Credit Scoring Series Part One: Introduction to Credit Scoring
Credit scoring is an integral part of today’s financial landscape, and credit scoring techniques are used in various industries and organizational departments. With the software tools available to today’s lenders, there’s never been a better time to learn more about how to optimize credit scoring, how to minimize risk, and what a streamlined, flexible, and powerful credit risk mitigation strategy can do for an organization. But first, we must ask: What is credit scoring, and how does it work?
Many financial and retail firms tell their customers they can “buy now and pay later” when trying to help them increase their customer base. However, both parties need to know the risks when taking such a credit decision. For both the lender and the customer, it’s paramount that lenders can be confident that customers will honor their credit obligations and pay the lender what they owe by the end of a loan term. Additionally, lenders must be able to assess each of its customers’ default risk so they can decide who they should offer loans to, and who they shouldn’t.
What is Credit Scoring?
Technological advances have given modern financial lenders an unparalleled capacity to reduce lending risk, now that lenders can utilize more customer data than ever to make informed, considerate lending decisions. Using statistical and machine learning techniques, lenders can analyze data and boil it down to a single value called a credit score, which represents a customer’s lending risk. The best part about credit scores is that they’re easy to use and interpret – the higher the credit score, the more likely a customer is to pay back their loans, and vice versa. Credit scoring is a form of artificial intelligence (AI), based on predictive modeling, that assesses the likelihood a customer will default on a credit obligation, or become delinquent or insolvent. The predictive model “learns” by utilizing a customer’s historical data alongside peer group data to predict the likelihood a customer will exhibit a defined future behavior.
One of the reasons credit scoring is so useful is because it gives lenders the ability to make fast, efficient decisions regarding whether to accept/reject a customer or increase/decrease loan values, interest rates, or loan terms. This speed and accuracy has made credit scoring a foundational risk management tool across diverse sectors like banking, telecommunications, insurance, and retail.
Credit Score Types and Customer Journey
Credit scoring is also flexible – lenders can utilize credit scoring throughout the customer journey, starting from the moment the relationship between lender and customer is established.
As depicted in Figure 1, different credit scores are utilized at different stages of the customer journey:
- Application score: Assesses new applicants’ default risk, informing lenders whether to accept or reject the applicant
- Behavioral score: Assesses the default risk of existing customers when making decisions relating to account management such as credit limit, over-limit management, and new products
- Collections score: Used in the collections stage to assess how likely a customer in collections is to pay back their debt
Figure 1. Credit scores throughout the customer journey
Credit Risk Scorecards
Over the years, several different techniques for implementing credit scoring have evolved. They range from parametric or non-parametric, to statistical or machine learning, to supervised or unsupervised algorithms. The most recent techniques include highly sophisticated approaches that utilize hundreds or thousands of different models, various validation frameworks, and ensemble techniques with multiple learning algorithms to obtain better accuracy.
But despite this diversity, there’s one modeling technique that stands out – the credit scorecard model. Usually referred to as “standard scorecard,” the model uses logistic regression as the underlying model. Easy to build, implement, use, tweak, and monitor, standard scorecard is the favored approach among practitioners and is used by nearly 90% of scorecard developers. As a statistical/machine learning hybrid, its prediction accuracy is on par with other sophisticated techniques and lenders can use its scores as probability estimates – thus giving lenders direct input for risk-based pricing. This is critical for lenders that comply with the Basel II regulatory framework. And since credit scorecards are easy to interpret and justify, regulators in some countries mandate them as the exclusive risk modeling technique.
A scorecard model result consists of a set of attributes (customer characteristics) that are usually displayed in a table (Figure 2). Within an attribute, weighted points (either positive or negative) are assigned to each attribute value in the range, and the sum of those points equals the final credit score.
|Age||Up to 25||10|
|26 to 40||25|
|41 to 65||38|
|66 and up||43|
|Income||Up to 20k||-10|
|21k to 40k||16|
|41k to 70k||28|
|71k and up||45|
|Bureau Score||Up to 300||-25|
|300 – 500||0|
|500 – 650||30|
|650 – 750||50|
|Total Score||Sum of Points|
Figure 2. Standard Scorecard Format
Credit scoring is a dynamic, flexible, and powerful tool for lenders, but there are plenty of ins and outs that are worth covering in detail. To learn more about credit scoring and credit risk mitigation techniques, read the next installment of our credit scoring series, Part Two: Scorecard Modeling Methodology.
And click here to learn more about Altair’s credit scoring, credit risk, and financial services solutions.