In particular, some studies following quality of governance research direction to study factors having impact on tax revenue are Makmun Syadullah & Tri Wibowo 2015, Dumisani Pamba 2023,
INTRODUCTION
The necessity of research
Tax revenue serves as the primary source of government budget collection for many countries, particularly in the ASEAN region Numerous studies have explored the factors influencing tax revenue globally, with notable research focusing on governance quality, such as the works of Makmun Syadullah & Tri Wibowo (2015), Dumisani Pamba (2023), and Epaphra and Massawe (2017) Other studies have approached taxation from a socio-political perspective, including research by Castaúeda Rodríguez (2018), Ayenew (2016), and Le Thi Huong Mai, Dinh Tran Ngoc Huy, and Tran Van Hung (2021) However, variations in natural conditions and socio-economic contexts limit the applicability of findings across different countries, often resulting in homogeneity in the analysis of tax revenue factors.
To build upon previous research, it is essential to investigate the factors influencing tax revenue in ASEAN countries from 1991 to 2021.
Research goals
The aims of this study consist of three main purposes:
- Applying tax knowledge learned and the understanding of tax theories to study and find out the factors affecting tax revenue in ASEAN countries in the period from 1991 to 2021
This study develops a research model by integrating established frameworks and innovative variables tailored to the research scope Utilizing econometric tools, the model analyzes the impact of various factors on tax revenue across ASEAN countries.
This study aims to analyze the influential factors affecting tax collection in ASEAN countries and provide actionable recommendations for enhancing tax collection policies Additionally, it will offer valuable insights and specific lessons for Vietnam to improve its tax collection system moving forward.
Scope of research
This research aims to analyze the factors influencing tax revenue in nine ASEAN countries—Vietnam, Singapore, Indonesia, Malaysia, Myanmar, Thailand, Cambodia, the Philippines, and Laos—over a period spanning from 1991 to 2021.
Methodology
Research methodologies used in the study include:
Structure of research
The structure of my research includes four main parts and each parts’content is presented below:
RESEARCH CONTENT
- Chapter 5: General discussion for ASEAN countries and recommendations for Vietnam
- Assessment of supportive level of research for taxation theory and practical
Literature Review
An overview of foreign research all over the world
Researching the factors that influence tax collection is essential, as taxes are a primary source of government revenue globally The Covid-19 pandemic and recent political events have heightened the need for studies in this area to develop more effective tax collection policies This section will provide an overview of international research trends regarding factors affecting tax revenue, the methodologies employed, the findings of previous studies, and a comparison of these results.
Research on the factors influencing tax collection in various countries gained significant traction in the mid-20th century and has continued to evolve in the 21st century, with new insights reinforcing earlier findings.
In the mid-20th century, specifically between 1970 and 1980, significant research was conducted on taxation trends in developing countries, notably by Cheliah (1971) and Chelliah et al (1975) Cheliah's study, titled "Trends in Taxation in Developing Countries," analyzed tax data from approximately 27 to 30 developing nations over a 15-year span, focusing on data collected from 1953 to 1955 and again from 1966 to 1968 The research emphasized two key factors: the tax ratio and the composition of tax revenue, which directly influenced tax collection activities and governmental policy adjustments For the cross-sectional analysis involving 50 developing countries, regression models were employed to explore the relationships between various factors—such as GDP, non-export income, and the shares of mining and agriculture in GDP—and the overall tax ratio, as well as the dynamics between direct and indirect taxes.
The author utilized a regression model to analyze the relationship between marginal tax rates and tax ratios, revealing key insights into tax collection in developing countries The findings indicated that the share of mining in GDP and non-mining exports positively influenced tax revenue, while agriculture's contribution had a negative effect Additionally, the research highlighted that higher per capita income in developing nations correlates with increased development and tax-paying capacity, though no direct link between per capita income and tax revenue was established Chelliah and colleagues furthered this inquiry in 1975 with their study “Tax Ratios and Tax Effort in Developing Countries,” which aimed to update the findings from their 1971 analysis.
A study involving 47 developing countries utilized regression models to examine the impact of tax ratio factors and revenue compositions on tax collection activities The findings indicated that the contributions of mining and agriculture to GDP continued to significantly influence tax revenue, consistent with results from a 1971 study Furthermore, the research revealed that trade per capita income (excluding exports) and exports (excluding mining) did not affect tax revenue.
From 1990 to 2000, there were also two notable studies by Stotsky and WoldeMarian
The study "Tax Effort in Sub-Saharan Africa" by Stotsky and WoldeMarian (1997) utilized regression models to analyze the relationship between six key variables—agriculture, mining, manufacturing, per capita income, exports, and imports—and tax collection in 46 Sub-Saharan African countries from 1990 to 1995 Findings indicated that both the share of exports and GDP per capita positively influenced tax revenue, whereas agriculture and mining shares negatively impacted it Additionally, manufacturing and import density showed no significant effect on tax revenue in the region.
A study examining the factors influencing tax revenue in 83 developing countries from 1978 to 1988 extended Tanzi's earlier findings from 1987, which identified a positive correlation between tax revenue and per capita income However, Tanzi's 1992 research indicated that this relationship had weakened, highlighting the importance of macroeconomic stability, debt servicing needs, and economic structure changes as more significant determinants of tax revenue His alternative model linked tax share in GDP to the agriculture share, imports share, foreign debt share, and per capita income, revealing that while per capita income did not significantly impact tax revenue, both the share of imports and foreign debt positively influenced tax revenue.
From 1970 to 2000, research primarily focused on developing countries, examining the correlation between economic factors—such as import share, agricultural contribution to GDP, and per capita income—and tax revenue While these studies successfully identified trends in this relationship, they fell short in providing actionable recommendations for improving tax collection policies.
In the 21st century, research has evolved significantly, with a notable study by Gupta and Briancon (2007) titled “Determinants of Tax Revenue Efforts in Developing Countries,” which examined the interplay of structural, institutional, and policy factors affecting tax revenue Utilizing a baseline regression model with fixed and random effects, the study analyzed key structural elements such as central government revenue as a percentage of GDP, per capita GDP in PPP, and various economic indicators including agriculture value added and imports Additionally, it explored policy variables like tax revenue from goods and services, income, and trade, alongside institutional factors including corruption and the rule of law, highlighting their collective impact on tax revenue in developing nations.
A study conducted on 105 developing countries from 1998 to 2007 identified key structural factors influencing tax revenue performance, including per capita GDP, agricultural share in GDP, trade openness, and foreign aid It was found that corruption adversely impacts tax collection, while political and economic stability also play a role, albeit in specific contexts Furthermore, the research revealed that developing nations relying primarily on tax revenue from goods and services tend to experience lower tax performance, whereas those with a higher focus on income, profits, and capital gains taxes achieve better collection outcomes This study builds on the foundational work of Chelliah (1971) and Chelliah et al.
Research since 1975 has focused on institutional factors affecting tax collection, including corruption, law and order, and political stability A significant study by Ajaz and Ahmad (2005) examined the impact of corruption on tax revenues in 25 developing countries from 1990 to 2005, utilizing the GMM regression model Their findings indicated that corruption negatively influenced tax revenue, while strong governance improved tax collection In contrast, Gupta and Briancon (2007) highlighted that various institutional variables also significantly impacted tax revenues More recent studies, such as Baunsgaard and Keen (2010), explored additional factors like globalization and inflation affecting tax collection globally Their research on 117 countries over 32 years revealed that trade liberalization positively influenced tax revenues through import taxation.
