
EV borrowers showed lower late-payment risk but paid higher rates
Atlas Public Policy matched 379,000 Capital One auto loans to EPA fuel-economy data. EV borrowers had 50 per cent lower predicted serious late-payment risk than the average group, yet paid higher interest rates.
By EV News Desk
Ask an electric car owner where the money goes and they will talk about electricity, tyres and insurance. Almost nobody names the interest rate. New analysis by Atlas Public Policy, developed with the Natural Resources Defense Council, found a mismatch worth examining: EV borrowers had the lowest predicted risk of serious late payment, yet paid higher rates than comparable petrol-vehicle borrowers.
Abbie Olson and Rachael Nealer worked from 379,000 auto loans originated between January 2019 and December 2025 by Capital One, one of the largest lenders in the country with no manufacturer attached. The loans come from Form ABS-EE filings to the Securities and Exchange Commission, which report borrower credit and income, loan terms and the make, model and year of the car in a consistent format. The authors matched each vehicle to its EPA fuel economy figure, held credit and loan characteristics steady, and asked whether the car's efficiency predicts whether the loan gets paid.
A 50 per cent gap in predicted risk
It does. Measuring the likelihood of falling 90 or more days past due across four efficiency bands, against the 25–35 MPG average, the thirstiest group at 10–25 MPG was 16 per cent more likely to reach serious non-payment. The 35–60 MPG band, mostly hybrids and plug-in hybrids, came in 17 per cent below. The top band, above 60 MPGe and composed entirely of battery electric vehicles, was 50 per cent below.
Then the interest rates. Across the low, average and high fuel-economy bands, rates barely moved: 4.5, 4.6 and 4.5 per cent for all borrowers. The very-high band, made up entirely of EVs, paid 4.8 per cent. Below the dataset's median income of $88,500, EV borrowers paid 5.0 per cent against 4.6 to 4.7 per cent across the other bands. On a 72-month term, Atlas calculated more than $300 in extra financing costs for the average EV borrower and more than $350 for those below the median income.
The authors are careful about what they have not shown. One lender, skewed towards good credit — the average score in the dataset is 778 against a national 713 in 2025 — and lenders price on proprietary factors nobody outside sees. Their framing stays narrow: “Our findings suggest that fuel economy is a meaningful predictor of repayment risk, one that current loan pricing does not appear to account for.”
What the study cannot explain
Nothing in the analysis shows that the vehicle caused the difference in repayment. The authors say unmeasured factors may explain the pattern, and their model does not identify why higher fuel economy travels with lower predicted delinquency. Lower running costs are one plausible explanation, but the fact sheet does not test it. Its narrower point is useful to lenders: fuel economy predicted repayment outcomes even after the researchers accounted for the credit, income and loan characteristics available in the filings.
The debt the same buyers are already carrying
Three hundred dollars over six years sounds trivial next to the rest of car finance. Edmunds recorded 29.6 per cent of second-quarter trade-ins towards new vehicles in negative equity, the highest second-quarter share since 2020, average shortfall $6,884. Monthly payments on loans carrying an underwater trade-in hit a record $944, and buyers rolling that debt forward are projected to pay $16,270 in interest. “Consumers are incurring more debt than ever when trading in vehicles that are underwater,” said Jessica Caldwell, Edmunds’ head of insights. “Buyers who financed at 2022's peak prices are starting to come back to trade in, and they're bringing thousands of dollars in old debt with them.”
The comparison is not evidence that EV finance caused the wider negative-equity problem. It shows why small differences in loan pricing deserve attention when buyers are already stretching terms and carrying old debt into new contracts.
Some lenders already do this
The fact sheet points to credit unions that have priced fuel economy for years, some programmes dating to at least 2012. Verity Credit Union discounts half a percentage point for higher fuel economy vehicles, JetStream Federal Credit Union a quarter point, and Forrit Credit Union a quarter point for high fuel economy plus a separate full percentage point for hybrids and EVs. A whole point dwarfs the quarter-point penalty in the Capital One data.
The question matters as the used electric market grows. Cox Automotive put the average used-EV listing at $37,832 in July, down 1.2 per cent on June but 8.3 per cent higher than a year earlier, with days’ supply at 46. Atlas has not shown that lenders are mispricing every EV loan, and its single-lender dataset cannot do that. It has shown a consistent gap that deserves testing across more lenders: the borrowers with the lowest predicted serious late-payment risk did not receive the lowest rates.
- https://atlaspolicy.com/wp-content/uploads/2026/08/Auto-Loan-Fact-Sheet.pdf
- https://atlaspolicy.com/fact-sheet-a-link-between-fuel-economy-and-auto-loan-repayment-risk/
- https://www.usatoday.com/story/cars/shopping/evs/2026/08/27/ev-drivers-default-car-loans/91494843007/
- https://www.fi-magazine.com/news/negative-equity-trade-ins-reach-new-heights
- https://www.coxautoinc.com/insights/ev-market-monitor-july-2026/
More from the file

A 234,000-mile Lightning gets a used battery. The repair matters more than the mileage
Out of Spec Renew replaces a high-mileage Lightning battery after a module fault. Ford warranty terms and fleet data explain what this repair can tell owners.

Audi Offers Up to $10,000 Off Lease Buyouts for Its EVs
A 1 September dealer bulletin cuts thousands from contractual buyout prices for Audi EV lessees and gives dealers $500 per completed deal. The discount is real, but lessees still need to compare it with market value.

EVs cost 33% less to drive in Europe, unless you have nowhere to park
The ICCT's EV Transition Check 2026 has a headline number and a footnote. Drivers who charge at home save a third on energy. Drivers who rely entirely on public chargers save five per cent.