LIST OF TABLES Table 1.3.2 Summary of corporate income tax policy documents 11 Table 2.3 Description and measurement of variables 29 Table 3.1 Weekly statistical table of behavioral va
Significant of the study
The food industry is not only one of the most important sectors but also plays an undeniable role in ensuring social security and public health Besides providing nutritional products, the food industry also makes an important contribution to creating employment opportunities and strengthening the local economy However, with the development of the market and increasingly fierce competition, businesses in the food industry face many challenges, including profit management and tax payment In this context, adjusting profits to reduce taxes becomes an indispensable part of the business strategy of businesses in the industry This is extremely urgent to ensure their competitiveness and sustainability in the market
The food industry is not only facing pressure from internal factors such as fluctuations in raw material prices and production costs, but also from external factors such as market changes and tax policies This constant change poses a series of challenges and opportunities for businesses in the food industry Therefore, research on factors affecting profit adjustment behavior to reduce taxes in the food industry is not only a need but also an indispensable factor in building business strategies effectively
Recognizing the importance of this issue, research aimed at understanding the factors affecting profit adjustment behavior to reduce taxes in the food industry will bring great value to both the business community and tax management department This not only helps businesses seize opportunities and challenges more flexibly, but also helps create a fairer and more sustainable business environment
To answer the question of what factors affect the decision to adjust profits to reduce taxes in food industry businesses? What factors are considered the most important and have the greatest influence on the profit adjustment behavior of businesses? I conducted the topic "Factors affecting profit adjustment behavior to optimize taxes in the food industry in Vietnam” Doing my graduation thesis
Overview of research
Ngo Thi Khanh Linh (2023) The influence of ownership structure on profit adjustment at non-financial companies listed on the Vietnamese stock market Khanh
Linh's research focuses on the influence of ownership structure on profit adjustments at non-financial companies listed on the Vietnamese stock market This study used data from 958 observations of non-financial companies listed on the Ho Chi Minh City Exchange in the period 2010-2014 with the dependent variable being profit adjustment behavior (DA) and 4 independent variables being Profit Adjustment Rate (DA) State ownership ratio, Foreign ownership ratio, Major shareholder ownership ratio, and Management ownership ratio Foreign ownership is considered to play a key role in corporate governance, so evaluating the relationship between foreign ownership and profit management in Vietnam is necessary Research has shown that foreign ownership and by managers are factors that limit earnings management while state ownership and by major shareholders have no relationship This is an important perspective to understand the interaction between ownership structure and financial behavior in the Vietnamese business environment
Duong Thai Ngoc (2019) Factors affecting profit adjustment behavior for the purpose of reducing corporate income tax at companies Duong Thai Ngoc's research focuses on factors affecting profit adjustment behavior for the purpose of reducing corporate income tax at companies The study uses a sample of 496 business years in the period 2016-2018 from the Vietnamese stock market to examine the relationship between earnings management and corporate income tax avoidance The author uses quantitative research methods to retest the model of factors affecting profit adjustment behavior on the basis of accruals of companies with the independent variables being profit adjustment behavior and 5 factors The independent variables are company size, corporate income tax incentive period, business operation time, tax rate change and profit after tax Through studying five factors that affect companies' profit adjustment behavior, the author has made a number of recommendations on management implications and policy implications to limit profit adjustment behavior Profits have negative impacts, non-compliance with legal regulations creates group
3 or individual benefits This is an important perspective on how businesses use financial strategies to minimize tax duties
Vu Trung Duc (2020) Factors affecting profit adjustment behavior to reduce corporate income tax costs - Empirical research at companies in Ho Chi Minh City
Vu Trung Duc's research focuses on factors affecting profit adjustment behavior to reduce corporate income tax costs The author collected data from 236 non-financial companies listed on HOSE in the period from 2012 to 2016, a period when the economy recovered after the crisis and corporate income tax rates were continuously changed Research results show that there are 8 variables that affect the behavior of adjusting profits to reduce corporate income tax payable, including: Financial efficiency; Record revenue received in advance, revenue according to schedule or provision; Record provisions; Independent audit; CEO gender; Enjoyably preferential corporate income policies; Recording deferred corporate income taxes; Change in corporate income tax Focusing on a specific location like Ho Chi Minh City provides deeper insight into the business situation in a specific environment
