ACCOUNTING
FINANCIAL AND ECONOMICAL IMPLICATIONS OF ONLINE BANKING
Course:
Tutor:
University:
Date:
CHAPTER THREE
METHODOLOGY
The study will be based on 275 banks in Switzerland covering a period of 20 years from the year 1996 to 2015. Secondary data for the task will be obtained from Swiss National Bank Website and World Bank website .From the Swiss National Bank database,all financial statements for 275 banks will be obtainedfrom the analysis.World Bank website will provide the data on online banking usage by Switzerland banks for the period under review. To examine the financial and economic implications of online banking, financial statements will be used for the analysis. This will involve obtaining the bank’sincome statements and Balance sheets from the identified websites (Halili, 2014)
This will be a census study whereby I am using the total population of banks in Switzerland as my sample. A census study is one which uses the entire population as a sample. A population is the total number of elements whose characteristics are being studied. It is of key importance in research as it forms the basis of inferencing the findings of a sample. The total number of banks will also form my sampling frame. Sampling frame is the elements of a population from which a researcher will draw his population from(Gujarati, 2003) The Census study was chosen because it is easy to obtain all required information from the Swiss National Bank website as it regulates all banks and it’s a requirement for the banks to file a return of their yearly audited financial statements with the regulator(SWiss.National.Bank, 2016)
Multiple regression modelswill be employed to establish the effects of online banking on banks performance, economical effects and risk. Multiple regression methodologies will be used to determine the relationship between internet banking and banks performance. This will consider 20 years period from the year 1996 to 2015. This will show the kind of relationship between the variables when the banks adopted the internet banking. The descriptive methodology will be used to compare the performance of banks prior to theadoption of internet banking which is from the year 1996 to 2000 and the post-adoption period which is the year 2001 to 2015(Halili, 2014).
Proxies will be used for both dependent and independent variables. The dependent variable for this study will be banks performance which will be proxies by ROA and ROE. Operating cost will be used as a proxy for economic effects which is the second dependent variable in the study and finally, the standard deviation of banking sector returns will be used as aproxy of the third dependent variable which is Risk. By conducting the regression analysis on the three dependent variables,one will be in a position to establish if indeed online banking in Switzerland affects the three variable under investigation(Caprio et al, 2008)
Five independent variables will be for the study. They include loan to asset ratio (LAR) will be obtained by Loans/Asset, deposit to asset ratio (DAR) which is Deposit/Assets, Overheads ratio (OHSR) which is Overheads divided by Net operating revenue , efficiency ratio (ER) which is non-interest expenses divided by assets and Internet banking (INT). The LAR, DAR, OHSR,ER will be calculated from the obtained balance sheets and income statements for the 275 banks while Internet banking variable will be a dummy variable represented by 0(zero) for years when there was no internet usage in the country and 1 (one) for years when banking industry adopted internet banking. Also, the banking industry performance for the period before and theperiod after the adoption of internet banking will be established and compared to establish whether indeed internet banking had impacts on the banking sector performance. This will employ a descriptive study methodology to establish the effects before adoption and after adoption (Halili, 2014)
Descriptive statistics is a summary of key measures of the variables of study. They give an overview of the sample and population characteristics at a glance. The measures contained in descriptivestatistics are mean, median, mode, range, variance and standard deviation. This is common descriptive measures of a sample. The mean is the average of all observations in a sample under study. The mode is the variable which appears most in the sample observations while median is the center observation of all members of a sample arranged in a systematic manner. The sample range is the difference between the highest value of a variable observed and the lowest value(Gujarati, 2003).