7 said that tax collection activities could be improved when trade expanded and there was an increase of credibility and competitiveness of the economy In 2010, another prominent study could be mentioned was the study by Profeta and Scabrosetti (2010) named “The political economy of taxation: Lessons from developing countries” The scope of this study was 39 countries including 11 Asian countries, 19 Latin American countries and 9 countries in the European Union in the period from 1990 to 2004 This study also researched the influence of institutional, policy and structural factors on tax revenue in developing countries The study concluded that GDP per capita and debt-to-GDP ratio were not important determinants for tax revenues in Asean countries but they were significant factors indicating tax collection in Latin American countries Besides, it clearly indicated the negative relationship between the share of agriculture in GDP and tax revenue in Latin American but in Asia, it did not have any influence on tax revenue In addition, there were new researched variables including the economy's openness; the indexes of democratic rights, civil liberties, political rights, education level; the proportion of the over-65 population; the percentage of female labor; the size of the underground economy; population density In Asian and European countries, the economy's openness had a positive impact on tax revenue whereas it negatively affected tax revenue in Latin America For the indexes of democratic rights, civil liberties, political rights, this study concluded that countries with higher indexes would have a better tax revenue and efficient tax system For the education level, the proportion of the over-65 population, the percentage of female labor, and the size of the underground economy, these factors positively and significantly influenced the tax revenue in Latin American countries whereas population density did not have any effect However, in Asian countries, the proportion of the over-65 population had a significantly negative influence on tax revenue while the high school graduation rate and the proportion of the urban population had no impact In 2012, another study by Didoa named “Structural determinants of tax revenue in Latin America and the Caribbean, 1990 - 2009” was a notable study in this period The scope of research was 32 Latin American and Caribbean countries in the period from 1990 to 2009 This study also had similar research variables as Profeta and Scabrosetti (2010) Besides, it also concerned the
The study identified eight new variables, including colonial heritage and internal conflicts, that clarify regional differences in taxation It concluded that factors such as economic openness, education levels, female labor force participation, and population density positively influence tax revenue Additionally, it found that higher civil liberties and political stability correlate with increased tax revenue, although no positive link was established between political rights and tax revenue, contrasting with previous research Notably, the political rights index positively impacts direct tax revenues but not indirect taxes The fiscal policies of the studied countries were shaped by colonial heritage and internal conflicts, revealing differences in tax burdens and compositions Specifically, Caribbean nations, particularly former British and Dutch colonies, exhibited higher tax revenues due to tax systems aligning more closely with Western standards, supported by greater civil liberties, political rights, and reduced conflict.
An overview of Vietnam research
In Vietnam, several studies have examined the factors influencing tax revenue in ASEAN countries Notably, research by Le Thi Huong Mai, Dinh Tran Ngoc Huy, and Tran Van Hung (2021) analyzed tax collection determinants from 2000 to 2019 using a static linear regression model Their findings indicated that GDP, trade openness, political rights, and civil liberties positively impact tax revenues Similarly, a study by Nguyen Minh Ha, Pham Tan Minh, and Quan Minh Quoc Binh (2022) investigated tax revenue factors in ASEAN from 2000 to 2016, employing both static and dynamic linear regression models Their results highlighted the significance of trade, foreign direct investment (FDI), and manufacturing in enhancing tax revenue.
In ASEAN countries, EXDEBT positively influenced tax revenue, while CIVLIB, SCHTER, and inflation showed no significant impact Conversely, the only factor that negatively affected tax revenue was Official Development Assistance (ODA).
Results inherited in the previous studies
All the studies reviewed share a commonality in that they build upon each other's findings to arrive at comparable conclusions This research will focus on analyzing the interconnected results from two primary research groups, highlighting the factors that have been inherited across these studies.
- 1 st group of factors: Quality of Governance factors
- 2 nd group of factors: Economy, Society, Administration factors
1.3.1 Results Inheritation of first group of factors: Quality of governance factors
Understanding the quality of governance is crucial before exploring the research trends of this group The World Bank’s 1992 report, "Governance and Development," initially defined good governance as the effective use of power to manage a country's economic and social resources for development Today, good governance encompasses minimizing corruption, considering minority opinions, amplifying the voices of the oppressed in decision-making, and addressing community demands This research group aims to investigate the link between tax revenue activities and the objectives of good governance, questioning whether effective tax collection is possible in nations with high-quality governance Their analysis is grounded in three theoretical frameworks: the theory of political legitimacy, the taxation mechanism, and the quality of governance mechanism The theory of political legitimacy, rooted in political science and sociology, pertains to the public's belief in the rightful authority of a ruler or institution.
The research group posits that tax compliance and collection are significantly influenced by citizens' trust in the government Utilizing the benefit theory, it asserts that individuals should pay taxes proportional to the benefits received from government services Conversely, the taxation theory examines the relationship between taxpayers' ability to pay and their perceived burden of taxation, particularly among poorer populations The study also incorporates the quality of governance mechanism, which encompasses regulations and best practices that assess governmental effectiveness and its impact on tax collection Notably, Dumisani Pamba's (2023) research on South Africa demonstrates that factors such as voice and accountability, regulatory quality, government effectiveness, political stability, and the rule of law negatively influence tax revenues, except for corruption control Additionally, the study by Makmun Syadullah & Tri Wibowo (2015) on ASEAN countries reveals that not all governance quality factors adversely affect tax revenue, using data from the World Bank spanning 2003 to 2012.
The study of ASEAN countries reveals that factors such as corruption control, political stability, and accountability negatively influence government tax revenue; as these factors improve, tax revenue tends to decline Conversely, elements like the rule of law and regulatory quality positively affect tax revenue, indicating that stronger enforcement leads to more effective tax collection Research by Epaphra and Massawe (2017) on 30 African nations from 1996 to 2016 highlights that government effectiveness, regulatory standards, and accountability enhance tax revenue, particularly for indirect taxes, while corruption significantly undermines tax generation.
1.3.2 Result Inheritation of second group of factors: Economy, Society, Administration factors
The second research group focuses on social-political theory regarding taxation, positing that social, political, and economic factors significantly influence tax revenues Key determinants include GDP, foreign direct investment (FDI), inflation (INF), and political rights An illustrative study by Castro and Camarillo (2014) examined the factors affecting tax revenue across 34 OECD member countries, providing insights into the dynamics of taxation influenced by these variables.
From 2001 to 2011, research indicated that GDP and manufacturing positively influenced tax revenue in OECD countries, while factors such as FDI, ARG, civil liberties indices, and life expectancy negatively impacted it A study by Imam and Jacobs (2014) on 12 Middle Eastern countries from 1990 to 2003 found that inflation positively affected tax income, whereas GDP per capita had a negative effect Additionally, Castañeda Rodríguez (2018) identified that long-term variables, including economic, social, political, and cultural aspects, significantly influenced tax performance, highlighting that taxation exhibits path dependence influenced by historical factors, including total tax burden and revenue from consumption and income taxes, as well as the level of progressiveness.
Tax revenue is significantly influenced by historical and structural factors, including the economic climate and other public income sources like inflation Research by Ayenew (2016) highlights the determinants of Ethiopian tax revenue from 1975 to 2013, revealing that manufacturing, GDP growth, and foreign aid positively impact tax income, while inflation adversely affects tax revenue.
Research gap
Research on the factors influencing tax revenue varies across different countries, yielding diverse conclusions from various study groups Rather than conflicting, these findings complement one another, providing a comprehensive understanding of the factors at play This collective insight aims to inform optimal solutions for enhancing tax revenue for governments globally.
However, these studies still have a research gap The research gap comes from:
Research on factors influencing tax revenue predominantly focuses on developing countries, particularly within the OECD region, South East Asia, Africa, and Sub-Saharan regions However, studies specifically targeting ASEAN countries are scarce, with significant research emerging only between 2009 and 2021.