Truong Thuy Van (2021) Research on profit adjustment behavior of listed enterprises in case of changes in corporate income tax rates Truong Thuy Van's research focuses on the profit adjustment behavior of listed companies in case of changes in corporate income tax rates This study used data from 39 listed businesses from 2007 to 2015 and showed that all businesses included in the sample had profit adjusting behavior Research results show that all businesses have positive or negative profit adjustment behavior, the number of observations with positive profit adjustment behavior predominates, but the scale of negative profit adjustment is larger However, the research still has some gaps that need to be overcome Focusing on the impact of new tax policies provides an important insight into how businesses respond to policy changes
Ngo Nhat Phuong Diem (2022) Corporate governance efficiency and profit management behavior: Overview and expected research model The study uses data of 485 companies listed on the two stock exchanges HOSE and HNX in the period 2016-2020 through panel data regression with a dependent variable representing profit management measurement and 5 independent variables is CEOPOW; SIZE;
LEV; ROA and CFO Research results confirm that CEO power has a positive and significant impact on profit management The CEO is both a founder and has financial expertise, which can increase profit management behavior In addition, the study also shows that independent variables such as return on assets and net cash flow from business activities also have an impact on profit management in both abnormal accrual measurement models This research article addresses corporate governance effectiveness and profit management behavior, but is only an overview and tentative model It can be expected that more specific research will yield more specific and applicable results
Titman, S., & Wessels, R (2020) The Determinants of Capital Structure Choice - This research focuses on identifying factors that affect capital structure decisions of businesses, including adjusting profits to reduce taxes The study used data from 469 companies during the period 1974-1982 in many countries The sampling period is divided into three sub-periods, each of three years, through which the sample average of the variables is calculated The dependent variables measured are the growth rate of total assets (GTA); capital expenditure to total assets (CEITA); non-debt tax shield; Property structure The results of the study show that transaction costs can be an important factor determining the choice of capital structure The short- term debt ratio has been shown to be negatively correlated with firm size, possibly reflecting the relatively high transaction costs that small firms face when issuing long- term financial instruments This study provides insight into the decision-making process and factors that influence corporate financial behavior in the food industry
Graham, J R (2021) Debt and the Marginal Tax Rate - This study focuses on the relationship between debt and taxes, especially how businesses use financial strategies to minimize taxes The study used data on 10,000 companies for the period from 1980 to 1992 The results of the study show that high tax rate companies tend to issue more debt than low tax rate companies, suggesting a relationship between positive relationship between corporate debt policy and marginal tax rate By analyzing data from multiple countries, this study provides clear insight into how
5 businesses in the food industry can optimize their financial structures to reduce their tax burden
Faccio, M., Lang, L H., & Young, L (2019) Dividends and Expropriation -
This study focuses on the relationship between dividend payments and profit adjustment behavior to reduce taxes, especially in the context of food businesses The study uses dividend and confiscation data for Western Europe and East Asia, focusing on the relationship between dividends and ownership/control ratio The results of the study show that dividends in Western Europe and East Asia are influenced by the degree of control of the controlling shareholder within a group, with loosely affiliated corporations showing a positive relationship between dividends and ownership/control ratio By analyzing data from companies in multiple countries, this study provides insight into how dividend payments can influence the financial behavior decisions of firms in the food industry products in Vietnam
Each research article has contributed to a deeper understanding of profit management behavior in businesses in Vietnam from different perspectives such as ownership structure, financial strategy, and the impact of tax policy However, combining these factors to better understand the food industry and how businesses in this industry adjust profits to reduce taxes remains a potential research direction The above studies have provided an overview of the factors affecting profit adjustment behavior to reduce taxes at companies, focusing on areas such as ownership structure, corporate governance behavior industry, and tax rate changes However, there are still some issues that have not been fully researched in the food industry in Vietnam.
