Faculty, Industry Experts, Research, Thought Leadership

Breaking ground in fintech research—it’s not “too new” to pursue

breaking-ground-fintech-research

Academic research on fintech was virtually nonexistent during the eight years Andrew Karolyi, the Harold Bierman, Jr. Distinguished Professor of Management and deputy dean of academic affairs for the Cornell SC Johnson College of Business, served as an editor of the Review of Financial Studies. Four years ago, when Karolyi become executive editor of the peer-reviewed academic journal, he set out on a mission to change that. “I couldn’t tell you how many times I would hear from my industry colleagues about the fact that the research being published in the top-tier finance journals is just not relevant enough to their work,” says Karolyi. The most commonly understudied topic cited by these industry practitioners was fintech, including blockchain currencies, peer-to-peer and online lending platforms, robo-advising, or big data. Frustrated by the lack of published research on the latest financial technology innovations and the overwhelming demand for such research, Karolyi put in motion an innovative process to generate a body of scholarly research on fintech.

Adapting the registered report

In order to address the root cause of the lack of fintech research, Karolyi first formulated a thesis about why scholars were not producing work on this exciting new topic area. “I just felt the perceived risks were too high among young scholars out there who would naturally be working on this topic, even if they had the capacity to do so,” says Karolyi. Whether they felt there was insufficient existing data to support their work, or they feared a lack of interest from editors, reviewers, and the industry as a whole—something was holding them back from producing their best work.

Then it dawned on Karolyi that maybe the concept of the registered report, which was invented by Professor Chris Chambers of Cardiff University as a way to create more transparency in neuroscience research, could be retrofitted to encourage scholars to explore new topic areas in finance that were previously considered too risky. Rather than the traditional approach of submitting research for review and publication once completed, registered reports invite scholars to submit a research proposal for acceptance before the execution and analysis has even commenced.

“The goal of these registered reports was to deal with the fact that they believed there was a crisis of confidence about research that’s going on in that field, that people are not fully transparent about the choices they make as scholars, and they’re going after results that pop off the page in order to gin it up for the referees and the editors to make it more palatable for them to publish,” says Karolyi.

By adopting the concept of registered reports for financial research,  Karolyi sought to “create a competition using the platform of a top journal in finance to encourage researchers to submit proposals, not final papers, using all the creativity they could conjure to formulate those proposals.” By committing to publishing the registered reports before the actual research had been conducted, the journal would be accepting some of the risk associated with conducting research in uncharted territory.

The first body of knowledge on fintech

Headshot of Andrew Karolyi
Andrew Karolyi, the Harold Bierman, Jr. Distinguished Professor of Management and deputy dean of academic affairs for the Cornell SC Johnson College of Business. Karolyi was the former editor of the peer-reviewed journal, Review of Financial Studies.

When the Review of Financial Studies first initiated a call for proposals in the registered reports format in January of 2017, Karolyi wasn’t sure what the response would be like. “What came back in time for our first conference was 156 submissions, which just blew our socks off,” says Karolyi. After that, it was off to the races. Submissions were anonymously reviewed by a 20-member committee and the top 10 were selected to be presented in person at a workshop held at Columbia University in May 2017. “We were judging them on the basis of what their question is, what they’ve proposed to do, and whether the design of the experiment had integrity,” says Karolyi.

Out of those 10 finalists, all of them were given principal acceptance to complete the research and present the results in the spring of 2018. “The researchers went off, were told they had about eight months to get to work and get this research done, and that papers were due back in January,” says Karolyi. Once the papers were complete, they went through another round of reviews and the scholars were invited to present their findings at the RFS fintech conference hosted at Cornell Tech in March 2018.

The conference, which sold out, was a huge hit. Karolyi says the conference was attended “primarily by scholars who wanted to see the first published body of work on fintech.” Comments from conference attendees were incorporated into another round of revisions, and the papers are now being finalized for publication in a special edition of the Review of Financial Studies this fall.

What topics are covered in this pioneering body of work? “One of the myths that we are demystifying is that blockchain and cryptocurrencies are everything that is interesting in fintech; in fact, they really are only a small fraction of what is fintech,” says Karolyi. The other papers to be included are focused on big data and machine learning techniques in the financial services industry, robo-advising, and disruptive online and peer-to-peer lending platforms.

Already, these learnings are having real-world implications for financial firms. For one of the sponsored papers on robo-advising, the research team partnered with a firm in India to conduct field experiments using their portfolio optimizer to advise individual clients. They were able to track how many clicks their clients were using this advisor with, how many online engagements there were, and how it changed their behaviors in terms of how well diversified their portfolios were. “I wouldn’t be surprised if that firm adjusts their optimizer as a result of this research that will be published,” says Karolyi.

To fintech and beyond

The insights gleaned from this experimental research approach are likely to leave a lasting impression on the financial services industry for years to come. According to Karolyi, more than 50 percent of the scholars that submitted proposals were from outside of the United States. “We have this inward-looking perspective that the US is the center of the world in financial services, but when it comes to fintech, it’s pretty clear that the United States is not at the center of this world and certainly not in the leadership position,” says Karolyi.

Interestingly, these were disproportionately younger scholars submitting proposals than typically seen in publications in top finance journals. In fact, 63 percent of the submitted authors were either assistant professors or PhD students. Two of the papers were solo-authored PhD student papers, which is an increasingly rare occurrence. “It led one of my co-collaborators at Columbia to say this was probably the most democratic process that we’ve seen in finance in that there weren’t seasoned, senior, well-established scholars who were already incumbents in a field challenging others trying to jump into it, because there really wasn’t much written on fintech to begin with,” says Karolyi.

The interest in this type of research and the success of the fintech experiments has given Karolyi even greater confidence in the process of registered reports and its place in the discipline of financial economics. “I’d like to think that some other entrepreneurial person affiliated with a journal will come along and pick it up and use it for some positive gain,” says Karolyi. “I think innovation along these lines, especially in the world of journals and scholarship, is a good thing. It’s a healthy thing to shake existing protocols up a bit.”


Johnson BusinessFeed > Thought Leadership > Breaking ground in fintech research—it’s not “too new” to pursue

0 Comments

Reply your comment:

Your email address will not be published. Required fields are marked *