MINISTRY OF EDUCATION & TRAINING STATE BANK OF VIETNAM HO CHI MINH CITY BANKING UNIVERSITY LÊ THỊ ANH THƯ THE IMPACT OF INFORMATION - COMMUNICATION TECHNOLOGY INNOVATION ON THE PERFORM The Impact of Information - Communication Technology Innovation on the Performance of Vietnamese Commercial BanksThe Impact of Information - Communication Technology Innovation on the Performance of Vietnamese Commercial BanksThe Impact of Information - Communication Technology Innovation on the Performance of Vietnamese Commercial Banks
INTRODUCTION
Problem Statement
The Fourth Industrial Revolution has become a crucial part of Vietnam's development policies and strategies, as evidenced by numerous important resolutions and decisions from senior leaders Notably, Resolution No 52-NQ/TW issued in
2019 explicitly stated the proactive participation in the Fourth Industrial Revolution, emphasizing the importance of applying information technology to the comprehensive development of the country Additionally, the identification of the finance and banking sector as one of the eight priority areas for digital transformation under Decision 749/QĐ-TTg approved by the Prime Minister's "National Digital Transformation Program to 2025, oriented towards 2030" has affirmed the importance of applying information technology to improve the performance of commercial banks in Vietnam
These policies and decisions have laid an important foundation for research on the impact of information technology innovation on banking operations Understanding the impact of investment and development of information technology on the performance of banks in the context of a business environment and banking management increasingly dependent on digital technology is a crucial part of determining the effectiveness and sustainability of the banking sector in the future.
The necessity of the Topic
The Finance and Banking sector plays an extremely important role in economic development and is considered the lifeblood of every nation's economy (Vo Xuan Vinh, Nguyen Huu Huan & Pham Khanh Duy, 2016) Many studies have highlighted the importance of the Information and Communication Technology (ICT) system in improving productivity and economic growth at both the enterprise and national levels, concluding that delays in investing and applying advanced techniques and technologies are among the main reasons for cultural, social, and economic regression (Brynjolfsson and Hitt, 1996)
Since the widespread adoption of the Internet, technological advancements have revolutionized the banking sector, with the proliferation of electronic banking leading the charge Today, ATMs, chip cards, and online banking are ubiquitous examples of ICT in banking This technological integration has expanded market reach, introduced new products and services, and streamlined distribution channels through online and mobile platforms Banks have embraced these innovations to enhance product features and reduce costs, gaining competitive advantages and fueling financial development by catering to diverse customer segments through digital transformation.
Currently, the importance of evaluating information technology (IT) assets in determining a company's competitive health and capabilities for future business activities is increasingly gaining attention from managers, consultants, financial analysts, and economic researchers Accordingly, IT resources and capabilities are decisive to a company's ability to survive and sustainably develop, as confirmed by Nolan, who stated that IT capability was a significant differentiator for banks with effective performance in the mid-1980s compared to banks with lower profitability (Nolan, 1994)
However, many studies have shown conflicting results when examining the relationship between IT investment and corporate profitability (Brynjolfsson, 1993) Researchers have argued that besides helping to increase enterprise productivity and product value, IT also allows removing entry barriers, eliminating monopoly positions, and promoting market competitiveness, thus affecting the goal of creating long-term profits for companies investing in IT systems (Hitt and Brynjolfsson, 1996) Additionally, some opinions suggest that the characteristics of the sample used, measurement errors, and the inability to control industry-specific factors also affect performance and are among the reasons for unexpected results (Hitt and Brynjolfsson, 1996)
Therefore, the author conducts the study "The Impact of Information and
Communication Technology Innovation on the Performance of Commercial Banks in Vietnam" to determine the impact of applying ICT systems on the performance of banks in Vietnam.
Objectives of the Research
The research focuses on examining the effectiveness of IT investment and application by commercial banks (CBs) through the financial efficiency indicators of these organizations Based on this, recommendations will be proposed to help managers decide on appropriate policies to optimize operating costs and maximize benefits for the banks
This thesis examines the impact of the ICT Index on the performance of 26 CBs surveyed during the period from 2006 to 2022.
Research Question
Research question: How does the ICT Index impact the performance of
Research Subjects and Scope
Research Subject: The research subject is the impact of ICT innovation on the performance of Vietnamese CBs through the Tobin’s Q ratio
Scope of the Study: The study focuses on 26 CBs in Vietnam during the period from 2006 to 2022 These CBs have fully participated in surveys on the readiness for
IT investment and application throughout the research period
Additionally, the period from 2006 to 2022 was chosen because the data, such as the ICT Index and financial performance indicators, can be fully collected to support panel data regression analysis.
Research Methodology
The data sources presented in the research are compiled from various reports, including governance reports, annual reports, and audited financial statements published on the official websites of the enterprises Additionally, the study uses historical data on stock trading prices at year-end collected from HOSE and HNX.
Research Sample
To serve the research objectives, the author selects a research sample comprising CBs in Vietnam for which the ICT Index is available, as recorded in the reports of the Ministry of Information and Communications (ICT) during the period from 2006 to 2022 to support panel data regression estimation This is a purposive non- probability sampling method
As per the SBV's (2023) annual report, Vietnam had 35 CBs as of December 31, 2023 To ensure representativeness, 26 CBs were selected for the research sample, resulting in 137 observations, which aligns with the principles outlined by Hair et al (2006).
