Speakers: Francesco D’Acunto is the A. James Clark Chair in Global Real Estate and Provost Distinguished Associate Professor of Finance at Georgetown University. His research interests are in the areas of households' beliefs and economic choices, housing and real estate, FinTech and inequalities. His work has been published in top academic journals such as the Journal of Political Economy, the Review of Economic Studies, the Review of Financial Studies, the Journal of Financial Economics, Proceedings of the National Academy of Sciences (PNAS), and the Journal of Economic Perspectives.
Recap：
Francesco D’Acunto at Boston College presented a full overview of research he has been involved in studying FinTech. D’Acunto presented four main papers that all touch on the use of digital technologies in finance, particularly on improving human decision marking.
The promises and pitfalls of robo-advising (2019)
Historically, people have relied on human investment advisors. But these advisors often show the same biases as their clients. Robo-advisors that launched aim to provide less behaviorally biased portfolios at lower fees. D’Acunto and co-authors sought to see if robo-advice can work with human advisors to provide better allocation advice to clients. They set up a study using a simple robo-advisor portfolio optimizer. They found that using a portfolio optimizer does increase diversification and thus, Sharpe ratios, for previously undiversified investors. But for already diversified investors, such a simple robo-advisor does not improve as much.
Crowdsourcing peer information to change spending behavior (2020)
An important issue societies are grappling with is the low rate of savings for retirement. A key reason is that households have little information about the optimal savings rate. Research has shown a visibility bias where people make inferences based on others, especially in an era of social media. The researchers tested whether social signals could shift behavior in a positive direction. For this, they used an income aggregator app called Status, to share information on more optimal behavior to the users. The study used two signals to give to participants. The results showed that indeed, those with deeper and broader social networks reacted to their perceived spending decisions. It also seemed to be a sharper change when people were exposed to signals towards less spending. These findings suggest the potential to use social networks and fintech to encourage better savings behavior.
Perceived Precautionary Savings Motives: Evidence from FinTech (2020)
During economic downturns, policymakers want to boost consumer spending through credit policy. This study aimed to evaluate the effectiveness of offering credit to consumers, through an overdraft facility. It used a European digital bank with over a million customers. The study aimed to see how much having access to such a facility would increase consumer spending. The results showed that those who reacted the most did not use the credit facility directly. Rather, what seemed to happen is that people have a “precautionary savings motive” where they need to keep a buffer of savings. The overdraft facility acts to reduce that motive, to allow consumers to keep less of a buffer.
How Costly are Cultural Biases? (2021)
Finally, discrimination is a critical topic these days and digital systems have the potential to improve or worsen the situation. The co-authors wanted to study discrimination, in particular they wanted to separate statistical vs. taste discrimination of systems. Statistical discrimination is when systems try to filter for attributes it wants. Taste is more discrimination based on personal biases. The scholars used a peer-to-peer lending platform in India. Since they are individual driven, these digital platforms have less scope for statistically discrimination. India sees two primary forms of discrimination. The first is in-group discrimination, most notably Hindu vs Muslim discrimination. The second is stereotypical discrimination as in the caste system. In this experiment, the P2P platform users can choose to use the automated tool, or choose their own loans. After the study, the scholars found a economically significant extent of discrimination where users to significantly preferred borrowers of their own religion. Because this may not have been the optimal borrowers, they paid in opportunity cost, with the average lender’s loss around $1,700.
In tying together this fulsome body of work, D’Acunto concluded by pointing out that the common takeaway from all these papers is that technology can help to improve people’s financial decisions and make finance more inclusive and less biased.
If you would like to
give a presentation in a future webinar, contact our Senior Economist Dr. Wen
Chen (wen.chen@luohanacademy.com). For
other inquiries, please contact: events@luohanacademy.com.
