Events

We collaborated with a credit union to conduct a randomized test of whether a loan product designed to improve credit scores does indeed improve consumer credit health and whether willingness to take-up the product provides otherwise unobservable information to lenders.  By randomizing the intensity of the loan application process (requiring some consumers to complete a financial education course prior to receiving product access), our experiment generated an 18 percentage point (60 percent) differential in participation rates in the loan product.  We do not find any causal effects of the product on average,  however we do find important heterogeneity – credit scores increased for consumers without any existing loans at baseline and decreased for those with existing loans due to increased delinquency on these existing credit obligations.  We also find that those who select into the product are more likely to increase their credit score (irrespective of the product).  Our results suggest that credit building loans reveal valuable information to the market and can help consumers if appropriately targeted.