Auto Loan Data for Research Insights

Accessing rich auto loan data obtained by the CFPB, this potential research aims to prioritize key questions for analysis. The dataset includes a diverse sample of financial institutions and extensive variables related to loan originations and servicing. Research questions span from specific inquiries to broader analyses, with the idea of studying the impact of high DTIs on borrowers being a focal point.

• Auto loans
• Research
• CFPB
• Data analysis
• Financial institutions

cassander Follow

Uploaded on Aug 22, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.

Presentation Transcript

1. Potential Research with New Auto Loan Data David Low| June 7, 2024

2. Disclaimer This presentation is being made by Consumer Financial Protection Bureau representatives on behalf of the Bureau. It does not constitute legal interpretation, guidance, or advice of the Consumer Financial Protection Bureau. Any opinions or views stated by the presenter are the presenter s own and may not represent the Bureau s views.

3. Introduction CFPB obtained auto loan data through 1022(c)(4) market- monitoring authority. This data is quite rich and could be used to study many important questions. Question: Given limited staffing resources, which research questions should OR prioritize? 3

4. Very Rich Data: Sample 3 banks, 3 captives, 3 finance companies Fairly representative cross-section of the market All loans originated or serviced 2018-2022 More than 30 million loans in total

5. Very Rich Data: Originations Lots of standard origination variables: Car price, down payment, loan amount, interest rate, monthly payment, borrower credit scores/income, etc. Lots of fairly unique ones, too: Direct vs. indirect Purchase vs. refi 11-digit VIN (make + model + trim + engine + factory) Dealer markup and add-ons

6. Very Rich Data: Servicing Dates of loan mods and account payoffs Lots of information about repos Dates assigned, completed, and redeemed Days past due, outstanding balances, recovery amounts Voluntary vs. involuntary Importantly, this is linked to originations information

7. Research questions? Some questions are (intentionally) narrow and/or quirky May be easier to answer well Other questions are larger / more ambitious May be harder to answer perfectly Range of options to publicize CFPB blog post, CFPB report, academic paper, etc.

8. Idea: High DTIs Anecdotally, some borrowers with limited income wind up with very high monthly payments What can we say about why? Effects on prepayment and repossession rates?

9. Idea: Inflation During COVID Car prices first collapsed, then surged during COVID. Why? Supply vs. demand? Great interest to economists broadly How did this affect consumers? Higher monthly payments more default? Subprime borrowers excluded from the market?

10. Idea: Auto Loan Term Lengths Previous CFPB blog posts have noted that auto loan term lengths have been growing since 2009, and continue to grow What can we say about why? Vehicle durability, underwriting, payment targeting, etc. What can we say about the effects? Affordability, delinquency, etc.

11. Idea: Direct vs. Indirect Loans Other datasets imply about 85% of auto loans are indirect, i.e. intermediated by dealers (though less in our data) Big contrast with other credit markets Lots of questions Why? What are the effects? How do direct vs indirect borrowers differ?

12. Idea: Auto Loan Refis Well-known that mortgage refis are rare, but auto loan refis are even rarer Can the same kinds of mechanisms explored in the mortgage refi literature also explain why auto refis are so rare? Huge rate variation in COVID good time period to use How the few refi loans differ from purchase loans?

13. Idea: Why is Auto Loan Prepayment so common? In the data, 30-40% of prime loans prepaid within two years What can we say about why? Trade-ins? Refis? Cash prepayments?

14. Idea: Drip Pricing in Auto Loans Junk fees and drip pricing top CFPB priority right now Anecdotally, lots in the auto market What are the implications? Cross-subsidization across consumer types? Or different consumers pay similar amounts in different ways? Repos, prepayment, etc.?

15. Idea: How are auto loans priced? 2021 CFPB Data Point showed that default risk explains some, but not all, variation Is this finding robust in newer and richer data? What else is it? Prepayment risk? ( we actually observe this!) Loss given default? ( we actually observe this!)

16. Idea: Why do High Credit Score Borrowers Get Loans from Finance Companies? Anecdotally, many used car dealerships assume all their customers are a significant credit risk and price accordingly. But 2021 CFPB Data Point provided evidence that many finance company borrowers could have gotten loans from a bank or credit union, and likely paid much lower rates. Does newer and richer data support this? Now we observe prepayment and repo rates What can we say about why? Geography and competition?

17. Idea: Why do Auto Borrowers Default? Lots of research on why mortgage borrowers default. Auto loans are both similar and different: Also secured loans, so equity plays a role Role for depreciation, maintenance, and insurance Repo process much faster than foreclosure Our data includes information on deficiency balances, consumer surplus, redemptions, voluntary repos, etc.

18. So what should OR do with the data? Given limited staffing resources, how can OR best make an impact with this data? Which research ideas to prioritize? Important research ideas I missed?