analyzing credit score migration

Analyzing credit score migration

Date: June 26, 2020

Study examines business-strategy implications of credit score changes over time

As surely as human behavior can change, so can consumer credit scores. But just how often do scores change? And by how much? The latest VantageScore white paper, Consumer Credit Score Migration, has the answers, and valuable insights into their implications for lender business strategies.

Each consumer credit score is a snapshot—a measure of an individual’s credit worthiness, as reflected in his or her credit file at a specific moment in time. The credit files maintained by the three national credit reporting companies (Equifax, Experian, and TransUnion) reflect consumer behaviors—timely (or late) payments, moderate (or excessive) use of available credit limits, and so on. If a consumer’s behavior changes, for the better or otherwise, his or her credit score will change accordingly.

To understand the frequency and extent of those changes, VantageScore studied anonymized credit file data for 2 million U.S. consumers. The files, furnished by Experian, were used to calculate VantageScore 3.0 credit scores once per quarter over a two-year period, from 2011 to 2013. Changes in scores over 90-day and 12-month intervals were analyzed and the following trends emerged:

  • After a three-month interval:

    • 49 percent of the consumers experienced credit score improvements, averaging 19 points; and

    • 30 percent of consumers experienced score decreases, averaging 24 points.

  • After 12 months:

    • 51 percent of consumers exhibited credit score improvements, averaging 27 points; and

    • 38 percent of consumers experienced credit score decreases, averaging 34 points.

To understand how these score changes could affect a lender’s business strategy, the study tracked how score migration affected consumers’ ability to meet, exceed, or fall below a hypothetical score cut-off of 620.

Sixty-five percent of consumers initially passed a 620 credit score cut-off. When re-scored three months later, three percent of the population saw scores drop from 620 or better to levels that fell below the cut-off, while seven percent saw sub-620 scores increase to a level that met or exceeded the cut-off.

After an interval of 12 months, six percent of the population that initially met the 620 credit score cut-off had scores below that threshold, while 13 percent of the total initial population saw scores rise to 620 or higher.

Accounting for score migration above and below the credit score cut-off, the study found that 6.4 percent of the total population would have received the opposite credit decision if they had been scored three months later.

The study also found that improving risk levels arising from a more stable economy partially offset the exposure related to a consumer with declining credit scores. The white paper cites an example of how the probability of default can shift from a 29 percent increase down to a 12 percent increase after factoring in the variance due to shifts in the economy.

“A central concern for lenders using credit scores is determining how the change in a consumer’s score will impact them over time,” said Sarah Davies, senior vice president at VantageScore Solutions for analytics, product management, and research. “The research demonstrates that a percentage of consumers are experiencing improvements in credit scores over a three- or 12-month period. While consumer scores will change with some degree of frequency, lenders should incorporate clarifications to help determine whether they represent meaningful changes. Such factors should include improved risk levels arising from a stabilized economy, which may present an opportunity to lower cut-offs and increase access to credit.”

Credit and risk strategies often include binary decisions on the strength of a consumer’s credit, as evaluated based on the consumer’s score value at the time of the loan application. The white paper suggests that the combination of insights from two components of credit score model design can provide a framework for defining whether there has been a meaningful change in a consumer’s credit score.

VantageScore also suggests that lenders would benefit from capturing consumer credit scores on a quarterly or monthly basis in order to separate consumers with substantial credit score volatility from those with more stable credit score fluctuation. Using these techniques to better understand score migration could help lenders refine processes such as account management, setting loan-loss reserves, and anticipating credit-counseling needs.

For more details, download the study here. A companion infographic, highlighting findings of the study, is also available here.