How to Identify Fraud Risk to Streamline Loan Approval
Now more than ever, lenders need a solution that creates more efficiency so they can better handle high volumes. HousingWire recently spoke with Robert Karraa, President of First American data and analyzes, on how lenders can better identify the risks of fraud and errors in mortgage loan applications.
HW: Fraud has been a hot topic lately, how does First American Data & Analytics help lenders identify fraud risks?
RA: Our company has been an established leader in fraud detection for over 15 years. Our FraudGuard alert-based solution uses the natural intelligence gained from millions of loan requests and the eventual results to help underwriters note anomalies in request data. While still extremely efficient, higher volume lenders have expressed the need to create even more workflow efficiency and help them focus their efforts on riskier applications. That’s why we introduced the AppIntelligence Score (AI Score) fraud solution earlier this year. AI Score is a powerful new fraud pattern recognition score that enables lenders and high-volume investors to more accurately identify the risk of fraud and prepayment default on new mortgage applications or on a portfolio.
This highly accurate score dramatically reduces review volume and false positives, and uses proprietary predictive modeling to paint a more accurate picture of where fraud and the risk of prepayment default are likely to occur. For example, loans with the highest AI score, typically around 10% of total application volume, account for 50% or more of the total fraud risk. This targeting allows lenders to focus their reviews on the most risky loans, while streamlining loan approvals and reducing operational costs.
AI Score uses our proprietary predictive fraud indices, using both natural and artificial intelligence as well as machine learning technologies. It simultaneously runs proprietary sub-models for risk including synthetic identity, income, employment, prepayment default (EPD), undisclosed debt, and risk review of loan participants.
The ability to analyze millions of alerts that underwriters have cleared and then feed that information back into the model is what makes AI Score one of the most sophisticated models on the market.
Housing wire: You recently announced the Data & Analytics division of First American. What does this rebranding mean for lenders?
Robert Karraa: At the end of last year, we renamed our data division to First American Data & Analytics. The rebranding was the culmination of a multi-year effort to create a world-class, property-centric data operation that could deliver critical risk management, analytics, valuation and marketing solutions. First American Data & Analytics solutions empower lenders to make better, and increasingly automated, decisions to manage their workflow.
To give you an idea of our scale, we curate and maintain the largest data set of real estate property information in the country. That’s more than 7 billion documents recorded. And every month we add over 5 million new document images, or over 20 billion data items, to the dataset.
In 2020, more than half of all mortgage origination transactions were based on our data or one of our analytical solutions. While mortgage is an important market for us, our division also serves other verticals: Title, Valuation, Government, FinTech, PropTech, Retail, Marketing, Real Estate, Real Estate Services and Education.
HW: What other products and services fall under First American Data & Analytics?
RA: Our flagship data platform, DataTree, would be at the top of the list. It is the industry’s most comprehensive source of property and housing data, along with offering solutions for customer acquisition and market share reporting. We also have a full suite of APIs and other custom solutions that allow our customers to access all of our market-leading datasets in the format that best suits their needs.
In addition to our data assets, we have put together a number of cutting-edge solutions that enable lenders and securities firms to make better and increasingly automated decisions to manage their workflow, audit operations. and compliance and their real estate research. Our other solutions include the RegsData loan compliance platform; TaxSource, one of the industry’s most comprehensive nationwide tax status reporting solutions; and a host of other analytics that provide solutions for assessment, assessment management and more.
On the securities side, we offer securities search and review automation solutions for securities and settlement service providers, as well as a nationwide offering of managed securities production services. All of our title solutions are supported by the industry’s largest geographic collection of title factories.
Finally, we offer a wide and expanding range of appraisal offerings, including AVMs and ACI workflow solutions used by appraisers, lenders and AMCs.
HW: What’s new and what’s on the horizon at First American Data & Analytics?
RA: In addition to new solutions, like AI Score, we’ve also focused on new ways to give users access to data. FinTech and PropTech companies, for example, tend to be heavy users of our data. More and more, these customers aggregate, enrich and consume their data via the cloud on innovative platforms, such as the Snowflake Data marketplace.
Now, a broader set of data consumers, including FinTech and PropTech customers, can easily leverage the industry’s largest set of property and ownership data to generate advanced insights, reduce risk, create new analytics, test new concepts and drive innovation in a cloud-hosted environment. By moving access to the cloud, we meet our customers where they want to work with our data, so they always have the most recent and up-to-date data.
First American data and analyzesCustomized fraud solutions detect fraud risks and mortgage errors early on, so lenders can more effectively handle high volumes.