Most research on tax revenue in developing countries, particularly in ASEAN nations like Vietnam, was conducted prior to the COVID-19 pandemic Consequently, these studies do not accurately reflect the pandemic's impact on tax revenue, limiting their ability to assess its effects or provide relevant recommendations If a similar pandemic occurs in the future, it could significantly harm tax revenue, highlighting the need for updated research to understand and mitigate such impacts.
13 challenge for most government when considering COVID 19 policies to improve tax revenue
The gap in research data poses significant challenges in analyzing tax revenue policies, particularly for developing countries Limited availability of crucial data, such as per capita income and tax-to-GDP ratios, from public sources like the IMF and World Bank hinders researchers' ability to provide accurate analyses and recommendations Consequently, this data scarcity impacts the formulation of effective tax revenue strategies globally.
Therefore, the basis of this research is to conduct research in ASEAN countries to understand and find more about the factors affecting tax revenue in ASEAN from
From 1991 to 2021, professional international organizations such as the IMF, WDI, and MOF have provided reliable data This research analyzes the impact levels of various factors to formulate tax revenue recommendations specifically for ASEAN and Vietnam.
Research questions
To answer the question of ways to turn taxes into a functional source of revenue, the research questions are:
1 What is definition of tax ?
2 What is tax administration ? What are criterias to assess an tax administration of a government ?
3 What are factors affecting tax revenue ?
4 What are factors affecting tax revenue in ASEAN countries ?
5 What are impact levels of each factor affecting tax revenue in ASEAN countries in the period from 1991 to 2021 ?
6 What are tax policies and recommendations for ASEAN countries and Vietnam in the future to have an efficient tax revenue source ?
Theoretical Framework
Tax Definition and Taxation Theories
According to the IMF's definition of government revenue in chapter 5 of
According to the "Government Finance Statistics Manual" (2001), tax revenue is the primary source of funding for government entities, consisting of compulsory transfers to the general government sector This revenue forms the majority of financial resources for various government operations Additionally, obligatory transfers such as fines, penalties, and most social security contributions are exempt from taxation Refunds and adjustments related to incorrectly collected tax funds are considered transactions that decrease the net worth of the collecting government entity, serving as corrections for previously recorded excesses in net worth, and are thus classified as losses.
Tax is a mandatory payment required by the government, imposed on individuals and entities that meet specific legal criteria Taxpayers are obligated to fulfill these payments, while tax collectors, as government representatives, are responsible for ensuring compliance and administering the tax collection process effectively.
Taxes are mandatory for all citizens, ensuring fairness among taxpayers and reflecting the government's authority Without tax revenue, the government lacks the economic capacity to fulfill its responsibilities, as 90% of budget revenue is derived from taxes Recognizing taxes as a source of power enhances effective tax collection, providing essential funding for the nation Furthermore, taxes are non-countervailing, meaning all eligible individuals must pay regardless of public benefits received, and they are indirectly refundable, as the government reinvests tax revenues into public services and infrastructure.
15 infrastructures such as building schools, roads, etc and society enjoys it, including tax payers
When examining tax collection policies, it is essential to consider three key theories: the cost of service theory, which emphasizes the expenses incurred in providing public services; the benefit theory of taxation, which suggests that taxes should correspond to the benefits received by taxpayers; and the social-political policy on taxation, which reflects the broader societal and political implications of tax systems.
The cost of service theory was first stated by Von Hock - an Australian economist This theory was supported by the statement “Building good taxation policy bases on
The principles of justice and equality suggest that taxation should correspond to the cost of services provided by the government Citizens collectively fund these services, implying that an individual's tax burden should reflect the benefits received However, this theory faces significant criticism from economists Firstly, accurately determining the total cost of government services poses a challenge, complicating the equitable distribution of tax responsibilities Secondly, in a modern welfare state, this approach could result in poorer individuals paying higher taxes due to their greater reliance on government benefits, while wealthier citizens may pay less, contradicting the principle of justice.
The benefit theory of taxation modernizes the cost of service theory by addressing its limitations This theory posits that individuals should pay taxes based on the benefits they receive from government services Essentially, citizens are expected to contribute to the costs of these services in proportion to the advantages they gain; thus, those who derive greater benefits are required to pay higher taxes.
The theory presents several drawbacks, primarily the challenge of quantifying individual benefits in monetary terms, which complicates the determination of corresponding taxes Additionally, it fails to resolve issues of equality, as individuals with lower incomes may face higher tax burdens despite receiving more government benefits, according to welfare state principles Lastly, the theory struggles to identify the specific proportion of general benefits that apply to individuals, as the value of governmental services remains inherently subjective and difficult to measure.
Adolph Wagner, a German economist, posited that tax determination—encompassing amount, rate, and policy—should align with social and political objectives rather than individual interests He argued that society transcends mere aggregation of individuals, possessing its own existence and mechanisms for preservation Consequently, the government's primary role is to ensure this societal existence, with the tax system designed to address the collective needs of society rather than catering to individual preferences Wagner's approach to tax policy emphasizes a modern philosophical perspective focused on the greater good.
This research will utilize the third theory due to its effectiveness in analyzing social and political factors such as GDP, FDI, and political leadership By examining these elements, the study aims to gather data that identifies their impact on tax collection policies in ASEAN countries Consequently, the findings are expected to yield objective results and provide improved solutions to address tax collection challenges in the region.
Tax administration
Research by the IMF (1997) in "Fiscal Federalism in Theory and Practice" defines tax administration as the government's establishment of mechanisms to uphold the rights and obligations of taxpayers and tax collection entities The study highlights various tax administration models across countries, emphasizing the distinction between centralized and decentralized tax management Centralized governments focus on tax policies aimed at stabilizing the economy, while decentralized governments argue that excessive control over tax implementation is counterproductive and indicative of weak governance Consequently, the research identifies four primary models of tax management based on these two government types.
- Centralized government tax management with the allocation of revenue sharing and transfers
- Centralized government tax management with tax assignment to various levels of government
- Multilevel tax administration with revenue transfers and sharing
- Each level of government managing the taxes allotted to it
2.2.2 Criteria for evaluating tax administration activities
According to Adam Smith (1723 - 1790), to build an effective tax administration system, there are four basis that government should consider including:
- The mobilizatoin of tax along with the ability and capacity of the population
- The transparency determination of tax rates and payment terms
- The suitable paying tax time for tax payers
- The lowest costs of organizing tax collection and payment
A well-functioning tax administration system, as outlined by the Association of British Public Accountants (ACCA), is essential for government oversight and must adhere to 12 fundamental principles These principles emphasize the importance of government control in preventing tax avoidance and evasion.
The tax-to-GDP ratio is a crucial metric that should be clearly defined to enhance transparency, accountability, and public understanding Simplification and stabilization of tax regulations are essential for ensuring competitiveness and efficiency within the economy It is important that regulations are regularly updated to maintain relevance, while establishing a direct connection between tax revenue and its allocation To foster a fair tax environment, double taxation must be avoided, and the rights and responsibilities of taxpayers should be explicitly defined Additionally, the implementation of green taxes can promote environmental sustainability.
International organizations, including the World Bank and the International Monetary Fund, have established six criteria to assess the effectiveness of tax administration systems, which are widely adopted by countries globally.
- Criteria for tax registration, tax return, tax payment, tax refund
- Criteria for debt collection management and tax debt enforcement
- Critieria for taxpayer information management
- Criteria for tax inspection, examination and handling of tax violations
- Criteria for settlement of tax complaints and denunciations
- Criteria for propaganda and support for taxpayers
My research will assess the effectiveness of a country's tax administration system using criteria established by the World Bank (WB) and the International Monetary Fund (IMF) Each criterion will be detailed and clarified in the following sections.
