THE STATE BANK OF VIETNAM MINISTRY OF EDUCATION AND TRAINING BANKING UNIVERSITY OF HO CHI MINH CITY TRA THANH THANH FACTORS AFFECTING THE CAPITAL STRUCTURE OF ENTERPRISES LISTED ON THE VIETNAMESE STOC[.]
INTRODUCTION
Reasons for choosing the topic
Capital structure is a crucial aspect of corporate financial management, attracting extensive research worldwide and domestically Studies aim to identify factors influencing a company's capital structure, assess its impact on corporate value, and develop models to determine the optimal capital structure Determining a reasonable capital structure is complex, as it varies according to the company's situation and growth cycle Optimizing capital structure is essential for financial managers to minimize the weighted average cost of capital (WACC) and maximize shareholder value.
Understanding the factors that influence a company's capital structure is essential for effective financial management Identifying these factors enables managers to optimize funding decisions and adjust capital sources accordingly A well-structured capital strategy supports the company's growth at different stages, as what is suitable during one phase may not be appropriate in another By analyzing the impact of each factor, managers can achieve a balanced and adaptable capital structure that aligns with the company’s development goals.
Vietnam has limited research on the optimal capital structure of joint stock enterprises, largely due to a lack of emphasis on financial management Many Vietnamese joint stock companies originated from state-owned enterprises, inheriting outdated obligations and interests that hinder focus on capital efficiency However, amidst economic renewal, equitization, and global integration, effectively utilizing resources, including capital, has become crucial for enterprise development Therefore, studying the factors influencing the capital structure of Vietnamese listed companies aims to identify key determinants, evaluate their impact, address existing shortcomings, and propose solutions to enhance financial leverage efficiency.
Research objectives
This research paper aims to assess the extent and direction of the impact of factors affecting the capital structure of enterprises listed on the Vietnamese stock market
In order to achieve the above general goal, the thesis needs to clarify the following contents:
- Survey to show the factors affecting the capital structure of listed enterprises in Vietnam
- Analyze the direction of influence and the impact of each factor on the use of financial leverage of the business
- Find out the outstanding problems of Vietnamese enterprises when using financial leverage
- Propose a number of solutions to improve the efficiency of the use of financial leverage for businesses.
Research question
The thesis raises four research questions based on the aforementioned thesis objectives:
1 What factors affect the capital structure of enterprises listed on the Stock Exchange of Vietnam?
2 What is the direction of influence of these factors?
3 What are the problems that still exist in the capital structure of enterprises listed on the Stock Exchange of Vietnam?
Research subject and scopes
This study focuses on the top 39 non-financial companies in Vietnam with the highest total capital as of 2021, listed on the Hanoi Stock Exchange (HNX) and the Ho Chi Minh Stock Exchange (HOSE) The research dataset comprises 380 observations collected between 2012 and 2021, providing comprehensive insights into these leading Vietnamese firms.
This study examines the impact of key factors on the capital structure of 39 non-financial companies listed on Vietnam's HNX and HOSE stock exchanges between 2012 and 2021 It provides valuable insights into how financial and non-financial variables influence companies' debt and equity decisions in the Vietnamese market The research highlights the significance of understanding these factors to optimize capital structure strategies for sustainable growth The findings contribute to the existing literature on corporate finance in emerging markets, emphasizing the importance of context-specific analysis for effective financial management.
This study focuses exclusively on sample data from publicly listed companies on the stock exchange, emphasizing large-scale enterprises where market volatility has a substantial impact Non-financial small-sized companies have been excluded to ensure the analysis centers on major players influencing overall market trends Due to limited time and resources, the research concentrates only on large enterprises, providing valuable insights for financial managers, investors, and investment funds seeking to understand market dynamics and volatility.
Research contribution
In terms of scientific significance, the thesis codifies and further clarifies the theoretical basis of capital structure including concepts and theories of capital structure
This article thoroughly examines the key factors influencing the capital structure of consumer goods companies listed on HNX and HOSE The study identifies critical determinants that impact financial decisions in these enterprises Based on the findings, practical recommendations are provided to optimize and improve the capital structure of these consumer goods companies, enhancing their financial stability and growth potential.
Research outline
The structure of the topic consists of 4 chapters:
Chapter 4: Research results and discussion of research results
Chapter 5: Some solutions to improve the effectiveness of the use of financial leverage for businesses in Vietnam
LITERATURE REVIEW
Definitions
Capital structure is a highly researched topic in finance, with its origins rooted in the influential Modigliani and Miller thesis published in 1958 Since then, the field has evolved to include various theories such as the static and dynamic trade-off theories, classification order theory, and market timing theory, each offering different insights into how firms determine their optimal capital structure.
The optimal capital structure is the one that maximizes the overall value of the business while minimizing the weighted average cost of capital (WACC) Achieving this balance ensures that the company is efficiently financed, enhancing shareholder value and reducing funding costs By optimizing the capital mix, businesses can improve financial performance and support sustainable growth.
Funding decisions directly impact the Weighted Average Cost of Capital (WACC), which is calculated by averaging the cost of equity and the cost of debt When a company adjusts its capital structure, it causes changes in WACC, potentially influencing the company's valuation However, it is essential to consider whether these changes in capital structure also affect the overall value of the business or the interests of shareholders.
The value of the firm approaches from the level of capital fees calculated by the current cash flow to firm price with a discount of WACC (Aswath Damodaran, n.d)
Optimizing the capital structure by reducing WACC can significantly increase a company's value To achieve this, identifying the optimal capital mix involves minimizing WACC, which is derived from the cost of equity and the cost of debt Since debt typically has a lower cost compared to equity, understanding which capital sources are less expensive is crucial According to Patrick Lynch (2009), the cost of debt is lower because debt is perceived as less risky than equity.
Interest payments are usually stable, mandatory (Patrick Lynch 2009) and are prioritized for prepayment of shares (Enterprise Law 2014)
When dissolving a business, the owner is prioritized for payments only after settling all debts and dissolution costs, ensuring creditors are paid first (Enterprise Law 2014).
From an industrial perspective, using debt financing is generally more cost-effective than equity because interest expenses are tax-deductible, providing a tax shield that reduces the company's taxable income Unlike equity, which involves paying dividends after profits are taxed, debt payments are primarily in the form of interest, offering a significant financial advantage This tax efficiency makes debt a more attractive financing option for companies seeking to optimize their capital structure (Patrick Lynch, 2009).
To reduce the average cost of capital (WACC), increasing the company's level of debt can be effective; however, this also raises interest expenses, leading to higher fixed costs that must be paid before dividend distributions As interest payments grow, the financial stability of shareholders diminishes, especially during periods of poor company performance, since debt obligations must still be met in full, impacting overall shareholding policy This heightened financial risk associated with increased debt levels can lead to stock market volatility and reduce dividend stability Consequently, higher financial risk demands greater profitability from shares, which subsequently increases the company's WACC.