Contributions of the Study
The study uses Tobin's Q ratio to evaluate the performance of CBs, an aspect that previous studies in Vietnam have not considered Using this ratio helps provide a more comprehensive assessment of CBs’ performance by reflecting market expectations about the company's future profitability through stock market prices This is particularly valuable in assessing the impact of IT investments on company performance, as it may take several years for these investments to convert into profits Therefore, using Tobin’s Q is appropriate and accurately reflects the impact of IT innovation on the performance of CBs (Bharadwaj, Bharadwaj, Konsynski, 1999).
Structure of the Thesis
The thesis is expected to be presented in 5 chapters following the structure of a quantitative study, detailed as follows:
- The necessity of the Topic
Chapter 2: Theoretical Basis and Empirical Evidence, including:
Definitions related to the study, including definition of ICT innovation, ICT innovation measurement indices, definition of bank performance, and performance measurement indices
Theoretical basis related to the study, including transaction cost theory, resource- based theory, and economic growth theory The study also investigates empirical evidence from around the world and Vietnam regarding the impact of ICT development on the performance of CBs
Chapter 3: Data and Research Methodology, including:
Research data extracted from financial reports, annual reports, market capitalization information, and other relevant data from companies listed on the stock exchange, as well as data collection and processing methods
Research model, explanation of variables, selection method of regression models, and model validation
Chapter 4: Research Results and Discussion:
Presentation of regression results, validation of results, analysis, and interpretation of findings
Conclusions and main findings of the research, recommendations, limitations, and suggestions for future research directions are presented in this chapter
Chapter 1 has established the context and framework for the entire research By identifying the research problem; reasons for choosing the topic; objectives and research question; research subjects and scope; data sources and research methodology; as well as the significance and contributions of the reasearch; Chapter
1 provides readers with a comprehensive overview of the thesis’s goals and scope Additionally, this chapter introduces the thesis structure, clarifying the organization and development of the subsequent contents.
THEORETICAL BASIS AND OVERVIEW OF EMPIRICAL
Definition of relevant terms
2.1.1 The definition of ICT innovation
The process of technological innovation is a complex concept, defined in various ways and from different perspectives "Technological innovation" can be seen as a process of enhancing and improving existing technology systems (Dosi, 1983) or transforming an opportunity into practical advantage (Pavitt, 1984) Comprehensively, the concept of "innovation" is viewed as a combination of pushing the limits of existing technology and transforming potential into the best commercial advantages, thereby expanding the distribution of an organization’s goods, products, and services in the market (Rothwell and Gardiner, 1985)
2.1.2 Indicator for Measuring ICT Innovation
According to the Ministry of Information and Communications (MIC) report,
"The ICT Index reflects the level of development and application of ICT across various sectors in each country" (MIC, 2022) The ICT Index report has been conducted by the Information Technology Department under the MIC since 2006 to rank and evaluate ministries, ministerial-level agencies, 8 agencies under the Government, provincial administrative units, CBs, and large economic conglomerates
The ICT Index was created with the following objectives:
- To assess the current status of ICT application and development in enterprises, CBs, provincial administrative units, and ministries in Vietnam;
- To provide a comprehensive and clear picture of the status of ICT development and application in Vietnam;
Recognizing the critical role of ICT in driving innovation and improving public services, financial institutions, and business entities should establish comprehensive policies that maximize the efficiency of ICT resources By doing so, organizations can better align their ICT strategies with broader economic, political, and social objectives, enhancing their ability to support sustainable development and meet the evolving needs of society.
The commercial banks will be ranked based on the results of this index This helps to reflect the capacity and operational efficiency of each organization in the banking sector Additionally, the overall ICT index and the component ICT indices of each commercial bank over three consecutive years will also be displayed to assess the level of innovation and improvement in the IT systems of the organizations over different periods It can be said that ICT is a comprehensive measure of the process of development, innovation, and modernization of IT at CBs and has been used in many studies in Vietnam to measure the development and innovation of the IT system (Ngo Van Toan and Pham Le Quang, 2023; Nguyen Van Thuy, 2021; Le Vu Toan Linh & Pham Duy Khanh, 2022; Nguyen Huu Manh & Vuong Thi Huong Giang, 2022)
After receiving the official dispatch from the MIC regarding the collection of data for the annual ICT Index Report, the IT department of each CB will collect and compile data in accordance with the indicator groups of each organization and fill in the data collection form according to the template provided by the MIC
Besides, the input data for calculating the ICT Index is also collected from reliable sources such as telecommunications service providers VNPT, FPT, Viettel, etc., when exploiting data related to telecommunications infrastructure Additionally, data from other official sources, such as the websites of surveyed units, data from the General Statistics Office, and Statistical Yearbooks, have been used as a basis for verification
Regarding CBs sector, there are four main component indicator groups used to calculate the ICT index, including: technical infrastructure, human infrastructure, internal banking applications, and online banking services, specifically:
- Technical Infrastructure, evaluated through criteria such as: o Server and workstation infrastructure; o Communication infrastructure; o ATM and POS infrastructure; o Implementation of information security and data safety solutions; o Data centers and disaster recovery centers
- Human Infrastructure, evaluated through criteria such as: o Ratio of IT-specialized staff; o Ratio of information security-specialized staff; o Ratio of IT-specialized staff with international IT certifications to the total number of IT-specialized staff
