Having a powerful default prediction model is not sufficient to guarantee the efficiency of the acceptance process for retail customers. Setting appropriate acquisition strategies is essential in order to ensure that the bank is accepting applicants with the desired credit profile and is minimizing the cost of the acceptance process. Especially in the last ten years, banks have done significant investments in models and tools to automate the acceptance process and reduce the number of referred applications and consequently the number of analysts needed to manually assess the quality of the applicants. However, if the right strategy is not set, these investments may prove to be useless and only add more complexity to the application process. In this paper, we analyze the different alternative methodologies that can be applied to set an optimal credit acquisition strategy. When a credit scoring model is available, setting the credit acquisition strategy mainly consists in defining the appropriate cut-off score and then implementing it in the correct way. We improve upon the existing literature analyzing the issue also from a business perspective as we are convinced that the optimum cut-off value cannot be found without a careful consideration of each particular bank peculiarities (e.g. tolerance for risk, profit-loss objectives, recovery process costs and efficiency, possible marketing strategies). Finally, we test our conjectures on a sample of credit card applicants collected during the year 2007 by an Italian bank.
When financial institutions lend funds to consumer or commercial clients, they follow a regular cycle (also called credit cycle) that can be divided into four phases: i.e. acquisition, account maintenance, collection/recovery and write-off. The first two phases are common for all customers, while the last two apply only to the delinquent/defaulted ones. The acquisition phase is when the credit quality of the lending portfolio is defined. Broad acceptance criteria will worsen the credit quality of the portfolio, tight acceptance criteria will improve it.
The acceptance (or approval) phase for a corporate client is a relatively length process leaded by a credit analyst that often needs to verify the information provided by the client, discuss it with the relationship manager and then take a decision. The larger the requested loan amount, the longer (and more expensive) will be the approval process due to the higher number of information to be assessed.
A similar process would not be efficient for retail customers. As a consequence of the high number of transactions to be processed per day and the relatively small amount of the loans, banks had to build a quicker and cheaper acquisition process for retail customers, consumers and small and medium sized enterprises (SMEs). Since the 80's, automated decision systems have started to be applied first to consumer applicants and later to SMEs. More recently, increasingly sophisticated statistical techniques have been applied to develop more powerful credit scoring models and set optimal acquisition strategies.
Unlock the rest of this paper by providing your email address below
You will receive an email with a link to download the paper.
Your request has been registered. Please check your email inbox for a link to the research paper. There is a possibility it may end up in your junk folder so please check there.
Other articles you might enjoy
A fifty-year retrospective on credit risk models, the Altman Z-score family of models and their applications to financial markets and managerial strategies
Abstract Fifty years ago, in 1967, I completed my PhD dissertation…Read more
Credit Risk Scoring Models
Abstract Credit scoring models play a fundamental role in the risk…Read more
Like these pieces? Register for our newsletter to get updates, learn when we publish more, and recieve the latest news from Wiserfunding straight to your inbox.
Thank you for subscribing to our newsletter you should receive an email shortly. Please check you junk folder if you cannot find it.
Join our newsletter and receive quarterly updates about Wiserfunding and our progress.