The key issue is determining whether a decrease in WACC, driven by higher levels of debt, has a more significant impact despite increased financial risks To address this, the analysis will incorporate relevant theories such as the Modigliani and Miller (M&M) theorem in both tax-free and taxed environments, alongside the hierarchical order theory These frameworks provide valuable insights into how capital structure adjustments influence overall firm value and risk considerations.
First, the Modigliani and Miller (M&M) theory in a tax-free environment (1958):
Under perfect capital market conditions, M&M asserts that a company's financing decisions do not impact its overall value While this may seem contradictory to the actions of CFOs who optimize capital structures to enhance corporate value, it holds true within the specific assumptions underlying the theory.
There is no CIT and personal income tax;
There are no transaction costs;
There are enough buyers and sellers on the meat, so there are no individual investors who have a great influence on the price of the stock;
Relevant information is available to all investors and does not have to lose money;
According to M&M theory, all investors can buy and lend at the same interest rate, eliminating any advantage for businesses over individual investors in using leverage This assumption implies that whether a business chooses to use leverage or not does not impact its value, as investors can replicate the same financial strategies independently.
All investors are rational and share a similar period of business profitability, based on the assumption that they have equal access to information Assuming that all investors possess the same reasonable thinking and calculations leads to consistent assessments of a company's performance This perspective results in the most homogeneous period of profitability.
Homogeneous risk assumptions: businesses operating under equal conditions will have the same level of business risk
Second, modigliani and miller theory (M&M) in tax environments:
In 1963, Modigliani and Miller published a pivotal study in the *Journal of Economics* examining the impact of tax factors on capital structure They highlighted that the use of debt can increase a company's value because interest expenses are tax-deductible, reducing taxable income This tax shield allows a portion of the company's income to be transferred to investors instead of being paid in taxes, thereby enhancing overall firm value.
The value of the business = Enterprise value if fully funded by equity + Show price (tax shield)
An increase in the tax shield price proportionally promotes higher borrowing within a company's capital structure, encouraging businesses to leverage more debt Consequently, a capital structure consisting of 100% debt is considered optimal, as it maximizes the company's financial efficiency by fully benefiting from the tax shield advantages.
In real-world conditions, increasing debt levels is unlikely to produce the expected benefits due to the costs associated with the tax advantages of debt These costs often offset the potential reduction in business prices, challenging the effectiveness of debt as a means to optimize capital structures This reflects the ongoing debate about the trade-offs involved in leveraging debt and the limitations of capital structure theories in practical applications.
Theories about the capital structure of the business
The foundational theory of capital structure was established by Miller and Modigliani in 1958, who argued that in a perfect market, a company's capital structure does not influence its overall value, implying there is no optimal capital mix However, their model's assumptions—such as no transaction costs, absence of taxes, perfect information, and risk-free loan interest rates—do not reflect real-world business environments Consequently, researchers have recognized that a company's value and performance are indeed impacted by its capital structure in practical settings.
Building upon Miller and Modigliani's foundational research, various capital structure theories have been developed to explain how businesses determine their optimal mix of debt and equity These include the static trade-off theory, which emphasizes balancing debt and equity to maximize firm value; the dynamic trade-off theory, highlighting adjustments over time based on changing conditions; the classification order theory, focusing on prioritizing different sources of financing; and the market timing theory, suggesting firms might time their capital raises based on market conditions to optimize valuation.
The static trade-off theory suggests that a company's optimal capital structure is achieved by balancing the benefits of tax shields against the costs associated with debt This includes evaluating the marginal tax advantages of debt financing while accounting for potential financial exhaustion and associated costs By carefully weighing these factors, firms aim to determine the most advantageous level of debt to maximize value without incurring excessive financial risks.
Miller & Modigliani (1963) expanded traditional capital structure theories by incorporating corporate income tax as a key factor, highlighting that debt can increase business value through tax shields They contended that, from a business perspective, maximizing debt levels can enhance overall value However, they also acknowledged that excessive debt introduces significant costs, including financial exhaustion and agency expenses, as discussed by Jensen & Meckling (1976) and Myers (1977).
The static trade-off theory explains that a company's optimal capital structure is achieved by balancing the tax benefits of debt against the associated costs, such as financial exhaustion and bankruptcy risks Increasing debt enhances tax shields and can improve business performance by reducing conflicts between shareholders and managers, as well as supporting optimal investment strategies However, higher leverage also raises the risk of financial distress and future repayment obligations, which can threaten business stability The optimal debt-to-equity ratio is where the present value of the tax benefits equals the present value of the costs incurred from debt Beyond this equilibrium point, additional debt costs outweigh the benefits, decreasing the company's value Therefore, corporate financial managers must carefully weigh the advantages and disadvantages of debt to determine an optimal capital structure that maximizes shareholder value.
The dynamic trade-off theory, introduced by Fischer et al (1989), emphasizes that the optimal capital structure is influenced by equity capitalization costs, leading to fluctuations in a company's debt ratio This theory suggests that businesses do not always operate at the optimal capital structure but instead adjust their debt levels over time Funding decisions are based on the expected marginal funding needs in future periods, allowing the company's capital structure to gradually move toward its optimal level.
According to this theory, more profitable businesses benefit more from the tax shield as they increase borrowing, suggesting that companies should leverage higher to maximize tax advantages While higher leverage also raises the risk of financial distress and bankruptcy costs, numerous studies reveal that these costs are relatively small compared to the benefits gained from the tax shield This explains the positive relationship between a company's capital structure and its operational efficiency, emphasizing the strategic use of leverage to enhance overall business performance.
502 Bad GatewayUnable to reach the origin service The service may be down or it may not be responding to traffic from cloudflared
502 Bad GatewayUnable to reach the origin service The service may be down or it may not be responding to traffic from cloudflared
502 Bad GatewayUnable to reach the origin service The service may be down or it may not be responding to traffic from cloudflared
The market timing hypothesis examines how businesses decide when to issue or repurchase shares under imperfect market conditions, providing valuable guidance for managers to identify optimal timing strategies Unlike traditional capital structure theories, this approach considers market imperfections and their influence on decision-making It also explores the relationship between market timing and a company's capital structure, analyzing how long the effects of these decisions typically last This understanding helps businesses optimize their capital structure in response to market dynamics for sustained financial health.
In market scenarios lacking an optimal capital structure, managers often operate under the belief that they can influence market conditions through their financial decisions These strategic financial choices serve as a means to guide the evolution towards an optimal capital structure over time Consequently, a company's current capital structure reflects a cumulative outcome of past efforts aimed at market timing and financial regulation, highlighting the dynamic interplay between managerial decisions and market adaptation.
Indicators for measuring capital structure
According to Circular No 200/2014/TT-BTC, liabilities are accounts that reflect the debts incurred during business operations, which enterprises are required to settle with creditors These liabilities encompass various obligations such as loan debts, payables to suppliers, the State, employees, and other payable amounts Enterprise liabilities are categorized into short-term and long-term debts, providing a clear overview of the company’s financial obligations.