- Internal Banking Application, evaluated through criteria such as: o Core banking implementation; o Basic application implementation; o Electronic payment implementation
- Banking Online Services, evaluated through criteria such as: o Bank's website; o Internet banking for individual customers; o Internet banking for corporate customers; o Other electronic banking services
Accordingly, the component indices of this index are “applied using the dual normalization method according to Z-Score and Min-Max when calculating the component indices and using independent expert evaluations of online public services to align with the calculation method of the United Nations E-Government Survey” Subsequently, the composite ICT Index will be calculated by averaging the values of these four component indices (MIC, 2022)
After being collected from CBs, the raw data is standardized using the Z-Score method before being used to calculate the component indices following the equation 2.1:
- A n : Value of criterion A after standardization using the Z-Score method;
After standardization, the values of the component indices are calculated using the formula 2.2:
- m: Total number of sub-criteria in group j;
- A n : Value of criterion A after standardization using the Z-Score method;
- Tk j: Value of the k-th component index in group j
Then, the component index T is normalized using the Min-Max method to bring it to a value range of |0 – 1| according to the formula 2.3:
- T n : Value of the component index T after standardization using the Z-Score method;
- Tmax: Maximum value of index T;
- Tmin: Minimum value of index T
After checking, adjusting, and supplementing the data, it is standardized and used to calculate the component indices The Principal Component Analysis (PCA) method is used to calculate the correlation coefficients of the component indices using the SPLUS Professional Release 3 software
After obtaining the correlation coefficients, the main ICT index is calculated by averaging the component indices, specifically as stated in equation 2.4:
- I: Main ICT index for CBs;
- THTKT: ICT component index for Technical Infrastructure;
- THTNL: ICT component index for Human Infrastructure;
- TUDNB: ICT component index for Internal Bank Applications;
- TDVTT: ICT component index for Online Services
This composite ICT index will be used to rank each unit within the same industry sector These data are considered very useful for regulatory agencies in directing, managing, and policy-making For organizations and individuals involved in research, policy consulting related to ICT development and application, this data serves as a valuable reference, sometimes being the only source available Therefore, in this thesis, the author chooses the ICT index to measure the level of ICT system innovation in the surveyed CBs
According to Ngo Dinh Giao (1997): The economic efficiency of an economic phenomenon or process is an economic category that reflects the level of resource utilization (human, financial, material, and capital resources) to achieve specific objectives It shows the relationship between the obtained results and all costs incurred to achieve those results, reflecting the quality of that economic activity The greater the difference between these two quantities, the higher the efficiency
Performance efficiency encompasses Cost Efficiency (CE), also known as Cost Economy, which includes Technical Efficiency (TE) and Allocative Efficiency (AE) TE measures how effectively output is maximized with available inputs, comprising Pure Technical Efficiency (PTE) and Scale Efficiency (SE) AE assesses the optimal use of inputs based on their prices, maximizing profit or revenue Economic efficiency demands minimizing costs or maximizing outputs, ensuring efficient resource allocation.
There are many indices reflecting the performance of enterprises, such as Return on Assets (ROA) and Return on Equity (ROE) However, in this thesis, the author uses Tobin’s Q to measure the performance of CBs due to its superior characteristics in measuring the impact of IT innovation
Tobin’s Q ratio, or simply Q, is defined as the market value of a company divided by its assets’ replacement costs This index was first introduced by James Tobin in
1969 as a measure to predict a company's future investments (Tobin, 1969, 1978) Since then, Q has been used in numerous studies for various purposes, such as (Bharadwaj và ctg, 1999):
- A proxy measure for business performance;
- A tool to predict profitable investment opportunities;
- Measuring the market capitalization of monopoly leases;
- Measuring the returns from diversification;
- Measuring the value of technological assets
Built on a robust theoretical and empirical foundation from the Efficient Market Hypothesis (Fama, 1970, 1991; Ball, 1995), Tobin’s Q possesses many superior characteristics, such as (Lubatkin và Shrieves, 1986):
- Providing an assessment basis for investors on the impact of management decisions;
- Being adjusted for market fluctuations, inflation, and market risk of the company;
- Comprehensively reflecting all aspects of a company's current operations;
- Being objectively recorded through the market stock price
Limitations of Traditional Accounting Indices
In-depth studies in finance and strategy, such as research by Montgomery and Wernerfelt (1988) has presented arguments against using performance evaluation indices based on accounting data due to the following limitations:
- Reflecting only past information without indicating future performance;
- Not being adjusted for risk;
- Being prone to distortions due to temporary imbalances, tax laws, and accounting conventions
Additionally, accounting indices of performance are ineffective in reflecting time-lagged impacts This is particularly detrimental when evaluating the impact of
IT investments on company performance, as it may take several years for these investments to convert into profits
DATA AND RESEARCH METHODOLOGY
Research Model
Based on previous studies such as Wang et al (2021), Gupta et al (2018), Pierri and Timmer (2022), particularly the research by Nguyen Huu Manh & Vuong Thi Huong Giang (2022), and Vu Thi Huyen Trang et al (2022) in Vietnam, the author proposes an estimation model for the impact of ICT system innovation on the performance of 26 CBs in Vietnam, as depicted in equation 3.1 as follows:
Yit = α0 + α1 ICTit + α2 BANKit + α3 MACROit + eit (3.1) Where: α0: Intercept coefficient α1-3: Coefficients of independent variables i, t: Bank i in year t
Yit: Performance of CBs measured by Tobin's Q index
ICTit: Level of application and development of ICT represented by the ICT index
BANKit: Control variables specific to banks (including SIZE representing bank size, LDR representing loan-to-deposit ratio, and LLP representing loan loss provision ratio)
MACROit: Macro-environmental factors (including GDP index representing economic growth rate and INF index representing inflation rate) eit: Model error term.
