Loan Pricing Model: Key Considerations for AutoLenders

When a borrower comes to an auto lender for a loan, the amount they are approved to borrow is only part of the decision. The lender also has to determine what rate makes sense for that particular loan. That means weighing the borrower’s risk and the deal structure against funding costs, expected losses, return requirements, and what comparable loans are priced at in the market.

A loan pricing model helps lenders weigh these factors consistently. This guide covers the key considerations for building and managing an auto loan pricing model and how to apply them in practice.

Auto Loan Pricing Model Considerations at a Glance

Once a lender has decided to approve a loan, the next question is what that loan needs to earn. The answer depends on more than the borrower’s credit. Lenders also need to consider their own costs, competing rates, pricing rules, and whether similar loans are performing as expected.

Key Consideration	What to Evaluate	Metrics to Track	Best Practice	Warning Signs
Individual loan risk	How borrower and deal characteristics affect expected risk	Credit score, LTV, term, vehicle type, borrower equity	Compare losses and recoveries across risk segments and use those relationships in pricing	A risk segment performs differently than the assumptions used to price it
Full cost of the loan	Whether the proposed rate covers costs, expected losses, and the required return	Funding cost, operating expense, expected credit loss, capital cost, required return	Put inputs on a comparable basis and calculate whether the proposed rate meets return requirements	The proposed rate no longer covers expected costs and losses while meeting the required return
Market alignment	How the proposed rate compares with similar loans in the market	Comparable APRs, pricing overrides, size and reason for adjustments	Compare similar loans, keep benchmarks current, and set limits on pricing adjustments	Employees routinely adjust model-generated rates to win business

These considerations work together, but pricing starts with the loan in front of the lender and the risk it carries.

1. Price for the Risk of the Individual Loan

Two borrowers can qualify for the same loan amount without presenting the same level of risk. One borrower might have stronger credit and more equity in the vehicle, while another has a higher LTV or longer loan term. Those differences can affect both the likelihood of a loss and how much the lender may recover if the borrower defaults.

That is why credit score should be treated as a starting point rather than the entire risk assessment. The Federal Reserve’s 2026 auto credit-risk model, for example, considers origination credit score alongside origination LTV, loan age, and whether the vehicle is new or used. Its analysis finds that lower starting LTVs generally correspond with higher recovery rates. Used vehicles, less borrower equity, and lower origination credit scores are also associated with higher loss severity.

For auto lenders, the practical takeaway is to look at several characteristics together and then compare them with the performance of similar loans already in the portfolio.

Metrics to Track When Pricing Individual Loan Risk

Risk Metric	How to Measure It	What to Compare
Credit risk	Track delinquency and losses by credit-score tier	Compare each tier with its historical and modeled delinquency and loss rates
LTV	Group loans by origination LTV	Compare defaults, losses, and recoveries across LTV bands
Loan term	Group loans by term, such as 48, 60, 72, and 84 months	Compare delinquency and losses across terms
Vehicle type	Separate new and used loans	Compare losses and recoveries for new and used vehicles
Borrower equity	Group loans by positive, break-even, or negative equity	Compare recovery rates across equity positions

Next, use those comparisons to find differences that may matter for pricing. Suppose, for example, loans above 110% LTV produce a 4.2% net loss rate, while loans below 90% LTV produce a 2.6% net loss rate within the same credit tier. If that pattern holds across enough loans, the lender can investigate whether the higher-LTV group needs a different risk premium or pricing rule. These figures are illustrative, but the process shows how portfolio data can turn a risk factor into a pricing decision.

The same process should continue after introducing the pricing rule. As newer loans mature, compare their actual losses and recoveries with what the model expected. If a segment starts performing differently, the lender can revisit the assumption before using it to price more loans.

For this approach to work, the underlying data needs to be reliable. Credit scores, vehicle values, amount financed, and other inputs should be captured and defined consistently so that loans are being compared on the same basis. Lenders also need enough observations within a segment before treating a difference as a meaningful pricing signal.

2. Account for the Full Cost of the Loan

Once lenders understand the risk of a loan, they need to determine what that loan needs to earn. The interest collected has to cover more than the cost of obtaining the money to lend. Origination and servicing expenses, expected losses, the cost of committing capital, and the lender’s required return all factor into the calculation.