Debts: Based on the nature of the debt maturity, the debts of the business are divided into short-term debt and long-term debt
Short-term debts are financial obligations due within one year, including bank loans, supplier liabilities, accounts payable to employees, and state budget payables To manage risks associated with short-term debt repayments, businesses often utilize short-term financing to meet immediate capital needs or invest in short-term assets, ensuring liquidity and financial stability.
Long-term debt refers to financial obligations with a repayment period exceeding one year, serving as a crucial source of capital for businesses It is commonly used to finance investments in both short-term assets and long-term fixed assets or projects In Vietnam, businesses primarily access long-term debt through two main channels, supporting their growth and investment strategies.
Medium- and long-term loans from credit institutions and the issuance of corporate bonds are common financing methods for enterprises According to Vietnam’s Accounting Regime, borrowing expenses, including interest costs, are classified as financial costs and are deductible when calculating corporate income tax, creating a tax shield that reduces tax payable Typically, the cost of debt (interest rates on loans or bonds) is lower than the expected returns demanded by investors, allowing businesses to lower capital costs and enhance investment efficiency However, excessive reliance on debt increases financial risks, potentially leading to financial distress, insolvency, and bankruptcy Overusing debt can also cause a conservative business mindset, causing firms to miss investment opportunities that could boost productivity and overall company value.
According to Circular 200/2014/TT-BTC, equity represents the owners’ capital in an enterprise that does not require repayment, originating from investments by owners or investors and accumulated business results In the context of joint stock enterprises listed on the Ho Chi Minh City and Hanoi stock exchanges, equity comprises contributed charter capital from shareholders upon establishment As the enterprise operates, additional capital sources under equity include retained earnings and other reserves, which support ongoing business growth without constituting debt.
Retained profits, corporate funds such as development investment funds and equity funds, constitute the primary sources of financing within a business, reflecting core equity capital As businesses grow, capital expansion through issuing new shares is often more stable than relying on debt, with stocks representing permanent ownership and serving as a key security Shareholders benefit from dividends linked to business performance and may see higher earnings when the company is more profitable, subject to decisions made at shareholders' meetings The market value of shares in reputable companies fluctuates based on supply and demand, investor sentiment, policy and political risks, and other market factors, with shares being tradable on official stock exchanges or through informal channels.
Under Vietnamese accounting regulations, dividends are paid to shareholders from after-tax profits, and increasing equity capital enhances a company's financial autonomy, although it does not exempt the enterprise from income tax on dividends From an investor's perspective, investing in stocks is riskier than providing debt, as shareholders’ returns depend on the company's operational performance Consequently, shareholders expect higher returns compared to interest on loans or bonds, making the cost of equity capital typically higher than borrowing costs.
When determining the optimal capital structure, businesses must carefully evaluate both quantitative and qualitative factors to effectively mobilize different types of capital The primary challenge is to select the most suitable form of capital that minimizes costs and reduces risks, ultimately maximizing business value Considering the characteristics and mobilization methods of each capital type is essential to making informed decisions Strategic capital choices enable companies to balance risk and return, ensuring sustainable growth and competitive advantage.
Factors affecting the capital structure of the business
Factors influencing a company's capital structure are of significant interest to scholars, though empirical studies often present conflicting views on the impact and extent of each factor These discrepancies can be attributed to differences in research methodologies and theoretical perspectives Key determinants include profitability, company size, tangible assets, growth rate, corporate income tax rate, non-debt tax shields, and years of operation, all of which play a crucial role in shaping optimal capital structure decisions.
There is a strong positive correlation between profitability and financial leverage, as higher profitability reduces the risk of bankruptcy, allowing businesses to prioritize borrowing According to the trade-off theory, companies are encouraged to increase debt leverage due to the tax shield benefits Additionally, Michael C Jensen (1986) highlights that effective businesses can benefit from increasing their loan share by investing in more affordable and secure financing options instead of costly infrastructure or negative NPV projects.
According to the theory of classification order, enterprises prioritize increasing internal capital, with debt and issuing shares regarded as last-resort financing options Profitable businesses tend to access and utilize more internal funds compared to inefficient ones, maximizing their internal capital for growth Sheridan Titman and Roberto Wessels (1988) support this view, arguing that, all else being equal, companies with high profit margins typically maintain lower levels of financial leverage.
The ratio of total tangible fixed assets to total assets is a fundamental measure of a company's capital structure, reflecting its relationship with asset composition Theories indicate that higher asset tangibility positively influences access to external financing, as tangible assets serve as collateral for loans, boosting investor confidence Companies with more tangible assets can secure larger loans and benefit from lower interest rates due to the reduced lending risk Numerous studies, including those by Chen (2003), Huang and Song (2006), A Shah and S Khan (2007), Merve Gizem (2018), Khaki and Akin (2020), and Le Thi Minh, have demonstrated a consensus on the positive relationship between capital structure and fixed assets.
Research by Nguyen (2016) and Cuong Thanh Nguyen (2019) highlights significant insights into business dynamics M Onofrei and colleagues (2015), along with Tran Viet Dung and Bui Dan Thanh (2019), identify an inverse relationship between the strength of certain business factors and overall business ratios These findings suggest that as specific business metrics weaken, the business ratios tend to increase, indicating a complex interplay affecting organizational performance Understanding this inverse correlation is crucial for strategic decision-making and improving business sustainability.
Corporate income tax is calculated based on the pre-tax profit of the enterprise and reflects the actual tax rate the company must pay According to the M&M theory (1963), there is a variable relationship between taxes and debt, as firms with higher tax rates often leverage debt to benefit from tax shields through interest expense deductions Le Dat Chi (2013) supports this view with findings that align with the positive relationship between tax rates and leverage However, Dang Thi Quynh Anh and Quach Thi Hai Yen (2014) present evidence of an inverse relationship, indicating that higher tax rates may reduce a company's debt ratio.
2.4.4 The size of the business
Large-scale businesses, measured by the logarithm of total corporate assets, tend to attract more investment capital and enhance owner trust due to reduced information asymmetry This increased trust facilitates easier access to bank loans and enables firms to maximize tax benefits through the tax shield Numerous studies, including those by Huang and Song (2006), Merve Gizem (2018), Khaki and Akin (2020), and others, confirm a strong positive relationship between business size and financial performance, highlighting the importance of asset scale in building credibility and financial advantage.
According to the 2020 theory of order, the business class typically prioritizes utilizing existing internal capital, such as retained profits, for funding needs When additional financing is required, these businesses prefer to obtain external loans to support their growth and operations Research by Chen further supports this approach, highlighting the strategic financial management practices of the business class in capital allocation.
(2003) and by M.Onofrei et al (2015) also shows that the larger the size of the business's assets, the smaller the ratio
Growth is primarily measured by the speed of asset or revenue increase over time, which boosts investor confidence and facilitates greater access to external loans during the business’s growth phase According to the theory of order, when a company's growth rate is high but its retained profits are insufficient to secure external financing, the enterprise will increasingly rely on external capital Empirical studies by Chen (2003), Khaki and Akin (2020), Cuong Thanh Nguyen et al (2019), and Tran Viet Dung and Bui Dan Thanh (2021) support this relationship Conversely, when a business is highly profitable with consistent growth, it tends to prioritize internal funding through retained earnings, resulting in a growth rate that is proportionally aligned with the business's expansion, as evidenced by research from Huang and Song (2006), A Shah and S Khan (2007), and M Onofrei et al (2015).
