The Federal Housing Finance Agency (FHFA) announced the approval of two credit score models: FICO 10T, an updated version of the current Classic FICO, and VantageScore 4.0, for use by the government-sponsored enterprises (GSEs). In addition, lenders would also have the option to move from tri-merge credit reporting to bi-merge reporting. This report analyzes the differences between the current credit score model (i.e., Classic FICO) and the new VantageScore 4.0 credit scoring model in mortgage lending and sheds some light on the differences between the bi-merge method versus tri-merge. Using a large dataset of mortgage loans, we examine how the Classic FICO and VantageScore 4.0 systems evaluate borrower segments and predict mortgage delinquencies. The data do not allow us to compare the more updated version of FICO 10T.

We find, on average, VantageScore 4.0 scores are higher than Classic FICO scores, especially for refinance loans and for investor properties and second homes. Both credit scoring models effectively distinguish between high-risk and low-risk borrowers. VantageScore 4.0 is marginally more effective at identifying high-risk borrowers from among those with the lowest credit scores, though the differences are small. Further analysis between bi-merge and tri-merge reporting shows that moving from tri-merge to bi-merge reporting does not make a significant difference in detecting loan default probabilities.

One big caveat is that the data do not allow us to see loans that were previously denied or loans that were made as Federal Housing Administration loans that could become eligible for GSE financing, so we cannot assess how many of these would be approved with the new scores and inputs. This limited analysis suggests that there would be little initial impact in the two transitions proposed for borrowers previously qualifying for GSE mortgages, though even there, we could see a larger impact over time because of the increase in competition this change is likely to bring.