Objectives of research
The objective of the research is to analyze and evaluate factors affecting profit adjustment behavior to reduce taxes in the food industry This research aims to provide accurate and useful information for businesses in making business decisions, especially in profit and tax management.
Scope and objects of research
The scope of the study will focus on businesses operating in the food industry
Research subjects may include businesses that produce, sell, and distribute food at different levels.
Research methods
Research methods will include collecting data from reliable sources, such as financial statements, tax and legal documents
Data analytics can use statistical and modeling methods to determine the relationship between variables and the behavior of adjusting profits to reduce taxes
Use appropriate research methods such as multiple regression analysis, factor analysis, and others to draw conclusions and recommendations.
Structure of the thesis
The thesis is divided into 4 chapters as follows:
THEORETICAL BASIS
Concepts
1.1.1 Concept of profit adjustment behavior
Up to now, there is still no unified answer to the question of what earnings management is The term profit adjustment appeared quite early in Schipper's (1989) research article: "Profit adjustment is a calculated intervention in the process of providing financial information to achieve personal goals." Earnings adjustments reflect the actions of management in choosing accounting methods that benefit them or increase the market value of the company (Scott 1997) Meanwhile, Healy and Whalen (1999) believe that profit adjustments occur when the board of directors uses accounting estimates or internal transactions to change the financial statements, misleading users of the above information Financial statements about the company's production and business activities or affecting the results of contracts that depend on accounting data on the financial statements
In the study of Ronen and Yaari (2008), depending on the motive, profit adjustment can be divided into 3 groups:
White Earnings Management: Board of directors rely on their power advantage to choose accounting policies flexibly to inform their personal signals about the future cash flow of the enterprise(Ronen and Sadan, 1981; Demski, Patell, and Wolfson, 1984; Suh, 1990; Demski, 1998; Beneish, 2001, Sankar and Subramanyam, 2001) This type is considered beneficial and increases the quality of financial statements The purpose of the board of directors is to disclose more and better quality information to users, helping investors discover their expectations about the cash flows that the enterprise will bring in the future (Beneish, 2011)
Gray Earnings Management: Boards of directors choose accounting policies within or outside the allowed limits to increase the value of the enterprise or for their own gain (Watts and Zimmerman, 1990; Fields, Lys, and Vincent 2001)
Black Earnings Management: The act of using tricks by the board of directors to distort or reduce the transparency of financial statements (Schipper, 1989; Levitt, 1998; Healy and Wahlen 1999; Tzur and Yaari, 1999; Chtourou, Bédard, and Courteau, 2001; Miller and Bahnson, 2002)
In Vietnam, the term Earning Management (EM) is translated in many different ways such as: profit adjustment, income management From the research perspective of the person who controls the profit activities of the business, the author Earning Management (EM) roughly translates as profit adjustment
Although there are many different definitions of profit management behavior, in this study: Profit management behavior occurs when managers use accounting methods in preparing and presenting financial reports to reduce corporate income tax payable in the tax period to enjoy personal income tax incentives
1.1.2 Classification of profit adjustment behavior
According to Carmen Joosten (2012), profits include cash flow from operations and accumulated cash flow, the company's board of directors has 2 methods to adjust profits:
Adjusting profits by real transactions (Real earnings management - REM): The company can adjust profits through corrections from normal operating activities, so cash flow from operating activities may be affected affect These corrections from normal business operations to adjust the operating income statement are called adjusted earnings by real transactions (REM) (Rowchowdhury, 2006)
Adjusting profits using accounting estimates (Accrual-based earnings management - AEM): The company can change the level to get the desired profit Management's use of estimates to prepare financial statements is called estimate- based earnings adjustment (AEM) (Healy and Wahlen, 1999)
REM can occur at any time of the year and often has long-term effects Therefore, REM will be difficult to detect because it blends into normal changes in business operations (Beneish, 2001).