2.2.2.1 Tax registration, tax return, tax payment, tax refund
A country has good tax registration, tax revenue, tax payment and tax refund activities when it is:
- Using a unique, high-integrity tax code, usually simple numbers
- Maintaining a complete, accurate and reliable tax registration database; simplifying declarations, including pre-filled declarations
- Promptly monitoring and taking appropriate measures for undeclared cases
- Providing and increasing the use of electronic forms of tax declaration, payment and refund for major taxes
- Using automated risk assessment software to evaluate all value-added tax (VAT) refund claims against risk criteria
- Paying a valid VAT refund, offsetting the deducted amounts with other tax obligations within a reasonable time frame
- Using automatic tax accounting system and ensure timely and accurate accounting of payments and other transactions for taxpayers
2.2.2.2 Debt collection management and tax debt enforcement
A country has good debt collection management and tax debt enforcement activities when it is:
- Strengthening the application of deduction at source; manage debts by amount, debt age and debt collection capacity
- Writing off debts for bad debts
- Using automatic risk assessment software to determine the priority order of tax debt cases, thereby giving the most effective measure in debt collection…
A taxpayer information system must prioritize accuracy, truthfulness, and completeness by incorporating essential details such as tax registration, declarations, payments, finalizations, and taxpayer accounts It should also include information on production and business results, tax compliance, and other relevant taxpayer operations and transactions To achieve this, data must be gathered from diverse channels and sources, including taxpayers themselves, tax administration agencies, state entities, and related organizations or individuals.
2.2.2.4 Tax inspection, examination and handling of tax violations
Tax inspection is based on risk analysis and management, including:
- Collecting and analyzing risk-related information from internal and external sources
- Identifying, assessing, and ranking risks within the framework of taxpayer classification, major taxes, major tax obligations
- Managing key risks through developing and implementing a compliance improvement plan
2.2.2.5.Settlement of tax complaints and denunciations
A clear and transparent dispute settlement mechanism is essential, outlining taxpayer rights and legal processes The government must actively monitor the root causes of disputes and implement corrective actions Additionally, if the resolution favors the taxpayer, prompt refunds of any overpaid taxes should be issued.
2.2.2.6 Propaganda and support for taxpayers
A country has good propaganda and support for taxpayers activities when it is:
- Diversifying forms of propaganda by designing user-friendly products
- Design information-providing products suitable for each group of requirements
- Regularly updating information products related to changes in tax management policies and processes…
Many indicators are utilized to evaluate the effectiveness of tax administration, with three key methods endorsed by international organizations such as the World Bank and the IMF, which have been widely adopted by various countries.
- Toolkit for assessing the effectiveness of TADAT tax administration
- Tax payment index in the Business Environment Report
- Assessing the satisfaction of taxpayers through surveys and surveys.
Factors affecting tax revenue
There are three main groups of factors having impact on tax revenue in coutries all over the world These three groups are economic, social and political factors
Economic factors such as GDP growth rate, trade openness, and value added in agriculture and manufacturing significantly influence tax revenue Research by Castro and Camarillo (2014) and Fox and Gurley (2015) indicates a positive relationship between these economic indicators and tax revenue, suggesting that fluctuations in these factors directly affect tax income levels.
Social factors significantly influence tax revenue, with educational attainment and life expectancy contributing positively, while a higher infant mortality rate adversely affects it.
Political factors such as political rights, civil liberties, and corruption significantly influence tax revenue, with research indicating a positive correlation between these elements and effective tax policies Conversely, the time required to prepare and pay taxes highlights the complexities of tax administration, demonstrating a negative impact on tax revenue; as tax processes become more cumbersome, citizens face longer payment times, ultimately leading to decreased government revenue.
Factor affecting tax revenue in ASEAN countries
ADB's research, "Tax and Development: Challenge in Asia and the Pacific" by Araki and Nakabayashi (2018), identifies three key challenges in tax collection: a low tax-to-GDP ratio, the effectiveness of tax policies, and the efficiency of tax administration.
2.4.1 Low tax to GDP ratio
Figure 2.4: Tax-to-GDP ratios in Asian and Pacific economies (total tax revenue as % of GDP), 2020
Source: Revenue Statistic in Asia and the Pacific (2022) According to ADB in its research namely “Revenue Statistic in Asia and Pacific”
As of 2020, the average tax to GDP ratio in 28 Asian and Pacific economies was 19.1%, with only Vietnam (22.7%) and Cambodia (20.2%) exceeding this average among ASEAN countries This situation arises from ASEAN governments encouraging export activities through supportive policies like preferential export tax rates, including tariff elimination Such strategies align with the export-led growth hypothesis, which suggests that boosting exports fosters economic growth However, an exclusive focus on economic growth may jeopardize government tax revenues, limiting budget collections essential for future investments To enhance budget revenue, governments must find effective solutions to increase domestic revenue, which poses a significant challenge.
2.4.2 Tax policies - Tax preferences and the difficulties in managing
Tax policies in ASEAN countries face challenges due to the rise of Information and Communication Technology (ICT) and the integration into the global economy through global supply chains While tax incentives attract foreign investment by creating a favorable business environment, they can also pose risks if not managed properly If the Ministry of Finance oversees tax management, governments can effectively control tax policies and preferences However, if tax administration is handled by Investment Boards or FDI promotion ministries, it becomes challenging to regulate tax incentive policies, potentially leading to revenue losses To mitigate these losses, ASEAN countries may resort to increasing other distortionary taxes Ultimately, while tax incentives can facilitate integration into the global economy, their effectiveness hinges on the choice of tax management structure within each country.
ASEAN countries encounter significant challenges in their tax administration systems due to political development issues and the persistent threats of corruption These factors contribute to an unstable and weak tax administration, ultimately undermining tax collection efforts across the region.
Political development involves enhancing institutions and governance structures to foster national growth across various sectors The interplay between political development policies and tax collection strategies is reciprocal; effective political institutions and strong leadership contribute to more efficient tax revenue systems, ultimately supporting the nation's overall progress.
Improving tax collection policies can significantly enhance government budgets and foster better citizen behavior regarding tax payments, strengthening the bond between the government and its citizens Conversely, a lack of innovative leadership in tax management can result in ineffective policies that fail to benefit the country and its people, ultimately leading to decreased tax revenue This shortfall can hinder government investment in future projects and create a strained relationship between the government and the populace.
Most ASEAN countries do not concentrate on developing national political apparatus There are several reasons for this neglect including:
- ASEAN countries’ primary goal is to improve economic growth and develop human resources and capital
ASEAN leaders often exhibit confidence in their leadership capabilities, believing they understand what is best for their nations This conviction can lead them to prioritize their decision-making authority over the democratic rights of their citizens, neglecting the importance of listening to public opinions to enhance political systems and policies.
Ruling parties in Vietnam, Malaysia, Singapore, and Myanmar face anxiety over political power dynamics as they maintain a single-party system The evolution of the political apparatus in these nations has led to the potential emergence of new political parties, posing a threat to the dominance of the current ruling party.
Without significant changes and improvements in the political apparatus and policies, the efficiency of tax administration will likely fall short of that seen in countries like Japan and Switzerland This inefficiency could lead to an increased reliance on foreign capital and heightened pressure on foreign debt.
2.4.3.2 Prevention and Detection of Corruption
Table 2.4: ASEAN Countries’ Corruption Perception Index (CPI) 2022
No Country Corruption Perception Index -
In the 2022 Corruption Perception Index (CPI) report by Transparency International, the average score for ASEAN countries was 43 Only three nations—Brunei, Singapore, and Malaysia—exceeded this average, while Myanmar recorded the lowest score This indicates that corruption remains a significant challenge for the majority of ASEAN countries, highlighting the urgent need for effective governance and anti-corruption measures.