Research Data
This study utilizes data collected directly from authoritative sources such as financial reports, management reports, and annual reports published on the official CBs’ websites
The author planned to collect data from 35 CBs operating as of December 31,
2023, according to statistics from the State Bank of Vietnam (SBV, 2023) However, to align with the research model using Tobin's Q as an indicator of performance, the author excluded samples that did not meet the criteria, including banks not listed on stock exchanges due to lack of market capitalization data Therefore, only 26 CBs met the conditions for this study
After collecting secondary data from the sources listed in Table 3.1, the author processed the data according to the methods outlined in Table 3.1 to create input data for the research model This model includes 01 dependent variable Q (Tobin's Q index) representing the performance of CBs, and 06 independent variables: ICT, LDR, SIZE, LLP, INF, GDP Among these variables, ICT is the most important explanatory variable in the model, representing the degree of readiness in applying ICT within the organization
Table 3.1 Summary of sources and preliminary calculation methods for variables used in the model
No Variable Meaning Calculation Sources
Performance of CBs indicated by Tobin’s
(Book value of debt + Market capitalization of the enterprise) / Total assets
2 ICT Degree of readiness in applying ICT - MIC
3 LDR Loan-to-deposit ratio
4 SIZE Scale of the bank Ln (Total assets) Financial statements
5 LLP Loan loss provision expense ratio
Loan loss provision expense / Total outstanding loans
7 GDP Economic growth rate - Worldbank
Source: Compiled by the author
Explanation of Variables and Research Hypotheses
In this study, Tobin’s Q serves as the dependent variable in the model to measure the performance of CBs in Vietnam, replacing traditional accounting ratios There are various methods to calculate Tobin’s Q, but research by Chung and Pruitt (1994) suggests no significant difference among these methods Therefore, the author chooses the calculation method proposed by Khanna and Palepu (2000) due to its simplicity and availability of financial and accounting data from credible databases The formula for Tobin’s Q is represented by Equation 3.2 as follows:
𝑄𝑄 = Market value of equity at year t+Book value of debt at year t
3.3.2 Explanatory Variables, Control Variables, and Research Hypotheses ICT Index - Degree of Readiness for ICT Application and Development
According to Agu and Aguegboh (2020), the application of ICT in banks is expected to enhance competitiveness and improve the efficiency of financial activities for those banks This not only helps banks save operational costs but also enhances service quality, improves customer experience, and introduces new financial products to meet market demands Additionally, banks equipped with modern ICT systems are better positioned to survive in volatile economic conditions and exhibit superior recovery capabilities through effective borrower screening, thereby reducing non-performing loan ratios and credit default risks (Dadoukis et al., 2021) Similar results were found in the studies by Pierri and Timmer (2022) Therefore, the author proposes the following hypothesis:
H1: ICT positively influences bank performance
The bank scale index (SIZE) is determined by taking the natural logarithm of total assets This index represents the competitive advantage of a business in terms of scale efficiency According to the economies of scale theory, larger financial institutions have cost advantages over smaller ones (Stigler, 1958) Scale expansion helps large banks improve performance by spreading production costs over a large quantity of products, thereby reducing average costs per unit This enables larger banks to offer services at lower costs, enhancing product competitiveness and attracting a larger customer base, thereby improving organization profitability (Khemani and Shapiro, 1993)
Furthermore, optimizing production costs according to economies of scale theory is achieved by reducing input costs through high-volume discounts from suppliers (Stigler, 1958)
In summary, the implications of economies of scale theory indicate that large banks have cost advantages over smaller banks and can achieve higher profits Therefore, the author expects bank scale to positively influence bank performance
Loan-to-Deposit Ratio - LDR
This index measures the extent to which borrowed funds are used to extend credit, reflecting a bank's liquidity A higher LDR signifies that the bank effectively fulfills its financial intermediary function As banks expand credit activities, they correspondingly increase profitability This aligns with the studies of Agu and Aguegboh (2020), Võ Xuân Vinh & Mai Xuân Đức (2017) Therefore, the author expects the loan-to-deposit ratio to positively influence bank performance
Loan Loss Provision Ratio - LLP
This measure represents the credit risk loss in lending activities, determined by loan loss provisions over total assets, indicating expected loss and serving as an early measure of actual loss from lending A high loan loss provision expense ratio significantly impacts a bank's profitability and reduces profit (Acharya et al., 2006; Berger et al., 2010) Therefore, the author expects the loan loss provision expense ratio to negatively influence bank performance
The economic growth rate index (GDP) measures the total market value of domestically produced goods and services within a specific territory over a defined period This is a critical index reflecting the development of an economy When this index increases, the market becomes more attractive to investors, thereby stimulating vibrant economic activities
The banking sector plays a crucial role in economic growth GDP expansion signals strong economic performance, leading to increased consumer spending, production, and credit activities Studies have shown that GDP growth positively impacts bank lending, market credit development, and profitability Consequently, it is predicted that economic growth will have a positive influence on bank performance.
Inflation is another macroeconomic factor that affects the overall economy, represented by the continuous increase in the general price level of goods and services over time, coupled with currency depreciation Therefore, during inflationary periods, depositors tend to increase investments or spending rather than keeping money in banks to protect the value of currency against continuous depreciation This leads banks to increase capitalizing costs and face higher financial instability Studies by Criste and Lupu (2014) and Fadzlan Sufian and Chong (2008) both show a negative correlation between inflation rates and bank profitability Therefore, the author expects the inflation rate to negatively influence bank performance.