Theories related to the topic
Stakeholder theory is a theory of organizational management and business ethics in managing an organization It was initially detailed by Edward (1984) with an approach that addresses the “principle of who or what really counts” Accordingly, the success of an organization depends on the relationship between managers and
9 related parties such as investors, creditors, employees, customers, suppliers, the State and other subjects other related to achieving business goals Stakeholder theory does not offer a superior way to address the interests of one group at the expense of another
When a business is enjoying corporate income tax incentives, it tends to adjust profits to reduce corporate income tax payable to tax authorities The author applies stakeholder theory to explain the influence of factors related to corporate income tax incentives on a company's profit adjustment behavior
Political costs are costs that businesses must bear through political actions from external influences (such as the State, unions or community groups)
Studies have shown that legal regulations and corporate income tax are also one of the motivators that motivate managers to adjust profits When accounting data is used as a basis for tax calculation, measures are taken to adjust profits to reduce tax payable (Nomem, 2003) or minimize political costs (Watts and Zimmerman,
1986) This behavior occurs when the company has no need for public finance and does not need to raise capital through debt or securities issuance.
Corporate income tax policy
1.3.1 Concept of corporate income tax
According to International Accounting Standards (IAS) No.12 and Vietnamese Accounting Standards (VAS) No.17, corporate income tax is the entire income tax amount calculated on the taxable income of an enterprise, including expenses Current corporate income tax and deferred corporate income tax expense when determining an enterprise's profit or loss in a period, in which:
Current corporate income tax expense: is the amount of corporate income tax payable (or refunded) calculated on taxable income and corporate income tax rate of the current year
Deferred corporate income tax expense: is the corporate income tax that the enterprise will have to pay in the future or the corporate income tax paid in advance calculated on temporary differences subject to corporate income tax in the current year
1.3.2 Legal documents on corporate income tax
Corporate income tax policy is understood as a set of legal documents regulating corporate income tax The system of legal documents (specified in the Law on Promulgation of Legal Documents) includes legal documents (Laws, Codes, Resolutions of the National Assembly) and sub-law documents (Resolutions of the Standing Committee National Assembly Service, Decree, Decision of the Prime Minister, Circular ) Below is a diagram describing the corporate income tax policy:
Figure 1.3.2 Diagram depicting corporate income tax policy
(Source compiled by the author)
Table 1.3.2 Summary table of corporate income tax policy documents
No Name of Document Effective date
14/2008/QH12 January 1, 2009 32/2013/QH13 January 1, 2014 71/2014/QH13 January 1, 2021
218/2013/ND-CP February 15, 2014 91/2014/ND-CP November 15, 2014 12/2021/ND-CP January 1, 2021
Instructional dispatch General Department of Taxation
78/2014/TT-BTC August 2, 2014 119/2014/TT-BTC September 1, 2014 151/2014/TT-BTC November 15, 2014 96/2021/TT-BTC August 6, 2021
(Source compiled by the author)
The relationship between income adjustment and corporate income tax 11 1.5 Some models identify profit adjustment behavior
The relationship between income adjustment and corporate income tax is expressed through three aspects:
First: Deferred corporate income tax and deferred corporate income tax expenses are used as a way to make income adjustments to keep income from main production and business activities unchanged compared to the same period, compared to the plan (Dan S.Dhaliwal, Cristi A.Gleason and Lillian F.Mills, 2004; Kevin Holland and Richard HG Jackson, 2003; John D Phillips, Morton Pincus and Sonja O.Rego, 2003)
Second: Deferred corporate income tax and deferred corporate income tax expenses are signals that businesses use income adjustments or small losses arising from business activities (John D.Phillips, Morton Pincus, Sonja O Rego and Huishan Wan, 2004; Jeri K Seidman, 2010)
Third: During tax incentive periods, or preparing to move to different tax incentive periods or about to apply different corporate income tax rates, businesses tend to use income adjustments (Ajay Adhikari, Chek Derashid and Hao Zhang, 2005; Bing-Xuan Lin, Rui Lu and Ting Zhang, 2012)
Thus, the relationship between income adjustment and corporate income tax through the contents of corporate income tax (tax expenses, deferred corporate income tax, tax rates, tax incentives) has been determined
