MReport September 2019

TheMReport — News and strategies for the evolving mortgage marketplace.

Issue link:

Contents of this Issue


Page 54 of 67

TH E M R EP O RT | 53 SERVICING O R I G I NAT I O N S E R V I C I N G DATA G O V E R N M E N T S E C O N DA R Y M A R K E T SPONSORED CONTENT The Freedom of Fair Competition By Tom Parrent F air competition in credit scoring can foster in- novation and expand the population of scored consumers. However, the credit bureaus' ownership of Vantage- Score creates anti-competitive in- centives that may ultimately harm both borrowers and lenders. With the Federal Housing Finance Agency considering comments on its Notice of Proposed Rulemak- ing regarding the Validation and Approval of Credit Score Models, this issue should interest anyone concerned about the safety and soundness of the $6.7 trillion con- forming mortgage market. The Status Quo T hree credit bureaus—Experian, TransUnion, and Equifax— collectively control nearly all consumer credit data available in the U.S. This balance has existed for decades and led to continu- ing improvements in data and services from each firm. However, in March 2006, the three bureaus jointly created VantageScore, a new entrant in the credit scor- ing market. While competition generally leads to better pric- ing and innovation, the creation of VantageScore represented a potential anti-competitive effect in the provision of both data and analytics. FICO's flagship product, the consumer credit score, uses data from the bureaus to build its models. FICO does not sell its scores directly to lenders, instead providing a distinct algorithm to each of the bureaus, who then sell the resulting score to lenders and other customers. Thus, the bureaus ultimately control the price lenders pay to receive a FICO Score. Anti-Competitive Effects B y forming VantageScore, the bureaus intend to jointly control both data and analytics, upsetting a beneficial balance that has existed for decades. In other words, the bureaus now have a financial stake in the outcome of competition between FICO and VantageScore and the ability to influence that competition through their control of the data and the role they play in provid- ing FICO Scores to customers. In fair competition, the superior product should eventually win. However, "competition" in the presence of supplier power over both input and output is not fair. The three bureaus can subtly en- gineer the success of VantageScore against FICO even if FICO's products are objectively superior. While the direct result of VantageScore's displacement of FICO as the primary supplier of consumer credit scores—higher prices to end customers—is itself concerning, the secondary impacts are far more serious. Stifled Innovation I f the bureaus are allowed to control the collection, analysis and pricing of consumer credit data, analytic innovation will suffer. If they did not jointly own VantageScore, they would have appropriate incentives to sell data to any firm with good ideas on improving consumer credit scoring. These new en- trants would likely use different approaches in their attempts to create more insightful models. However, if VantageScore were to dominate the field, these new firms could be shut out of the market, either through data access restrictions or anti-competitive pricing. They could still build models with non-bureau data but would be at an enormous disadvantage compared to VantageScore, with its unfettered access to all bureau data. Effect on Consumers and Lenders A nti-competitive control of both data and analytics is not just about one firm against another. Important systemic ef- fects result from a lack of access and innovation, including quality control, inclusivity and pricing. Independent modelers have incentives to uncover new insights in consumer data that expand the population of consumers accurate- ly identified as creditworthy. FICO uses both bureau and non-bureau data to predict creditworthiness of young borrowers, consumers reen- tering the market, and others for whom the bureaus may have little or no information, thus adding data with potential explanatory power. By contrast, VantageScore has focused on using only bureau data but loosening the standards by which consumers are scored. VantageScore's reliance on less data, rather than more, is a direct result of their ownership structure. Non-bureau data may be the single most promising area for credit modeling advances. Banking, transaction and real es- tate data can measurably increase the power of scoring models, especially for consumers who are new to credit or who have had credit problems. These consumers generally have less information in their bureau files, so new mod- els can use additional data and advanced techniques that are well suited to relatively unstructured data. Conforming mortgage is unique among consumer credit markets. Not only are the bureaus respon- sible for the pricing and sale of both credit reports and scores, the FHFA requires mortgage origina- tors to purchase a consumer's credit report from each bureau. Without competition between the bureaus as a result of this requirement, allowing them to offer VantageScore—which they own and control—would enable them to use their market domi- nance to engineer the adoption of VantageScore over the FICO Score. The FHFA's proposed rule wisely requires scoring model providers be independent of data repositories. Separation of VantageScore from the bureaus would foster healthy competition, leading to new ideas and greater insight, not only from exist- ing players, but also from new entrants, ultimately better serving the marketplace." TOM PARRENT has deep executive experience across multiple specialties in financial services. His particular strengths include risk management, quantitative analysis, process improvement, and behavioral economics. Parrent has served as Chief Risk Officer at three major mortgage firms and has extensive experience in government, rating agency and regulatory relations, crisis management, and innovative statistical analysis. He has a bachelor's degree in economics from Princeton University and an MBA in finance and statistics from the University of Chicago.

Articles in this issue

Archives of this issue

view archives of TheMReport - MReport September 2019