Technical underpinning of Credit scoring for Saccos in Kenya
“Credit Scoring” means allocating default probability to every credit applicant.
Many SACCOs in Kenya have shifted their operations from manual to digital. With most of them having scaled up their operations with the help of technology. Systems are supposed to facilitate the integration of internal risk scoring as well as external credit scores which provide intelligence on a customer’s probability of loan default.
SACCOs that have embraced technology have witnessed high trust and reliability on the financial services they offer. Engineering a credit scoring system is a more highly technical and specialized area of work. For application scoring, the applicant’s form is the main primary source of data whereas the credit firm can get other sources of information from the credit reference bureau, bank statements and any predictor variables of default of the customer.
It is very important to note that scoring systems do not provide credit decisions. They only give the credit firms a score and a probability of default. And therefore, it is the responsibility of the credit management team to ensure their ability to move from the probabilities of default to the actual decision on “accept” or “reject”.
This leads us to, engineering the Score.
- Subjective scoring; This relies on the experts input, the loan officer and the organization to produce a qualitative judgment on the customer’s credit score.
- Statistical scoring; This credit scoring model relies on quantified characteristics of the prospect’s portfolio history recorded in a database for a period of time. It quantifies set of rules and statistical techniques to forecast risk as a probability. Credit scores are brought together to a scorecard by collection of data variables with customer characteristics such as; age, source of income, industry, credit bureau history, family responsibilities, residential status just to mention a few.
- Predictors and their stability effectively predict the credit behavior of the credit customers. Assigning weights to each predictor of a new credit applicant depending on the importance. The sum of the weights becomes the applicant’s score.
Engineering a good credit scoring system is a highly technical and specialized area that requires high level of management and technical expert involvement.
So what’s next after Credit Scoring??
Look out for the next blog.
Why a “Quality Collateral” should be part of a good credit management procedure for SACCOs in Kenya.