Matthew A. Bruckner of Howard has written The Promise and Perils of Algorithmic Lenders' Use of Big Data, 93 Chicago-Kent Law Review (2018). Here's the abstract:
Like many new technologies, algorithmic lenders’ use of Big Data holds great promise but may also be perilous. At the most basic level, Big Data is simply a toolkit for “creating, refining, and scaling financial solutions for consumers.” A company’s decision to use Big Data is “neither inherently good nor bad.” Instead — as with any other tool — it can be used to help or to harm consumers. The Janus-faced nature of emerging financial technology (“fintech”) firms is particularly noteworthy, and lies at the heart of this Article.
Appropriate regulation will likely be key to delivering on Big Data’s promises in the financial services sector. All financial services companies are potentially subject to a significant amount of regulation. But while regulators have paid attention to fintech’s development, regulations “have not kept pace with modern Big Data capabilities.” This presents challenges both for regulators and “for companies looking for firm legal guidelines as they build” their companies. Indeed, the Consumer Financial Protection Bureau (CFPB) noted in a recent Request for Information that it needs to better understand these technologies to “encourage their responsible use and lower unnecessary barriers, including any unnecessary regulatory burden or uncertainty that impedes such use.” Impliedly, it will also seek to prevent fintech’s irresponsible use.
This Article proceeds as follows. Part I provides a (very brief) overview of Big Data, machine learning/predicative analytics, and their use in making credit determinations. Part II discusses the promising and perilous nature of “algorithmic lending 2.0.” Its major promise is to bring the so-called “credit invisibles” into the credit markets by using non-traditional credit measures. Its primary threat is the possibility that it will exacerbate financial services discrimination. Part III discusses several major pieces of the current regulatory regime, where it fails to adequately address the worst threats, and how we might improve oversight of algorithmic lenders.