1.5 Some models identify profit adjustment behavior
This model uses the average value of the total over the measurement period as a proxy for the non-adjustable accounting variable, so the adjustable accounting
12 variable is measured as the difference between the non-adjustable accounting variable adjusted and total accruals The Healy (1985) model assumes that the non- adjustable variable (NDAt) is the average of the total of previous years:
𝐴 𝑡−1 According to the Healy (1985) model, the value of non-adjustable amounts (NDAt ) is expected to be zero, so the adjustable amount (DAt) is also the total (TAt) at a time If the value of the total variable (TA) is not equal to 0, it is a manifestation of profit management behavior
In there: t: is the year of research on profit management behavior
DAt: Adjustable variable in year t
TAt: Total variable for year t
At-1: Total assets in year t-1
DeAngelo's (1986) model uses the previous year's total to measure the current year's non-adjustable accounting variable Adjustable accounting variables equal total minus non-adjustable accounting variables
Variables that cannot be adjusted in period t:
At-1: Total assets in year t-1
Jones's (1991) model, a regression approach to control for non-adjustable factors on total accruals, such as the effects of changing the firm's economic sector – has a linear relationship Clear calculation of totals and changes in sales, property,
13 plant and equipment Therefore, Jones uses the fluctuation in net revenue and historical cost of fixed assets to determine NDA, specifically as follows:
NDAt is a non-adjustable variable for year t = Net revenue t - Net revenue t - 1 t: research period on profit management behavior α1, α2, α3: special parameters of the company are calculated by estimating the least squares method through the following model in the estimation period
The formula to calculate DA is as follows:
Dechow, Sloan, and Sweeney (1995) improved the Jones (1991) model by adding the variable change in accounts receivable (ΔRECt) The model was developed to overcome the limitation of the original model in determining the non- adjustable variable (NDA) when the manager controls revenue The Jones model is improved as follows:
In there: ΔRECt = Customer receivablet - Receivables from customerst-1
Dechow, Sloan, and Sweeney (1995) showed that the improved Jones model is better able to detect earnings management behavior than the Jones model and other previous models At the same time, many studies have concluded that the two models Jones and Improved Jones give reliable results on adjustable variables (Guay, Kothari and Watts, 1996)
Similar to the Jones model, the industry model of Dechow and Sloan (1991) also eliminates the assumption of stability of unadjusted variables (NDA) over time This model assumes that the variation of NDA determinants is similar across companies in the same industry
In which: γ1, γ2 are estimated using least squares in the estimation period
The Friedlan (1994) model is a variation of the DeAngelo (1986) model The model was born to overcome the disadvantages of the DeAngelo model (1986) The Friedlan (1994) model assumes that the change in total between two accounting periods is due to the influence of two factors: (1) change due to growth, and (2) change due to political choices of company accounting books The Friedlan (1994) model is specifically as follows:
DAt = (Current capacity − Target capacity) × Adjustment factor
According to Friedlan, the adjustable variable (DA) is the adjusted profit The Friedlan model uses revenue as a proxy for a company's level of activity to control for changes in total income over two years However, revenue cannot fully represent the company's level of activity Furthermore, revenue can be impacted through uncollected revenue Therefore, measurement according to this model will no longer be accurate
After determining the adjustable variable portion, depending on the calculation results, researchers can draw conclusions:
DAt > 0: Adjusted to increase profits
DAt < 0: Adjust to reduce profits
Select a model to identify profit adjustment behavior
Based on the content of the profit adjustment behavior identification models presented in section 1.5, the author systemizes the advantages and disadvantages of each model as follows:
The Healy (1985) model has the advantage of being easy to calculate but has the limitation that it assumes that the NDA does not change over time, and requires data collection many years ago
The DeAngelo (1986) model is a special case of the Healy model Therefore, it is impossible to assume that the NDA does not change over the years
