The Skinny–Mar 2026
Auto loan delinquencies remained elevated in the latest reporting set, which now extends through January 2026. For all accounts (300–850), 30–59 DPD was 4.10% and 60+ DPD was 2.65% in January, keeping both early-stage and late-stage performance high at the start of the new year.
Relative to pre-pandemic benchmarks, delinquency rates remain meaningfully higher: in January 2019, 30–59 DPD was 3.12% and 60+ DPD was 1.50%.
Looking ahead, delinquencies should improve meaningfully over the next few months as seasonal tax refunds arrive, with this year’s refund season expected to be stronger (higher average refunds).
Auto ABS Data
This section summarizes loan-level data from monthly U.S. auto ABS filings with the Securities and Exchange Commission. The dataset provides a comprehensive view of the securitized auto loan market:
Source: 15,500+ SEC ABS-EE monthly trust filings (2017–present)
Loans: ~27 million unique loans
Fields: 72 data elements per loan, covering borrower, loan, vehicle, and performance characteristics
Issuers: 24 issuing trusts, 22 currently active
Series: 446 issuance series, 240+ currently active
Data Methodology
Loan-level data is sourced directly from SEC ABS-EE filings and processed as follows:
Loan counts include only obligors with a valid numeric credit score. Loans with missing, blank, or non-numeric score values are excluded.
Credit score range is capped at 900 for issuers reporting FICO Auto Score (the applicable scale for auto lending) and 850 for issuers reporting classic FICO. Scores outside the applicable range are excluded.
Commercial loans are identified and excluded on a best-efforts basis using the obligor credit score type field reported by each issuer. Where issuers commingle consumer and commercial loans within a single trust, only consumer-scored obligors are counted.
Residual balances below $10 are excluded from loan counts to eliminate administrative close-out artifacts that do not represent economically active loans.
Duplicate filings are resolved using row-level deduplication within each reporting period to ensure each loan is counted once.
All figures reflect the reporting period end date as filed. Where an issuer's reporting date falls within a calendar month but not on the last calendar day, the data is attributed to that month.