Bio

 
Photo of

Santiago Restrepo Escalante is a Colombian Business Administrator with some experience in data monetization in the telecommunications industry. I’m a tech fanatic who will introduce you to some of the technologies that are shaping the present, so you will be relevant in the future.

 

A study conducted by Queen Mary University of London of more than 2,600 banks in 86 countries, showed that high levels of financial inclusion contribute to bank stability [1]. But, no matter how vital financial inclusion is for the development of the sector and the fight for inequalities, entities cannot take the risk of bringing unviable people to their credit scoring models. Information asymmetries make the credit market inefficient, since lenders do not have access to the information they need for their decision-making processes [2]. This imposes stricter measures to counteract the risk associated with the lack of information in the placement of loans, leaving millions of people excluded from the financial sector. However, a branch of Artificial Intelligence promises to solve this problem that has prevented both the expansion of markets by banks and the reduction of poverty through access to credit.

Fintechs (companies that apply new technologies to financial activities) have seen Machine Learning algorithms (branch of Artificial Intelligence) as their best ally. This is because these algorithms carry in-depth analysis that reduces the risk of information asymmetries. They have the potential to link all the information and evaluate it from multiple perspectives, to obtain new knowledge that aims to reduce the information gap between the lender and the person applying for a loan.

These companies have been in charge of collecting large amounts of alternative data that feed the Machine Learning algorithms, so that they can discover correlations that were not previously considered. These alternative data include information on applicants for a loan, which at first glance seems to be unrelated to the ability to pay the loan, but have proven otherwise. That is why Fintech companies collect data such as the frequency in which capital letters are used and the speed with which the user moves the mouse in online credit forms [3]. This is because Artificial Intelligence has determined that this type of data can significantly explain the ability to pay from the people who need a loan. Consequently, it becomes possible to offer credit to new consumers or those with short credit histories, who cannot meet the demands of traditional banking and therefore have had to suffer from financial exclusion.

Machine Learning can more accurately capture non-linear relationships between data and determine levels of significance never before discovered in certain variables [4]. This allows Fintech companies to grant loans to those potential consumers, but “invisible” to traditional banks. To which they can potentially grant loans without incurring in additional risks to the expected levels of loss in the loans granted to the average consumers of the traditional financial system.

For example, a study from the Federal Reserve Bank of Philadelphia says that the Machine Learning model of a famous Fintech called LendingClub, has achieved that people previously neglected by the traditional financial sector can now have access to credit. Since this model can represent a more complete and accurate picture of people’s financial life and their solvency [5]. This ensures that many of the people with optimal payment intentions are not going to be neglected due to lack of information or risks associated with it.

But the financial inclusion that Artificial Intelligence brings is not only due to the reduction of information asymmetries. This methodology to grant credits, based on the decisions of Machine Learning algorithms, also offers lower interest rates. Where the savings differentials for customers ranged approximately between 8% and 10+%, between the credits provided by LendingClub and traditional credit cards in loans that originated between 2014 and 2015 [5]. Which makes us conclude that Machine Learning not only reduces information asymmetries in order to reach people who have normally been neglected, but also makes it possible for LendingClub to reach the market with more affordable rates than those offered by the traditional financial system. And this undoubtedly fights against one of the most important barriers to financial inclusion: the high cost of financial products.

In conclusion, the innovative characteristics of Fintechs compared to the traditional banking system have allowed and will allow breaking the barriers of access of users to financial products in the world, where, with the help of Artificial Intelligence, greater financial inclusion will be generated for the general population. These technology solutions represent an excellent business opportunity, but also a clear path to humanitarian good. If you have ever received a NO requesting a loan from a traditional financial system, perhaps the next time you request it, Artificial Intelligence will say otherwise.

Footnotes

  • [1] Ahamed, M Mostak and Mallick, Sushanta (2019) Is financial inclusion good for bank stability? International evidence. Journal of Economic Behavior & Organization, 157. pp. 403-427. ISSN 0167-2681
  • [2] Greenwald & Stiglitz (1990) “Asymmetric Information and the New Theory of the Firm: Financial Constraints and Risk Behavior” American Economic Review.
  • [3] Moldow C (2015) A Trillion Dollar Market By the People, For the People: How Marketplace Lending Will Remake Banking As We Know It, Foundation Capital White Paper., Retrieved from: https://foundationcapital.com/wp /content/uploads/2016/07/TDMFinTech_whitepaper.pdf
  • [4] Aggarwal, N. (2018). Machine Learning, Big Data and the Regulation of Consumer Credit Markets: The Case of Algorithmic Credit Scoring. Autonomous Systems and the Law (Beck 2019).
  • [5] Julapa, J & Catharine L (2018). “The Roles of Alternative Data and Machine Learning in FinTech Lending: Evidence from the LendingClub Consumer Platform” Federal Reserve Bank of Philadelphia Working Paper, April 15.

Leave a Reply