The liquidity of a business is determined by the ratio of short-term assets to total short-term debt, indicating its ability to meet short-term obligations High liquidity ratios boost investor confidence and enable businesses to access additional borrowing when needed According to the trade-off theory, maintaining high liquidity allows businesses to make timely payments to employees and suppliers, supporting continuous operations When companies possess abundant short-term assets, they can finance their activities internally, reducing reliance on external debt The theory of order suggests that to sustain high revenue levels, companies tend to utilize their internal capital more efficiently Additionally, the ratio of short-term assets to short-term debt correlates with leverage, a relationship supported by empirical studies in Vietnam and other countries, including research by M Onofrei et al (2015), Merve Gizem (2018), Khaki and Akin (2020), and Le Thi Minh Nguyen.
The multiplier is measured by the ratio of the capital of goods sold to the business’s net revenue According to Frank and Goyal (2008), businesses that produce unique products or operate in specialized industries tend to have less debt, as they face difficulties liquidating assets in the event of bankruptcy This indicates that the lack of liquidity in specialized assets results in an inverse relationship between the asset's specific factor and the leverage ratio Additionally, Cuong Thanh Nguyen and colleagues’ (2019) research confirms this relationship, highlighting that the unique nature of a company's products and services influences its financial leverage.
Asset turnover (AT), measured by net revenue to average assets, was included in the model based on the research of Ahmad et al (2012), Hoque et al (2014), Karaca et al (2012), and Le Thi Phuong Vy (2013) These studies highlight the significance of asset efficiency in evaluating overall firm performance and its impact on financial outcomes Incorporating asset turnover into the model enhances its robustness by reflecting the company's ability to generate revenue from its assets This metric is a vital indicator for investors and analysts seeking insights into operational efficiency and asset management effectiveness.
To optimize financial strategies amidst changing borrowing conditions, it is essential to increase financial leverage when interest rates are expected to rise and borrowing becomes more difficult Conversely, if interest rates are anticipated to fall, implementing a plan to reduce or postpone current borrowing is advisable These adjustments help businesses effectively navigate the financial landscape and manage risk accordingly (Lâm Đức Anh, 2021).
2.4.10 Fluctuations in the season, business cycle
Industries experiencing significant seasonal fluctuations in business rely heavily on short-term debt to manage financial needs during specific periods Choosing an appropriate capital structure is crucial at each stage of a business’s lifecycle: in the early stages, high failure rates make equity the primary source of capital since borrowing options are limited; during the development phase, substantial capital is required for expansion and scaling operations; and in periods of peak growth, businesses must prepare for seasonal and cyclical sales fluctuations To mitigate risks during potential recessions, it is essential to develop a financial structure that facilitates flexible capital management, ensuring resilience against economic downturns (Lâm Đức Anh, 2021)
Access to capital is greatly influenced by regulations governing enterprise funding Implementing a favorable mechanism with preferential fees can significantly enhance access to loans, particularly for small and newly established businesses Such supportive regulations create conducive conditions for business growth and financial stability.
EXISTING EMPIRICAL STUDIES
Foreign studies examining the factors influencing the financial performance of companies have garnered significant scholarly interest Recent research in this area predominantly employs quantitative methods and models to analyze key determinants affecting firm performance These studies highlight the importance of data-driven approaches in understanding the various internal and external factors that impact corporate financial results, ensuring robust and reliable insights for researchers and practitioners alike.
Jean J Chen's 2003 study of 77 large companies in Shanghai, China, revealed that profitability and business size have an opposite effect on a company's capital structure, while growth rate and tangible fixed assets influence the debt ratio positively Additionally, the research highlighted significant differences between Chinese companies and developed countries in debt preferences, with Chinese firms favoring short-term debt over long-term financing options.
China and Song's (2006) study of 1,086 Chinese enterprises from 1994 to 2003 identified key factors influencing capital structure, including company size and tangible fixed assets, which showed a strong correlation The research also highlighted that profitability, tax shields, growth opportunities, and industry sector differences significantly affect capital structure decisions Notably, whether an enterprise is state-owned or corporate property does not impact its capital structure Additionally, Chinese firms tend to use less debt compared to companies in other economies The study employed three estimation models—OLS, FEM, and REM—but did not perform necessary tests to validate these models' assumptions.
M.Onofrei et al (2015) studied the capital structure factors of 385 micro and small businesses based in Romania between 2008 and 2010 The paper uses a method of estimating fem fixed impact patterns with causality: profitability, tangibleness, liquidity, scale and growth opportunities The result is that financial leverage is critical to tangibleness, profitability and liquidity, while corporate tissue factors and growth opportunities also have a negative impact on leverage but at a greater level
Merve Gizem (2018) investigated the capital structure factors of non-financial enterprises, applying financial theory to explain their determinants The study analyzed data from 111 Turkish non-financial firms between 2009 and 2016 using regression methods and FEM model selection Results indicate that profitability, debt-free tax shields, and liquidity negatively influence leverage, while scale and tangibility have a weak positive effect Risk factors do not significantly impact leverage The findings suggest that decisions on the capital structure of non-financial enterprises in the United States align more with the pecking order theory than with the trade-off theory.
Jan and Mateus (2008) examined the dynamic relationship between financial leverage and the performance of U.S banking companies, highlighting how high financial leverage can positively influence business performance Their study utilized performance metrics rooted in the theoretical framework of intermediary costs, concluding that increased leverage is associated with improved bank performance.
In 2007, a study examined the relationship within New Zealand's small and medium-sized enterprises, employing the distance function to assess business effectiveness The findings aligned with the intermediate cost theory, supporting the idea that operational efficiency is influenced by certain cost structures in these enterprises.
The study by Dang Thi Quynh Anh and Quach Thi Hai Yen (2014) studied the impact of a number of factors on the capital structure of enterprises listed on the
Ho Chi Minh Stock Exchange (HOSE), with the use of fem (fixed effect model) and data collected from 180 non-financial companies listed on HOSE between 2010 and
In 2013, research showed that business size and profitability are positively correlated, indicating that larger firms tend to be more profitable Conversely, the study found that taxes are negatively correlated with a company's capital structure, suggesting that higher taxes may influence firms to adjust their leverage These findings highlight the interconnected relationship between enterprise size, profitability, and tax considerations in financial decision-making.
The author uses the research sample is an enterprise by industry such as: Phan Thanh Hiep (2016) with the sample set is industrial production enterprise
Le Dat Chi’s 2013 research examined the key factors influencing the capital structure decisions of Vietnamese listed companies from 2007 to 2010, integrating traditional capital structure theories with behavioral finance The study identified six significant factors: taxes, financial leverage, and management behavior all have a positive impact, while inflation, the market-to-book ratio, and ROA exert opposite effects Findings indicate that during this period, companies’ capital structure planning was not significantly aligned with trade-off theory but was strongly supported by classification order theory.
