IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.
Sponsor Content
What does this mean?

Fast, Accurate Identity Verification Helps California Homeowners

A person in a business suit sitting at a table typing on a laptop. There are symbols hovering above the laptop including "AI" and locks, peoples and buildings.
Shutterstock

The state used AI/ML-driven identity verification software to fight fraud and deliver nearly $1 billion in post-COVID-19 mortgage relief.

With nearly a billion dollars in federal funds available to help homeowners hit hard by the COVID-19 pandemic, the state of California’s Mortgage Relief Program needed technology to approve assistance applications quickly, easily and accurately — while stopping criminals looking for lax controls and easy money.

When homeowner relief funding from the American Rescue Plan Act arrived in March 2021, program administrators were on high alert due to reports of massive fraud in other COVID-related programs.

“Identity fraud is a major risk that requires a robust mitigation strategy,” says Stephen Johnson, director of finance for the California Mortgage Relief Program and vice president of Outreach Solutions as a Service LLC (OSaaS), a Sacramento-based technology firm. “We took this risk very seriously in the design of our administrative processes and system controls.”

The California Housing Finance Agency (CalHFA) set up a nonprofit entity called the CalHFA Homeowner Relief Corp. (CalHRC) to distribute the relief funds. CalHFA then contracted OSaaS to administer the Mortgage Relief Program with oversight from the CalHRC board of directors.

CalHFA Director of Enterprise Risk Management and Compliance Rebecca Franklin, who served as president and board chair of CalHRC, struck a careful balance. “We wanted to get the funds to the right people in a compliant manner, while also making sure the processes we used to verify application information weren’t burdensome to the homeowner,” she says.

Leveraging Socure and other Amazon Web Services cloud-based technologies, the program’s system automatically and almost instantly verifies homeowners’ identities in most cases. It also reviews and approves their eligibility for the program, while rejecting fraudulent applications. The agency’s experience underscores the benefit of using artificial intelligence and machine learning (AI/ML) to streamline identity verification and minimize fraud.

HARD-PRESSED HOMEOWNERS AND MONEY-HUNGRY FRAUDSTERS

Many California homeowners suffered financial hardships during the pandemic such as job loss or business closure, potentially facing foreclosures if they could not make payments. The California Mortgage Relief Program awarded as much as $80,000 to eligible homeowners for overdue mortgage payments, unpaid property tax bills and other housing costs.

The program was designed to assist low- and moderate-income California homeowners. Franklin and her team were reluctant to impose cumbersome identity verification requirements on people who were already suffering. “We did not want to do it at the cost of extremely slow processing times or very manual and taxing application processes.”

The California Mortgage Relief Program also needed an identity verification solution that would let staff focus their time and energy on helping struggling homeowners.

“We were looking for real-time, accurate identity verification that allows us to improve our operational efficiencies,” Johnson says. “If we are able to have our staff spend more time on processing and decisioning applications rather than validating an applicant’s identity, it not only improves productivity but also saves costs for the program.”

AUTOMATED IDENTITY VERIFICATION AND FRAUD CONTROLS

“Our priority was to design processes and incorporate system controls that ensure the assistance is awarded to applicants who are who they claim to be,” Johnson says. Socure’s primary assignment was to verify applicants’ identities after they submitted applications to the California Mortgage Relief Program.

Socure’s solution uses AI/ML algorithms that analyze more than 30,000 variables to create fraud models that accurately predict the veracity of an applicant’s identity. These models calculate risk scores and reveal risk factors that may be important when making decisions about an application. There are more than 800 risk factors that, along with the risk scores, are available to automate a decision. Examples include the risk associated with an email address or a phone number, whether an address is residential or commercial, or the geolocation of IP addresses (foreign IP addresses often point to fraud). Socure’s solution leverages these scores and risk factors in custom identity logic specific to the Mortgage Relief Program’s needs to validate an individual’s identity.

The technology is accessed through a secure, standards-based application programming interface (API), simplifying integration with resident-facing applications. The API returns responses in less than 400 milliseconds, providing a completely frictionless experience for applicants.

The solution also provides monitoring and alerts to help the agency tweak its processes in response to new grant programs or unexpected threats. “We have to tailor it to what we think is most important for our needs,” Franklin says. “A housing assistance program is going to be different than, say, a social services grant program.”

Socure’s technology also met the state’s need for fast deployment. “We kicked off the implementation in January 2022 and within a couple of weeks we were looking at live results,” Johnson says.

SPEEDY APPROVALS, ROBUST RISK MANAGEMENT 

The California Mortgage Relief Program positively verified the identity of applicants nearly 94 percent of the time through Socure’s automated solution, Johnson says. Without the solution, agency staff may have spent five to 10 minutes verifying each applicant’s identity even in the simplest of cases.

“If you extrapolate that to the entire population of applications, we are talking about thousands of hours of manual review that has been eliminated by the solution,” Johnson says.

The software also provides transparency to the agency’s reviewers. “When it flags an application for review, it clearly identifies why it was flagged so that we can remediate it quickly,” Johnson says.

From an executive perspective, the Socure solution achieved the agency’s goals. “It was very seamless for me,” Franklin says. “I was given a level of assurance that the applications we were approving had an extra layer of fraud detection. It’s not just that an applicant’s income is being calculated correctly, but that the person we are helping and funding is a real person with a real home.”

Socure is the leading provider of digital identity verification and fraud solutions. Its predictive analytics platform applies artificial intelligence and machine learning techniques with trusted online and offline data intelligence to verify identities in real-time. The company has more than 2,200 customers across the financial services, government, gaming, healthcare, telecom, and e-commerce industries, including four of the five top banks, the top credit bureau and more than 400 fintechs. Organizations including Chime, SoFi, Capital One, Robinhood, Gusto, Public, Poshmark, Stash, DraftKings, Prizepicks and the State of California trust Socure for accurate and inclusive identity verification and fraud prevention. Learn more at www.socure.com.

OSaaS rapidly creates scalable digital government solutions that connect everyone anytime, anywhere.  Focused on Salesforce and AWS, we help clients transform and modernize through cloud adoption and migration.  Our services include strategy, architecture, design, application development, artificial intelligence, integration, readiness, training, etc. Learn more at www.osaasllc.com.