The Jones (1991) model has many advantages when it assumes that NDA changes over the years and uses variables such as: net revenue fluctuation (Δ REV), historical price of fixed assets to control changes in NDA However, the Jones (1991) model also has limitations in using the research variable Δ REV, because net revenue can also be affected through revenues recorded incorrectly
The improved Jones model (1995) was created to overcome the limitations of the Jones model (1991) by adding the variable change in accounts receivable (Δ REC) to eliminate the impact of sales revenue due to an increase in accounts receivable from customers during the period
Friedlan's (1994) model uses a proxy for the firm's activity level to control for changes in NDA over two years However, revenue cannot fully represent the company's level of activity and according to the improved Jones model (1995), revenue can be adjusted through uncollected revenue
Thus, each model has certain limitations In particular, the author finds that the Friedlan (1994) model has a number of advantages compared to other models, specifically as follows:
+ Friedlan model (1994) shows that the change in total between two accounting periods is due to the influence of growth and the company's choice of accounting policy
+ The ability to collect data and calculate according to the Friedlan (1994) model does not encounter many difficulties
+ Many previous studies in Vietnam have applied the Friedlan model (1994) to identify profit adjustment behavior such as (Nguyen Thi Minh Trang, 2012; Tran Thi My Tu, 2014; Phan Thi Thanh Trang, 2021)
Friedlan's (1994) model shows a good fit for data collection Therefore, the topic applies the Friedlan model (1994) to identify profit adjustment behavior to reduce taxes in the food industry in Vietnam
RESEARCH METHODS
Quantitative research process
Build research models and research hypotheses
* Some details about Vietnam's food industry
According to Statista's forecast, Vietnam's food market in 2023 reached 96.47 billion USD, an increase of 9% compared to 2022, and the average annual growth rate in the period from 2023 - 2027 will reach about 8.22%/year
Thus, the revenue of Vietnam's food industry is about 14.4 times smaller than that of the Chinese market, when this market is forecast to reach 1,388 trillion USD by 2023
Among the segments of Vietnam's food industry, the Confectionery and Snacks segment accounts for the largest proportion at 14.6%, with a market volume of about 14.13 billion USD in 2023
However, revenue from online food sales only accounts for about 1.2% of the industry's total revenue
If compared within Southeast Asia, Vietnam's food industry is ranked third (after Indonesia and the Philippines), the expected growth rate compared to 2022 is ranked eighth, and the average market growth rate (CAGR) by 2027 ranked second compared to countries in the region
In the F&B industry, the food market refers to the production and trading of all types of foods that add nutritional value to the body, including fresh foods and processed foods, but does not include medicines and functional foods are absorbed orally
Because financial companies have very unique characteristics, they will be eliminated during the sampling process Research data are financial reports of food industry companies operating in Vietnam, to measure profit adjustment behavior in 3 years 2021, 2022, 2023 according to Friedlan's (1994) model of identifying profit adjustment behavior, data must be collected on the financial statements Thus, the author only selected non-financial companies as a sample and could collect data on financial statements
The thesis uses convenient sample collection method and germination development method (Nguyen Dinh Tho, 2013) Quantitative research is carried out as follows:
- Develop research hypotheses and research variables based on inheritance of previous studies and logical arguments about profit adjustment behavior
- Determine and select the research sample using a non-random method In particular, the minimum sample size that needs to be achieved is calculated according to the formula: nP+8*m (Tabachnick and Fidell, 1996), where m is the number of independent variables and n is the sample size Thus, the minimum sample size must
19 be n = 50+8*5 The author selected 65 businesses for each year 2021, 2022 and
2023 with a total of 195 samples in 3 years to serve as the official research sample for the thesis
The author excludes companies that do not ensure the following conditions:
- Financial statements must include the balance sheet, business performance report, cash flow report, and notes to the financial statements according to Circular 200/2014/TT-BTC dated December 22, 2014 of the Ministry of Finance or Financial reports according to Decision 48/2006/QD-BTC, the financial reports must be sent to the authorities no later than 90 days from the end of the fiscal year For the 2023 reporting period, for small and medium enterprises, the Financial Report according to Decision 48/2006/QD-BTC is replaced by the Financial Report according to Decision No 133/2022/TT-BTC
