Menu

New Coordinates Articles Details

 

Alternative Data Sources: Reading the Signals and Warning Signs - Eric Reusch and Mark Westmoreland

Eric Reusch and Mark Westmoreland
New Coordinates
Spring 2017

Lenders are mining new alternative data sources today to expand access to credit, make their credit models more predictive, and improve servicing, collection, and product development. Digital payment platforms, mobile wallets, social media networks, "real world" credentials such as college degrees and professional accreditations-these and other sources are increasingly explored for their usefulness in establishing creditworthiness beyond credit bureau scores and other traditional inputs. At the same time, however, financial services companies and regulators are realizing that all this data should be handled with great car with consumer protection, customer satisfaction, and financial performance all at stake.

Pros and Cons
The business case for using alternative data is quite strong Companies can get more out of their marketing dollar-casting a broad net and potentially converting or approving a higher percentage of applicants without increasing risk. Access to credit could be created or enhanced for a population of consumers the CFPB refers to as "credit invisible." New data points could improve loan performance, helping to earlier identify and reach out to current borrowers at risk of default. It could help build a better case for predicting future performance when packaging and selling loans to underwriters or investors. It is a competitive tool in a market being reshaped by online lenders and other financial technology companies (FinTechs), who are pioneers in alternative data use.

However, simply "throwing data at the problem," as the saying goes, doesn't always provide better outcomes for lenders and their customers. There are two important provisos: It's about whether the data is meaningful and how you use it-on its own or in combination with other inputs.

Data Is As Data Does
Data is not neutral. There is always a signal in personal data that may tell you something about a customer's or prospect's ability and/or willingness to pay. In this way, using more data can expand consumers' access to credit if it is positive. On the flip side, however, it can restrict access, if negative.

The Consumer Financial Protection Bureau (CFPB) is concerned about the latter, as it made clear in a recently issued "Request for Information Regarding Use of Alternative Data and Modeling Techniques in the Credit Process." 1 However, the CFPB is also interested in encouraging innovation, pointing out that today, "26 million Americans are 'credit invisible,' meaning that they have no file with the major seven credit bureaus, while another 19 million are 'unscorable' because their credit file is either too thin or too stale to generate a reliable score from one of the major credit scoring firms."

Also spurring alternative data usage are the FinTechs catering to millenniaIs. Many consumers in this next generation of home buyers may use new and different emerging tools to manage their personal finances, which can obscure their credit history if viewed on a traditional basis.

Regulatory Risk

Yet there is a danger inherent in using alternative sources of data or even using traditional data in new ways. A lender may not fully comprehend how its decision-making constructs or processes fit the existing regulatory structure. Ultimately, it's not clear whether the regulatory structure is going to change over time to fit business models or whether business models will have to change.

In the meantime, a lender using alternative data sources has to manage inherent regulatory risks. One example would be compliance with fair lending rules. If a regulator tests a loan portfolio using proxy data, developed in the context of widely available data sources, an examiner could find unintended violations if you are using alternative data-and you won't know that you have a problem until it's too late. You should consider the aggregate impacts of the data you're using on decisions to pre-approve or approve/decline customers as well as on the terms of the credit you extend.

Data Integrity
Lenders have to be able to trust any data source they use for its accuracy, validity, and completeness. Traditional credit scores and data sources have been developed and have evolved over decades and are relatively reliable. But new alternative data sources should be carefully vetted and tested up front-for instance, back-test a historical loan vintage to see whether the predicted default rate would go up or down using the new data. Also, keep an eye to the fair lending concerns of whether the new data would create the potential for disparate impact.

Business Complexity

The CFPB raised the issue of complexity in another way: Would using alternative data make credit decisions more complex for both consumers and industry? One issue that could arise, for example, is data ownership. Is a consumer's financial transactional data portable--do consumers, effectively, own their own data or does it belong to the financial institution? If so, asking a third party to release that data to a lender should be relatively seamless. But what if the third party disagrees?

Security and Privacy
Security and privacy are top-of-mind across the financial services industry and the agencies regulating it, and alternative data provides a case in point. When using "other people's data," your customers need to know what data you are collecting on them. That data also needs to be protected against security breaches as the pace and severity of cyberattacks are increasing.

The Final Analysis
Alternative data can be beneficial across the life cycle of a loan-from customer acquisition to underwriting and credit to servicing to collection. But lenders would be well advised to proceed cautiously. Data for data's sake is not necessarily good; only truly predictive data is good.

And it's still early in the alternative data game. Companies are still struggling to figure out how to harness the power of all the data that's available. Regulators are also trying to understand exactly what lenders are doing and how they should address it. The CFPB's recent request for information underscores this fact. Regulation nearly always lags innovation, but companies shouldn't get too far out in front of the regulators, due to compliance risks.

The best course would be to stay aware of your compliance obligations in a changing landscape. The current slate of federal consumer protection laws will continue to be enforced. As your business model emerges, you have to stay compliant with existing rules but be flexible enough to take advantage of new opportunities.

Fundamentally, companies should maintain a compliance mindset. Understand that while your business may move at light speed, regulations don't necessarily keep pace. As long as you always analyze your business decisions' impact on consumers and compliance with the law, that's a good start. 

View as PDF

Treliant Risk Advisors, Compliance, Risk Management, and Strategic Advisors to the Financial Services Industry, brings to you New Coordinates, a quarterly newsletter offering insights and information regarding pertinent issues affecting the financial services industry. This article appeared in its entirety in the Spring 2017 issue. To subscribe to our quarterly newsletter, please Contact Us.