How to reduce repurchase risk through automation

Getting accurate data the first time reduces your risk downstream

Originating loans in the still-stagnant housing economy is tough enough — no one wants to have to repurchase those loans due to process mistakes that could have been prevented.

The pressure on lenders around buybacks has never been greater. In 2013 alone, banks bought back $37.8 billion in mortgage loans. While many lenders have resolved legacy representation and warranty liabilities with the GSEs, Fannie Mae and Freddie Mac have turned their attention to potential problems with performing loans, and banks are on high alert to avoid even more repurchases.

“Mortgage lenders have suffered staggering losses and gone out of business because of the dreaded loan repurchase,” Mark Greene wrote in a Forbes article. “Avoiding defaults and loan buybacks has become the primary goal of mortgage lenders.”  

As lenders look for ways to reduce their risk, many have sought to automate manual processes that can introduce errors into the origination process. One significant technology innovation — automated data capture — eliminates the need for borrowers or lenders to manually enter data from mortgage documents, which can needlessly compromise data integrity.

Automated capture technology doesn’t just send pictures of the documents, but extracts data from photo IDs, financial statements, pay stubs, and other documents, and then matches, validates and corrects that data instantly.

In a September article in Fannie Mae’s Housing Industry Forum, Jeff Bounds wrote, “When loan data is inaccurate, it costs money — especially if it’s found later in the process of making loans or selling loans to entities like Fannie Mae or to other private investors.

“The rule of thumb is that verifying loan data upfront costs $1 per record. If that information is wrong and has to be ‘cleansed’ later, the cost rises to $10 per record. And, if the problem surfaces much later, that cost rises to nearly $100 per record, as the impact from inaccuracies is felt over and over again,” Bounds wrote.

Part of those escalating costs is the time it takes for staff to recognize the inaccuracies and try to correct them. Automated processes can compare borrower data faster than a manual process, and quickly identify incomplete information, alerting a loan officer or underwriter immediately. By identifying blank fields, incorrect dates and unsigned documents right away, automated processes can save lenders time and tremendous pain later on.

As Bounds noted, “If data is entered correctly and verified the first time around, the result is shorter time frames to close and fund loans. It also means less money that needs to be set aside for loan losses, and an easier time for servicers in buying mortgage portfolios and making them profitable.”

The specter of buybacks makes mortgages more costly to originate, as lenders are forced to check and recheck information that should be settled. Using automated processes, including a document capture solution, helps banks and other lenders to minimize repurchase risk.

What other ways are you reducing repurchase risk? Share your thoughts in the comments section.

Most Popular Articles

FHA, VA join Fannie, Freddie in relaxing some standards

With the coronavirus continuing to reshape the face of the country and the economy, the biggest players in the mortgage business are moving to try to make it easier to lend. Last week, it was Fannie and Freddie. Now, it’s the FHA and VA’s turn.

Mar 30, 2020 By
3d rendering of a row of luxury townhouses along a street

Log In

Forgot Password?

Don't have an account? Please