Conclusion of Chapter 2
Chapter 2 provides an overview of capital structure and factors affecting the capital structure of banks Next, the fundamental theories of capital structure in general are synthesized such as: capital structure theory from a traditional point of view, M&M theory, trade-off theory, classification order theory and representative cost theory Experimental studies at home and abroad on the factors affecting the capital structure of the bank are summarized, analyzed, giving the direction of research methods for the thesis in the next chapters.
METHODOLOGY AND DATA
Experimental research model of factors affecting the capital structure of
of enterprises on the Vietnamese stock market
The thesis uses the following quantitative methods to study and analyze factors affecting the capital structure of enterprises in Vietnam:
This study employs synthesis, statistical analysis, and comparative methods to collect and examine data on the factors influencing capital structure The research focuses on assessing the current capital structure of Vietnamese enterprises between 2010 and 2021, providing valuable insights into trends and determinants affecting corporate financing strategies in Vietnam during this period.
This study conducts a comprehensive panel data regression analysis to examine the factors influencing the capital structure of Vietnamese enterprises The analysis utilizes multiple models, including the Pooled OLS, Fixed Effects Model (FEM), and Random Effects Model (REM), to ensure robust results By applying these models, the research identifies key determinants that significantly impact enterprise financing decisions in Vietnam, providing valuable insights for policymakers and business leaders aiming to optimize capital structure strategies.
This article explores the factors influencing a business's capital structure through both qualitative and quantitative analysis Qualitative analysis draws on global financial theories proposed by leading scholars, providing theoretical insights into capital structure determinants Meanwhile, quantitative analysis employs statistical methods and econometric models to empirically validate these factors Combining these approaches offers a comprehensive understanding of the key influences shaping a company's capital decisions, ensuring an SEO-friendly overview of capital structure analysis.
Y i is the dependent variable or the variable explained
X n is an independent variable or the variable that explains
α i as the blocking factor in the model
The dependent variable in this analysis is the debt-to-total capital ratio, commonly referred to as financial leverage, which measures a company's level of debt relative to its overall capital This ratio includes both short-term and long-term debt, providing a comprehensive view of the company's leverage To fully understand the financial structure, all three related dependent variables must be examined, highlighting the importance of analyzing different aspects of a company's debt profile for accurate financial assessment.
Table 3 1 Dependent variables in the regression model
Book value of total short-term debt and long-term debt divided
Book value of total short-term debt and long-term debt divided
Book value of short-term debt divided by total capital
Table 3 2 Explanatory variables in the regression model
Independent variables Ampersand Formula calculation Correlation of wonders Return on total assets ROA
Weighting of fixed assets TANG
Characteristics of the company's assets
Enterprise size SIZE Log (revenue) +
Based on the general model, the author extends to a simple multivariate regression model defined as follows
Y it = α + β 1 ROA it + β 2 ROE it + β 3 SIZE it + β 4 TANG it + β 5 GROW it + β 6 TAX it + β 7 UNIQ it + β 8 LIQ it + β 9 AT it + u it
H1 hypothesis: Profitability on assets of the enterprise has the opposite effect with the ratio of corporate debt
Profitability, measured by return on total assets (ROA), indicates a company's financial performance While some theories suggest that profitability and debt ratio are directly related, many studies, including those by Dang Quynh Anh & Quach Thi Hai Yen (2014), show that ROA has an inverse relationship with a company's debt ratio Experimental research on this topic produces varied results, but the majority of evidence supports the notion that higher debt levels tend to negatively impact profitability.
H2 hypothesis: The profitability on equity of the enterprise has the opposite effect with the ratio of corporate debt:
The inclusion of these variables in the supported research model of experimental research authors such as Karaca et al (2012); Le Thi Phuong Vy (2013); Asiri (2014)
H3 hypothesis: The size of the enterprise has a favorable relationship with the debt ratio of the enterprise
Business size (symbol: SIZE) is measured by the total asset value, which is often converted to natural logarithms to manage large value differences There is generally an expected relationship between business size and debt ratio, which can be either negative (classification order theory) or positive (trade-off theory) Recent studies, including Wahab and Ramli (2014) and Dang Thi Quynh Anh (2014), have found a positive correlation between business size and debt ratio, indicating that larger businesses tend to have higher levels of debt.
H4 hypothesis: The fixed assets of the business have a favorable relationship with the debt ratio
The fixed asset ratio (TANG) reflects the company's asset structure, representing the proportion of fixed assets to total assets According to the theory of representative costs and trade-off costs, holding more fixed assets can enhance a company's borrowing capacity by providing secured collateral Secured loans are perceived as safer, which typically results in lower borrowing costs, incentivizing businesses to borrow more This positive relationship between fixed assets and borrowing costs is supported by empirical research.
H5 hypothesis: Growth is in the same direction as the debt ratio of the business
The business growth rate (symbol: GROW) is primarily reflected through the annual revenue increase, calculated by dividing the difference in net revenue between consecutive years by the previous year's net revenue Domestic research indicates a positive correlation between enterprise growth rates and debt ratios, as evidenced by studies conducted by Dang Thi Quynh Anh and Quach Thi Hai Yen (2014).
H6 hypothesis: The relationship between the corporate income tax rate and the debt ratio is an inverse relationship with each other
The corporate income tax rate (TAX) is calculated as the ratio of corporate income tax payable to pre-tax profits While the M&M Theory suggests a positive relationship between the corporate income tax rate and the debt ratio, studies by Jan Bartholdy and Cesario Mateus (2008) support this view However, recent research in Vietnam, such as the study by Dang Thi Quynh Anh and Quach Hai Yen (2014), indicates an inverse relationship between income tax rates and the extent of business borrowing Consequently, this study expects an inverse relationship between the corporate income tax rate and the debt ratio in Vietnamese companies.
H7 hypothesis: Whether the specific characteristics of the product has the opposite effect on the ratio corporate debt
Liquidating specialized assets can be challenging as bankruptcy costs increase, depending on the liquidity of the company's assets When an enterprise possesses highly liquid business assets, its specialized assets can be more easily liquidated This aligns with Frank and Goyal's 2008 research, which highlights that the unique characteristics of corporate assets are inversely correlated with their liquidity ratios.
H8 hypothesis: The liquidity of the business has the opposite effect on the ratio of corporate debt
Businesses with abundant short-term assets can effectively support their production and operational activities To sustain high revenue levels, companies often prioritize utilizing their internal capital Research both internationally and within Vietnam, including studies by M Onofrei (2015), Merve Gizem (2018), Khaki and Akin (2020), and Le Thi Minh Nguyen (2016), consistently highlight this perspective Consequently, liquidity plays a crucial role in the financial health and performance of billions of businesses worldwide.
Research methods
The study analyzes 380 observations from 39 enterprises listed on Vietnam's two major stock exchanges, the Ho Chi Minh City Stock Exchange (HOSE) and the Hanoi Stock Exchange (HNX), over a 10-year period from 2012 to 2021.