The variance of a variable is the squared deviations from its mean observation. It is used as a measure of the volatility of the variable over the period under review. The variable standard deviation is the square root of its variance. It is a more objective measure of the volatility of a variable as compared to variance. In another word, variance and standard deviations are the measures of risk of a variable under study(Gujarati, 2003)
Table1: Descriptive statistics of variables(SWiss.National.Bank, 2016)
ROA | ROE | LAR | DAR | OHSR | ER | INTER Dummy | NPA | |
Mean | 2.53% | 16.37% | 48.81% | 47.41% | 7.60% | 0.13% | 0.78947368 | 13.42% |
Standard Error | 0.55% | 4.77% | 7.05% | 1.49% | 2.86% | 0.00% | 0.09609168 | 1.56% |
Median | 2.88% | 19.53% | 25.76% | 43.77% | 4.61% | 0.13% | 1 | 12.00% |
Mode | #N/A | #N/A | #N/A | #N/A | #N/A | #N/A | 1 | 15.00% |
Standard Deviation | 2.42% | 20.80% | 30.74% | 6.49% | 12.47% | 0.02% | 0.41885391 | 6.81% |
Sample Variance | 0.06% | 4.32% | 9.45% | 0.42% | 1.56% | 0.00% | 0.1754386 | 0.46% |
Kurtosis | 636.78% | 1133.30% | -189.55% | -110.89% | 1516.70% | -91.12% | 0.41911765 | -33.33% |
Skewness | -206.70% | -302.62% | 44.31% | 78.02% | 376.00% | 17.63% | -1.54483177 | 59.95% |
Range | 10.96% | 97.84% | 71.89% | 17.90% | 58.89% | 0.06% | 1 | 23.00% |
Minimum | -5.50% | -60.78% | 19.60% | 41.24% | -2.18% | 0.10% | 0 | 5.00% |
Maximum | 5.46% | 37.06% | 91.49% | 59.15% | 56.72% | 0.16% | 1 | 28.00% |
Sum | 48.15% | 311.02% | 927.36% | 900.71% | 144.46% | 2.46% | 15 | 255.00% |
Count | 19 | 19 | 19 | 19 | 19 | 19 | 19 | 19 |
The model used for this study was also used by Khwarsh (2011), Huizih (1999), Aburime (2008) and (Hyeon-Wook Kim, 2003)in their studies on impacts of internet banking on banking industryperformance in their respective countries. The General regression model for the analysis will be:
Y_{SWt}=β_{0}+β_{1}X_{SW1}+β_{2}Xsw_{2}+β_{3}X_{SW3}+β_{4}Xsw_{4}+β_{5}Xsw_{5}
Where:
Y_{SWt=}Swiss banking industry performance
β_{0=}Y intercept which is a Constant
β_{1,}β_{2,}β_{3,}β_{4,}β_{5=}Independent Variable Coefficients
X_{SW1=}Loan to Asset ratio for Swiss banking industry(LAR)
X_{SW2=}Deposit to assets ratio for Swiss banking industry (DAR)
X_{SW3=}Overheads ratio for Swiss banking industry (OHSR)
X_{SW4=}Efficiency Ratio (ER)
X_{SW5=}Internet banking
X_{SW1,}X_{SW1,}X_{SW3,}X_{SW4,}X_{SW5 }are the independentvariables in the study.
EFFECTS OF INTERNET BANKING ON PERFORMANCE OF SWISS BANKING INDUSTRY
Two measures which were used by (Hyeon-Wook Kim, 2003) in their study will be used to as a measure of the Swiss banking industry performance for the period 1996 to 2015. They include ROE and ROE. Earnings before taxes will be used for the computation of the two measures as tax legislations in place in Switzerland differs with company’s nature of operations. This means use of Industry net tax will show fluctuations as different banks will make realignments in attempt to minimise the tax effects on their operations hence ends up distorting the real banking industry income in Switzerland (Hess, 2016)
The hypothesis being tested is that internet banking improves Swiss banking industry performance represented by ROE and ROA. Two models will be established and compared to check whether the hypothesis holds during the period under study. Two Dummy variables will be used to represent internet usage. Dummy variable with avalue of Zero (0) will represent non-internet usage in the industry. According to World bank, Switzerland banking industry started using internet banking from the year 2001.A dummy variable of one (1) will be used to represent the internet banking usage and improvements after the adoption from the year 2001 (WorldBank, 2016)
The LAR and ER in the model will explain the systematic impacts of ROE on the Swiss Banking industry income and costs from operations. The ER in the model shows the level of managerial efficiency in managing banking cost. This should have a negative relationship with ROE and ROA in the banking industry(Hyeon-Wook Kim, 2003). The Data used is converted to log–form to improve the results(Gujarati, 2003)