Corruption significantly impacts tax collection and government budgets, as highlighted by Tanzi and Davoodi's research Implementing 26 policies to monitor and address corruption is essential to mitigate its adverse effects on tax revenue.
(2000) There are several important conclusion in this research:
- Higher corruption is consistent with lower revenues of all types with the exception of non-tax revenues
- Corruption has a statistically significant negative correlation with individual income taxes
- Higher corruption is also associated with lower taxes collected from VAT, sales tax and turnover tax
ASEAN countries face significant challenges in enhancing tax collection policies due to three primary factors: a low tax-to-GDP ratio, inefficiencies in tax administration, and outdated tax policies.
As a result, it is necessary to study about these factors, then use these them to suggest better solutions to overcome difficulties for ASEAN countries.
The Empirical Model of Taxation: Castro and Camarillo (2014)
Research on factors influencing tax revenue globally can be categorized into two main groups: the impact of governance quality and the effects of economic, social, and administrative factors Castro and Camarillo (2014) focused on the latter, employing a research model that analyzed the influence of economic, structural, social, and institutional elements on tax revenue Their study utilized static and dynamic panel data techniques, demonstrating the significance of these factors through an analysis of 34 OECD countries from 2001 to 2011.
Castro and Camarillo developed a research model based on the original framework that illustrates the connection between tax revenue and various factors, including economic conditions, productive specialization, social influences, and institutional elements.
In this model, tax revenue serves as the dependent variable, while the independent variables, represented by X kit, encompass economic, productive specialization, social, and institutional factors The unobservable individual effects, denoted as N, vary across different countries, and the error term U is assumed to adhere to white-noise conditions The parameter Alpha allows the intercept to differ for each country, capturing specific variations among them, with I and t representing the respective country and year.
In their study of unbalanced panel data from 34 OECD countries, Castro and Camarillo employed two research methods: static regression models and dynamic regression models, to identify the most suitable analytical approach.
Static methods in econometrics encompass the Ordinary Least Squares (OLS) method, Fixed Effect Model (FE), and Random Effect Model (RE) These methods involve a comparative analysis of the results from the three models, alongside critical tests such as Multicollinearity, Autocorrelation, and Heteroskedasticity tests, to identify the most suitable static model Ultimately, the static regression model reflects similarities to the original model.
Dynamic methods were employed to analyze the impact of past tax revenue on current tax revenue, allowing researchers to determine if the variable could be self-explanatory These methods also addressed the issue of autocorrelation However, a challenge arose due to the correlation between the lagged dependent variable and the model's error term, necessitating the use of various estimation and testing techniques to validate the model's suitability The dynamic regression model is outlined below.
Research Methodology
Hypothesis
Figure 3.1: Relationship between economic, social and political factors and tax revenue in ASEAN countries from 1991 to 2021
In this section, my research will state the expectations about the results of my studies for three groups of studied factors
For economic factors, following the results of studies by Castro and Camarillo
In 2014, GDP and IND are anticipated to positively influence tax revenue, whereas ARG and EXDEBT are expected to have a negative effect According to Iman and Jacobs (2014), INF is predicted to positively impact tax revenue, while EXDEBT is also expected to contribute positively Additionally, given that ODA represents another borrowing source in ASEAN countries, this research hypothesizes that its influence will be similar to that of EXDEBT However, there is a lack of prior studies exploring the relationship between TRADE and tax revenue, leading to the expectation that TRADE does not significantly affect tax revenue.
Political factors, as highlighted by Castro and Camarillo (2014), indicate that POLRIG and CIVILB negatively affect tax revenue, leading my research to predict similar adverse effects in my model Additionally, Dumisani (2023) suggests that corruption, represented by the CPI, does not influence tax revenue, prompting the expectation that the corruption variable in my model will also have no impact on tax revenue.
29 is a new variable measuring the cumbersome level of tax administration systems in ASEAN countries so it is expected to have a negative impact on tax revenue
Social factors such as SCHTER and INFMOR are anticipated to have no effect on tax revenue In contrast, research by Castro and Camarillo (2014) suggests that LIFEXP is likely to negatively impact tax revenue.
My research anticipates that GDP, IND, INF, and LIFEXP will positively influence tax revenue, while TPPT, EXDEBT, ODA, ARG, POLRIG, and CIVILB are expected to have negative effects Additionally, SCHTER, INFMOR, TRADE, and CPI are projected to have no significant correlation with tax revenue.
Research Model
My research follows the Static Regression Model in Castro and Camarillo (2014) study to study the factors influencing tax revenue in ASEAN countries in the period from 1991 to 2021
This model will show the relationship between 15 independent variables which can be separated into three groups of studied factors namely:
- Economic factors: GDP, Trade Openness (TRADE), Agriculture Value Added (ARG), Manufacturing Value Added (IND), Inflation (INF), ODA, Foreign Debt (EXDEBT)
- Political factors: Political Right Index (POLRIG), Index of Civil Liberty (CIVLIB), Corruption perception index (CPI), Time to prepare and pay taxes (TTPT)
- Social factors: Educational Attainment (SCHTER), Life Expectancy (LIFEXP), Infant mortality rate (INFMOR)
I opted for the static regression model proposed by Castro and Ramirez (2014) and introduced two additional variables for several reasons Firstly, this model is widely recognized and utilized by numerous researchers in similar fields, providing a solid foundation for reference Secondly, my research aims to focus specifically on social, political, and economic factors, allowing me to gain valuable insights and formulate effective policy solutions.
This research addresses the critical issues of corruption and bureaucratic inefficiencies in tax collection for ASEAN countries, particularly Vietnam By incorporating the Corruption Perception Index (CPI), the study aims to analyze the impact of corruption on tax revenue and propose government policies to mitigate these challenges Additionally, the research will examine the Time Taken to Prepare Taxes (TTPT) to identify ways to streamline the tax administrative system, ultimately benefiting the government, citizens, and private sectors by saving time and enhancing tax revenue.
The form of my static regression model is showed below:
TaxRevenue= 𝛽 1 *GDP+ 𝛽 2 *TRADE+ 𝛽 3 *FDI+ 𝛽 4 *ARG+ 𝛽 5 *IND+ 𝛽 6 *POLRIG+ 𝛽 7 *CIVILB+ 𝛽 8 * SCHTER+ 𝛽 9 *LIFEXP+ 𝛽 10 *INFMOR+ 𝛽 11 *EXDEBT+ 𝛽 12 *ODA + 𝛽 13 *INF+ 𝛽 14 *CPI+ 𝛽 15 *TPP
GDP: The annual growth rate of GDP in ASEAN countries (%)
TRADE: Trade volume measured by the total exports and imports as the percentage of GDP (% GDP)
FDI: Foreign direct investment and it is measured by the net inflows to GDP (% GDP) ARG: Agriculture value added as the percentage of GDP (% GDP)
IND: Manufacturing (Industrial) value added as the percentage of GDP (% GDP)
The Political Right Index (POLRIG) evaluates a country's performance in key areas such as freedom of expression and belief, rights to association and organization, adherence to the rule of law, human rights, personal autonomy, individual and economic rights, and the independence of the judiciary.
