auto loan risk assessment

AUTO LOAN RISK MANAGEMENT FOR YOUR PORTFOLIO

The defi Team auto loan origination software, defi INSIGHT

auto loan risk management

With auto sales in decline in 2018, lenders are looking for options to build a profitable portfolio. There’s always the temptation to assume greater loan volume and risk. Lenders have become more aggressive in pricing new originations, offering longer loan terms to help borrowers with student or household debt, and increasing the volume of subprime loans to build business.

To successfully balance risk with profitability requires both art and science. Lenders with years of experience may have mastered the art by developing an innate ability to fine-tune credit policies in response to market conditions. Lenders new to the auto market should employ recent advances in fintech capabilities to guide auto loan risk management.

We’ve identified three areas of the lending cycle—applications, decisioning, and servicing—where the use of advanced fintech capabilities can positively impact auto loan risk management.

  1. Applications: eliminate risk at the outset to avoid the future burden of delinquency or default.
  2. Decisioning: enable better quality lending decisions that minimize the risk of delinquencies and defaults.
  3. Servicing: regularly analyze portfolio performance to proactively identify risk factors and adjust credit policies and practices accordingly.

A lending cycle approach to auto loan risk management lets you use technology-powered tactics to identify and reduce or eliminate risk as early as possible.

3 Tactics for Auto Loan Risk Management

Identify Fraudulent Applications

Auto thefts are down, thanks to ever-more sophisticated anti-theft technology installed in vehicles. The preferred method of auto thievery is now accomplished via fraud. Key methods of fraud committed on a loan application include:

  • Income misrepresentation—as many as 20% of applications are not accurate;
  • Employment misrepresentation—google ”fake pay stubs” to see how easy this can be;
  • Synthetic identity—a blend of disparate and fictitious data, including Social Security numbers, names, and addresses used to create a fake identity;
  • False collateral—utility bills, “proof” of identification or residency; and  
  • Straw borrower—an individual whose name, social security number, and credit history hide the identity of fraudsters.

Based on extensive analysis of millions of loan applications, machine learning can now be applied to analyze applications in real-time and recognize information misrepresentation, falsification, and other subtle indications of fraud. The analysis returns a risk score—1 to 999, low to high—and a risk reason code. A low-risk score and high applicant credit score could be immediately moved to the auto-approval queue. Mid-range scores could be moved to an underwriting queue for further review. High-risk scores could be auto-declined, with adverse actions automatically generated and sent electronically to the applicant.

The best way to manage risk is to prevent it in the first place. Fintech analytic capabilities now enable lenders to accomplish this by identifying and declining fraudulent applications before they impair your lending practice.

Minimize Risk with Data-driven Lending Decisions

Fraud-free applications are not necessarily risk-free. Bureau scores and attributes provide one means of determining creditworthiness and are useful for validating the creditworthiness for applicants at the extremes—exceptional or very poor scores. As applicants move from very good to fair, a bureau score may not be entirely indicative of creditworthiness. In these ranges additional sources of consumer data help provide a more detailed and accurate picture of creditworthiness.

These alternative credit data (ACD), captured, aggregated and provided as a service can include payment records for electricity, gas, water, cable, and mobile phone; rental or lease locations and payment history, and real estate ownership or liens. Alternative credit data helps lenders make better quality lending decisions in the following ways:

  • Provide an accurate, current picture of creditworthiness when alternative credit data is combined with a bureau score;
  • A high bureau score supported by strong ACD credentials gives lenders confidence in offering the best rates and terms;
  • A high bureau score contrasted with weak ACD credentials indicates a recent decline in financial strength. In such case, a lender may use decision rules to determine the appropriate rate and terms, or have an experienced underwriter review the application to determine creditworthiness;
  • A low bureau score contrasted with strong ACD credentials indicates improving financial strength and low-risk; and
  • A low bureaus score supported by weak ACD credentials represents a risk to be avoided.

The rich data sources now available—bureau and alternative—enable better quality lending decisions based on a more accurate and current assessment of applicant creditworthiness.

Use Reporting and Analytics to Continually Identify and Mitigate Risk

Modern lending systems provide integrated analytics, allowing business users to conduct nearly infinite variations of portfolio analysis. Analytics provides the ability to identify and evaluate risk factors that influence portfolio performance and make adjustments to credit policies and practices to mitigate risks. Regular portfolio risk analysis and reporting allow lenders to:

  • Identify underwriters whose overrides result in an unacceptable number of defaults;
  • Determine which borrower attributes or credit policies contribute to delinquencies for loans with 5-year terms;
  • Investigate any correlation between delinquencies and particular dealers;
  • Tighten underwriting criteria for applicants with credit scores below their defined thresholds to reduce delinquencies in that segment;
  • Calculate the optimum advance rates for borrowers seeking extended terms; and
  • Decline applications from dealers whose loans during the past year show an increasing number of delinquencies.

Analytics allows lenders to make data-driven decisions with the goal of continually improving portfolio performance through risk mitigation.

Fintech Capabilities

Fraud prevents risk at the outset. Data-driven lending decisions reduce the risk of delinquency and defaults. Analytics enables lenders to identify and eliminate risk factors by continually fine-tuning credit policies. Together these advanced fintech capabilities are essential tools for auto loan risk management.

 

Getting Started

defi SOLUTIONS has been identified as one of the Top 50 Most Promising Fintech Providers for 2018 by CIO Review Magazine. We’d welcome the opportunity to show you how our loan origination and analytics solutions address your auto loan risk management needs. Contact our team today or register for a demo.

Curious?

Get in touch with us today and get a demo!

REQUEST A DEMO

(Visited 19 times, 1 visits today)