School of Business Research Collection

Permanent URI for this communityhttps://research.unilus.ac.zm/handle/123456789/424

Browse

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Examining the Credit Risk Techniques and its Effect on Performance for Enhanced Productivity at Selected Financial Institutions in Lusaka, Zambia
    (University of Lusaka, 2025) ALUFEYO, Mazanga
    The purpose of this study was to examine the credit risk techniques and its effect on performance for enhanced productivity at selected financial institutions in Lusaka district. The study was guide by three objectives; to ascertain the best techniques of credit risk modelling of selected financial institutions in Lusaka district for enhanced productivity, to examine the effects of credit risk modeling techniques of selected financial institutions in Lusaka district for enhanced productivity and to evaluate the performance of credit risk modeling techniques of selected financial institutions in Lusaka district for enhanced productivity. The Agency theory served as the theoretical foundation of this study. A parallel convergent design was used in this study. The sample size comprised of 104 respondents selected through simple random sampling and purposive sampling. The data which was collected using questionnaires and interview guides was subjected to descriptive analysis and thematic analysis. The findings revealed that the effects of credit risk modeling include proactive risk management, enhanced decision making, enhanced customer satisfaction, faster loan approvals and Improved risk assessment accuracy. The study findings indicatively show that commonest or frequently utilised credit risk modeling technique is loan restructuring. This was followed by risk-based pricing (27%) and collateral management. The credit risk modelling techniques employed by a subset of financial institutions in the Lusaka district were evaluated for their ability to improve productivity using Key Performance Metrics (KPM). The results indicate that the institutions' credit risk models had an average accuracy of approximately 83.1%. In 83.1% of the situations with an average precision rate of 72.4%, the models accurately classified 72.4% of the loans that were recognised as high-risk. About 71% of the real high-risk loans were accurately recognised by the models, according to the recall rate of about 71%. The qualitative findings further indicated that there has been a noticeable decline in non-performing loans. The key informants reported an average decrease in nonperforming loans (NPLs), indicating a notable enhancement in their capacity to evaluate and handle credit risk. The study concluded by showing that financial institutions in Lusaka can improve their capacity to evaluate and responsibly manage credit risk by implementing suitable credit risk modelling approaches. Based on the findings, the study recommended that to improve their credit risk assessment procedures, financial institutions ought to think about incorporating cutting-edge machine learning models. Financial institutions ought to automate repetitive operations and incorporate model outputs straight into decision-making workflows in order to optimise the credit risk assessment process. Key words; Credit Risk Techniques, Performance, Productivity, Financial Institutions