CIVILB: Civil Liberties Index measures country performance on the quality of the electoral process, political pluralism and participation, government corruption and transparency, and fair political treatment of ethnic groups
SCHTER:Government expenditure on education as the percentage of GDP (% GDP)
LIFEXP: Life expectancy which is average life expectancy of population and is measured by the number of years (Years)
INFMOR: Mortality infant rate which is measure by the number of deaths per 1000 live births of children under the age of one (%)
EXDEBT: Foreign debt which is measured by cumulative external public debt as a percentage of GDP (% GDP)
ODA: The inflows of foreign aid which is net ODA received as a percentage of GNI (% GNI)
TPPT, or Total Time to Prepare and Pay Taxes, refers to the annual hours dedicated to the preparation, filing, and payment of taxes This encompasses three primary tax types: corporate income tax, value-added or sales tax, and labor taxes, which include payroll taxes and social security contributions.
Data Collection Methods
This research analyzes factors influencing tax revenue in ASEAN countries from 1991 to 2021 using a secondary data collection method Secondary data refers to information that has already been statistically analyzed and sourced from existing literature or other researchers Essentially, it is second-hand information, primarily consisting of numerical data.
To collect the secondary data, my research uses data presented by:
- Asian Development Bank Institute (ADBInstitute)
Table 3.3: The information of data sources of variables
Variables Research Sources Data Sources
(MOF) Trading Economic Asian Development Bank (ADB)
2 Nguyen Minh Ha, Pham Tan Minh
1 Le Thi Huong Mai, Dinh Tran Ngoc Huy and Tran Van Hung
2 Nguyen Minh Ha, Pham Tan Minh
Country and Territory Ratings and
Worldwide Web: https://freedomhouse.org/report/fre edom-world#Data
Country and Territory Ratings and
Statuses, 1973-2023 Worldwide Web: https://freedomhouse.org/report/fre edom-world#Data
2 Le Thi Huong Mai, Dinh Tran Ngoc Huy and Tran Van Hung
1 Amin et al (2014) 2.Castro and Camarillo (2014)
Transparency International Worldwide Web: https://www.transparency.org/en/cp i/
TTPT Worldwide Bank Indicators (WDI)
Data Analysis Method
In this research, STATA software is utilized to execute static regression methods, including Pooled OLS, Fixed Effects (FE), Random Effects (RE), and Generalized Least Squares (GLS) models The analysis aims to compare the results across these four models, employing essential model tests to identify the most suitable model for the data.
Results and Analysis
Descriptive statistics of the study
Figure 4.1: The summary table of data descriptive results
Source: Results from STATA software
Figure above describes characteristics of data including:
Data includes dependent variable which is tax revenue and 15 independent variables namely GDP, Trade Openness (TRADE), Foreign Direct Investment (FDI), Agriculture Value Added (ARG), Manufacturing Value Added (IND), Inflation
This article examines key indicators that influence a country's development and governance, including Official Development Assistance (ODA), Foreign Debt (EXDEBT), and the Political Rights Index (POLRIG) It also highlights the significance of the Civil Liberty Index (CIVLIB), Educational Attainment (SCHTER), Life Expectancy (LIFEXP), and Infant Mortality Rate (INFMOR) in assessing societal well-being Additionally, the Corruption Perception Index (CPI) and the Time to Prepare and Pay Taxes (TTPT) are crucial metrics for evaluating economic transparency and efficiency.
The analysis focuses on six key variables with complete observation data: Tax Revenue, LIFEXP, INFMOR, POLRIG, CIVILB, and SCHTER Other factors are excluded from the analysis due to incomplete data, as the necessary information is not accessible on public websites.
Concenrning the information about mean values, max values and min values, it can be observed that:
Vietnam's tax revenue has an average rate of 13.00438%, peaking at 25.3% in 2010 In contrast, Cambodia recorded its lowest tax revenue in 1991, with only 2.3%.
- GDP growth has the mean value of 5.514434% Singapore’s GDP growth in 2010 was the highest value which was 14.51975% whereas Cambodia’s GDP growth in
1994 was the smallest value and had the negative value of -34.80864%
- TRADE has the mean value of 130.0322% Singapore’s TRADE in 2008 was the highest value which was 437.3267% whereas Myanmar’s TRADE in 2012 was the smallest value which only reached to 11.8554%
- FDI has the mean value of 5.478184% Singapore’s FDI in 2019 was the highest value which was 29.69044% whereas Indonesia’s FDI in 2000 was the smallest value and had a negative value of -2.75744%
- ARG has the mean value of 17.82513% Myanmar’s ARG in 2001 was the highest value which was about 58% whereas Singapore’s ARG in 2018 was the smallest value which only dipped to 0.0301281%
- IND has the mean value of 20.4461% Indonesia’s IND in 2002 was the highest value which was about 32% whereas Laos’s IND in 1991 was the smallest value which was only 5.213625%
- INF has the mean value of 7.083835% Laos’s INF in 1999 was the highest value which was about 126% whereas Vietnam’s INF in 2000 had the negative value of around -1.7%
- ODA has the mean value of 2.696432% Laos’s ODA in 1991 was the highest value which was 17.5198% whereas Thailand’s ODA in 2003 was the smallest value and had a negative value of -0.6426207%
- EXDEBT has the mean value of 68.62152% Myanmar accounted for the highest and smallest EXDEBT value The highest value was 292.8527% in 1991 while the smallest value was about 17% in 2018
- POLRIG has the mean value of 5.086022 points while CIVILB has the mean value of 4.83871 points
- SCHTER has the mean value of 3.671292% Vietnam’s SCHTER from 2000 to
2007 had the maximum value of 15.6% while Myanmar’s SCHTER in 2011 was the smallest value which was only 0.85032%
- LIFEXP has the mean value of about 70 years The highest value belonged to Singapore in 2020 while the smallest one was about 54.46 years belonging to Laos
- INFMOR has the mean value of 31.57 % The country had maximum value of 102.9% was Laos in 1991 while the smallest value was 1.7% belonging to Singapore in 2021
- CPI has the mean value of 18.55502% In 2008, Myanmar had the smallest value of CPI which was only 1.3% wheras in 2012, Singapore had the maximum value of 87%
The average time for tax preparation and payment (TPPT) across ASEAN countries is 293.8421 hours Notably, Vietnam had the highest TPPT, taking approximately 1,050 hours from 2005 to 2009, while Singapore recorded the lowest at just 49 hours.
Research Results
4.2.1 Test of Stationarity for the database
To test the stationarity of data, my research uses a Fisher-type test with a reliable level of 95% My hypothetical pair is:
H0: All panels contain unit root
H1: At least one panel is stationary
With a reliable level of 95%, the variable is stationary when P-value < 0.05 while variable is not stationary when P-value > 0.05
Variable P-Value Stationary/Not Stationary
TTPT 0.9782 Not Stationary d.TRADE 0.0000 Stationary d.IND 0.0000 Stationary d.SCHTER 0.0000 Stationary d.CPI 0.0000 Stationary d.TTPT 0.0000 Stationary
Source: Results from STATA software
Table 4.2.1 presents the findings of the stationary test, revealing that five variables—TRADE, IND, SCHTER, the Corruption Perception Index (CPI), and Time to Prepare and Pay Taxes (TPPT)—are non-stationary To address this issue, my research analyzes the difference values of these variables, which indicates that they become stationary.
4.2.2 Test of Cointegration for database
This research employs the Westerlund test to evaluate the cointegration properties of each variable in the model, with a confidence level set at 95% The focus of the analysis is on a specific hypothetical pair.
With a reliable level of 95%, variable is not cointegrated when P-value > 0.05 while it is cointegrated when P-value < 0.05
Table 4.2.2 presents the findings from the cointegration test, revealing that the variables ARG, LIFEXP, and INFMOR are cointegrated To address this issue, it is essential to remove these variables from the model to ensure its validity and avoid misleading conclusions.
