FintechTechnology

Stepping up the fight against fraud in mortgage lending

As instances of attempted fraud proliferate, fintech can help in hiring processes and operations

Wherever there is commerce, there will be fraud, and the mortgage industry is no exception. In recent years, we have seen a significant uptick in fraudulent activity ranging from the high tech — intercepted wire transfers and electronic title phishing scams — to low-tech, such as applicants submitting falsified or doctored bank statements. The transition to a largely remote workforce during the COVID-19 pandemic only gave criminals more of an incentive to attempt to deceive lenders or misrepresent qualifications.

The full extent of fraud in the mortgage industry may be impossible to fully quantify, but the 2020 True Cost of Fraud study by LexisNexis Risk Solutions estimated that the cost of fraud has risen 7.3% across U.S. retailers and e-commerce merchants, and every $1 of fraud now costs $3.36, up from $3.13 in 2019. The COVID-19 pandemic, coupled with skyrocketing refinance applications due to a low interest rate environment, have created a perfect storm for fraud.

With e-signatures enabled for virtual closings, it has become more difficult for lenders to authenticate income and asset information, while the incentive to cheat is boosted by desperation or perceived opportunity.

Underwriters with hidden agendas

One emerging form of fraud we have seen plaguing our industry has nothing to do with the loan manufacturing process, but rather with the labor used to generate the loans.

Given the historic volatility of 2020 (driving loan demand) and the search for elusive industry experience, underwriters have become a prized catch. Companies are offering exorbitant salaries, benefits, signing bonuses and hefty referral bonuses; it’s no wonder everyone suddenly decided they could become an underwriter! However, the real prize, for some, was not the cash made by originating loans but rather the value of the private, personal information stored in companies’ systems.


How lenders can prepare for growing fraud threats

As origination volumes hit record highs in 2020, Truework’s verification experts saw a spike in fraud in mortgage lending, and expect that trend to continue this year. HousingWire recently spoke with Jeffrey Morelli, General Manager of Truework Verifier, about what lenders can do to prepare for and overcome the growing threat of fraud and data inaccuracy.

Presented by: Truework

Even as employers get wiser to the deceitful tricks, the perpetrators get more sophisticated. The going price for a social security number is about $4, but based on credit score, that price tag rises significantly. Although these fraud rings may not be getting rich just selling SSNs, coupled with all of the other banking and sensitive information available, the risk/reward is more than worth it. And so, as more people become aware of the potential financial gain of working in the mortgage industry, there have been numerous fake websites set up with the sole intent of capturing individuals’ personal data.

For the sake of integrity and security, lenders need to refine the recruiting and hiring process to include additional “scrubs” on candidates. These mechanisms include verifying that applicants are listing real employment history and that their social media presence matches the resume. In addition, all interviews should be conducted via video and any testing monitored on video to ensure that the candidate is in fact the person taking the test.

Data-driven analysis

Now is the time for lenders to tighten their procedures not only on recruiting and hiring practices, but also lending operations. Fortunately, fintech innovation is helping lead the way with methods to detect inconsistencies in the due diligence process, such as drawing data directly from banks (with the borrower’s consent) and eliminating the need for written statements. Data-driven analysis also flags questions lenders should be asking, such as whether a deposit sum came from the borrower’s assets or represents a gift from an outside source (which should be calculated in the debt-to-income ratio).

In a pilot program we developed, a technology solution was able to successfully scan a small batch of loans to detect undisclosed property liabilities such as tax liens and child support/alimony payments. Artificial intelligence can read specific loan documents and wage information to determine if there were additional liabilities that were not listed on Form 1003. There were enough significant findings to make these loans unsalable. Using such technology reduced the review time for an underwriter and caught information that was not caught during the normal quality control process.

Due diligence to verify the destination account of an escrow or down payment is also essential, since such transfers can’t be undone, and there is little recourse for seller or buyer once the sum is gone. At the same time, consumers should do their part by verifying numbers to make sure any emails they receive are actually from the lender, and verifying account numbers before transferring funds.

Lenders have little recourse if a funded loan is later found to have been approved on the basis of faulty or fraudulent data. Only borrowers have the right of rescission. Banks have no choice but to honor these loans and, in the 2007 financial crisis, we saw the devastating consequences. There is no reason to repeat the same mistakes again.

Covid challenges

In addition to document and identity fraud, the economic devastation and displacement caused by the pandemic presents new challenges to lenders related to the loan modification and forbearance processes. We learned from the Great Recession to be aggressive about verifying claims of income changes in such requests. Now we need to adapt again.

Today, thousands of borrowers, while still employed, may be asking for forbearance in anticipation of future hardship, or even just to take advantage of an opportunity to skip or lower payments.

Further, in the verification of employment process, lenders must be cautious that a positive response may represent an employee who has been given notice of termination or furlough, but is still working at the time of inquiry.

Lenders must also reverify the automated valuation model if an application has been pending longer than six months, and commission a new title search to see if borrowers were added or removed following the initial application.

Credit reports are valid for the application process for up to 120 days, meaning those with lowered income or maxed-out credit cards may present an outdated credit score in the application process.

Fintech providers must continue developing solutions to make mortgages safe, on the front end and back end, while educating originators and lenders and protecting buyers and sellers. As fraud in mortgage becomes more sophisticated, so must we.

As originators step up and get smart, the industry will be better and safer for it.

This column does not necessarily reflect the opinion of HousingWire’s editorial department and its owners.

To contact the author of this story:
Kimberly Lanham at [email protected]

To contact the editor responsible for this story:
Sarah Wheeler at [email protected]

Leave a Reply

Your email address will not be published. Required fields are marked *

Most Popular Articles

3d rendering of a row of luxury townhouses along a street

Log In

Forgot Password?

Don't have an account? Please