Time series evolution of credit scores is a great tool to understand the effects of removing certain credit characteristics or adding them. These traits can play a major role in a person’s credit scores. This article discusses the effects of dropping certain credit attributes and the impact of high-cost credit scores on credit scores.
Time series evolution in credit scores
In many credit decisioning models, time series data is an important component. This data allows lenders to assess the risk of a borrower by tracking how much a consumer has paid over time. Lenders may be able to see more detail about borrowers' past history of late payments by using time series data from credit card balances.
This data is generally positive but can show a downward trend. This is especially true for consumers in lower risk and lower scoring segments. A recent drop in hard credit inquiries might be due to increased consumer attention on reducing spending and decreasing debt.
The impact of dropping credit characteristics related to groups
One study examined the impact of removing credit characteristics that are related to a credit score. Dropping the credit characteristics in question raised the average credit score by 2.5 points. That's about one-fifth. The differences were larger in people with younger credit scores compared to people with older credit scores.
Dropping a single characteristic from a credit score had very little effect on the mean score for blacks. The largest change in black credit score average was 0.1 points. This small change can be attributed to the high correlation between these characteristics and the scoring model. These differences held across the three scorecards.
Effects of adding other characteristics
Traditionally, analysis of credit scores has only examined the effects of a single characteristic, such as age. Although the effects of adding another characteristic aren't well understood, an additional characteristic may have a significant impact. The model for each scorecard was reevaluated with the additional characteristic. It was then compared to the FRB base modeling.
While the addition of race or ethnicity did not change the mean score, the inclusion of those characteristics would have an effect on the predictive value. These attributes can be removed, but it would have a significant impact on model predictiveness.
High-cost credits have negative consequences
A negative credit score can result from several factors. It sends lenders a message that the borrower poses a risk to their credit. High-cost borrowing can lead to more defaults which can have negative effects on overall financial health. The third effect of high-cost credit is on the borrower’s social reputation.
High-cost credit can reduce the demand for standard sources of financing and can restrict future access to those sources. A second reason is that high-cost borrowers may choose to take out high-cost financing, which can be more risky. This may be a good option for short-term financial problems, but it can also limit the availability of traditional sources of financing.