instrument, which in turn ensures that the resulting data collections are effective. These techniques are meant to reduce the total public burden of the information collection by ensuring that the full study information collection is optimized.

5. Provide the name and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze the information for the agency. The name and telephone numbers of these individuals will be provided in the clearance request for each specific data collection.

Thomas Conkling (202-450-9801)

Brian Bucks (202-365-7730)

Christa Gibbs (202-309-8830)

Yan Lau (202-451-1301)

Marie Rush (202-297-5514)

1 This research has been published in: Bucks, Brian, Mick Couper, and Scott Fulford, 2020, “A Mixed-Mode and Incentive Experiment Using Administrative Data,” Journal of Survey Statistics and Methodology, 8(2): 352-369.

Page 2 of 2