Simplifying Credit Risk Scoring for All.
You might have come across various materials, articles, books on the subject of credit scoring/ risk scoring or credit risk scoring; Most of the authors must have applied certain algorithms, statistical terms and applications or mathematical approach to construct the models that predicts an outcome- Right?
Yes! that is what credit score in its applicability sense entails. However, I can predict your reading of those works got boring and you quitted quite early. The objective here is not far-fetched from what other authors were trying to explain but this author aims to give you an adequate understanding by getting your attention level on the high to the end of this read.
Credit Risk Scoring is a predictive model tool that is used to assess and evaluate the level of credit risk associated with a prospective applicant/customer. This model is a derivative of factors/criteria as defined by a firm which also are attached with weighted scores referred to as score cards. The results of the credit score model does not detect “good” (a positive behavior expected) or “bad” (a negative behavior expected) applications on an individual analysis. However, it provides statistical odds, probabilities that an applicant with any given score will turn out “good” or “bad”.
The score model does not only predict these outcomes, it also accurately assigns expected interest rates to be applied, expected profit margin/loss which are used for decision making. Simply put, credit score cards are group of characteristics that predicts good and bad account in a statistical approach.
The data uploaded to the score model is made available to the lender at the time of application. The sources of this data are not limited to- age, time at residence, duration/time on present job, payment performance, delinquencies, gender, amongst others.
Here is how it work….. Considering age as one of the score card characteristics, the options under “age” is considered attributes, e.g. 23–25/26–28/9–31 are considered attributes. The score model takes into consideration the various factors such as the predictive strength of the characteristics, strength of the attributes and the relationship of the characteristics/attributes with operational factors. The total score of an applicant is the sum of the scores for each attribute present in the scorecard for the applicant.
Risk score information as well as other business considerations (approval rate, revenue/profit potential at each risk level) can be used to develop new application strategies that will maximize revenue and minimize bad debt. The risk score will suggest approach in dealing with high-risk applicant which can include-
· Charging a higher interest rate on a loan
· Charging a higher insurance/premium fee on the loan
· Adjusting the repayment structure plan for business/SME customers.
· Reducing the obligor limit i.e offering a lower starting credit limit for high risk customers.
· Asking high risk applicants for further documentation like guarantorship/collaterals
· Declining credit if the risk is too high.
The action points above relates to NTB (New-to-Bank) clients. The credit risk scoring could differ after relationship has been established, i.e behavioral data with the company is ascertained which is used to predict the probability of ongoing positive behavior. The credit score model will suggest the following approach for existing client-
· Increasing or decreasing credit limits.
· Upgrade products to better customer
· Allowing better customers to use credit cards even in delinquency, while blocking the high-risk ones immediately.
· Offering better pricing on loan
· Pre-qualifying clients for express approvals on reapplied credit.
Risk scoring, in addition to being a tool to evaluate levels of risk, has also been effectively applied in other operational areas, such as:
· Forecasting
· Compare quality of credit per sector/industry/region or channel
· Reducing TAT (Turn-around-Time) for processing loan application by applying automations in the decision making process thereby increasing customers’ satisfaction
· Rationalization of the decision-making process. Through recognizing the higher risk client, application can be shared to more experienced staff for more scrutiny and “detailedness” while low-risk application are assigned to junior staff. This also can be done not just in risk department but at the collection department, branch level (customer on-boarding phase) and remedial unit.
In conclusion, we have been able to understand that credit risk scoring procedure provides lenders with an opportunity for consistent and objective decision making, based on empirically derived information. The credit score model provide risk managers with added efficiency and control over the risk management process.
Professional Advice-
“The only virtue of being an aging risk manager is that you have a large collection of your own mistakes that you know not to repeat.””
Donald Van Deventer
Today’s Challenge: Should the functions of risk scoring be outsourced to digital firms or in-built on company’s CBA?
© Timothy Akinyomi ¦ timothyakinyomi@gmail.com