The specific model which will describe the relationship will be as bellow
ROEswt=β_{0}+β_{1}LAR_{SWt}+β_{2}DARsw_{t}+β_{3}OHSR_{SWt}+β_{4}EFsw_{t}+β_{5}INTDUMMYsw_{t}………Model 1
ROAswt=β_{0}+β_{1}LAR_{SWt}+β_{2}DARsw_{t}+β_{3}OHSR_{SWt}+β_{4}EFsw_{t}+β_{5}INTDUMMYsw_{t}………Model 2S
Where:
ROEswt_{=}Swiss banking industry Return on equity
ROAswt_{ =}Swiss banking industry Return on Assets
β_{0=}Y intercept which is a Constant
β_{1,}β_{2,}β_{3,}β_{4,}β_{5=}Independent Variable Coefficients
LAR_{SWt=}Loan to Asset ratio for Swiss banking industry (LAR)
DAR_{SWt=}Deposit to assets ratio for Swiss banking industry (DAR)
OHSR_{SWt =}Overheads ratio for Swiss banking industry (OHSR)
ER_{SWt=}Efficiency Ratioof Swiss banking industry (ER)
INTDUMMYsw_{t =}Internet banking dummy variable
Table 1: Model 1 (Effects of internet banking on Swiss banking industry ROA)
SUMMARY OUTPUT | ||||||||
Regression Statistics | ||||||||
Multiple R | 1 | |||||||
R Square | 0.9999 | |||||||
Adjusted R Square | 0.9999 | |||||||
Standard Error | 0.0001 | |||||||
Observations | 20 | |||||||
ANOVA | ||||||||
do | SS | MS | F | Significance F | ||||
Regression | 5 | 2.7033 | 0.5407 | 37120214.6059 | 0.0000 | |||
Residual | 14 | 0.0000 | 0.0000 | |||||
Total | 19 | 2.7033 | ||||||
Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | Lower 95.0% | Upper 95.0% | |
Intercept | 0.0022 | 0.0019 | 1.1553 | 0.2673 | -0.0019 | 0.0063 | -0.0019 | 0.0063 |
LAR | 0.0009 | 0.0003 | 2.9807 | 0.0099 | 0.0002 | 0.0015 | 0.0002 | 0.0015 |
DAR | -0.0039 | 0.0015 | -2.6334 | 0.0197 | -0.0072 | -0.0007 | -0.0072 | -0.0007 |
OHSR | -1.0000 | 0.0001 | -11223.4739 | 0.0000 | -1.0002 | -0.9998 | -1.0002 | -0.9998 |
ER | 0.9996 | 0.0007 | 1511.8634 | 0.0000 | 0.9981 | 1.0010 | 0.9981 | 1.0010 |
INTER Dummy | 0.0000 | 0.0001 | 0.3874 | 0.7043 | -0.0002 | 0.0002 | -0.0002 | 0.0002 |
Table 2:Model 2 (Effects of internet banking on Swiss banking industry ROE)
SUMMARY OUTPUT | ||||||||
Regression Statistics | ||||||||
Multiple R | 0.987138 | |||||||
R Square | 0.974441 | |||||||
Adjusted R Square | 0.965312 | |||||||
Standard Error | 0.072038 | |||||||
Observations | 20 | |||||||
ANOVA | ||||||||
df | SS | MS | F | Significance F | ||||
Regression | 5 | 2.769843 | 0.553969 | 106.7486 | 1.227E-10 | |||
Residual | 14 | 0.072653 | 0.005189 | |||||
Total | 19 | 2.842496 | ||||||
Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | Lower 95.0% | Upper 95.0% | |
Intercept | -4.59976 | 1.131317 | -4.06584 | 0.001157 | -7.026188 | -2.17332 | -7.02619 | -2.17332 |
LAR | -0.72365 | 0.176255 | -4.10569 | 0.00107 | -1.101679 | -0.34562 | -1.10168 | -0.34562 |
DAR | 1.558228 | 0.895274 | 1.740504 | 0.103693 | -0.361944 | 3.478399 | -0.36194 | 3.478399 |
OHSR | 0.988268 | 0.053186 | 18.58145 | 2.91E-11 | 0.874196 | 1.10234 | 0.874196 | 1.10234 |
ER | 0.817002 | 0.394644 | 2.070226 | 0.057403 | -0.029425 | 1.663428 | -0.02942 | 1.663428 |
INTER Dummy | 0.003503 | 0.056702 | 0.061786 | 0.951607 | -0.11811 | 0.125117 | -0.11811 | 0.125117 |
EFFECTS OF INTERNET BANKING ON ECONOMIC IMPLICATIONS OF SWISS BANKING INDUSTRY
Most researchers concur on the argument that internet banking has contributed to cost reduction through the multi channels it creates in the industry(Halili, 2014). It has changed to a great deal the existing banking business models which have led to change in banking industry cost structure. This is because internet banking has eliminated the physical contact of thecustomer with the bank hence reducing the queuing cost and bank branch manpower to service clients in traditional brick and motor banking models. Some banks like the Standard bank are currently operating branchless banking which has been enabled by internet banking(Caprio et al, 2008)
Also, internet banking has been associated with cost reduction as a result of paperless banking in Switzerland banks (WorldBank, 2016). The expected outcome is that internet banking should contribute to areduction in Non-interest expense efficiency. In this case, the Efficiency Ratio will become the dependent variable in determining whether internet banking has economic implication as proxies by banking industry efficiency ratio. However, the effects should be lagged because of the initial heavy investment in ICT infrastructure by banks. The real effects of cost reduction and increased performance in banking industry accrues with time(Hyeon-Wook Kim, 2003)
