Borrowers with low credit scores are falling behind on car payments at a pace not seen in more than three decades, and vehicle repossessions have climbed past 1.73 million. The headline numbers paint a picture of widening stress at the bottom of the auto lending market, but Federal Reserve researchers point to a less obvious force behind the spike: a pandemic-era reshuffling of credit scores that is distorting how “subprime” gets measured in the first place. That distinction matters for millions of households facing tighter lending terms and higher rates on their next car loan.
Why a 32-year high in subprime auto delinquencies demands closer reading
Measured subprime auto delinquency rates have surged past levels recorded at any point since the early 1990s. On the surface, that suggests a wave of borrower distress concentrated among lower-score consumers. But the Federal Reserve’s own research complicates that reading. A January 2024 FEDS Notes paper published by the Board of Governors examined how pandemic-era score changes reshaped the composition of the subprime pool itself. Government stimulus payments, student loan forbearance, and other relief programs temporarily lifted millions of credit scores. When those supports ended, scores drifted back down, and many borrowers who had briefly exited the subprime category re-entered it, often carrying loans originated when their scores were higher.
The practical result is a statistical illusion layered on top of real stress. The subprime bucket now contains a different mix of borrowers than it did before the pandemic, and comparing today’s delinquency rate to the rate from 2019 or 2005 is not an apples-to-apples exercise. Lenders and bond investors who rely on subprime benchmarks to price risk may tighten credit or raise interest rates faster than actual default patterns justify. That response hits hardest for lower-income households already squeezed by elevated used-car prices, higher insurance premiums, and the broader rise in living costs.
For policymakers and market participants, the nuance matters. If rising delinquencies are interpreted purely as a deterioration in borrower quality, regulators might push for stricter underwriting just as some families are relying on cars to reach jobs in a still-adjusting labor market. If, instead, part of the increase reflects a reshuffling of who counts as “subprime,” the appropriate response may lean more toward targeted relief, better disclosure of score dynamics, and closer monitoring of how lenders adjust their models.
Fed research and CCP data trace the delinquency surge through 2025 Q3
The strongest primary evidence comes from two FEDS Notes papers published by the Board of Governors of the Federal Reserve System. The first, dated January 12, 2024, focused on how credit score migration inflated measured subprime delinquency rates for both auto loans and credit cards. The second, released on November 24, 2025, used the New York Fed panel data to track post-pandemic delinquency dynamics through the third quarter of 2025. Together, these papers establish that auto delinquencies have been rising faster than credit card delinquencies in the Consumer Credit Panel/Equifax (CCP) data, though raw volumes still sit below the peaks recorded before the 2008 financial crisis.
The CCP dataset, drawn from Equifax credit files, allows researchers to follow individual borrowers over time rather than relying on aggregate snapshots. That granularity reveals how score migration works in practice. A borrower whose score rose above a common subprime cutoff during the pandemic might have taken out an auto loan that was priced as near-prime. As temporary supports faded, that same borrower’s score could slip back into subprime territory without any change in the original loan’s terms. When researchers later calculate delinquency rates for today’s subprime borrowers, that loan – and any missed payments on it – are counted in a different risk bucket than when it was originated.
Fed economists find that this reclassification effect is large enough to noticeably inflate measured subprime delinquency rates, particularly for auto loans where balances are relatively high and repayment schedules are long. At the same time, the CCP data confirm that there is genuine deterioration beneath the statistical noise. More borrowers are rolling from current status into 30- and 60-day delinquency, and the share of accounts transitioning from late payment to charge-off has edged higher as well. The pattern is most pronounced among younger borrowers and those with thinner credit files, groups that were more likely to see their scores swing during and after the pandemic.
What the spike means for borrowers, lenders, and policymakers
For households, the convergence of higher borrowing costs, elevated car prices, and tighter underwriting standards creates a difficult trade-off. Some borrowers who would have qualified for affordable used-car financing in 2018 now face steeper rates or outright denials, even if their underlying income and job stability have not materially changed. Others may turn to longer loan terms to keep monthly payments down, increasing the risk of ending up “upside down” on a vehicle that is worth less than the remaining balance.
Lenders, meanwhile, are recalibrating. Those that rely heavily on traditional score cutoffs to segment borrowers may be overestimating risk in parts of their portfolios, potentially leaving profitable business on the table. Others are experimenting with more nuanced models that incorporate income volatility, payment histories across multiple products, and alternative data. The Fed research suggests that paying closer attention to how borrowers move between score bands over time – rather than treating those bands as fixed categories – could improve both risk management and access to credit.
For regulators and policymakers, the challenge is to distinguish between cyclical stress that warrants macroprudential caution and measurement artifacts that call for clearer communication. The recent spike in subprime auto delinquencies is real enough to merit concern, especially for the households behind the statistics. But the underlying data also show that part of the surge reflects who is being counted as subprime, not just how those borrowers are behaving. Recognizing that distinction will be critical as officials weigh the health of the consumer sector, the resilience of auto lenders, and the broader implications for financial stability in the years ahead.



