Research Results For Dependent Variable Scale Efficiency

shows that economies of scale mean that as the scale increases, the profits of banks increase. Thanks to the scale of operations, banks develop a wide network and operating system throughout the country. This allows banks to develop capital mobilization activities for all classes of people in the areas, thereby reducing input costs. Banks take advantage of costs of scale by saving on fixed costs per transaction volume. This result is consistent with the research of Ongore (2013), Ayadi (2014).

The variable CV has a negative impact on the HQKTT for group 2 banks at a statistical significance level of 5%. In fact, in recent times, Vietnamese commercial banks have expanded the credit market, so they have been more open in appraising lending projects. However, the ability to manage and control credit quality and the ability to manage risks are still poor, making loans more risky and reducing the efficiency of capital use, increasing the rate of overdue debt. The variable TTKT has a negative impact on the HQKTT for group 2 banks at a significance level of 5% because in the research period from 2008-2014, although the gross domestic product increased in quantity, the economy still had many instabilities, so it had a negative impact. This research result is consistent with the research of Ongore (2013), Nguyen Minh Sang (2014). In 2008, the global financial crisis that started in the US spread throughout the world and had a significant impact on Vietnam, the economic growth rate was only 5.66% and the ROA of banks was only 1.24%. By 2010, the financial crisis had subsided but the echoes of the crisis still remained and to ensure the safe development of the banking system, the State Bank applied stricter regulations to banks in the direction of meeting international standards, Vietnam's economic growth rate reached 6.42% but the average ROA was only 1.13%. And in 2013 and 2014, the GDP growth rate decreased to 5.42% and 5.98% but the average ROA was 0.54% and 0.55% as shown in Figure 2.3. This shows that if economic growth does not go hand in hand with stable and sustainable development, the components of the economy will be negatively affected, which is an obstacle to the development of social resources, including the banking industry.

4.2.6. Research results on dependent variable scale efficiency

4.2.6.1. Selection and testing of regression models for dependent variables of scale efficiency

Table 4.22: Regression model results with dependent variable HQQM

Variable

All samples

FGLS Model

Group 1 NH

REM model

NH group 2

FGLS Model

CVNT

-0.0062644

(0.550)

-0.0229074

(0.589)

0.0050155

(0.595)

TSNNT

-0.0317596

(0.255)

-0.082754

(0.277)

0.0124882

(0.633)

VCSH

0.028787

(0.347)

0.3495524

(0.151)

0.0803187 **

(0.024)

QMTS

0.0048155

( 0.161)

0.0010299

(0.935)

0.0137401 *** (0.000)

-0.082662 ***

(0.000)

-0.3542843 ***

(0.001)

-0.0326891 *

(0.071)

VHDCV

-0.0187937 ***

(0.009)

-0.0606122 *

(0.090)

-0.007626

(0.155)

TTKT

-0.1710881

(0.696)

-0.1336482

(0.934)

0.1790037

(0.616)

-0.0696219 **

(0.032)

-0.3777144 ***

(0.001)

-0.0359849

(0.273)

CONS

0.981067 ***

(0.000)

1.251849 ***

(0.000)

0.7489874 ***

(0.000)

Maybe you are interested!

F-test

F(37, 196) = 4.31

Prob > F = 0.0000

F(10, 57) = 3.56

Prob > F = 0.0010

F(26, 131) = 2.20

Prob > F = 0.0020

Hausman test

chi2(8) =29.96

Prob>chi2 = 0.0002

chi2(8) =9.70

Prob>chi2 = 0.2865

chi2(8) =3.86

Prob>chi2 = 0.8692

Modified

Wald test

chi2 (38) = 4880.16

Prob>chi2 = 0.0000

Breusch-Pagan

Lagrangian test

chi2(1) = 0.94

Prob > chi2 = 0.3333

chi2(1) = 8.03

Prob > chi2 = 0.0046

Wooldridge test

F( 1, 36) = 4.653

Prob > F = 0.0377

F( 1, 10) = 4.084

Prob > F = 0.0709

F( 1, 25) = 1.806

Prob > F = 0.1910

Source: Synthesis of results from Stata 11.1 with data from 38 surveyed commercial banks. For the research sample of all banks: Through F-test (p-value=0.0000) and Hausman test (p-value=0.0002), the selected model is the fixed-effects model FEM. However, when testing for heteroscedasticity (p-value=0.0000) and autocorrelation (p-value=0.0377), the model has heteroscedasticity and autocorrelation. To overcome heteroscedasticity and autocorrelation, the FGLS regression model is selected for the entire sample of banks.

study.

