Fair, Isaac & Co., a pioneer of credit scoring, has been stumped like everyone else when it comes to the causes for this latest spike in bankruptcy filings.

In a recent proprietary study comparing a six-month period in 1994 with the same time frame in 1995, the California-based company couldn't pinpoint the causes for a 25% rise in filings.

Helen Wilby, project manager, for Fair, Isaac's credit bureau and promotional services, said the research was designed to answer the questions: "Is the new wave of bankruptcies seen in 1995 a new breed?" and "Are the current tools that we have to manage bankruptcy still the best?"

In looking at the bankruptcy data - specifically how it played out for bank card issuers - Fair, Isaac found the main distinction was the increase in the number of recent bank card openings.

On accounts open less than two years, Fair, Isaac saw an increase in total balances on all cards. Also, it found more bank cards with credit lines more than 75% utilized. Accounts opened two years or more had an increase in bank card and revolving credit lines.

At the time of bankruptcy, consumers in 1995 carried an average of $4,100 on bank cards - a $300 increase from 1994.

Ms. Wilby said that the company's customers have expressed concerns that credit bureau scores wouldn't work to predict bankruptcy. However, in this study, Fair, Isaac found that the best predictors of bankruptcy in 1994, including those scores, remained the same for 1995.

From the information that credit card lenders keep, those predictive characteristics are utilization, amount currently past due, payments as percentage of debits, time on books, and cash advances. From the credit bureau, they include: revolving utilization, number of bank cards 75% or more utilized, number of bank cards with a balance, number of finance trade lines, worst delinquency in last 12 months.

Fair, Isaac compared three products - two of its own, and a behavior score based on lender data - and found all predict bankruptcy, but not as well they did the year earlier.

The bankruptcy-specific score did identify a few more bankruptcies, Fair, Isaac found, but it also eliminated revenue.

In trying to steer clear of problem borrowers, lenders run the risk of rejecting safe and potentially lucrative accounts, Ms. Wilby said.

"We looked at each of those lenders individually looking at chargeoff rates," Ms. Wilby said. "We found different things will predict for different lenders. I think it's very complex. It depends on where you are in the marketplace."

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.