A Comparative Study on Non-Performing Loans of Some Selected State-Owned and Private Commercial Banks in Bangladesh

Abstract

This thesis utilizes the data gathered from 5 state-owned commercial banks along with the 16 private commercial banks for the time period of 2009 to 2017 to perform a comparative analysis. Dynamic panel data model (GMM estimation) has been employed with time dummies to determine the effect of non-performing loans, efficiency, loan recovery and loan write-off. Furthermore, questionnaire survey with five-point LIKERT scale has been conducted to formulate the important comparative factors under which the state-owned and private banks can be clearly distinguished. Before performing the comparative analysis, this thesis also used the data gathered from 16 private commercial banks for the time period 2006 to 2017 to investigate the relationship between non-performing loans and its determining factors. Dynamic panel data model (GMM estimation) has also been employed in this case with time dummies to control for macroeconomic events to determine the effect of bank-specific factors on the performance of loans. Questionnaire survey with five-point LIKERT scale had been conducted to formulate the perception of bankers and borrowers about the prevailing factors affecting the condition of non-performing loans. Before the comparative analysis between the state-owned and private commercial banks, this thesis analyzes the impact of the cost of fund, average lending rate and excessive loans from a long- run perspective. In this regard, lagged per-capita GDP and real interest rate have been used to control for macroeconomic factors. Time dummies were used to replace the macroeconomic factors to find out the overall time-specific fixed effects on the condition of nonperforming loans. Additionally, this paper uses the banks size and individual bank’s loan to total loans of the banking industry as a measure of bank size and market power. This is a novel research in Bangladesh which specifically identifies the impact of lagged credit lending by banks up to four years. Moreover, this paper uses the lagged per-capita GDP up to two periods to investigate the impact GDP on the performance of bank’s loan quality. Empirical analysis in this study has found a significant relationship between lagged credit growth and the future non-performing loans. Loans which are disbursed today out of competition or aggressiveness have a huge probability of becoming default in the long-run. Furthermore, this research also uses inflation-adjusted cost of fund and average lending rate to find a significant long-run association with bad loans. Another finding of this research indicates that an increase in the cost of fund induces a positive change in the lending rate in the following period which ultimately affects the credit quality in the long-run. Empirical results have shown that there is a significant positive relationship between the banks’ “cost of fund” and “non-performing loans” in the long-run but the relationship is not significant in the shorter time frame. The results highlight a strong long-run causality between banks’ NPLs and COF in the PCBs of Bangladesh. Furthermore, results also suggest that higher NPLs for some banks can induce in the rise in banks’ COF in a bidirectional form. Moreover, the results have shown that higher COF leads to the growth of NPLs and as a result, the provision for bad loans is also increased which ultimately reduces the income of the banks. This process forces the banks to write-off loans from the balance sheet to reduce the non-performing assets from the books. In terms of the lending rate, the study found some interesting situations where higher lending rate causes the non-performing loans to fall and rise gradually. It has been found in the study that higher lending rate causes a decline in the demand of the loans which makes the current non-performing loans smaller. This situation does not prevail for a long period of time because banks ultimately increase the loan growth abruptly to maintain their earnings which causes the bad loans to increase in the long-run. Based on the questionnaire survey, this thesis brings out the perception of bankers and borrowers in identifying the bank-specific and macroeconomic factors of non-performing loans in Bangladesh. This investigation used questionnaire survey on 75 bankers and 25 borrowers in 2018 to identify the effect of the cost of fund and average lending rate on the non-performing loans. The data have been collected through face to face interview, telephone and emails. The study found political stability, loan evaluation, banking corruption, energy-crisis and borrowers honesty to be significantly related to non-performing loans. The survey used face to face interview, emails and phone calls to question about the factors of bad loans. The study used SPSS software to find the reliability of the survey. The study used the variable cost of fund and lending rate to formulate a link in analyzing the relationship between cost of fund and nonperforming loans. Finally, the comparison between the state-owned and private commercial banks revealed important comparing factors under which the distinction between these different bank groups have been put forward. GMM estimation results found that cost efficiency, loan write-off and loan recovery (which was previously written off ) does not affect the state-owned commercial banks in the same manner as they affect the private banks. Non-performing loans were found to be significant for both bank groups that affected their profitability negatively. The effect of operating cost (measure of efficiency )was insignificant on the profitability of the state-owned commercial banks unlike the negative impact it posed on the income of the private banks. The comparative analysis found loan write-off to pose a negative impact on the pre and after-tax profit of the private commercial banks but the effect was not significant for the public banks. Loan recovery had a positive consequence on the profit of the private commercial banks as expected but the recovery of the written off loans did not increase the profit of the public banks. Aside from the comparative analysis through secondary data, the study also employed another questionnaire survey through face to face interview to compare the pubic banks with the private banks. The study found non-performing loans, capital injection, use of advanced technology, corruption, branding and marketing, willful defaulters, cost inefficiency, loan against letter of credit acceptance (inland ), customer service and political influence to be the important factors of comparison.