Credit scoring in the mortgage market: a new ball game
Content sponsored by Partner Insights by VantageScore
A little-known secret in lending circles is that unlike in any other consumer loan category, nearly every single consumer mortgage loan requires a credit score that was built on data from prior to the turn of the century. To this day, what many consider an outdated version of the FICO score enjoys a government-sanctioned monopoly for applications submitted to the Fannie Mae and Freddie Mac automated underwriting systems, which, for all practical purposes, set the standard for the mortgage industry
But thanks to a new rule issued by the Federal Housing Finance Agency (FHFA), that is about to change.
In August 2019 the FHFA, which acts as the regulator and the conservator for both Fannie Mae and Freddie Mac (also known as Government Sponsored Enterprises, or GSEs), issued a final rule to address the lack of credit score model competition in the mortgage market. The new rule requires the GSEs to put in place a process by which newer and more predictive and inclusive scoring models can be considered.
Establishing competition in the mortgage marketplace is the proverbial win-win. More creditworthy borrowers become eligible for mainstream credit can achieve the American Dream of sustainable homeownership, and lenders have an unprecedented growth opportunity. And by establishing a process by which new models can be tested and adopted, the mortgage industry will be able to keep pace with changing consumer behaviors and adopt the models that are most suitable for the time period in which they are used.
The introduction of more sophisticated reporting and data analysis techniques has ushered in a new era of credit scoring, bridging the gap between access to mainstream credit and those consumers with limited credit histories, who can be accurately scored with a more advanced credit scoring model. To be more specific, there are 40 million consumers, of whom 10 million are creditworthy enough to be eligible for a mortgage, using more modern credit scoring methodology alone. And of these consumers, 2.4 million are African-American and Hispanic consumers.
This is extremely important because U.S. households are becoming more racially diverse. In fact, the minority share of U.S. households is projected to rise from 34 percent in 2018, to 37 percent in 2028, and to 41 percent in 2038, according to the State of the Nation’s Housing 2019 Report by the Joint Center for Housing Studies of Harvard University. Moreover, the Urban Institute expects that in the next 10 years, minorities (i.e., African-Americans, Hispanics, among other groups) will account for 90 percent of new household formations.
One new model that allows lenders to accurately assess millions more consumers is VantageScore 4.0. Phil Bracken, VantageScore’s Managing Director of Government and Mortgage Industry Relations, explains how their model may help transform the mortgage industry under FHFA’s new rule.
“From a credit score perspective, we know we can score a lot more people. Many of those would be of the demographics that are in that tsunami of changing demographics coming at us. That’s why it is so important, from a public policy standpoint, to have competition in credit scoring and allow mortgage lenders and others to advance this opportunity for those qualified consumers,” Bracken said.
Unlike conventional credit scoring models, which typically contain model attributes that incorporate only one or two dimensions of behavioral credit data, VantageScore 4.0 leverages machine learning techniques to develop multidimensional attributes for consumers with limited credit histories. With an eye toward explainability and transparency, all these attributes once developed remain static so that easy-to-understand reason codes can be assigned.
VantageScore 4.0 also leverages trended credit data, which goes back up to two years to reveal patterns in credit behaviors, such as the number of loan balance increases/decreases over time or growth/decline in borrower’s utilization versus a point-in-time snapshot. The new insights drive predictiveness and allow those without the benefit of deep and diverse credit histories – namely Millennials and Generation Z – to have more accurate credit scores that are reflective of their relative risk.
Since emerging on the scene in 2006, VantageScore has enjoyed increased adoption. According to global management consultant Oliver Wyman, in the 12-month period between July 1, 2018 and June 30, 2019, approximately 12.3 billion VantageScore credit scores were used by more than 2,500 users. This is approximately a 20% increase in the number of scores versus the prior year.
So now that the rule is final, what comes next?
In general, credit scoring models that respond to yet-to-be released “solicitations” from Fannie Mae and Freddie Mac will be evaluated on the model’s integrity, reliability, accuracy, historical record of measuring and predicting borrower credit behaviors (such as default rates), among other measurement criteria.
The final rule defines a four-phase process for GSEs to validate and approve credit score models. The process begins with the Credit Score Solicitation, which is a solicitation by GSEs of applications from credit score model developers. This is followed by the Submission and Initial Review of Applications phase, which is an initial review by GSEs of the submitted applications. The third phase is a Credit Score Assessment by the Enterprise, which is a detailed evaluation of the accuracy, reliability and integrity of a credit score. The fourth phase is an Enterprise Business Assessment, which is an evaluation of the potential impact of using the credit score model within the GSEs’ proprietary business systems and processes.
As stated in the final rule, FHFA expects GSEs will “evaluate applications received in response to the initial Credit Score Solicitation and may submit to FHFA a proposed determination to approve one or more of those credit score models for use, either to replace Classic FICO or in addition to Classic FICO.”
FHFA anticipates the process of approving alternative credit score models will take as many as 26 to 28 months, and the forecast for “industry adoption” is somewhere between December 2022 and June 2023. A more detailed description of FHFA’s timeline follows this article.
“We are extremely excited about this, not just because we believe that the VantageScore model is a great risk management tool, and that we can score more people and can get more people fair and responsible access, but because this is an opportunity for constituents,” said Bracken. “This changing demographic is quickly evolving, and the need for homeownership in America has never been greater. It has never been more compelling for constituents who need and deserve to be served.”
To learn more about the new FHFA-issued rule and a pilot program using VantageScore, visit www.vantagescore.com/GSEnext.