This study processes collected data through Excel to ensure accuracy and removes heteronormative observations and unrelated businesses, resulting in a balanced dataset of 39 occupations over a 10-year period (2012-2021) with 380 observations The researcher employs Stata 13.0 for regression analysis, applying various methods such as Pooled OLS, Fixed Effects (FEM), Random Effects (REM), and Generalized Least Squares (GLS) to achieve reliable results After conducting these regressions, the study compares the methods to identify the most suitable approach aligned with the research objectives Finally, the author interprets the results by integrating relevant economic theories to explain the correlation between explanatory and control variables, including magnetic sub-variables, thereby highlighting their economic significance.
Regression analysis measures how independent variables influence sub-variables, indicating both the direction and strength of their relationship In panel data regression, variables are modeled to assess these effects over time and across entities, providing valuable insights into variable interactions This approach helps identify significant predictors and quantify their impacts within the dataset, enhancing understanding of variable dynamics in complex analyses.
N is the ith cross unit t=1, 2, ,
Yit: the value of Y for object i at point t
The variable Xit represents the value of X for object i at point t, serving as a key indicator in the analysis The error term uit captures deviations or inaccuracies associated with object i at time t, ensuring the robustness of the model Additionally, αit denotes the shock or unexpected influence on the estimate for object i at point t, reflecting dynamic changes over time The parameter βit measures the degree of regulation system influence, quantifying how much the regulatory framework impacts the outcome Y and the personal effect of object i at the time of death These elements collectively provide a comprehensive understanding of the factors influencing Y in the studied context.
According to Baltagi (2008), various methods such as pooled OLS, Fixed Effects Model (FEM), Random Effects Model (REM), GLS, and FEM are commonly used to estimate regression models The choice of the most appropriate method depends on evaluating each model's suitability through diagnostic tests Ultimately, the thesis selects the most suitable regression technique based on these assessments to ensure accurate and reliable results.
This study focuses solely on analyzing how various factors influence capital structure, without examining the interactions between these variables This approach simplifies the model construction process by concentrating on the direction of each factor's impact, streamlining the analysis of determinants affecting capital structure.
The study employs the Ordinary Least Squares (OLS) method to estimate regression coefficients across three models, ensuring straightforward computation with STATA 13 software Model selection between Fixed Effect Model (FEM) and Random Effect Model (REM) is determined using the Hausman test, which helps identify the most appropriate approach However, this method relies on several assumptions; violations of these hypotheses can compromise the reliability of the results.
3.3.2.1 The smallest method of regression of the commonest squared (PooledOLS)
According to Gujarati et al (2009), the Pooled OLS method combines both cross-sectional and time series data, providing a robust approach for panel data analysis It assumes the absence of omitted variables and relies on constant parameters across individuals and time periods The model integrates the system of coefficients (β), where 'i' represents the cross-sectional units, and 't' denotes the time periods This method is widely used for its simplicity and effectiveness in estimating relationships in panel data datasets.
Regression equation as follows:Y it = αit + β 1 X 1it + + β k X ki t uit
According to Gujarati et al (2009), Pooled OLS estimates can distort the true relationships between variables due to the neglect of the data structure When a model includes many explanatory variables, this approach may lead to issues such as multicollinearity or multicollinearity, which can compromise the accuracy of the estimates.
Therefore, Pooled OLS is very easy to violate the statistical significance leading to inefficiencies
3.3.2.2 fixed impact regression method FEM
According to Gujarati et al (2009), regression of impact does not eliminate effects in time series and across units, as the model assumes varying degrees between units while maintaining a constant system This suggests that the regression approach may have limitations in fully accounting for effects in dynamic data Understanding these constraints is essential for accurate analysis in time series and cross-unit studies, especially when considering impact regression methods.
Regression equation as follows:Y it = α it + β 1 X 1it + + β k X ki t uit
The error term in the model, μ_it, consists of two components: the α_i component, which captures unobserved factors that are constant over time and do not vary between objects, and the u_it component, representing unobserved factors that vary over time but remain consistent across objects Understanding these components is essential for accurately analyzing the sources of variability in the data and improving model precision.
This article discusses how individual characteristics influence the variability observed in different coin tosses, with each person's inherent traits contributing to the outcome consistency over time It highlights that individual toss degrees remain stable across periods, while identifying limitations of the fixed impact model, such as decreased autonomy when too many false variables are included, and the risk of multicollinearity and incorrect parameter estimates when numerous explanatory variables are considered Additionally, since i represents cross-sectional observations and t refers to time-series data, the model must account for various possibilities like first-order autocorrelation, changing variances, and other complex dynamics over time.
3.3.2.3 Random impact regression method REM
Y it = α it + β 1 X 1it + + β k X ki t uit
Instead of being constant, we assume that this is a random variable with an average value of α (without i) Since the value for a single business can be expressed as follows: αi= α + εi, i = 1.2, ,N
In this context, εi represents the number of false lives with an average value of zero and a specific variance Each individual sample is modeled to have a common normal distribution characterized by a certain degree of variation (α) The differences among individuals are captured by the error term εi, which quantifies the individual-specific variability in the data This framework highlights the importance of understanding both the average behavior and the variability within individual measurements for accurate statistical analysis.
Y it = α + β 1 X 1it + + β k X kit + ε i + u it = α + β 1 X 1it + + β k X kit + W it
Wit's wrong number comprises two key components: the εit component, representing the cross-digit or individual error, and the uit component, which accounts for the cross-digit error within the combined time series The term "shape of the error components" refers to the random impact model (REM), highlighting its basis in the combination of multiple error components that influence the overall error structure in the data This model effectively captures the complexity of errors arising from various sources within time series analysis.
3.3.2.4 Generalized Microscopic Regression Method (GLS)/ FEM solid estimate
The GLS (Generalized Least Squares) or FEM (Finite Element Method) solid estimation technique is an extension of the Ordinary Least Squares (OLS) method, adapted to meet the standard minimum squared hypothesis assumptions These estimates, known as GLS estimates, are BLUE (Best Linear Unbiased Estimators), ensuring optimal accuracy This methodology effectively addresses issues related to heteroskedasticity and self-correlation within the research model, leading to more reliable and precise econometric results.
3.3.2.5 Check the suitability of the model
Collect data
This study utilizes secondary metrics sourced from Repupanel data management software, ORBIS, and Data Stream, offering significant time savings by avoiding primary data collection methods like surveys and interviews While leveraging existing data from suppliers and online sources enhances efficiency and reduces research costs, it also introduces challenges related to data accuracy and relevance, as secondary data may have been collected for different purposes Additionally, using secondary data can entail costs such as registration fees paid to suppliers, which may represent a considerable expense.
In relation to the observation sample, the criteria for selecting the analytical company in this study include:
(1) Companies listed on HNX and HOSE as of December 2021
Companies in the financial sector, including banking, insurance, and investment funds, are excluded from the observation form due to the complexity of their balance sheets and varying reporting standards This exclusion ensures a more accurate and consistent analysis of business results across different industries.
