Loan Loss Provisions and and Income Smoothing-evidence From Commercial Banks in Kenya

Abstract

According to CBK's 2017 report, Loan Loss Provisions (LLPs) accounted for 12% of the total expenses incurred by commercial banks in Kenya in 2017. Given the magnitude of the expense and their role, it is clear that LLPs play a critical role in indicating the banking sector's stability and soundness. Regulators demand that for the expected losses on the loan portfolio, sufficient LLPs should be kept but there is no agreement for what sufficient or adequate LLPSs really are. The guidelines for estimating LLPs allow for managers to exercise their own discretion in estimating what they would consider sufficient LLPs. This provides room for the managers to use the LLPs estimate to achieve other objectives which are not related to the credit. This study sought to determine the relationship between loan loss provisions and income smoothing among commercial banks in Kenya. The population for the study was all the 43 commercial banks operating in Kenya. The independent variable for the study was income smoothing as measured by EBTP divided by beginning total assets. The control variables were non-performing loans as measured by the ratio of NPLs to beginning total assets, loan growth as measured by change in outstanding loans, economic growth as measured by change in GDP growth rate, being listed at the NSE as represented by a dummy and being audited by a big 4 firm also measured using a dummy. Loan loss provisions as measured by provisions for losses divided by beginning total assets was the dependent variable. Secondary data was collected for a period of 11 years (January 2007 to December 2017) on an annual basis. The study employed a descriptive cross-sectional research design and a multiple linear regression model was used to analyze the relationship between the variables. Statistical package for social sciences version 21 was used for data analysis purposes. The results of the study produced R-square value of 0.239 which means that about 23.9 percent of the variation in loan loss provisions of commercial banks in Kenya can be explained by the six selected independent variables while 76.1 percent in the variation of loan loss provisions was associated with other factors not covered in this research. The study also found that the independent variables had a weak correlation with loan loss provisions (R=0.489). ANOVA results show that the F statistic was significant at 5% level with a p=0.000. Therefore, the model was fit to explain the relationship between the selected variables. The results further revealed that only non-performing loans and choice of an auditor produced positive and statistically significant values for this study. Income smoothing, loan growth, GDP growth rate and being listed at the NSE were found to be statistically insignificant determinants of loan loss provisions among commercial banks. This study recommended that adequate measures should be put into place to control and regulate prevailing levels of non-performing loans as they significantly influence loan loss provisions among commercial banks. The study further recommends that banks should make an informed decision before settling on an auditor as this too has an effect on loan loss provisions.