University of Salford

ELM’s customers tend to have poor credit score ratings and can be vulnerable to exploitation by money lenders and loan sharks. The company’s aim is to provide ethical finance to tackle financial exclusion and reduce the risk of bad debts.

However, ELM faced the dilemma that many of the people who applied for its loans failed at the credit scoring stage of the application process. The company approached the University to evaluate the factors that influence a customer’s ability to pay and to develop a new system for loan approvals.

University researchers used data-mining methods to develop a new risk assessment system based on probability mathematics rather than credit scoring. The research was inspired by the probability systems used by the NHS, for example to assess the likelihood of a patient developing a particular illness.

Data-mining detects patterns in data which can be used to inform decision-making models. Previous clients’ data is used to identify good and bad performance which can then be used to predict the likelihood of a new customer paying back their loan.

By using factors such as income, expenditure patterns, age of dependents, the time of year and amount of money applied for, customers benefit from a fairer and more realistic system.

The new system enabled ELM to expand its business by 50 per cent between 2011 and 2012, increasing access to affordable finance amongst poorer members of society. Total lending by ELM to people on low incomes will top £10 million this year.

Diane Burridge, Chief Financial Officer of ELM, said: “There has been a significant improvement in the collection, storage and analysis of client data. There is also increased appreciation by the underwriting staff that social factors can have a significant impact on the client’s ability to pay.

“The project has helped us gain the confidence to move our business into more geographical areas and, as a result, more people now have access to affordable finance than would previously have been possible.”