Credit Risk Assessment For Customers Of Kenyan Commercial Banks: A Case Of Co-operative Bank

Abstract

Over the last two decades, credit assessments made by commercial banks have been evolving. Instead of the traditional assessment of the banks’ credit experts which is subjective, increased credit risk means that comprehensive mathematical and statistical models must now be used. However, credit risk scoring applied for most commercial banks is not very effective, since a lot of defaulting exists which leaves the banks disadvantaged. The main purpose of this study is to conduct comprehensive modeling for effectively assessing credit risk. Primary data was collected through interviews with personnel related to credit control to establish the existing credit scoring methods and their viability. The target population included 50 staff members of Co-operative banks in Nairobi who comprised of credit officers, staff in risk management and staff ICT departments. The primary data was supplemented by secondary data gathered from bank records, company statistics, financial periodicals, books, journal articles and reports. The data was analyzed using descriptive and inferential statistics. The descriptive statistics include frequency distribution tables and measures of central tendency and measures of variability. Different inferential methods are tried and tested, leading to a conclusion that principal components analysis and logistic regression provide a suitable set of methods. Principal components analysis is used to identify significant variables among the many variables that can be used to assess credit risk. With fewer and effective measures of model performance, model development becomes a much more efficient process, the same goes for variable selection. Since the data used is only a small sample of the population, a resampling method is applied that is used to get stable estimation of credit scoring using a dataset of reasonable size. The developed model for credit risk scoring will inform management for decision making and provide predictive information on the potential for delinquency or default that may be used in the loan approval process and risk pricing. The study is expected to be of value to the various stakeholders who will include the management of commercial banks in Kenya and other financial institutions; to the CBK as the regulator, to the borrowers and to scholars and researchers.