Study: Basic Numeracy Good Predictor of Future Loan Default

The National Academy of Science has published a study that links a lack of numeracy to increasing odds that a borrower could default on their subprime mortgage. According to Khadeeja Safdar of The Wall Street Journal, deficiency in simple mathematics skills could have led to the rash of defaults that ushered in the 2008 financial collapse and a global economic slowdown.

Researchers asked those who originated subprime home loans in the period between January 2006 and August 2007 to take a test that would assess not only their math skills, but also their financial knowledge, ability to communicate and their “cognitive ability.” Only those with loans in good standing were asked to participate. Borrowers who were in foreclosure or were seeking loan modifications were excluded.

Kristopher Gerardi, Lorenz Goette and Stephan Meier, authors of the report, concluded that a lack of basic numeracy skills best predicted whether participants would default on their loan in the future.

Survey questions ranged from calculating the cost of an item on sale at half price to evaluating the effect of inflation on money. Among the categories tested, the ability to perform basic math calculations was most closely associated with mortgage default. “Our analysis raises the possibility that limitations in numerical ability may have significantly contributed to the massive amount of defaults on subprime mortgages in the recent financial crisis,” write the authors.

In a phone interview with the journal, Goette dispelled the myth that the type of mortgage issued played a role in the odds of foreclosure. According to Goette, lack of basic math knowledge served as a predictor even when the loan wasn’t one of the notorious adjustable rate mortgages that made calculating the monthly payments more difficult. The study concluded that the type of mortgage was less likely to predict a default than the numeracy scores attained by participants.

In total, 339 individuals responded to the survey. The economists had information such as debt-to-income ratio, terms of the loan and credit score for all subprime mortgage borrowers (even the ones who didn’t respond). Using the data at hand, the economists concluded that there was little evidence of selection bias among the sample of respondents.

People who could not accurately add, subtract, multiply or divide were more likely to get in trouble with their repayment than those with good math skills and lower incomes.

The study also suggests that a borrower’s income is not useful for predicting mortgage default. The household income among the survey respondents varied widely with an average at $80,000. “It’s not that subprime mortgage borrowers were low-income,” said Mr. Goette. “It’s more that they had huge mortgages in comparison to their incomes.”