RESEARCH RESULTS AND DISCUSSION
Descriptive Statistics of Variables in the Model
The descriptive statistics of the dataset from 26 CBs in Vietnam, collected during the period 2006 – 2022, are presented in Table 4.1:
Table 4.1 Descriptive Statistics of Variables in the Research Model
Var Obs Mean Std Dev Min Max
Source: Compiled by the author from STATA 16.0
Table 4.1 provides data on the number of observations, mean value, standard deviation, minimum value, and maximum value of the variables in the model Accordingly, with 137 recorded observations, the values of the dependent and independent variables in the model are as follows:
- Tobin’s Q: The mean value is 1.03, ranging from a maximum of 1.3316 (Asia
Commercial Bank – ACB, in 2006) to a minimum of 0.9575 (An Binh Commercial Joint Stock Bank – ABB, in 2020) with a standard deviation of 0.0538 Overall, although ACB was the most efficient bank during the survey period, it did not maintain this position Instead, Vietcombank (VCB) took over ACB’s position in 2009 and has maintained its leading efficiency for 17 years This could be because ACB leveraged its early listing advantage on the Vietnamese stock exchange to attract significant domestic and foreign investor interest, resulting in the outstanding growth recorded in 2006 Regarding VCB, after the official listing in 2009, VCB also leveraged this advantage to take the lead, combined with its predominantly state-owned capital structure and innovative operational policies, helping VCB sustain top efficiency among CBs
- ICT Index: The mean value is 0.5477, ranging from a maximum of 0.8200
(Bank for Investment and Development of Vietnam – BIDV, in 2015) to a minimum of 0.2654 (National Citizen Bank – NCB, in 2017) with a standard deviation of 0.1211 The data also shows that BIDV maintained the highest ICT index throughout the survey period However, in 2022, Tien Phong Commercial Joint Stock Bank (TPB) and Techcombank (TCB) surpassed BIDV with scores of 0.71
- SIZE: The mean value is 19.3981, ranging from a maximum of 21.4750
(BIDV, in 2022) to a minimum of 16.8121 (NCB, in 2010) with a standard deviation of 0.9505 Generally, the size of banks tends to increase steadily over time, with BIDV, one of the "Big 4" banks, maintaining the largest size throughout the study period
- LDR (Loan to Deposit Ratio): The mean value is 0.8476, ranging from a maximum of 1.3083 (Vietnam Prosperity Joint Stock Commercial Bank – VPB, in 2017) to a minimum of 0.4778 (ACB, in 2006) with a standard deviation of 0.1449 The LDR of credit institutions does not follow a common trend, indicating that each bank has unique goals and development strategies to balance liquidity risk and credit limits, thus setting different LDR ratios suitable for their developmental strategies over different periods
- LLP (Loan Loss Provisions): The mean value is 0.0133, ranging from a maximum of 0.0327 (VCB, in 2009) to a minimum of 0.0035 (ACB, in 2006) with a standard deviation of 0.0049 Provisions are reserves set aside by banks to cover bad debts and potential credit risks The general trend shows that banks increasingly focus on financial safety and risk management through the rising LLP ratio from late 2020 This might be a consequence of the COVID-
19 pandemic, which severely impacted the global economy, leading to an increase in bad debts, thus prompting higher provisioning ratios to protect banks from credit risks and ensure the stability and sustainability of the financial system
- INF (Inflation): The mean value is 4.1794, ranging from a maximum of
18.6777 (in 2011) to a minimum of 0.6312 (in 2015) with a standard deviation of 2.9451 Macroeconomic factors such as fuel prices, monetary policies, and global commodity price fluctuations are among the main factors affecting this inflation index
- GDP (Gross Domestic Product): The mean value is 6.3729, ranging from a maximum of 8.0198 (in 2022) to a minimum of 2.8654 (in 2020) with a standard deviation of 1.6670 This can be explained by the severe impact of the COVID-19 pandemic on the Vietnamese economy in 2020, with lockdown directives and border closure decrees halting the global economy After efforts to curb the pandemic, 2022 saw economic recovery and improvement, leading to a GDP nearly three times higher than in 2020.
Correlation Analysis of Variables in the Model
The correlations among the research variables were examined by using STATA 16.0 software, based on the data of 26 CBs from 2006 to 2022 The correlation coefficients of the variables are shown in Table 4.2
Q ICT SIZE LDR LLP INF GDP
Source: Compiled by the author from STATA 16.0 software
The results of the correlation coefficient analysis between the variables in the model, as shown in Table 4.2, align well with the author’s expectations The coefficients indicate that ICT, SIZE, LLP, and INF positively correlate with Tobin’s
Q Conversely, the LDR and GDP negatively correlate with this performance indicator
Additionally, the correlation matrix of the independent variables was constructed to detect any signs of high correlation among variables in the model According to Hair, Black, Babin, Anderson, and Tatham (2006), multicollinearity exists when the absolute value of the correlation between independent variables is 0.9 or higher However, Pallant (2020) suggests that 0.7 should be the threshold for multicollinearity among independent variables Table 4.2 shows that all absolute values of the correlation coefficients between variables do not exceed 0.7 Therefore, there is no high correlation between the selected variables, and multicollinearity does not exist in the research model.
Multicollinearity Test
A multicollinearity test is conducted to examine the correlation among variables in the model using the Variance Inflation Factor (VIF) method The results of the multicollinearity test are shown in Table 4.3
Source: Compiled by the author from STATA 16.0 software
According to Hair et al (2006) and Pallant (2020), if the VIF value is greater than
10 and the 1/VIF value is less than 0.1, there is potential multicollinearity in the model The results from Table 4.3 show that the mean VIF value is 1.30, and the VIF values of the independent variables in the model are all less than 10, while the Tolerance values (1/VIF) of the variables are all greater than 0.1 Thus, there is no multicollinearity in the model.