The new model will be as follows
EFsw_{t}=β_{0}+β_{1}LAR_{SWt}+β_{2}DARsw_{t}+β_{3}OHSR_{SWt}+ β_{4}INTDUMMYsw_{t}………Model 3
Where
EFsw_{t=}Non- interest expense efficiency ratio
Table 3: Effects of internet banking on cost reduction (Model 3)
SUMMARY OUTPUT | ||||||||
Regression Statistics | ||||||||
Multiple R | 0.699141 | |||||||
R Square | 0.488799 | |||||||
Adjusted R Square | 0.352478 | |||||||
Standard Error | 0.047131 | |||||||
Observations | 20 | |||||||
ANOVA | ||||||||
df | SS | MS | F | Significance F | ||||
Regression | 4 | 0.03186 | 0.007965 | 3.585664 | 0.030436 | |||
Residual | 15 | 0.033321 | 0.002221 | |||||
Total | 19 | 0.065181 | ||||||
Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | Lower 95.0% | Upper 95.0% | |
Intercept | 2.815175 | 0.139676 | 20.15497 | 2.82E-12 | 2.517461 | 3.112888 | 2.517461 | 3.112888 |
LAR | 0.084291 | 0.113244 | 0.744334 | 0.468176 | -0.15708 | 0.325665 | -0.15708 | 0.325665 |
DAR | -0.1448 | 0.584546 | -0.24772 | 0.807713 | -1.39073 | 1.101129 | -1.39073 | 1.101129 |
OHSR | 0.010944 | 0.034682 | 0.315548 | 0.756693 | -0.06298 | 0.084867 | -0.06298 | 0.084867 |
INTER Dummy | 0.096913 | 0.027388 | 3.5385 | 0.002979 | 0.038536 | 0.155289 | 0.038536 | 0.155289 |
EFFECTS OF INTERNET BANKING ON SWISS BANKING INDUSTRY RISK
The Swiss banking industry risk will be proxies by the ratio of Non-performing loans to Net loans issued to customers. The higher the ratio of NPA the higher the banking industry risk. Internet banking has been associated with both increase in risk and decreases in risk by conflicting findings obtained by scholars. Therefore, there are two existing schools of thoughts pertaining to the peak effect of internet banking to banking risk. In this case, the ratio of Net non-performing loans to Net loans advances will be the dependent variable when determining the effects of internet banking on industry risk(Halili, 2014)
The internet banking introduction hasthe effect of modifying the existing risk structure of banking industry. This involves increasing the traditional banking risks which consist of credit risk, forex risk and interest risk. Credit risk is the probability of a borrower not paying back the loans interest and principle. This is inherent to banks loan assets and as borrowers are prone to loans default. The forex risk is a risk associated with volatilities in exchange rates which leads to transactional translation losses. The third element of risk is associated with volatility of interest rates in the market hence affecting negatively the bank’s operations and performance(Sathye, 2003). Another kind of risk banks face is the operational risk. This is associated with the bank’s internal systems failure leading to poor performance of the banks. Therefore, NPA can have both positive effects and negative effects on banks profitability . This is because, banks can realise high returns by assuming thehigh risk and at the same time, it can realise huge losses or even bankruptcy as a result of taking high risk. However, banks are highly regulated by central banks and they are supposed to assume a reasonable level of risk.(Sathye, 2003)
The new model will be as follows
NPAsw_{t}=β_{0}+β_{1}LAR_{SWt}+β_{2}DARsw_{t}+β_{3}OHSR_{SWt}+ β_{4}INTDUMMYsw_{t}………Model 4
Where
NPAsw_{t=}is the ratio of Net non-performing loans to Net loans advanced to customers
Table 4: Effects of internet banking on banking industry risk
SUMMARY OUTPUT | ||||||||
Regression Statistics | ||||||||
Multiple R | 0.471491 | |||||||
R Square | 0.222303 | |||||||
Adjusted R Square | 0.014918 | |||||||
Standard Error | 0.229627 | |||||||
Observations | 20 | |||||||
ANOVA | ||||||||
df | SS | MS | F | Significance F | ||||
Regression | 4 | 0.226085 | 0.056521 | 1.071932 | 0.40471 | |||
Residual | 15 | 0.790925 | 0.052728 | |||||