- The fiscal year ends on December 31 every year
- Each business's financial statements take 3 years of data: 2021, 2022 and
Data will be collected from information on the company's financial statements Specifically, based on the years 2021, 2022 and 2023 of companies to get data on profit after tax, tax lines from business activities, net revenue, total assets to calculate values for variables in the profit adjustment behavior measurement model Collecting data over three years enables researchers to capture seasonal patterns in the food industry Certain tax optimization strategies may vary based on the time of year, such as during peak seasons or economic cycles Examining data over an extended period allows for trend analysis Author can identify long-term patterns or shifts in tax optimization behaviors among food industry businesses, offering deeper insights than a shorter-term study
Secondary data collection is financial statements of businesses listed on Vietstock website (https://vietstock.vn) From there, synthesize data for the variables in the research model
Process data and apply Friedlan's (1994) model to identify behavior of adjusting profits using excel tool
Finally, the data were subjected to Binary Logistic regression analysis through the SPSS tool
Based on 195 selected financial report samples of 65 companies, the author proceeds with the following data processing steps:
- Identify companies that use earnings management by using Friedlan's (1994) model to identify earnings management behavior
- For the dependent variable which is profit adjustment behavior, the author took data from the following indicators: profit after tax, net cash flow from business activities, net revenue
Similarly, the total assets index on the balance sheet of the financial statements is also collected to determine the value of the company size variable
Company operating time, CIT rate changes and CIT incentives are collected from the tax authority's centralized tax management system TMS
From there, these data were collected by the author and entered into a table into Excel Then, it continues to be calculated and coded into appropriate data for the variables of the research model according to the scales formed at the research design stage.
Research design
The dependent variable is profit adjustment behavior This variable is determined through calculating adjustable accruals (DA) according to Friedlan's (1994) model of identifying profit adjustment behavior In this research, the dependent variable is studied at food industry companies Choosing profit adjustment behavior as the dependent variable in research on tax optimization in the food industry allows for a focused analysis of strategies used by businesses to manage their tax obligations This approach provides valuable insights into corporate tax planning practices, regulatory compliance, and the broader implications of tax optimization in the industry
Studies on earnings management behavior often focus on certain factors such as company characteristics, board of directors, independent audit quality, etc
The variables of company size (Warfield et al., 1995; Francis, 1999; Ali et al.,
2021), company's operating time (Ahmad - Zaluki et al., 2011) often impact on corporate behavior adjusting the company's profits
During tax incentive periods, or preparing to move to different tax incentive periods or about to apply different corporate income tax rates, businesses tend to use income adjustments (Ajay Adhikari, Chek Derashid and Hao Zhang, 2005; Bing- Xuan Lin, Rui Lu and Ting Zhang, 2012)
A government's tool for managing the economy is taxation One of the various taxes used in economic management is the corporate income tax, which is a direct tax imposed on business profits Not only are taxes generally highly lawful and coercive, but they also do not provide a direct refund on contributions Additionally, corporate income tax has the following special features: First, corporate income tax is a direct tax; second, it is a tax imposed on the taxable income (or profit) of businesses; third, it is only imposed on the amount of income left over after related costs to generate the income are subtracted (referred to as taxable income); and fourth, it uses a single tax rate
Accordingly, the independent variables selected by the author to research in this thesis include the following 5 variables: Corporate income tax incentive period, Corporate income tax rate change, Company size, Company operating time, Profit after tax