Regression Results and Model Selection
Table 4.4 Regression Results of Variables by OLS, REM, FEM Methods
F (6,130) = 2,30 Wald chi2(6) = 15,66 F (6,105) = 5,87 Prob > F = 0,0382 Prob > chi2 = 0,0157 Prob > F = 0,0000
Source: Compiled by the author from STATA 16.0 software Note: The symbols (*), (**), (***) in Table 4.3 represent statistical significance levels of 10%, 5%, and 1%, respectively
The author conducted regression analyses using STATA 16.0, utilizing common regression models such as Pooled OLS, REM, and FEM, and obtained the results shown in Table 4.4 Subsequently, the author performed various tests to select the Best Linear Unbiased Estimator (BLUE), specifically as follows:
F-test to choose between OLS and FEM
The author conducted the F-test with the null hypothesis H0: The OLS model is appropriate The test result was F (25, 105) = 5,98 with a Prob > F value of 0.0000 < 5% Therefore, the null hypothesis H0was rejected This means the FEM model is more appropriate than OLS for estimating the research model
Breusch-Pagan test to choose between OLS and REM
The author conducted the Breusch-Pagan test with the null hypothesis H0: The model's random errors have zero variance The test result was chibar2(01) = 49,77 with Prob > chibar = 0.0000 < 5% Therefore, the null hypothesis H0 was rejected This means the REM model is more appropriate than OLS for estimating the research model
Hausman test to choose between FEM and REM
The author conducted the Hausman test with the null hypothesis H0: There is no correlation between the residuals and the independent variables of the model The test result was chi2(7) = (b-B)'[(V_b-V_B) ^ (-1)] (b-B) = 107,74 with Prob > chi2
= 0.0000 < 5% Therefore, the null hypothesis H0 was rejected This means the FEM model is more appropriate than REM for estimating the research model.
Model Diagnostic Tests
To test for heteroscedasticity in the model, the author used the Wald test for the FEM model and obtained the results chi2 (26) = 1.4e+33, Prob>chi2 = 0.0000 < 5%, thus rejecting the null hypothesis (H0) This indicates that heteroscedasticity is present in the model
To test for autocorrelation in the model, the author used the Wooldridge test and obtained the results F (1, 13) = 82,508 with Prob > F = 0.0000 < 5%, thus rejecting the null hypothesis (H0) This indicates that first-order autocorrelation is present in the model
The Durbin-Wu-Hausman test was used to detect endogeneity in the research model The result was p = 0.0806 < 0.1, thus rejecting the null hypothesis (H0) This indicates that endogeneity is present in the model.
Model After Correcting for Defects
Due to model defects such as heteroscedasticity, autocorrelation, and endogeneity, the System Generalized Method of Moments (S.GMM) is the most suitable estimation technique (Roodman, 2009) To account for these defects, the researcher employed the S.GMM method and conducted the Arellano-Bond, Sargan, and Hansen tests during model estimation.
Table 4.5 Regression Results of Variables Using S.GMM Estimation Variable Coefficient P-value Expected Previous Studies
Ngo Van Thuy (2021); Nguyễn Hữu Mạnh & Vương Thị Hương Giang
Võ Xuân Vinh & Mai Xuân Đức (2017)
Arellano-Bond test for AR (2) in first differences: z = 0.57 Pr > z = 0,569
Sargan test of overid restrictions: chi2(3) = 1,58 Prob > chi2 = 0,664
Hansen test of overid restrictions: chi2(3) = 0,52 Prob > chi2 = 0,915
Source: Compiled by the author from STATA 16.0 software Note: The symbols (*), (**), (***) in Table 4.3 represent statistical significance levels of 10%, 5%, and 1%, respectively
The data in Table 4.5 shows that the S.GMM model is suitable for the research because it meets the conditions and corrects the defects of the model Specifically:
- The number of instruments < number of groups (11 < 16) (Roodman, 2009);
- The Arellano-Bond test has a Pr > z value of 0,569 > 1%, indicating no second-order autocorrelation in the model (Arellano and Bond, 1991);
- The Sargan and Hansen tests have Prob > chi2 values of 0,664 and 0,915, respectively Both values are > 5%, proving that the instruments used in the model are appropriate (Sargan, 1958; Hansen, 1982).
CONCLUSIONS AND RECOMMENDATIONS
Conclusions
In the context where digital transformation is becoming increasingly prevalent and crucial across all industries, the adoption and implementation of digital technologies are emerging as key factors for businesses and organizations to enhance productivity, optimize processes, and create new values Therefore, stemming from the desire to examine the impact of IT investment and innovation on the performance of banks in Vietnam, the author conducted the study titled "The Impact of Information – Communication Technology Innovation on the Performance of Vietnamese Commercial Banks" Based on data from 26 CBs in Vietnam during
2006 – 2022, collected from highly reliable official sources, the author used the System Generalized Method of Moments (S.GMM) to estimate the research model The research results presented in Chapter 4 provide a basis for accepting hypothesis H1: ICT has a positive impact on the performance of banks By demonstrating a positive relationship between the ICT Index and the performance of CBs, the research indicates that enhancing investment in IT systems is not just a temporary trend but a critical factor in improving performance and competitiveness for CBs Additionally, the study reveals a positive correlation between equity size, organizational size, provisioning ratio, and the performance of surveyed banks Conversely, there is a negative correlation between loan-to-deposit ratio, inflation rate, economic growth rate, and the performance of these banks
To assess the impact of IT on financial performance, this study employs Tobin's Q as a measure, aligning with Nolan's (1994) emphasis on future revenue generation potential Tobin's Q, derived from market valuation, overcomes limitations of traditional book measures, providing a comprehensive representation of IT investment effectiveness This approach contributes to the understanding of the positive influence of the ICT Index on the potential future performance of Vietnamese commercial banks.