(3) Companies meet business standards and top 500 in Vietnam
Based on those criteria, there are 39 satisfied businesses, presented in Appendix 1 The data was collected between 2010 and 2021 with a total of 456 observations.
Research sequence
The author uses STATA 13 software as a tool to support data processing and will be presented as a panel data, in which the data series runs from 10 years from
2012 to 2021 with 380 observations The author conducts the study in specific steps as follows
Step 1: Statistics of research data
Statistical analysis of variables enables researchers to clearly and comprehensively describe their research data by utilizing key metrics such as average values, standard deviations, and the maximum and minimum values within the research model.
Multicollinearity occurs when independent variables in a study are highly correlated and interdependent, leading to the multilinear phenomenon that undermines statistical validity This issue results in unreliable estimates and makes it difficult to justify rejecting the null hypothesis To address multicollinearity, researchers can utilize the Variance Inflation Factor (VIF) and correlation matrices; specifically, multicollinearity concerns diminish when the correlation coefficient between independent variables is below 0.8 and the VIF value is less than 4, thereby ensuring the model remains reliable without multicollinearity issues.
This article analyzes the impact of independent variables on business performance using different regression models, including pooled OLS (POLS), fixed effects (FEM), and random effects (REM) The FEM model assumes that each enterprise has unique, time-invariant characteristics that may influence the independent variables, allowing for precise separation of these effects to accurately estimate their true impact on dependent variables In contrast, the REM model views differences between businesses as random and uncorrelated with the independent variables, providing a different approach to understanding business performance variations.
Step 4: Choose the right regression model
After analyzing the regression results, the author conducts model selection tests to identify the most appropriate econometric approach The F-test is employed to compare pooled OLS and Fixed Effects Models (FEM), while Hausman testing helps determine the suitability between FEM and Random Effects Models (REM) Additionally, the Breusch-Pagan Lagrange Multiplier (LM) test is used to decide between Pooled OLS and REM, ensuring robust model selection based on statistical evidence.
Step 5: Check out the defects in the model
The correlation between the components of an observed sequence over time or space is a common phenomenon that can significantly impact regression analysis When such similarities are present, traditional pooled OLS methods with the smallest squared estimation become ineffective The presence of uncorrelated random errors (Ui) in the overall regression function is a fundamental assumption of classical linear regression models Detecting this phenomenon can be achieved through various techniques, including graphical methods and statistical tests like the Berusch-Godfrey test, which help identify autocorrelation and ensure model accuracy.
Variance changes, or heteroscedasticity, occur when the variance of estimated errors is unequal, often caused by outliers or measurements on different scales Detecting this phenomenon can be achieved through graphical methods or statistical tests like Breusch-Pagan and White inspections To address heteroscedasticity in the Fixed Effects Model (FEM), employing robust standard errors is effective, while in Random Effects Models (REM), solutions include using Feasible Generalized Least Squares (FGLS) or random effects estimators to correct for variance inconsistencies.
In Chapter 3, the author details the research data, variables, methods, and experimental models used in the study, providing a comprehensive overview of the research framework The chapter sets the foundation for understanding the methodology behind the findings In Chapter 4, the author will present and discuss the results of the study, analyzing the data to draw meaningful conclusions This structure ensures a clear and logical flow from research design to data interpretation, enhancing the article's coherence and SEO relevance.
CHAPTER IV RESEARCH RESULTS AND DISCUSSION OF RESEARCH
Chapter 4 presents a detailed analysis of the research model estimation, including statistical descriptions and regression analysis based on the collected data The findings are compared with previous experimental studies to highlight similarities and differences Ultimately, the chapter draws comprehensive conclusions on the factors influencing the capital structure of publicly listed companies on the Vietnam Stock Exchange, providing valuable insights for investors and policymakers.
Descriptive statistical analysis
Descriptive statistical analysis is essential for summarizing research data characteristics and facilitating the examination of key indicators such as the mean, standard deviation, minimum, and maximum values These statistical measures provide valuable insights into data distribution and variability, supporting accurate and comprehensive data interpretation The results of this analysis are presented to offer a clear understanding of the data’s central tendency and dispersion, aiding in informed decision-making and further research analysis.
TABLE 4 1 The list describes the variables used in the research paper
Variable Obs Mean Std Dev Min Max
(Source of results from STATA 16 software)
Understanding key statistical indicators is essential for data analysis Mean represents the average value within a dataset, providing a central tendency measure Standard Deviation (Std.Dev) indicates the variability or dispersion of the data points around the mean Min and Max denote the smallest and largest values observed, respectively, offering insights into the data range Obs (Observation) refers to the total number of data points collected, ensuring the reliability of the analysis These statistical metrics are fundamental for accurate data interpretation and effective decision-making.
Based on the results of table 4.1 we can comment as follows:
The debt-to-equity (TLEV) ratio of hose and HNX-listed enterprises from
2012 to 2021 has an average of 57.98%, the highest being 1 of Dabaco Group in
2016 and the lowest being 0.0404083 of The Vegetexco Corporation (VGP) in 2015
Between 2012 and 2021, the short-term debt to equity ratio (STLEV) for HBX and HNX-listed enterprises averaged 46.13%, indicating moderate leverage levels across these companies The highest STLEV was recorded at 0.95708 by The Vegetexco Corporation (VGP) in 2018, reflecting increased short-term debt relative to equity during that year Conversely, the lowest ratio was 0.04041 in 2015, demonstrating a significant reduction in short-term liabilities compared to equity for The Vegetexco Corporation, highlighting variability in financial leverage over the analyzed period.
Between 2012 and 2021, the average Return on Assets (ROA) for companies listed on HOSE and HNX was approximately 6.95%, with a standard deviation of 0.073 Notably, Khang Dien House Trading and Investment (stock code KDH) achieved its highest ROA of 0.7837 in 2015, while its lowest ROA of -0.078486 occurred in 2013, highlighting significant fluctuations in the company's profitability over the period.
The Return on Equity (ROE) of enterprises listed on Hose and HNX from 2012 to 2021 averages 14.84%, indicating generally healthy profitability levels The data exhibits a standard deviation of approximately 0.1285, reflecting moderate variability in ROE over the years The highest ROE was recorded at 98.22% by Khang Dien House Trading and Investment (stock code: KDH) in 2015, while the lowest was -50.11% for SMC Trading Investment (stock code: SMC) in the same year, highlighting significant fluctuations in company performances during this period.
From 2012 to 2021, the tangible assets (TANG) of enterprises listed on HOSE and HNX averaged 19.83%, with a standard deviation of approximately 0.146, indicating moderate variability The highest TANG value was recorded at 1.0 for Dabaco Group (stock code DBC) in 2016, reflecting full asset utilization at that time Conversely, the lowest TANG value was 0.0003718 for Petroleum General Distribution Services JSC (stock code PSD), highlighting significant disparities in asset tangible levels among listed companies over the examined period.