For the research sample of group 1 banks: Through F-test (p-value=0.0010) and Hausman test (p-value=0.2865), the selected model is the random effect model REM. However, when testing for heteroscedasticity (p-value=0.3333) and autocorrelation (p-value=0.0709), the model does not have heteroscedasticity and autocorrelation. Therefore, the REM regression model is selected for group 1 banks.

For the research sample of group 2 banks: Through F-test (p-value=0.0020) and Hausman test (p-value=0.8692), the selected model is

random effects model REM. However, when testing for heteroscedasticity (p-value=0.0046) and autocorrelation (p-value=0.1910), the model showed heteroscedasticity. To overcome the heteroscedasticity, the FGLS regression model was selected for group 2 banks.

4.2.6.2. Discussion of research results and hypothesis testing for the dependent variable of scale efficiency

The CVNT variable has a negative impact on the entire sample and Group 1 banks, showing that increasing foreign currency lending does not increase HQQM. On the contrary, for Group 2 banks, increasing foreign currency lending will increase HQQM. However, this result is not statistically significant, so hypothesis 5 is rejected.

The variable TSNNT has a negative impact on the entire sample and group 1 banks, proving that when banks increase foreign currency capital mobilization activities, it does not increase HQQM. On the contrary, for group 2 banks, increasing foreign currency capital mobilization will increase HQQM. However, this result is not statistically significant, so hypothesis 6 is rejected.

Table 4.23: Summary of research results for dependent variable HQQM

Variable

Impact on HQQM
All samples	Group 1 NH	NH group 2
CVNT	Inversely correlated but not significant	Inversely correlated but not significant	Same direction but no significance level
TSNNT	Inversely correlated but not significant	Inversely correlated but not significant	Same direction but no significance level

Source: Synthesis of results from Stata 11.1 with data from 38 surveyed commercial banks. In addition, other control variables such as equity are estimated to have a positive impact on HQQM with a significance level of 5% for group 2 banks. Thus, for group 2 banks, equity becomes a solid cushion for stable business operations of commercial banks. Therefore, the rapid increase in charter capital of commercial banks

In recent times, to meet the minimum capital regulations of the State Bank used to compensate for credit risks, market risks and operational risks, it has also contributed to increasing the operational efficiency of commercial banks. For commercial banks with low equity capital, increasing charter capital is very necessary to increase liquidity, asset quality and increase bank market share, ensuring capital safety regulations to maintain normal operations of a commercial bank.

The QMTS variable has a positive relationship with HQQM at a significance level of 1% for group 2 banks. So, if these banks have more resources used to expand their scale, they can generate more profits and minimize input costs compared to other banks. This also reflects a reality in Vietnam recently when bank mergers have occurred frequently, causing small-scale commercial banks to become larger and operate more effectively thanks to increased efficiency according to scale. Group 2 banks are banks with assets of less than VND 100,000 billion and equity of less than VND 10,000 billion, so if banks have the conditions to expand their scale, it will increase the operating efficiency of this group of banks. In September 2013, the State Bank of Vietnam issued Circular 21/2013/TT-NHNN regulating the operational network of commercial banks, which stipulates that one of the conditions for establishing branches abroad and 100% foreign-owned banks of commercial banks is to have total assets of VND 100,000 billion or more according to the audited consolidated financial statements as of December 31 of the year immediately preceding the year of application. This may be one of the motivations for banks that have not yet met the conditions to accelerate the growth of their total assets in the coming time.

