ABSTRACT
It has often been said that credit management is an art, not a science. It is true that in consumer credit, much of the decision-making process, and the predictability of accounts being good or bad, has been increasingly performed in the last 40 years by a range of scorecard products and services. It is not difficult to see that statistical probability, the basis of consumer credit scoring, is comparatively accurate founded as it is on definitive criteria such as age, employment, marital status and so on – evidence of stability, in other words. It is also not difficult to see (though not always easy for us individual humans to accept) that we consumers can be ‘categorised’. Men of a certain age in a certain social group are more likely to ‘x’, while women of the same age in the same group will most probably ‘y’. Car drivers over 45 are safer than car drivers under 25, and so on. The accuracy of predictions regarding the consumer comes from the millions of items of data that can be processed, researched, analysed and experienced, and that science is now well established.