Suggestions and Recommendations
Based on the research findings, the thesis proposes the following suggestions and recommendations:
5.2.1 Enhance Investment, Innovation, and Application of ICT Systems
Based on the research results on the impact of ICT innovation on the performance of CBs in Vietnam, the thesis offers several policy implications to optimize the benefits from IT and improve banking performance
Generally, CBs have actively engaged in digital transformation activities, not only in internal operations like digitizing internal document approval systems at leading banks such as VCB, BIDV, HDBank, etc., to minimize paper documents and expedite processing, but also in professional operations like electronic identification (eKYC), cloud computing, big data, artificial intelligence (AI), etc However, this digitalization process still faces limitations in terms of research and development (R&D) investment According to World Bank statistics, Vietnam’s R&D investment has been modest, maintaining at 0.4% of GDP from 2017 to 2021, compared to 3% in the US and 2.2% in China To keep pace with global developments, Vietnamese CBs need to establish dedicated fintech R&D centers to test and implement new initiatives This will not only maintain competitive positions but also improve banking services to meet the increasing customer demands
Moreover, labor structure shifts are creating opportunities and challenges for the banking workforce While many technology-related positions are opening up, repetitive jobs are being automated by AI, leading to an increasing need of 8-9% in the quality of this workforce towards specialization and adaptation to modern technology requirements (Manyika et al., 2017)
According to the World Economic Forum (WEF), 65% of new jobs related to digital transformation will appear in Southeast Asia within the next 20 years, causing 56% of the total workforce in the region to face unemployment risks This underscores the importance of equipping traditional workforce with digital skills
Banks need to organize continuous training programs to enhance employee skills in new technologies, including cybersecurity, data analytics, and advanced banking software A well-trained tech-savvy workforce will help banks operate more efficiently and quickly adapt to continuous changes in the financial industry According to Khanna (2014), 86% of users stop using services after experiencing poor service, thus promoting digital technology application is also crucial Banks should expand and strongly implement digital banking services such as mobile banking, internet banking, and contactless payment services These services not only enhance customer experience but also reduce operating costs and expand service reach to remote areas
Additionally, to drive banking digitalization, the amended Law on Credit Institutions
No 32/2024/QH15, passed by the National Assembly on January 18, 2024, includes key provisions to promote digital banking activities This demonstrates strong commitment to accelerating digital transformation and modernizing banking services It is crucial to enact specific mechanisms and policies that support financial incentives and create favorable conditions for technology innovation projects These policies will serve as a catalyst for banking sector development and encourage banks to invest in technology (Hoàng Công Gia Khanh & Trần Hùng Sơn, 2020) However, given the rise in high-tech crimes, ensuring safety, security, and customer data privacy remains a significant challenge According to the MIC, the banking-finance- securities sector has the highest spending on information security measures, leading in IT risk management expenses This highlights the urgent need for a comprehensive legal framework, including specific regulations on data management and protection, enhanced collaboration between credit institutions and regulatory bodies, and increased awareness and skills among employees and customers on cybersecurity Only then can the banking sector develop sustainably in the digital era
5.2.2 Policies Related to Bank Scale
As of April 30, 2024, the banking sector in Vietnam reported total assets of VND 17,270 trillion, representing a slight slowdown in growth compared to the previous year State-owned commercial banks held assets of VND 8,321.8 trillion while joint-stock commercial banks accounted for VND 8,948.2 trillion This growth moderation is attributed to a 0.05% and 0.43% decrease in asset growth for state-owned and joint-stock commercial banks, respectively.
The government has encouraged the expansion of the Vietnamese commercial banking system to promote economic development through the “Restructuring the system of credit institutions for the 2011-2015 period" project, as per Decision No 254/QD-TTg This policy continues to be reinforced through the “Restructuring the system of credit institutions associated with bad debt settlement for the 2021-2025 period” project, as per Decision No 689/QD-TTg dated June 8, 2022, by the Prime Minister
Research results show that bank scale positively impacts performance Therefore, Vietnamese CBs need to focus on growth measures to enhance performance This includes expanding branch networks and transaction points, especially in rural and remote areas, to reach more customers Additionally, mergers and acquisitions (M&A) play a crucial role in bank scale growth Merging or acquiring banks not only creates larger organizations but also combines resources and capabilities to form stronger financial bases This improves service and product offerings, optimizes operating costs, and enhances overall performance Consequently, SBV and regulatory agencies need to facilitate M&A activities The government can offer support policies, tax incentives, and streamline legal procedures to promote this process M&A not only allows banks to grow rapidly but also enhances competitiveness, optimizes resources, and expands market share
5.2.3 Policies Related to Provisioning Ratios
Due to heightened concerns over rising non-performing loans (NPLs), commercial banks (CBs) in Vietnam have significantly increased their provisioning expenses In response to the State Bank of Vietnam's (SBV) extended debt restructuring policies, banks like HDBank, ABBank, ACB, and Techcombank have doubled their provisioning compared to Q1 2024, indicating their commitment to addressing bad debts and supporting the economic recovery by the end of 2024.
Provisioning ratios significantly impact Vietnamese commercial banks' (CBs) performance, underscoring the importance of managing credit risk through appropriate provisioning To ensure sustainable banking development, policies controlling provisioning ratios are crucial The State Bank of Vietnam (SBV) should establish clear regulations and guidelines for provisioning, which should be regularly revised based on economic conditions and international standards Banks must adopt advanced credit risk assessment methods, utilizing predictive models and big data analysis, to optimize provisioning decisions Periodic monitoring and inspection by the SBV and supervisory agencies are essential to ensure compliance and mitigate risks Additionally, banks should invest in training programs to enhance the skills of risk management staff, thereby improving the effectiveness of provisioning decisions and contributing to overall banking performance.
