Most of us have heard about risk scores whether we are seeking to rent a property or applying for a loan. A risk score measures the likelihood of a customer defaulting within a certain time frame. It acts as a tool to help in understanding someone’s probability of missing payments and eventually ending up in collections.
However, the less common, yet still very significant scoring model which we tend to overlook is the bankruptcy score. So, what is it exactly?
The bankruptcy score measures the likelihood of filing for bankruptcy within a period of time. For lending organizations, identifying customers who may be planning to file bankruptcy ahead of time will mitigate future losses by reducing risk and lowering acquisition costs.
Having developed both risk and bankruptcy models in the past, I was able to get insight into the predictive characteristics that represent the behavior of these accounts. Surprisingly, some of the characteristics that bankrupt accounts depict are the same characteristics as those with a good risk score. It is common to see good standing customers declare surprise bankruptcies while those that are delinquent do not. Customers with a high bankruptcy rate tend to have higher utilization rates, have recently been highly seeking credit and have fewer accounts in collections.
The process of developing bankruptcy scores can be a timely and complex project. Fortunately, this is achievable with the help of data scientists. Take action by developing predictive bankruptcy scores and don’t get caught by surprise!