Hi! Affirms scores (evaluates the risk) of each transaction in real time, which allows then to have reasonably low bad rates. Very few lenders can do it.
Until there's a credit crunch or a recession, in which case all that "real-time" risk management will be moot and losses will be massive.
I am curious why you think 'real-time' risk modelling in particular might be deficient? For the record I agree, but my issues are with fintech lenders desire to lend out money too freely (models that are designed to pump lending #'s, not repayment).
Banks have been saying that about online lenders for over decade now…and yet LendingClub, Upstart and SoFi loan books are still fine (and this three companies are taking 1/3 of all personal loan originations in the US)
We haven’t had a real recession in over 10 years, so that makes sense. Unless you count the very brief one back in mid 2020, which was mitigated by a flood of federal and state dollars, mortgage and student loan moratoriums, etc.
Why is this an advantage? The factors in assessing the risk are the creditworthiness of the customer and the amount. Also, I find it hard to believe that the creditworthiness of a customer actually fluctuates fast enough for Affirm to be able to do some arbitrage on it.
Time of purchase, purchased goods, user behavior during the purchase, etc are additional inputs into a scoring model
The ol’ Algorithm-as-Patsy strategy. CFPB continues to sleep at the wheel, as designed.