In the context where Vietnam in particular, and the world in general, is facing macroeconomic challenges such as geopolitical conflicts, rising inflation, prolonged high interest rates, and volatile prices of essential commodities, it is crucial to issue policies that closely follow macroeconomic developments to ensure the sustainable and safe development of financial institutions
Understanding the impact and importance of macroeconomic factors will help Vietnamese CBs capture the overall economic picture, thereby devising appropriate business and investment strategies Macroeconomic factors such as GDP and inflation rates not only affect the operating environment but also influence banks' decision-making in ICT innovation in various aspects Firstly, GDP growth reflects the overall economic health; when GDP grows well, income and consumer demand also increase, creating opportunities for banks to expand lending and investment activities Understanding GDP trends will help banks forecast and plan financially, as well as seek investment opportunities in ICT to improve service quality and performance For example, when the economy develops, the demand for digital banking services and online payments increases, requiring banks to invest in advanced technologies to meet this demand Secondly, controlling inflation is crucial for maintaining economic stability Stable inflation helps banks accurately predict costs and profits, thereby building effective investment strategies Conversely, high and unstable inflation can pose numerous risks, reducing the real value of loans and increasing operating costs Understanding the SBV's inflation control policies will help CBs take preventive measures to maintain stable operations and continue investing in ICT innovation Finally, a deep understanding of macroeconomic factors will help CBs adjust their business strategies in line with economic fluctuations For instance, in the context of changing interest rates, banks can adjust financial products to optimize profits and manage risks Additionally, understanding economic trends will guide banks in directing ICT investment projects to align with actual market needs, thereby enhancing competitiveness and performance In conclusion, understanding the impact and importance of macroeconomic factors will help Vietnamese CBs build effective business strategies, make reasonable ICT investments, improve performance, and better meet customer needs in an increasingly volatile economic environment.
Limitations of the Research and Suggestions for Future Research
Despite the significant contributions of this study through the obtained results, there are still some limitations as follows:
Tobin's Q has limitations as a performance measure It gauges capital market efficiency rather than firm performance (Shepherd, 1986) Its formula relies on a subjective market definition, leading to varied Q values (Roll, 1977) It overlooks market noise and information asymmetries (Copeland et al., 2005) These limitations impact the reliability of Tobin's Q as a measure of firm value.
Data for the ICT Index is self-reported by enterprises, potentially introducing errors due to data collection and sampling issues Despite this, data standardization methods enhance reliability However, comprehensive reporting on ICT indices is limited, presenting challenges for data collection and estimation Furthermore, Vietnam's emerging stock market leads to limited data on bank stock values, posing obstacles for research.
To continue exploring and clarifying the impact of IT factors on bank profitability, future research can be conducted using other independent variables representing IT investment levels, such as IT investment costs, or studying the component impacts of the ICT Index (technical infrastructure index, human resources infrastructure, internal bank applications) on performance through Tobin’s
Q As banks increasingly focus on digitalization, reports on ICT indices and related indicators are becoming more common, facilitating future research access This will expand the scope of research, enhance the reliability of data sets, and improve research outcomes.
1 Bộ Chính trị (2019) Nghị quyết số 52-NQ/TW của Bộ Chính trị về một số chủ trương, chính sách chủ động tham gia cuộc cách mạng công nghiệp lần thứ tư, ban hành ngày 27/09/2019
2 Bộ Thông tin và Truyền thông (2006), Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2006, Hà Nội
3 Bộ Thông tin và Truyền thông (2009), Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2009, Hà Nội
Theo Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2010 của Bộ Thông tin và Truyền thông (2010), Hà Nội, Việt Nam đã đạt được những bước tiến đáng kể trong lĩnh vực công nghệ thông tin và truyền thông (CNTT&TT).
5 Bộ Thông tin và Truyền thông (2011), Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2011, Hà Nội
6 Bộ Thông tin và Truyền thông (2012), Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2012, Hà Nội
Theo báo cáo chỉ số sẵn sàng phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2013 của Bộ Thông tin và Truyền thông, Việt Nam đứng thứ 117 trên 133 quốc gia về chỉ số này Đây là một kết quả chưa thực sự khả quan, đòi hỏi cần nỗ lực nhiều hơn nữa để cải thiện năng lực tiếp cận và ứng dụng công nghệ thông tin và truyền thông trong quá trình phát triển kinh tế - xã hội của đất nước.
Bộ Thông tin và Truyền thông công bố Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2014 nhằm đánh giá toàn diện và khách quan thực trạng phát triển CNTT-TT tại Việt Nam Báo cáo này là nguồn tham khảo quan trọng cho các nhà hoạch định chính sách, doanh nghiệp và viện nghiên cứu trong việc đưa ra các quyết định và triển khai các hoạt động liên quan đến CNTT-TT tại Việt Nam.
9 Bộ Thông tin và Truyền thông (2015), Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2015, Hà Nội
Theo báo cáo của Bộ Thông tin và Truyền thông năm 2016, Việt Nam đã đạt được những tiến bộ đáng kể trong việc phát triển và ứng dụng công nghệ thông tin và truyền thông (ICT) Báo cáo này cung cấp một đánh giá toàn diện về tình trạng ICT của Việt Nam, bao gồm cả sự phát triển của cơ sở hạ tầng, mức độ tiếp cận, và sử dụng các dịch vụ ICT.
11 Bộ Thông tin và Truyền thông (2017), Báo cáo chỉ số sẵn sàng cho phát triển và ứng dụng công nghệ thông tin và truyền thông Việt Nam năm 2017, Hà Nội