Variable P-Value Cointegrated/Not Cointegrated
GDP 0.2045 Not Cointegrated d.TRADE 0.2363 Not Cointegrated
ARG 0.0105 Cointegrated d.IND 0.2426 Not Cointegrated
CIVILB 0.1282 Not Cointegrated d.SCHTER 0.1437 Not Cointegrated
INFMOR 0.0047 Cointegrated d.CPI 0.3833 Not Cointegrated d.TTPT 0.4360 Not Cointegrated
Source: Results from STATA software
This research focuses on evaluating the impact of 12 key variables on tax revenue in ASEAN countries from 1991 to 2021, following tests for stationarity and cointegration The variables analyzed include GDP, TRADE (its differential), FDI, IND (its differential), INF, ODA, EXDEBT, POLRIG, CIVILB, SCHTER (its differential), CPI (its differential), and the differential value of Time to Prepare and Pay Taxes (TPPT).
Pooled OLS Model is the first model that my research will observe with aims:
- To assess the statistical significance and the impact level of each independent variable on a dependent variable
To evaluate the reliability of the model, I will determine whether to adopt it based on its performance If the model proves inadequate and exhibits significant flaws, I will consider alternative approaches such as the Fixed Effects (FE) or Random Effects (RE) models To make an informed decision on which model to implement, my research will utilize the Hausman Test to compare the FE and RE models.
Figure 4.2.3: Pooled OLS model’s results
Source: Results from STATA software
The pooled OLS model results, illustrated in Figure 4.2.3, demonstrate a trustworthy adjusted R-Squared value of 0.6140 (61.40%) At a 95% confidence level, the analysis identifies seven statistically significant variables—FDI, INF, ODA, EXDEBT, POLRIG, CIVILB, and d.TPPT—that contribute to model development.
TaxRevenue = 0.3839995*FDI + 0.2760503*INF - 0.8488348*ODA + 0.3977445*EXDEBT + 1.020302*CIVILB - 0.310687*d.TPPT + 𝜀 𝑖
This model indicates that Foreign Direct Investment (FDI), Inflation (INF), External Debt (EXDEBT), and Civil Behavior (CIVILB) positively influence tax revenue, while Official Development Assistance (ODA) and tax policy (d.TPPT) have a negative effect Among these, CIVILB and ODA exert the strongest impact on tax revenue, with the other variables showing a lesser influence To evaluate the effectiveness of this model, three specific tests must be conducted.
Figure 4.2.3.1a shows the results of VIF of each variable and we can see that every VIF value is smaller than 10 So, there is no multicollinear phenomenon in this model
Figure 4.2.3.1a: Results of Multicollinearity Test for Pooled OLS Model
Source: Results from STATA software b) Heteroskedasticity or Homoscedasticity Test
My research uses White Test in STATA software to check whether this model’s variance changes or not My hypothetical pair is:
At a 95% confidence level, if the significance value of the variable exceeds 0.05, we accept the null hypothesis (H0), indicating that the model exhibits homoscedasticity Conversely, if the significance value is less than 0.05, we accept the alternative hypothesis (H1), suggesting that the model demonstrates heteroskedasticity.
Figure 4.2.3.1b : Result of White Test for Pooled OLS Model
Source: Results from STATA software
The sig value of this model is 0.3923 which is higher than 0.05 so we can conclude that this model is homoscedasticity c) Autocorrelation Test
To test the autocorrelation of this model, my research uses Wooldridge Test with the hypothetical pair
H0: Model do not have autocorrelation
At a 95% confidence level, if the significance value of the variable exceeds 0.05, we accept the null hypothesis (H0), indicating that the model exhibits no autocorrelation Conversely, if the significance value is below 0.05, we accept the alternative hypothesis (H1), suggesting that the model does exhibit autocorrelation.
Figure 4.2.3.1c: Pooled OLS model - Autocorrelation Test’s Results
Source: Results from STATA software
The result shows that the sig value of this model is 0.0238 which is smaller than 0.05 Thus, we will accept H1 which means pooled OLS model has autocorrelation
After conducting three tests, we found that the pooled OLS model exhibits no multicollinearity and maintains consistent variance over time However, it does display autocorrelation, which may lead to unreliable and biased conclusions when evaluating the impact of the selected independent variables on tax revenue.
To address this issue, my research examines two models: the Fixed Effects (FE) Model and the Random Effects (RE) Model Subsequently, I will apply the Hausman Test to determine which model is more appropriate for my data Additionally, I will conduct three tests to ensure that the selected model is both meaningful and reliable.
4.2.3.2 FE Model and RE Model a) FE Model
The FE Model results, illustrated in Figure 4.2.3.2a, demonstrate a reliable statistical significance at a 95% confidence level, employing the t-student test to evaluate each variable The P-value column indicates that six variables—FDI, d.IND, INF, ODA, EXDEBT, and d.SCHTER—are statistically significant, as their P-values are all below 0.05.
Tax Revenue = 0.2903758*FDI - 0.5183338*d.IND + 0.1185292*INF - 0.7632534*ODA + 0.1187394*EXDEBT + 0.4259362*d.SCHTER + 𝜀 𝑖
Figure 4.2.3.2a: Results of FE Model
Source: Results from STATA software
Foreign Direct Investment (FDI), Infrastructure (INF), External Debt (EXDEBT), and d.SCHTER positively influence tax revenue, while d.IND and Official Development Assistance (ODA) exhibit a negative relationship Notably, d.IND, d.SCHTER, and ODA significantly affect tax revenue, indicating that changes in these variables can lead to substantial fluctuations in tax income Although other variables also impact tax revenue, their effects are comparatively smaller than those of d.IND, d.SCHTER, and ODA.
The RE Model, illustrated in Figure 4.2.3.2b, employs the z test to assess the statistical significance of seven key variables: FDI, INF, ODA, EXDEBT, POLRIG, CIVILB, and d.TPPT Both the RE and FE Models identify FDI, INF, ODA, and EXDEBT as influential on tax revenue, though the impact levels differ between the two models Notably, the RE Model introduces three additional variables: POLRIG, CIVILB, and d.TPPT.
Tax Revenue = 0.500094*FDI + 0.3041032*INF - 1.498617*ODA +
The RE model reveals that foreign direct investment (FDI), inflation (INF), and external debt (EXDEBT) positively impact tax revenue, while official development assistance (ODA) has a negative effect Among three new variables, political risk (POLRIG) is positively associated with tax revenue, whereas civil liberties (CIVILB) and tax policy transparency (TPPT) negatively influence it Overall, the RE model indicates that POLRIG, CIVILB, ODA, FDI, and INF significantly affect tax revenue levels.
In contrast, d.TPPT and EXDEBT just have a small impact on tax revenue
General Discussion for ASEAN countries and recommendations for Vietnam 54 5.1 General Discussion for ASEAN countries
CONCLUSION
- Assessment of supportive level of research for taxation theory and practical
REFERENCES
II RESEARCH CONTENT Chapter 1: Literature Review 1.1 An overview of foreign research all over the world
Researching the factors influencing tax collection is essential, as taxes are the primary revenue source for governments worldwide The Covid-19 pandemic and recent political events have intensified the need for studies aimed at improving tax collection policies This section will provide an overview of international research trends regarding the factors affecting tax revenue, the methodologies employed, the findings of previous studies, and a comparison of those results.
Research into the factors influencing tax collection in various countries gained significant traction in the mid-20th century and has continued to evolve in the 21st century, incorporating new perspectives that reinforce earlier findings.
In the mid-20th century, particularly between 1970 and 1980, significant studies emerged, notably Cheliah's (1971) research titled “Trends in Taxation in Developing Countries.” This study analyzed tax data from approximately 27 to 30 developing countries over a 15-year span, utilizing data from 1953 to 1955 and from 1966 to 1968 Additionally, a cross-sectional analysis included tax data from 50 developing countries during 1966 to 1968 The research focused on two key factors: tax ratio and the composition of tax revenue, which directly influenced tax collection activities and government policy adjustments Cheliah employed regression models to illustrate the relationships between various factors, such as GDP, non-export income, and the shares of mining and agriculture in GDP, and their impact on tax ratios, as well as the dynamics between direct and indirect taxes.
