This are results of a dissertation data analysis highlighting stationary of the data tested, co-integration test, granger -Causality test and data envelopment analysis (DEA) for testing efficiency of bank performance please. It focuses on data from Portuguese and Greek banks. These results are based on data derived from a panel data using GLS (Generalized least square)
Stationarity of Data
The panel data stationarity test has a severe size distortion inconsistent with the null hypothesis. Stationary is vague since the mean and variance of the data is not constant. The most appropriate resolution in this case is to merge ailing banks or let strongly financed banks purchase bad debts in accordance with market mechanisms and securitization in Portugal. In view of that, weak banks should be merged or be acquired by stronger banks that appear in the panel data.
Causality Test
The influence of NPL on technical competence is close to zero and not significant. On the other hand, the effect of technical efficiency on loan Loss is positive and significant at 5% level, once more indicating that the causality would run from bank efficiency to non-performing loans. Turning to the allocative efficiency case, the performance is poor and there is no significant response to variation in NPL. Consequently, the loan losses react negatively to transformations in this bank efficiency, with a significance level of 10%. Finally, looking at the economic efficiency, it appears to have a negative impact on problem loans and to be despicably affected by the credit risk variable. Nonetheless, none of the coefficients are significant.
Co-integration test
The results of co-integration between Bank Capital, Loan and GDP at 5% significance level for all commercial banks in Portugal suggest the presence of co-movements among the variables, indicating long-run stationarity. The Error Correction Model (ECM) corrected the deviation from the long-run equilibrium by short-run adjustments.
In particular, the banking sector in Portugal frequently realizes bad debts that bring weigh down at the market. This may be because of high inflation and instability in economy especially during 2005 to 2011.Thus; short-term transactions have a strong influence in financial markets.
This situation increases fund costs and affects investors negatively. Similarly, unstable economy causes small investors withdraws from financial markets by increasing their risk.
Eigenvalue
Likelihood ratio (Qmax)
Critical Value (5%)
0 . 2 8
4 0 . 3 6
2 9 . 6 8
0 . 2 4
1 9 . 6 0
1 5 . 4 1
0 . 0 4
2 . 4 0
3 . 7 6
Data Envelopment Analysis
DEA compares each bank with all other service units, and identifies those units that are operating inefficiently compared with other units. It accomplishes this by locating the best practice or relatively efficient units. It also measures the magnitude of inefficiency of units compared to the best practice units. The best practice units are relatively efficient and are identified by a DEA efficiency rating. The inefficient units are identified by an efficiency rating of less than or equal to 1. The upper limit is set as 1 or 100% to reflect the view that a unit cannot be more than 100% efficient.
The large number of commercial banks in Portugal, the high branch density, the slow technological change and stiff competition hasn't added any pressure enough to improve the performance. Data Envelopment Analysis system necessitated no specification of any meticulous functional shape or error configuration.
Generally, based on the nominal values of the input and output variables, the total productivity of commercial banks in Portugal slightly increased from 2006 to 2011, at a rate equal to 1.5% (TFP=1.055). However, based on the natural logarithm of the data the total productivity decreased by 3.6% (TFP= 0.974). It is obvious that the TEC index and the TC index are not moving towards the same direction in both cases. Conversely, the TEC index improved by 0.5% and the TC index was reduced by 3% based on the natural logarithm approach of DEA.
Granger Causality Test
From the causality test, all the parameters are jointly zero, portraying that change in loan does not granger-cause change in BC. Accordingly, change in loan does not granger-cause change in GDP and vise versa. This is further explained as follows:
Change in BC does not granger-cause change in loan, in the first regression,
Change in LOAN does not granger-cause change in BC, in the second regression,
Change in LOAN does not granger-cause change in GDP, in the third regression,
Change in GDP does not granger-cause change in loan, in the fourth regression.
Apparently, any contradiction in the results of the bank efficiency hypothesis tests, among the commercial banks in Portugal may be attributable to differences in financial structures and the effectiveness of the accounting system. Additionally, banks differ widely in their relative reliance on loan vs. market finance. The table below show the result of the variances.
Variance and direction
Probability
AIC
BC LOAN
0 . 5 2
- 6 . 9 7
LOAN BC
0 . 4 9
- 6 . 9 7
LOAN GDP
0 . 2 6
- 9 . 7 2
GDP LOAN
0 . 13
- 9 . 7 2
General Analysis
Output growth rate is realized to be significant for commercial bank at 1% significance level throughout the years used in the panel data. This portrayed a positive relationship with the dependent variable showing income elasticity of 0.197 -- 0.21.
You’re 82% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.