The Effects of Collateralization on Loan Performance of Commercial Banks in Kenya

Abstract

In the recent years, banking industry in Kenya has witnessed unprecedented rise in default rates which has in turn led to massive increase in the stock of nonperforming loans across all the banks’ portfolio thus reducing profitability. The main source of banks’ revenue is the interest earned on loans and advances, and as such, portfolio quality is the greatest driver of a bank’s profitability and growth. In view of the above, banks have adopted different ways of credit risk mitigation strategies, inter alia, collateral lending. The objective of the study was therefore to establish the effects of collateralization on loan performance of commercial banks in Kenya. The dependent variable in the study was loan performance measured using NPL ratio. The main independent variable was collateralization. Other control independent variables were bank size and financial performance. Data was extracted from secondary sources which included published audited annual financial reports of commercial banks in Kenya and Central Bank of Kenya annual reports and surveys for the period between 2016 and 2020. The study adopted census method whereby all the commercial banks in Kenya were studied. However, only 38 banks out of 40 were in operation during the study period translating to 95% response rate. The data was subjected to other tests like normality, autocorrelations, Collinearity and the results were impressive and conformed to the required standards. Regression model was used in the study to determine the effects of collateralization on loan performance. R-square was found to be 26.1%. This represented the percentage of the variation in the dependent variable that is explained by predictor variables. F statistics value was found to be greater than F critical value, and that led to a rejection of the null hypothesis. The study established F statistics value by way of ANOVA of 4.246 and F critical value of 3.26 at degrees of freedom of 3 and 36. P-value established was 0.011 which was less than alpha value=0.05. Null hypothesis, therefore, was rejected and the conclusion was that Regression model was significant hence fit for predicting effects of collateralization on loan performance. The finding of the study shows that collateralization was statistically significant in predicting the performance of loans of commercial banks in Kenya, with a p-value of 0.006. The study revealed negative significant statistical relationship between Collateralization and loan performance with a determinant of -3.239 established in the model. This means that increasing loan collateralization by one unit would lower non-performing loans by 3.239 units.