"This research is done to find a more objective and comprehensive formula for performance evaluation of banks listed in Jakarta Stock Exchange. This is based on the statistical theory of how to find some key variables used to evaluate banks performance, and the finance/banking management theory to formulate the model and to further evaluate those key variables found. If we already have those variables which we get from the statistical evaluation, we do not have to consider all the financial ratios. All we can have is just those key variables. Hopefully, the findings can be applied for all banks in the Indonesian economy.
Most researches are merely univariate, so that they are quite subjective. In this research I try to eliminate eveything that is based on subjectivity. Although I still use financial ratios to find the key variables (as the independent variables), I also use three more variables to formulate the model (as the dependent variables). Those are the growth of the stock prices, the growth of the (quality) assets and the ratio of Non Interest Income to Total Asset (a measure of fee-based income). These three variables are used as the indicator of the bank performance. In other words, I would like to know whether the groupings which is based on the three performance variables is right and consistent. I also would like to know whether the ratios chosen can be used to describe the bank performance.
To find those key variables, I use data from the Jakarta Stock Exchange. There were twenty two banks listed from January to December 1996 and used as the models (the other two banks were not considered in the analysis because they were just listed on November dan December 1996). I then devided the banks into two groups, each of which consists of nine and thirteen members. The first group has a bad performance indicating that the members can not generate fee-based income and has a much lower growth of the stock prices compare to the growth of the asset. The second group has a good performance which means that the group has a higher level of fee-based income and has a little lower growth of the stock prices compare to the growth of the asset during the same period. (Cluster analysis is used in making the groupings).
By using a Multivariate Discriminant Analysis, I get three key variables succesfully. Then, I will say this function as The Alternative Model for Bank Performance Evaluation, and is expressed as :
A A A
Z=1.2188X1+0.8153X2+1.2721X3
X1 : Non Interest Income to Total Asset (NOIITA)
X2 : Loss Reserve to Total Loan (LRTL)
X3 : Loan to Deposit Ratio (LDR)
The results tells us the following points :
The formula can be applied to all bank in the economy for the performance evaluation as they find no misclassification in the groupings, and are supported by a good statistical test result.The banks have a tendency to operate outside their main business. The banks are also exposed to risk. They have given more loans to the public higher than their deposits and have more loss reserve in their operation. The banks performance is sensitive to LDR and then to NOI1TA, shown by the relatively high elasticity of LDR (1.2721) and NDIITA (1.2188). Any changes in those two variables will significantly affect the bank performance. Each bank has a relatively high growth of its asset compare to growth of its stock prices.
The only thing that will cause a little effort in applying this formula is the fact that the bank financial statements are not in good standard, so that people have to look thoroughly in those statements. This research can be followed by further research concerning extension of the period and extension to non-banks business operation. It might also be done to test the findings of any further situations such as whether the banks have a tendency to operate outside their main business (or not), whether the banks are likely to operate exceeding their maximum credit limit (or not), and whether the risk factor is dominant in the banking business (or not)."
Depok: Fakultas Ekonomi dan Bisnis Universitas Indonesia, 1998
