When people ask to borrow money, the words they use can predict whether they will pay it back – Quartz
The words people use say a lot about their personality, emotions and thinking. And the ones they use when asking to borrow money, at the end of the day, also say a lot about their likelihood of repaying others.
Borrowers who try to appeal to the emotional side of their potential lenders – by mentioning God, divorce, or the needs of family members, for example – are less likely to meet their loan obligations (something to keep in mind when our buddies or family members try to make fun of us).
On the other hand, people who mention aspirations like higher education or marriages are more likely to repay their loans.
The findings were presented to the Boulder Summer Conference on Consumer Financial Decision Making this week and comes from research by Columbia University professors Oded Netzer and Alain Lemaire, and University of Delaware professor Michal Herzenstein. Research is ongoing and has not yet been published.
The team analyzed the text of more than 18,000 loan applications made between 2007 and 2008 via the US peer-to-peer lending website. Prosper to measure which words appeared most often in the 33% of loans from that period that eventually went into default. (Remember, the United States was on the verge of a major recession then, and Prosper was just getting started.)
They found, for example, that the word God was 2.2 times more likely to appear in a loan application where the borrower ultimately defaulted. Borrowers who used the term “payday loan” in their applications were 2.7 times more likely to default than average.
Being nice (using words and terms such as “hello”, “thank you” and “to your blessings”) correlated with higher default rates, as did future use (“will not” , “Will have”). The two tend to correlate with making promises that a person doesn’t intend to keep, Netzer told the conference. Meanwhile, people who referred to financial stability signals and demonstrated financial literacy with phrases like “excellent credit” and “lower interest rates” were more likely to repay their loans.
The group found that analyzing words used in initial loan applications could improve the ability to predict defaults by 4% to 5.7% compared to information traditionally used like credit scores and others. financial measures, which the researchers said could have saved Prosper’s lenders from losing over $ 1.4 million in loans gone bad.
Discoveries take on greater importance in a world where a an increasing amount of lending is starting to occur on the internet, without a human loan officer who once could have adopted questionable body language.
Instead of real people, we can find meaningful traces of personality in the words people use, Netzer said.
Certainly, making lending decisions in part based on a borrower’s grammar, spelling, and word choice could unintentionally bias lenders against groups like immigrants with poor English.
“One of the perks of being an academic is the ability to study things without thinking about the ethical issues of implementing those processes,” Netzer said. “When it comes to people, banks and lenders actually using these techniques, ethical issues absolutely have to be taken into account. “
|Words more likely to be mentioned by defaulters||Words more likely to be mentioned by payers|
|Payday loan||Excellent credit|
|Give me a chance||University|
|Will not do|