Case Study

Improving Just One Algorithm Can Help Tens Of Millions of Americans

Improving Just One Algorithm Can Help Tens Of Millions of Americans

Pages 2 Pages

Improving Just One Algorithm Can Help Tens Of Millions of Americans To comply with federal fair lending laws, banks and credit unions must prove they don’t discriminate based on race and other protected statuses. But lenders cannot ask for an applicant’s race except in mortgage lending, and even in mortgage lending almost a third of applicants put nothing down. In the absence of data, lenders, regulators and credit bureaus have to guess. The de facto way to do that is with a 30-year-old formula called Bayesian Improved Surname Geocoding (BISG) that combines last name and geographic data to estimate the race. BISG brought objectivity to fair lending and its predictions work well in racially homogenous areas. But our country is not homogenous. BISG’s accuracy has declined a

Join for free to read