As Income Risk Rises, Income Validation Alert Targets the Problem

The mortgage market is improving. More borrowers are seeking loans as interest rates decline and taking advantage of market timing. While that is good news, income-related mortgage application fraud risk holds the potential to increase as borrowers attempt to qualify for loans.

First American Reports Increased Income Fraud Risk in Mortgages

First American recently published that employment misrepresentation risk has been on the rise for quite some time and also advised that income related fraud risk could rise as well.

“Purchase loan applications typically are more likely to have fraud than refinance transactions,” said First American Chief Economist Mark Fleming in a press release. “Furthermore, in the strong seller’s market we experienced in 2018, borrowers had more motivation to misrepresent income on a loan application in order to qualify for the bigger mortgage necessary to win the bidding war for a home.”

“As affordability improves and demand increases going into the spring home buying season, we expect the seller’s market conditions to return, potentially elevating income misrepresentation risk as well,” Fleming said.

While income-related defect risk stabilized over the past few months, the risk of defects due to employment misrepresentations soared. In February, the risk of potential employment fraud was 175, the highest ever, up from 161 in January and 130 in October.

Point Predictive Analysis Suggest Income Misrepresentation is on the Rise As Well

Point Predictive’s findings are similar to those of First American’s analysis. In fact, Point Predictive tracked income misrepresentation across auto and mortgage lending and noticed a disturbing trend of increased income risk over time.

Point Predictive analyzed over 1.2 million historical applications from its proprietary data repository to analyze and quantify the effect of income misrepresentation on lenders. Each of the stated incomes on those applications was verified by the lender against income documentation, database checks or other methods. The results of the analysis showed an increase in risk over time as well.

Additionally, Point Predictive determined that income misrepresentation was surprisingly common among some lenders — up to 33% of their applicants had inflated their incomes.

Income Validation Alert Addresses Lenders’ Growing Concerns with Income Misrepresentation

To counter the rising tide of income misrepresentation, Point Predictive is actively integrating machine learning technology to precisely identify when a borrower’s stated income is inflated.

Income Validation Alert offers a real-time predictive assessment of an applicant’s stated income. If that income appears to be overstated by 15% or more, the lender is notified so they can perform additional verification. For borrower incomes below the 15% threshold, the lender can streamline the underwriting process to ensure good loans are not impacted.

Lenders use Income Validation Alert for a variety of purposes:

  • Pre-screen online mortgage and automotive applications.
  • Streamline automotive application underwriting stipulations at dealers.
  • Enable faster credit decisions for consumers.
  • Reduce the lender’s cost of utilizing more time consuming, less accurate, and more expensive income verification solutions.

Income Validation Alert analyzes an applicant’s stated income against millions of reported incomes and salaries from several diverse sources. Then, using the borrower’s employer, occupation, job title, residence, and estimated years of experience, a sophisticated machine-learning model predicts the applicant’s likely income. When the stated income exceeds what the model predicts by 15% or more, the lender is alerted to the discrepancy and can further scrutinize the borrower’s income.

What makes Income Validation Alert different than other approaches is the breadth and depth of salaries and reported incomes available to the validation process. Comparing a borrower’s stated income against many different sources simultaneously, and then using machine learning to cascade through those sources to determine the most reliable income for that borrower.

In laboratory testing with lenders, we have been able to successfully clear the stated income data on more than 85% of applications while successfully identifying more than 80% of the applications with known, overstated incomes.

If you are concerned with income misrepresentation, contact for more information on how we can help.