The CV variable has a negative impact on HQQM for the entire sample of banks and group 1 banks at a significance level of 1%, for group 2 banks at a statistical significance level of 10%, indicating that increasing lending activities will reduce HQQM. The expansion of credit activities but the quality of loans is not controlled causes the bad debt ratio to increase, according to table 2.8, the bad debt ratio of the entire banking system is 3.25% as of December 31, 2014. The high bad debt ratio of banks reflects the decline in the asset quality of the banking system and banks have to use part of their resources.

to reserve for risks. The variable VHDCV has a negative impact on HQQM for the entire sample of banks at a significance level of 1%, for group 1 banks at a statistical significance level of 10%. This means that if banks do not manage mobilized capital well, it will reduce HQQM, commercial banks will be under pressure on liquidity, increase interest expenses, thereby reducing the efficiency of resource use of Vietnamese commercial banks.

LP variable has a negative impact on HQQM for the entire sample of banks at the 5% significance level, for group 1 banks at the 1% significance level, this proves that increased inflation will reduce HQQM. When inflation increases, in order to mobilize capital or not want capital from their own banks to flow to other banks, commercial banks raise their mobilization interest rates close to the developments of the capital market, but how much is reasonable to raise is always a difficult problem for each bank. An unexpected race for mobilization interest rates at most banks creates a new mobilization interest rate level, some banks bring mobilization interest rates close to credit interest rates. With high inflation, the State Bank must tighten monetary policy to reduce the amount of money in circulation, but the demand for loans from business customers is still very large, banks can only meet the needs of a few customers with signed contracts or projects that are truly effective with an acceptable level of risk. On the other hand, due to the high mobilization interest rate, the lending interest rate is also high, which has worsened the investment environment of banks, and moral risks will appear. Due to the decrease in purchasing power of Vietnamese Dong, the increase in gold and foreign currency prices, the mobilization of medium and long-term capital has become really difficult for each bank while the demand for medium and long-term loans for customers is very large, so the use of short-term capital for medium and long-term loans in the past time at each bank is inevitable. This has affected the liquidity of banks, leading to maturity risks and exchange rate risks. Therefore, commercial banks need to pay more attention to forecasting the inflation situation and have appropriate interest rate product policies for each stage of the economy.

Chapter 4 Conclusion

From the overview of theories and experts' discussion opinions, the thesis has built a research model of international financial services affecting the performance of Vietnamese commercial banks to

see the importance of international financial services. In chapter 4, the author analyzes the results of the regression model on the impact of international financial services on the performance of Vietnamese commercial banks in the period from 2008 to 2014, giving comments on the factors and the level of impact of each specific factor, thereby identifying the direction of impact of the factors in the model. With the final research results, we have: TSNNT has a positive impact on performance for the research sample of all banks, group 1 banks and group 2 banks. The variable CVNT has a negative impact on performance for group 1 banks. Some other control variables such as equity, QMTS have a positive impact but CV, VHDCV, TTKT, LP have a negative impact on performance.

CHAPTER 5: SOLUTIONS TO DEVELOP INTERNATIONAL BANKING SERVICES TO IMPROVE THE OPERATIONAL EFFICIENCY OF VIETNAMESE COMMERCIAL BANKS

Through the empirical research process in chapter 4, through the DEA quantitative method and regression model, the topic has found a suitable model and explained the factors reflecting international financial services such as CVNT, TSNNT and some other factors such as: Equity, QMTS, CV, VHDCV, TTKT, LP have the same or opposite impact on the performance of the business with different levels of significance. On that basis, the author proposes some solutions to develop international financial services to improve the performance of Vietnamese commercial banks in the coming time.

5.1. Orientation for developing international banking services to improve the operational efficiency of Vietnamese commercial banks in the context of integration

The long-term goal in the development strategy of the Vietnamese banking system in the context of integration is to build a commercial banking system that develops stably, is diversified in depth, and increasingly plays an important role and influence on the domestic, regional and world economies, fully meeting the diverse needs of the economy for financial products and services. The project "Development of the Vietnamese banking industry to 2010 and orientation to 2020" according to Decision 112/2006/QD-TTg dated May 24, 2006 of the Prime Minister stated that the banking system develops stably and sustainably with a scale at the average level of the world and the region, ensuring the stability of the financial market; the motto of credit institutions is "Safety - Efficiency".

- Sustainable development - International integration" with the following orientations for developing international banking services and improving the operational efficiency of the commercial banking system:

- Vietnamese commercial banks need to have strong innovations in their organizational models, expand their cross-border activities and, if strong enough, can gradually establish a number of financial groups; expand agency relationships, business cooperation, and product development.