The enterprise size (SIZE) of enterprises listed in HOSE and HNX from
From 2012 to 2021, the average stock price was approximately 15.92, with a standard deviation of 1.27, indicating moderate variability over the period Vingroup (VIC) reached its highest value of 19.88 in 2021, reflecting strong performance, while The Vegetexco Corporation (VGP) recorded its lowest value of 12.00 in 2016, highlighting notable fluctuations within the decade.
The growth rate (GROW) of hose and HNX listed enterprises from 2012 to
In 2021, the average value recorded was approximately 0.254, with a standard deviation of 3.510, indicating variability in the data The highest recorded value was 59.55 for Khang Dien House Trading and Investment (stock code KDH) in 2012, reflecting a significant peak in its trading performance Conversely, the lowest value was -0.99 for The Vegetexco Corporation (stock code VGP) in 2017, highlighting a substantial decline during that period.
Corporate income tax (TAX) of enterprises listed in HOSE and HNX from
Between 2012 and 2021, the average return was 21.32%, with a standard deviation of 0.2411, indicating moderate variability The highest return during this period was 3.4021% in 2021 for Vingroup (VIC), while the lowest was -1.2878% in 2021 for Ho Chi Minh City Infrastructure Investment JSC (CII).
From 2012 to 2021, the liquidity (LIQ) of enterprises listed on HOSE and HNX averaged approximately 1.736, with a standard deviation of 1.312 The highest liquidity value observed was 13.536 for The Vegetexco Corporation (VGP) in 2015, indicating a significant spike in liquidity that year Conversely, the lowest liquidity was recorded at 0.569 for Vingroup (VIC) in 2017, reflecting a period of lower asset liquidity among listed companies during that time.
Product specificity (UNIQ) of enterprises listed at HOSE and HNX from
2012 to 2021 has an average of 0.823859 and the standard deviation is 0 1271469, the highest value was 1.34996 of Khang Dien House Trading and Investment in
2012, the lowest value was 0.2906995 of Ho Chi Minh City Infrastructure Investment JSC (stock code is CII) in 2012
The turnover rate of total assets (AT) of enterprises listed in HOSE and HNX from 2012 to 2021 has an average of 1.290178 and the standard deviation is
In 2018, Vietnam Germany Steel Pipe JSC (stock code VGS) reached its highest stock value of 4.873003, indicating strong performance in that year, while Khang Dien House Trading and Investment (stock code KDH) experienced its lowest value of just 0.0017819 The fluctuations in stock prices highlight the volatility within these companies, with KDH's lowest point occurring in 2012 and VGS's peak occurring in 2018 Understanding these significant value changes is essential for investors analyzing market trends and company performances.
Correlation analysis
TABLE 4 2 Correlation Matrix between variables
TLEV ROA ROE TANG SIZE GROW TAX LIQ UNIQ AT TLEV 1
(Source of results from STATA 16 software)
Table 4.2 presents the correlation matrix of regression match variables, revealing the direction and strength of their linear relationships This analysis helps identify potential multicollinearity issues, which can occur when two explanatory variables are highly correlated A high pairwise correlation coefficient indicates redundancy, as one variable's information overlaps with another, potentially compromising the reliability of the model's estimates Addressing multicollinearity ensures more accurate and dependable regression results.
The correlation matrix analysis reveals very low correlation coefficients (less than 0.8) between variable pairs and among independent variables and LTEV variables This indicates a minimal risk of multicollinearity issues within the model, ensuring the reliability and validity of the regression analysis.
TABLE 4 3 VIF Variance Magnification Coefficient
(Source of results from STATA 16 software)
Testing the Variance Inflation Factor (VIF) provides crucial evidence regarding multicollinearity among variables in the model As shown in Table 4.3, all VIF coefficients for the independent variables are below 4, with an average VIF value of 1.63, indicating a low likelihood of multicollinearity Therefore, it can be concluded that the model does not suffer from multicollinearity issues, allowing all variables to be included confidently in the regression analysis.
Selection of regression model
This study investigates the inherently variable structure by developing three regression equations The author systematically replaces the TLEV and STLEV variables into equations (1) and (2) to analyze the models effectively As a result, specific models are derived, providing insights into the relationships between variables This approach enhances the understanding of the variable structure, offering valuable findings for related research.
Model 1: TLEV sub-variables and independent variables: ROA, ROE, TANG SIZE, GROW, TAX, LIQ, UNIQ, AT
TABLE 4 4 Regression Results by Pool OLS, FEM, REM for TLEV
(Source of results from STATA 16 software)
When selecting the appropriate model for analysis, the Hausman test plays a crucial role in determining whether to use a fixed impact model (FEM) or a random impact model (REM) This test evaluates if the unique individual effects are correlated with the explanatory variables, guiding researchers to choose the most suitable model for accurate results.
The Hausman test tested the hypothesis:
H0: Ui and non-correlated independent variables=> REM model is appropriate
H1: Ui and independent variables are correlated=> FEM model is appropriate
When the P-value (Prob > chi2) is less than 0.05, we reject the null hypothesis (H0), indicating a significant relationship between Ui and the correlated toxic variable, which warrants the use of a fixed impact model (FEM) Conversely, if the P-value exceeds 0.05, we accept H0, suggesting no significant correlation between Ui and the independent variable, and thus the random impact model (REM) is appropriate.
Choose between FEM and REM
Statistical value (F test that all u_i=0) chi2(9)$3.88
(Source of results from STATA 16 software)
According to the data from Stata 13.0, with a meaningful level of 1%, we have Prob = 0.0000 < 5% should reject H0 So the FEM model is suitable
The author went on to perform the F-test to help select the right model between Pooled OLS and FEM with the hypothesis:
H0: There is no difference between objects or different times => pooled OLS model is appropriate
H1: There are differences between objects or different times => fem model is appropriate
According to Gujarati et al (2009), the test results for the P REM model is appropriate
H1: Ui and independent variables are correlated=> FEM model is appropriate
When the P-value (Prob>chi2) is less than 0.05, the null hypothesis (H0) is rejected, indicating that the impact of Ui and the correlated toxic variable should be modeled using a fixed effects model (FEM) Conversely, if the P-value exceeds 0.05, the null hypothesis is accepted, and a random effects model (REM) is appropriate for analyzing Ui alongside the non-correlated independent variables Proper selection between FEM and REM based on the P-value ensures accurate and reliable statistical modeling.
TABLE 4 8 HAUSMAN-TEST CHOOSE BETWEEN FEM AND REM
Statistical value (F test that all u_i=0) chi2(9) = 55.01
(Source of results from STATA 16 software)
According to the data from Stata 16.0, with a meaningful level of 1%, we have Prob = 0.0000 < 5% should reject H0 So the FEM model is suitable
The author continues to perform F-testing to help select the right model between Pooled OLS and FEM with the hypothesis:
H0: There is no difference between objects or different times => pooled OLS model is appropriate
H1: There are differences between objects or different times => FEM model is appropriate
According to Gujarati et al (2009), the test results for the P