The author utilized a regression model to analyze the relationship between marginal tax rates and tax ratios, revealing significant insights into tax collection in developing countries The findings indicated that both mining's contribution to GDP and non-mining exports positively influenced tax revenue, while agriculture had a negative effect Additionally, the research highlighted that higher per capita income in developing nations correlates with increased development and tax-paying capacity, yet no direct link between per capita income and overall tax revenue was established In 1975, Chelliah and colleagues conducted further research titled “Tax Ratios and Tax Effort in Developing Countries” to update previous findings from their 1971 study, focusing on data from 1969 to 1971.
A study involving 47 developing countries utilized regression models to analyze the impact of tax ratio factors and revenue composition on tax collection The findings revealed that the contributions of mining and agriculture to GDP continue to significantly affect tax revenue, consistent with results from a 1971 study Furthermore, the authors noted that trade share per capita income (excluding exports) and exports (excluding mining) do not influence tax revenue.
From 1990 to 2000, there were also two notable studies by Stotsky and WoldeMarian
A study by Stotsky and WoldeMarian (1997) titled "Tax Effort in Sub-Saharan Africa" utilized regression models to analyze the relationship between various economic variables and tax collection in 46 Sub-Saharan African countries from 1990 to 1995 The findings revealed that the share of exports and per capita GDP positively influenced tax revenue, while the share of agriculture and mining negatively affected it Additionally, the study indicated that the share of manufacturing and import density had no significant impact on tax revenue in the region.
A study examining the factors influencing tax revenue in 83 developing countries from 1978 to 1988 built upon Tanzi's earlier findings from 1987, which indicated a positive correlation between tax revenue and per capita income However, in a subsequent 1992 analysis, Tanzi noted that this relationship had weakened and emphasized the significance of macroeconomic stability, debt servicing needs, and the evolving economic structure as more critical determinants of tax revenue His alternative model related tax share in GDP to agriculture share, import share, foreign debt share, and per capita income, revealing that while per capita income had no effect on tax revenue, both import share and foreign debt share positively influenced it.
Between 1970 and 2000, research primarily focused on developing countries, examining the correlation between economic factors—such as import shares, agricultural contributions to GDP, and per capita income—and tax revenue While these studies successfully identified trends in these relationships, they fell short in providing actionable recommendations for improving tax collection policies.
In the 21st century, research has evolved significantly, with a notable study by Gupta and Briancon (2007) titled “Determinants of Tax Revenue Efforts in Developing Countries.” This research utilized a baseline regression model with fixed and random effects to explore the interplay between structural, institutional, and policy factors influencing tax revenue Key structural factors included central government revenue as a percentage of GDP, per capita GDP in PPP, agriculture value added, imports, aid, and debt ratios The policy variables examined encompassed various tax revenues, including those from goods and services, income, profits, and trade, as well as the highest marginal tax rates for individuals and corporations and average tariffs Additionally, institutional factors such as corruption and legal frameworks were analyzed to understand their impact on tax revenue efforts in developing nations.
A study conducted on 105 developing countries from 1998 to 2007 identified key determinants of tax revenue performance, emphasizing the significance of structural factors such as per capita GDP, agricultural share in GDP, trade openness, and foreign aid The research found that corruption adversely impacts tax collection, while political and economic stability also influences tax revenue, albeit in specific contexts Furthermore, it revealed that developing nations relying primarily on tax revenue from goods and services often experience poorer tax performance, whereas those with substantial income, profit, and capital gains taxes achieve better tax collection outcomes This study builds on the foundational work of Chelliah (1971) and Chelliah et al.
Since 1975, research has increasingly focused on institutional factors like corruption, law and order, political stability, economic stability, and government stability A significant study by Ajaz and Ahmad (2005) analyzed the effects of corruption on tax collection in 25 developing countries from 1990 to 2005, revealing that corruption negatively impacted tax revenues, while effective governance improved tax collection In contrast to Gupta and Briancon (2007), Ajaz and Ahmad's findings indicated that various institutional variables also significantly influenced tax revenues From 2010 onwards, numerous studies have explored additional factors affecting tax collection, including globalization and inflation Notably, Baunsgaard and Keen (2010) examined the relationship between globalization and tax revenue across 117 countries over 32 years, concluding that international trade positively affected tax revenues through import taxes.
Tax collection can be enhanced through expanded trade and increased economic credibility and competitiveness A significant study by Profeta and Scabrosetti (2010) examined 39 countries, including 11 in Asia, 19 in Latin America, and 9 in the European Union, from 1990 to 2004, focusing on the impact of institutional, policy, and structural factors on tax revenue The findings revealed that GDP per capita and the debt-to-GDP ratio were not crucial for tax revenues in ASEAN countries, while they were significant for Latin America Additionally, a negative correlation between agricultural GDP share and tax revenue was noted in Latin America, which was absent in Asia New variables studied included economic openness, democratic rights, civil liberties, education levels, the aging population, female labor participation, the underground economy, and population density Economic openness positively influenced tax revenue in Asia and Europe but had a negative effect in Latin America Higher indexes of democratic rights and civil liberties correlated with better tax revenue and efficiency In Latin America, factors like education level and female labor participation positively affected tax revenue, while the aging population negatively impacted tax revenue in Asia A subsequent study by Didoa (2012) on 32 Latin American and Caribbean countries from 1990 to 2009 echoed similar research variables as the earlier study.
The study identified eight new variables, including colonial heritage and internal conflicts, that explain the differences in taxation across various regions It concluded that factors such as economic openness, education levels, female labor force participation, and population density positively influence tax revenue Additionally, it found that higher civil liberties and political stability correlate with increased tax revenue, while political rights showed no positive correlation with tax revenue, contrary to previous findings by Profeta and Scabrosetti (2010) However, the political rights index significantly impacted direct tax revenues but did not affect indirect taxes The research highlighted how colonial heritage and internal conflicts shaped the fiscal policies of the countries studied, revealing differences in tax burdens and compositions Specifically, Caribbean nations, particularly former British and Dutch colonies, exhibited higher tax revenues, attributed to their tax systems' alignment with Western standards, enhanced civil liberties, and reduced conflict.
1.2 An overview of Vietnam research
In Vietnam, several studies have explored the factors influencing tax revenue in ASEAN countries Research by Le Thi Huong Mai, Dinh Tran Ngoc Huy, and Tran Van Hung (2021) analyzed data from 2000 to 2019, utilizing a static linear regression model based on Castro and Ramírez (2014) Their findings indicated that GDP, trade openness, political rights, and civil liberties positively impacted tax revenues Similarly, a study by Nguyen Minh Ha, Pham Tan Minh, and Quan Minh Quoc Binh (2022) examined tax revenue determinants in ASEAN from 2000 to 2016, employing both static and dynamic linear regression models as proposed by Castro and Camarillo (2014) This research highlighted that trade, foreign direct investment (FDI), and manufacturing also significantly affected tax revenue.
In ASEAN countries, EXDEBT positively influenced tax revenue, while CIVLIB, SCHTER, and inflation showed no significant impact Conversely, the only factor that negatively affected tax revenue was Official Development Assistance (ODA).
1.3 Results inherited in the previous studies
APPENDIX
1 Results of Stationary Test for Database
2 Results of Cointegration Test for Database
3 FE model’s Modified Wald Test Result
4 FE model’s Autocorrelation Test Result