Choose your words carefully when you apply for a loan

Banks are very interested in predicting whether or not a debtor will repay their loan. Typically, banks value your assets or income (ability to repay), and they also check if your "demographic factors" are preferable. However, these methods have proven to be inaccurate. Economists Oded Netzer, Alain Lemaire, and Michal Herzenstein have developed a much more accurate method. The article "When words sweat: Identifying signals for loan default in the text of loan applications," published in 2019,  states that the words used in a loan application project the likelihood that a person will repay her loan.

The researchers analyzed (computer-assisted) more than 120,000 loan applications sent to Prosper (an online crowdfunding platform). The potential borrowers wrote a brief description of why they need the loan and why they can repay it. The researchers combined the loan applications with loan repayment data. The study found that the language used by potential borrowers is good at predicting how likely they will repay their loans. People unknowingly leave clues about their intentions, circumstances, and personality traits in the text they write when applying for a loan. These linguistic cues predict whether they will repay their loans much more accurately than the analysis of economic or demographic factors traditionally used.

The study found that the use of certain words is revealing. People who have failed to repay their loans used words related to their family, God, or short-term plans in their loan applications. The stories of financial or other difficulties also correlate negatively with the loan repayment. The study also found that the more you appeal to the need for help or promise to repay the loan, the more likely you are not to pay. Some of the words of those who did repay their loans indicated some knowledge about the economy. Yet other words indicate a positive outlook for the future, such as graduating, etc.

I have listed below ten typical terms used in loan applications and what they indicate according to the study:

I do not know if any bank, in reality, uses this type of method when evaluating loan applications. If not yet used, I think a similar approach will be applied in the future. After all, the indicator of the loan repayment is financially important information. So don't promise or thank too much in your loan application. Don't tell sobbing stories about family difficulties and hospital trips. Instead, describe what actions will help you repay the loan and how you generally handle your finances. Sounds reasonable.

Sources

Netzer, Oded, Alain Lemaire, and Michal Herzenstein. "When words sweat: Identifying signals for loan default in the text of loan applications." Journal of Marketing Research 56.6 (2019): 960-980.

 

Stephens-Davidowitz, Seth, and Andrés Pabon. Everybody lies: Big data, new data, and what the internet can tell us about who we really are. New York: HarperCollins, 2017.

ps. Thanks for the picture Adamr at Free digital photos