Total | 19 | 1.01701 | ||||||
Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | Lower 95.0% | Upper 95.0% | |
Intercept | 0.473035 | 0.68051 | 0.695118 | 0.497609 | -0.97744 | 1.923508 | -0.97744 | 1.923508 |
LAR | 0.191608 | 0.55173 | 0.347287 | 0.733201 | -0.98438 | 1.367593 | -0.98438 | 1.367593 |
DAR | 1.072475 | 2.847935 | 0.37658 | 0.711759 | -4.99775 | 7.142703 | -4.99775 | 7.142703 |
OHSR | -0.12786 | 0.168974 | -0.75669 | 0.460957 | -0.48802 | 0.232298 | -0.48802 | 0.232298 |
INTER Dummy | 0.244783 | 0.133436 | 1.834459 | 0.086497 | -0.03963 | 0.529195 | -0.03963 | 0.529195 |
CHAPTER THREE
RESULTS ANDDISCUSSIONS
From the regression results obtained interesting observations have been made from thestudy of Swiss banking sector in regard to internet banking in theperiod covering the year 1996 to 2015 with a sample of 275 banks . The impacts of internet banking on Swiss banking industry performance presents some contradicting results from the hypothesis that , internet banking leads to the industry performance improvement. The R-squaredmeasure which shows the correlation between thedependent variable and theindependent variable is 99%. This is an indication that the model 1 developed is a good measure predictor of the industry return on assets.(Kennedy Okiro, 2013)
Also, the model developed has F statistic of 0.00 which means that the depended variables used are good estimators of Swiss industry ROA. The p-values of LAR, DAR, OHSR and ER are statistically significant meaning that they affect the ROA of the industry in the period under review. However, INTDUMMY variable which represents the internet banking in Switzerland is not statistically significant in the model. This means that the internet banking has no effect on theperformance of the Swiss banking industry. This concurs with the finding of (Hyeon-Wook Kim, 2003) and (Caprio et al, 2008) who found that there was no relationship between internetbanking and banks performance using secondary data (Sullivan, 2000)
The coefficients of the dependent variable of the analysis give a conflicting finding compared to earlierresearch done by (Hyeon-Wook Kim, 2003). The LAR coefficient indicates a zero correlation on Swiss banking industry performance. This means that the change in aloan to asset ratio which can be used to indicate the industry risk does not affect the industry performance. This can be attributed to the fact that banks are highly regulated especially after the world financial crisis and regulators are keener on these fundamental ratios to ensure the banking industry stability. Lack of relationship means that the industry maintains a set level of the Loan-asset ratio hence it cannot affect the industry performance.In other industries where there are no regulations on key ratios, a reduction in LAR could lead to increased performance as measured by ROA of the company(SWiss.National.Bank, 2016)
The depositsto –asset ratio is expected to have negatively correlated with the banking industry performance. This is because banks are not limited by the regulator on deposits receipts hence the negative relationship. This implies that an increase in DAR is beneficial to banks as increased level of deposits means the banking industry can invest more and lend more from the increased deposits. This has the effect of increasing banking industry profitability hence its performance. This is the scenario depicted by the model 1(Sullivan, 2000)
Overheads ratio is a measure of how adequately the management in the banking industry managed to control thecost of operations relating to salaries and administration cost.A lower ratio is desirable for the industry and will be expected to yield increased performance due to the cost cut. This is expected to be negatively correlated with the industry performance as shown in the coefficient of DAR in table 1.The model has supported the hypothesis that increase in overheads efficiency levels leads to increase in ROA in thebankingindustry. Also, managers will try to cut operating cost in their endeavour of pursuing the goal of shareholders wealth maximization(Hyeon-Wook Kim, 2003) .
