March 2008 working paper no. 330 Issn 1975-5163 Joon-Ho Hahm
Non-Interest Income Ratio and Bank Profitability and Risks
Download 1.4 Mb. Pdf ko'rish
|
Non-Interest Income of Commercial Banks
3. Non-Interest Income Ratio and Bank Profitability and Risks
Now we turn to our main empirical question of whether diversification and expansion toward non-interest income contributes to a higher and more stable profit for commercial banks. In order to answer this question, we adopt four distinct dependent variables: return on asset (ROA) as a measure of profitability, standard deviation of ROA as a measure of profit variability, equity-asset ratio as a measure of capital adequacy soundness, and z-score as a measure of bank insolvency risk. The z-score is constructed by dividing the sum of average ROA and average equity-asset ratio by standard deviation of ROA. It can be understood as a measure of insolvency risk since it represents the number of standard deviations that negate mean earning plus capital. It evaluates a bank’s capital and profit-generating capacity against its risk-taking level. A lower z-score value indicates that the bank is exposed to higher insolvency risk. Note that, in actual estimations, we need to define intervals during which, the average and standard deviation variables are constructed. Our first approach is to exploit three distinct sub-periods: 1992-96, 1997-2001, and 2002-06. For each bank, we computed sub-period mean ROAs, standard deviations of ROA, mean equity-asset ratios, and z-scores. We also computed the sub-period mean values for bank specific factors as well as macroeconomic variables, and estimated the following regressions: 18 t i i t t i t i t i t i CT PD MCF BSF NIIR BPF , , , , , ' ' (2-1) t i i t t i t i t i t i CT PD MCF BSF NIIR BPF , 1 , 1 , 1 , , ' ' (2-2) i i i i i i CT MCF BSF NIIR BPF ' ' (2-3) Where, BPF denotes bank performance variables including ROA, standard deviation of ROA, equity-asset ratio, and z-score. NIIR is non-interest income ratio of individual banks, and BSF and MCF are vectors of bank specific and macroeconomic variables as before. PD and CT are period and country dummy variables respectively. The regressions are estimated using the pooled OLS method. Note that, while we use contemporaneous explanatory variables in (2-1), we lag explanatory variables by one sub-period in (2-2) to avoid potential endogeneity problems. Note also that, in regression (2-3), we estimate the simple cross-section regressions across banks using the mean and standard deviation variables computed over the whole sample period of 1992-2006. The bank-sub-period regression estimation results on (2-1) are summarized in Table 3. Note that the table is organized by three different sets of regressions – first, bank specific variables only, second, with macroeconomic variables included, and third, by adding an interaction variable between the stock market capitalization to GDP ratio and non-interest income ratio. This interaction variable should capture any non-linear accelerating effect of non-interest income through capital markets. First, consider the regression results with bank specific variables only. Bank 19 profitability as measured by ROA is positively associated with non-interest income ratio after controlling for the effect of bank size, equity capital to asset ratio and loan ratio. Note also that the non-interest income ratio is significantly positively associated with profit variability as measured by standard deviation of ROA, while the bank asset size is negatively associated with standard deviation of ROA. The non-interest income ratio is also significantly positively associated with equity-asset ratio, while it has no impact on the z-score. In sum, the expansion toward non-interest income seems to enhance bank profitability and capital adequacy. However, since it increases profit variability as well, it has little impact on bank insolvency risk. When we include macroeconomic variables, non-interest income becomes insignificant in the ROA regression, while it remains significant in the ROA standard deviation and equity-asset ratio regressions. It is interesting to note that real GDP growth and stock market capitalization are positively associated with bank ROA, while inflation and real interest rates are negatively associated with ROA. As for the profit variability, Real GDP growth is negatively associated with standard deviation of ROA, while inflation and real interest rates are positively related to the standard deviation of ROA. Hence, economic growth seems to favorably shift banks’ return-risk tradeoff, while higher inflation shifts the tradeoff negatively. When we include the interaction dummy, non- interest income becomes insignificant while the interaction term becomes significant. In sum, the evidence in this table indicates that the expansion toward non-interest income tends to increase profit variability rather than stabilizing bank revenue. However, 20 since the expansion also raises equity-asset ratio and possibly ROA of banks, there seems to be no significant impact on the insolvency risk of banks as measured by z-score. Moreover, in the channel through which non-interest income tends to be associated with higher profit variability, stock market developments seem to play an important role. Regression results with explanatory variables lagged one period are summarized in Table 4. In general they are consistent with the results reported in Table 3 despite the fact that we lost almost a third of observations. Two major differences are that the interaction term now becomes insignificant, and that the size of bank assets tends to be negatively associated with z-score. Hence, the non-linear effect of non-interest income through stock markets discovered above should be interpreted with due caution. Table 5 reports estimation results for the cross-section regressions across banks based upon whole period average and standard deviation variables. In general, the evidence is consistent with our previous findings. The non-interest income ratio is significantly positively associated with ROA, standard deviation of ROA and equity-asset ratio even in the presence of macroeconomic controlling variables. The regression also shows a non-linear, positive correlation between the non-interest income ratio and ROA variabilitiy potentially through stock markets. Note again, since the non-interest income raises both the numerator (ROA and equity-asset ratio) and the denominator (standard deviation of ROA), it has no significant impact on z-score of banks. |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling