For any key decision, a minimum amount of information is required to make an informed judgment. Consider an airplane pilot readying for takeoff. Suppose that the radar is out, or the instruments are flickering. A responsible pilot wouldn’t proceed down the runway until having all vital information to ensure a safe takeoff.
Similarly, a minimum amount of current data about a prospective borrower is required before making a responsible credit decision. We’ve previously explained the logic behind our “minimum score criteria” for the FICO® Score. But of course, that leaves millions of US consumers without scores.
We recently completed research on how to safely extend credit to these “unscorable” consumers (and in my last post, I shared insights into their credit behaviors). In a key part of our study, we analyzed the 28 million unscorable consumers with bureau files having sparse or old bureau data. We wanted to find out if it’s possible to calculate a meaningful, reliable score using credit bureau data alone.
Spoiler alert: it’s not. Research results consistently showed that scoring models relying solely on sparse or old credit data were weak and did a poor job forecasting future performance.
Consider scoring based on sparse bureau data. We developed a research score for the approximately 7 million consumers (about 25% of the group) with one or more collections or adverse public records but no other credit account data. We calculated several standard predictive measures to evaluate score performance. For these scant-file consumers, the Gini index of the score was 0.147, significantly less than the 0.600 to 0.800 Gini indices for scorable consumers. A lower Gini index means the score is less predictive of future behavior and less able to separate good from bad credit risk.
We also looked at scoring consumers with older bureau data. Using a research model with a recent national credit bureau sample, we scored consumers with no credit account updated in the last six months. We compared the odds-to-score alignment of this group against a baseline of scorable consumers with at least one credit account updated in the last six months.
The results showed that the older the data, the less reliable the implied odds of the score relative to the benchmark odds-to-score relationship observed on the traditionally scorable population. That means the risk level associated with a particular score, such as 700, will not be the same across successively more stale segments of the population.
To understand the implications, think about auto loans. Lenders setting an underwriting strategy among borrowers at a given score cutoff could be accepting consumers with markedly different repayment risk, depending on how long it’s been since the credit file was updated. For example, our research showed that a 640 score based on files that have not been updated in 21 months or more exhibits repayment risk roughly in line with a 590 score for the traditionally scorable population—an odds misalignment of about 50 points.
The bottom line is that risk discrimination is weak when scoring on sparse or old bureau data. For lenders, use of a weak score could mean declining applicants they should have accepted, and vice versa—producing higher levels of delinquency and lower lending volume than necessary. For consumers, it could mean receiving lower credit lines/loans than requested and needed or higher than they can handle.
Moreover, for most of these 28 million consumers, scoring would not make it easier for them to establish credit. About 65% have a negative item and no active account. With no positive data flowing into their files to offset the negative, they would likely score too low to obtain credit. These consumers are stuck in a catch-22: To obtain credit, they have to be using credit—but without a reliable way to assess creditworthiness, lenders may not take a chance on them.
Similarly, scoring from bureau-only data won’t help the 25 million with no credit files. They’re stuck in the same catch-22.
So what’s a better approach? I’ll answer that in my next post…but here’s spoiler #2: bureau data must be supplemented!