The model also shows that the efficiency ratio relating to non-interest expense and banking industry assets has a positive correlation. This operating cost relates to costs of acquisition of ICT infrastructure which consumes a considerable banks resource. The increase can be explained the increased cost associated with the ICT infrastructure, depreciation andamortisation costs associated with the increasing asset base of the banking industry.(Halili, 2014)Argues that, the cost of internet banking incurred does not benefit the industry immediately. It has a lagging effect whereby, banks will realize increased costs of operations immediately after the ICT infrastructure acquisition. However, in the long run, the industry will benefit from the reduced cost benefits associated with internet banking. This is the scenario depicted by the Swiss banking sector during the period under review(Halili, 2014)
The dummy variable coefficient in table one analysis is zero. This indicates that there is no relationship between the internet banking and Swiss banking industry performance during the period under review. This is supported by the increased efficiency ratio effects discussed above. Therefore , my study on the effects of internet banking on Swiss banking industry agrees with the findings of (Caprio et al, 2008) and(Hyeon-Wook Kim, 2003) who found that there is no relationship between internet banking and banks performance as measured by ROA. The findings conclusion will be to reject the null hypothesis (Sullivan, 2000)
On the analysis of Swiss banking industry performance using ROE, it brings a rather interesting discussion. The model for predicting Swiss banking industry ROE using the identified independent variable is a good model since it has R-squared of 98.7% and F-statistic of 0.00 which is significantly statistic, This is an indication that all changes in Swiss banking industry ROE can be explained by the predictor variables to a great extent. Also, it is an indication that, the variables used are sound and are logical to explain the relationship(Sathye, 2003)
The p-values for the intercept, loan-asset ratio, Overheads efficiency and non-interest expense efficiency are statistically significance. However, deposits to asset ratio are statically insignificant contrary to the findings when using ROA. Return on equity is highly affected by the macroeconomic factors prevailing in the economy. There is a positive correlation between DAR and ROE whichis attributed to the fact that when banks take up more deposits, it will pay more interest on the increased deposits .On the other side, increased deposit taking translates to more business for the banking sector which will lead to anincrease in performance. This is an indication that, increase in DAR will lead to increase in performance of banking sector when measured by ROE(Hyeon-Wook Kim, 2003)
The dummy variable representing the internet banking likewise is not statistically significant as was in the case of ROA. This means that in Swiss banking industry, internet banking does not lead to increase in the industry performance as measured by ROE. Thisis the same findings as evaluating the impacts of internet banking using ROA. Also, the coefficient of variation of the dummy variable is 0.00 which means that the two variables are not correlated in any way. These findingslead to rejection of the null hypothesis that internet banking positively affects banking industry performance as measured by both ROE and ROA(Kennedy Okiro, 2013)
Another interesting observation from the regression in model 2 is that both efficiency measures proxies by overheads efficiency ratio and non-interest expense efficiency ratio are increasing as ROE increases. This can be explained by the school of thought which points out that, an increasing investment in ICT infrastructure in the mind term period’s leads to increase in cost as well as an increase in banking industry performance. This is because, in the mind term, some benefits of internet banking leggings to accrue as operations becamestreamlined and less costly. Likewise, the firm will witness increased profitability but at a low rate due to efficient systemsand less costly operations.(Hyeon-Wook Kim, 2003)
The loan assets ratio is negatively correlated to ROE as a measure of performance in the banking industry in Switzerland. This is because as the loan asset ratio decreases, the bank will be investing the funds in less risky investments which pay good returns leading to increasing in net incomes hence increase in ROE. This contrary to when ROA is used as a performance measure of the banking industry where we saw that the regulatorimposes limits if this key fundamental ratio. This is because an increase in the Ratio is an indication of increase in risk levels of the banking sector(Halili, 2014)
The impact of internet banking in reduced cost to represent economic benefit is represented in table 3, model 3. The model is relatively weak to predict cost cutting as it has 48% r-squared and 0.03 F-statistic which is statisticallysignificant. The LAR, OHSR and ER are statistically insignificant thus they don’t explain the reduction in cost of operations in Swiss banking industry. Likewise, loan to asset and overhead efficiency ratio are positively correlated to cost thus increase in the two independent variables leads to increase in the cost of operations in the banking industry. This is because if loan to asset ratio increases, the banking industry is expected to realize some increase in default cost(Sathye, 2003)
Incraese in overheads logically leads to increase in operating cost as the relationship between the two variables suggest. Also, DAR is negatively related to cost of Swiss banking sector as the deposits increases, the risk associated with default reduces hence the industry realises cost savings. The hypothesis that internet banking leads to reduced cost of operations is supported in the model by the fact that p-value of the dummy variable is 0.03 which is statistically significantly and low coefficient of 0.09 which indicates that in the mind term period of internet banking , cost of operation increases at a decreasing rate(Hyeon-Wook Kim, 2003)
This model which is developed shows a weak relationship between the dependent and independent variable because only internet banking predictor variable included in the model can predict thecost of banking industry objectively(Sullivan, 2000)
Model 4 which is developed to explain the relationship between internet banking and Swiss banking industry risk is a weak model. This is evidenced by the low R-squared in the model and high value of F-statistic of 0.41%. All p-values of predictor variables are statistically insignificant in the model. It shows that increase in LAR and DAR leads to increase in nine Firms risk. This is true as when firms increases its loans, there will be probability of default risk by the borrowers. Also, when the industry increases deposits taking, the interest rates have probability of increasing too(Hyeon-Wook Kim, 2003)
The dummy variable in the model is statistically insignificant meaning that internet banking in Swiss banking industry does not reduce the industry risk as one of the opposing Hypothesis states. This leads to rejection of the hypothesis that internet banking leads to risk reduction in banking industry as proposed by some scholars. However, the findings supporting the otherconflicting hypothesis that internet banking in Swiss leads to increase in banking industry risk. This is because the coefficient of variation of dummy variable is a positive 0.244(Hyeon-Wook Kim, 2003)
From the analysis of the variablesusing multiple regression methodology, it evidenced that internet banking in Swiss banking sector does not lead to increase in banking industry performance as measured by ROE and ROA. However, it leads to reduction in cost of operations in Swiss banking industry and hasled to increase in risk in the Swiss banking industry during the 20 years period which I conducted the research covering year 1996 to 2015 for 275 banks. The study concurs with studies done by (Halili, 2014)(Hyeon-Wook Kim, 2003)and(Caprio et al, 2008) which evaluated on aspect of internet banking using a smaller sample and shorter period. My study differ in that in evaluates different aspects relating to cost, performance and risk at ago with long time span of 20 years. Also, I have used all banks in Switzerland for the task unlike other researchers who used a sample of banks.
References
Caprio et al, 2008. The impact of internet banking on profitability -Casestudy of Turkey. Oxford Business & Economics Conference Program, 1(1), pp. 1-16.
Gujarati, D., 2003. Basics Econometrics. 4Th ed. New York: McGraw Hill.
Halili, R., 2014. Impacts of Online banking on banks performance. Charles University in Prague, 1(1), pp. 1-73.
Hess, J., 2016. Switzerland Tax Highlights 2016. Deloitte, 1(1), p. 3.
Hyeon-Wook Kim, C.-G. P., 2003. Impact of internet Banking on performance of korean banking industry: An imperical Analysis. Journal of Banking & Finance, 1(1), pp. 1-26.
Kennedy Okiro, J. N., 2013. Impact of Mobile and Internet banking on performance of finacial institutions in Kenya. European Scientific Journal, 9(13), pp. 146-162.
Sathye, M., 2003. Impact of internet banking on performance and risk profile: Evidence from Australian credit union. University of Canberra, 1(1), pp. 1-24.
Sullivan, R., 2000. How has the adoption of internet banking affected performance and risk of banks. Fideral Reserve bank of Kansas, 1(1), pp. 1-26.
SWiss.National.Bank, 2016. Swiss National Bank. [Online]
Available at: https://www.snb.ch/
[Accessed 23 November 2016].
WorldBank, 2016. International Telecommunication Union, World Telecommunication/ICT Development Report and database, and World Bank estimates.. [Online]
Available at: http://data.worldbank.org/indicator/IT.NET.USER.P2?locations=CH
[Accessed 23 November 2016].