• Auto lending has the highest first-party fraud (FPF) rate of any industry: 5.31% of applications in 1H26, more than twice the average.
• First-party fraud passes identity checks, gets booked as ordinary defaults, and its warning signs often sit outside any single lender's records.
• In private auto studies, SentiLink's FPF Score flagged 0.73% of funded loans, which accounted for ~14% of losses.
• Network-wide signals are the strongest risk signals of first-party fraud, especially frequent applications to many institutions.
• Lenders should use high scores to trigger closer review at application, when they can still act.

What is First-Party Fraud in Auto Lending?

In first-party fraud, the applicant uses their own name, date of birth, and Social Security number but otherwise misrepresents themselves by lying about their income, their credit history, or their intent to repay. In auto lending, first-party fraudsters may visit dealerships in person and drive a car away with no intention to make future payments, or use mules to obtain vehicles for them. Fraud rings can then use title-washing schemes to remove liens and resell the vehicles.

In our analysis of 170 million applications across industries in the first half of 2026, auto lending has the highest first-party fraud (FPF) rate of any industry we measure — 5.31%, more than twice the overall average. Several of our auto lending partners saw first-party fraud rates above 10% of applications over the same period. Rates also rise around holiday sales events, when dealerships and underwriters are busiest — we recorded a first-party fraud spike in late February, for example, coinciding with Presidents’ Day promotions and the arrival of tax refunds. It is also the only industry where we found first-party fraud outpacing identity theft: 5.31% against 4.51%.

Why auto lending attracts first-party fraud

Several features of auto lending make it especially susceptible to first-party fraud:

• A valuable payoff. A vehicle is a portable, resalable asset worth far more than many unsecured credit lines.
• Pressure to sell. Dealerships have a financial incentive to close sales, which can compete with scrutiny of suspicious applications.
• An in-person process. Applying in person makes stolen or synthetic identities harder to use, favoring schemes that rely on genuine identities.
• Organized support. Fraud rings recruit borrowers and help conceal vehicles, remove liens, and resell cars.

Common Fraud Schemes in Auto Lending

First-party fraud in auto lending can take many forms, depending on what the borrower misrepresents and how they exploit the loan process. Forms of first-party fraud in auto lending include:

Bust-outs: A borrower obtains a vehicle with no intention of repaying in full, then keeps or sells it without paying off the debt.

Loan stacking: A borrower obtains multiple auto loans before the new debts appear on credit reports — sometimes pledging the same car to multiple lenders.

Straw buying: A borrower recruits someone else to apply using that person’s name and credit, concealing their own role to obtain financing they might not qualify for themselves.

First-party synthetic fraud: A borrower uses their own name and birth date but a different Social Security number, concealing their credit history.

Why Conventional Controls Miss First-Party Fraud

• First-party fraud can slip through every checkpoint in the lending process, from identity verification and underwriting to collections and credit reporting. Here’s why:

• Intent cannot be measured. A borrower can afford to repay a loan—and still plan to bust out.
  Identity checks only confirm who is applying. Traditional fraud controls test whether an applicant’s name, date of birth, Social Security number, and other personal details are real and belong together. A first-party fraudster uses their own identity, so the check comes back clean.

   

• Underwriting is only as sound as its inputs. A first-party fraudster can fabricate documents, inflate income, or buy access to authorized-user tradelines to misrepresent their credit profile. Verifying each claim independently adds cost and delay.
  

• Collections is built for borrowers who fall behind, not fraudsters who default on purpose. The fraudulent borrower never intends to repay, meaning the lender wastes time and money on notices and payment plans for a borrower who never intended to pay, costs mount and the recovery window narrows.
  

• Credit reporting mislabels first-party fraud—or misses it altogether. Fraudsters either falsely claim identity theft, or the account joins the pile of ordinary defaults. Soft inquiries used for prequalification generally aren’t visible to other lenders, making it harder to spot a borrower seeking credit at several institutions in quick succession.

What the Warning Signs Look Like

No single data point proves first-party fraud in auto lending. In a recent study of 25 applications that later defaulted and led to lawsuits, we used our risk model to examine the factors that contributed with elevated first-party fraud scores. The strongest differences came from application activity across the lending network, including:

• A burst of recent applications: In the previous 90 days, these applicants had submitted 4.8 applications on average across our partner network, compared with 1.2 in the benchmark.
• More institutions approached: Over the previous year, they had applied to 5.8 institutions on average, versus 2.9. An individual lender might see only one piece of that activity.
• Less established address history: Their addresses had first appeared in our application data an average of 2.5 years earlier, versus five years for the benchmark. That measures the history of an address in our network, not how long someone has lived there.
• Phone and contact signals: Longer ties between an applicant and a phone number were reassuring in the benchmark. Our broader network also lets us examine evidence of disposable numbers and whether contact details change across applications.

Recommendations

For lenders, the challenge is to identify first-party fraud early without adding unnecessary friction to every application.

• Treat first-party fraud as a distinct risk. Identity checks ask whether applicants are who they claim to be. Underwriting assesses their ability to repay. First-party fraudsters can pass both checks, so lenders need controls designed to identify that risk.
  
  
• Look beyond your own records. The key predictors of fraud depended on data from outside an individual institution’s records. Lenders need a broader view of that activity to recognize patterns their own data cannot reveal.
  
  
• Use high scores to prompt closer looks. An elevated FPF score is not proof of fraud and should not be treated as one. It can identify applications that warrant added scrutiny including document verification or other steps. This allows lenders to apply friction selectively while speeding approvals for other applicants.
  
  
• Set thresholds using your own portfolio. A lender should test the model against its own applications and losses, and factor in its own costs and processes in order to determine optimal cutoff scores for manual review.

Applied effectively and paired with appropriate treatment strategies, this approach can help lenders approve more legitimate borrowers while reducing fraud losses.

Want to learn more?

• Reach out to learn how SentiLink can help reduce first-party fraud losses in your auto lending portfolio and approve more good borrowers.
• Read our white paper on the signals that predict first-party-fraud in auto lending.
• Review our latest Fraud Report to find out what’s happening with first-party fraud rates in auto lending and other industries in 2026.