At a high level, lenders need to account for:

Funding cost + operating expense + expected credit loss + capital cost + required return

Each component answers a different question:

• Funding cost: What does it cost to get the money being lent? Use the current cost of funding new loans rather than relying only on the average cost of existing funding, which may reflect older rates. Also consider the lender’s funding mix and the expected loan duration.
• Operating expense: What will it cost to originate and manage the loan? Include relevant acquisition, origination, servicing, payment processing, and expected collection costs. Because these expenses can vary by product and channel, using one portfolio-wide average may hide important differences.
• Expected credit loss: How much does the lender expect to lose on loans with a similar risk profile? Use the performance of comparable loans based on factors such as credit tier, LTV, term, vehicle type, and vintage. Recent performance may deserve more weight when underwriting or economic conditions have changed.
• Capital cost: What does committing capital to the loan cost the institution? Use the lender’s applicable capital-allocation method and account for the economic cost of putting that capital into this loan rather than another use.
• Required return: What does the loan need to earn after those costs and expected losses? Use the same return measure the institution uses to evaluate risk-adjusted profitability rather than mixing measures such as Return on Assets (ROA), Return on Equity (ROE), return on allocated capital, and loan-level margin.

The important part is to put these inputs on a comparable basis. For example, a one-time origination cost and a lifetime expected credit loss cannot simply be added to an annual funding rate. The model needs to account for when each cost occurs and how long the lender expects the loan to remain outstanding.

Once the inputs are comparable, lenders can see whether the proposed rate is expected to cover the loan’s costs and losses while still producing the required return. If one input changes, the rate does not automatically need to increase. Instead, recalculate the loan’s economics and determine whether the proposed price still meets the lender’s requirements.

3. Keep Pricing Aligned With the Market

Once a lender determines what a loan needs to earn, the next question is whether that rate can compete in the market. The key is to compare it with similar loans rather than a broad industry average.

Experian’s Q1 2026 data shows why. Average new-auto APRs ranged from 4.55% for super-prime borrowers to 16.01% for deep-subprime borrowers, while used-auto APRs ranged from 6.30% to 21.77%. With differences that large, an average APR across all auto loans provides little context for pricing a specific borrower and deal.

Instead, lenders should compare their proposed rate with loans that have similar characteristics. Start with the credit tier and whether the vehicle is new or used, then narrow the comparison using factors such as term, LTV, and channel when reliable market data is available.

Once lenders have a useful comparison, they need rules for responding to it:

• Keep benchmarks current: Review comparable market rates regularly so pricing decisions are not based on outdated conditions.
• Set adjustment limits: Define how far employees can move the model-generated rate before additional approval is required.
• Protect the minimum return: A competitive adjustment should not automatically override the lender’s risk and return requirements.
• Track overrides: Record how often rates are changed, by how much, and why. Repeated discounts within the same loan segment can signal that the model’s pricing no longer matches the market.

This creates a feedback loop between the model and what happens at the point of sale. If a particular segment repeatedly requires large discounts to win business, lenders can investigate the cause and decide whether the model, adjustment limits, or pricing strategy needs to change.

Managing Auto Loan Pricing With the Right Platform

An effective loan pricing model gives auto lenders a consistent way to account for borrower risk, loan economics, market conditions, and internal pricing requirements. The work doesn’t end when a rate is set. Reviewing actual loan performance against the assumptions used at origination can help lenders identify where their pricing strategy may need to evolve.

defi SOLUTIONS supports auto lenders with technology designed to connect and manage processes across the lending lifecycle. Book a demo with defi to learn how its lendtech solutions can support your pricing and origination strategy.

Frequently Asked Questions About Loan Pricing Models

What is the difference between a loan pricing model and an underwriting model?

An underwriting model helps determine whether a borrower and loan meet the lender’s credit requirements. A loan pricing model determines the price and terms needed for an approved loan to meet the lender’s risk and economic requirements. The two can use some of the same borrower and deal data but serve different purposes.

Can two borrowers with the same credit score receive different auto loan rates?

Yes. Credit score is only one factor that may inform auto loan pricing. Differences in LTV, loan term, vehicle characteristics, amount financed, and other permitted factors can result in different pricing even when two borrowers have the same credit score.

What is a pricing exception in auto lending?

A pricing exception occurs when an authorized user changes the rate or other pricing terms produced under the lender’s standard rules. Lenders can document the reason, size, and approval of each exception so they can later evaluate whether exceptions are producing acceptable results.

Should new and used auto loans use the same pricing model?

They can use the same overall pricing framework, but the assumptions do not necessarily need to be identical. New and used vehicles can differ in depreciation, collateral value, borrower characteristics, recovery experience, and market rates, so lenders may need separate assumptions or pricing rules within the model.