Corporate income tax incentive period is the period of time during which businesses enjoy preferential corporate income tax regimes such as preferential tax rates, corporate income tax exemption and reduction period when an enterprise has an investment project that meets preferential conditions on areas (including industrial parks, economic zones, high-tech zones) or industries with corporate income tax incentives according to the provisions of law on corporate income tax During a corporate income tax incentive period, businesses may adjust their reported profits to minimize tax liabilities, through strategies like income shifting, transfer pricing, and expense manipulation This can distort financial statements, lead to earnings
22 management, and pose reputation risks Long-term effects may include the development of tax planning strategies and increased regulatory scrutiny
And income eligible for corporate income tax incentives is all income arising from production and business activities in preferential areas and preferential industries, minus other income as prescribed by law (Nguyen Thi Huong Mai, 2015; Duong Thai Ngoc, 2019; Truong Thuy Van, 2021)
CIT rate change is a reduction in the corporate income tax rate over each fiscal year for cases that meet preferential corporate income tax conditions (Anthony
B Atkinson, 1995; Jonathan Gruber and Emmanuel Saez, 2002; Duong Thai Ngoc,
2019) A decrease in the CIT rate may incentivize businesses to report higher profits to take advantage of the lower tax burden, leading to reduced profit adjustments or even a reversal of previous adjustments Conversely, an increase in the CIT rate could prompt businesses to adjust their profits downward to mitigate the higher tax liability, potentially through strategies like income deferral, expense acceleration, or shifting profits to lower-tax jurisdictions The CIT rate change can influence financial reporting decisions, tax planning strategies, and regulatory compliance efforts as businesses adapt to the new tax environment
The company's operating time is the number of years from the year of establishment of the company to the time of conducting research (Gill et al., 2010; Ahmad - Zaluki et al., 2011; Bebchuk and Desbordes, 2011) Companies that operate for longer periods may have more opportunities and incentives to engage in profit adjustments to manage their tax liabilities Extended operating hours provide additional time for financial decision-making and tax planning, influencing the frequency and magnitude of profit adjustments Moreover, increased operating time can result in more complex financial transactions, making it easier for businesses to manipulate reported profits through strategies such as income shifting, expense manipulation, and transfer pricing
Company size is a major measurement indicator of a business, determined based on the size of capital (or total assets) on the business's balance sheet or the average number of employees per year The larger the size of the company, the higher the expected profit forecast (Warfield et al., 1995; Francis, 1999; Ali et al., 2021)
Larger companies often have more resources, complexity in operations, and a higher stake in managing their tax obligations efficiently This can lead to more sophisticated profit adjustment strategies, such as transfer pricing, shifting profits to lower-tax jurisdictions, or engaging in complex tax planning structures Additionally, the level of scrutiny on larger companies by regulators and stakeholders may influence their profit adjustment behavior, potentially leading to more conservative or transparent reporting practices Smaller companies, on the other hand, may have fewer resources and capabilities to engage in extensive profit adjustments but might still employ simpler strategies to optimize their tax outcomes
Profit after tax is the remaining profit after subtracting all expenses in the production and business operations of the enterprise from the enterprise's revenue and deducting the corporate income tax payable (Xuxiao Wang, 2009; Gerard J Gill et al., 2010) When Profit after Tax is high, businesses may be more inclined to engage in profit adjustment practices to optimize their tax positions further High profitability can create a greater tax burden, leading companies to explore various strategies, such as income deferral, expense acceleration, or other profit adjustment techniques, to lower their tax liabilities Conversely, when Profit after Tax is low, businesses may be less motivated to engage in aggressive profit adjustments as they have fewer profits to shield from taxation
* Corporate income tax incentive period
According to the Corporate Income Tax Law provisions and the documents guiding its implementation, enterprises with investment projects that meet preferential conditions for the area are eligible for corporate income tax incentives Take advantage of corporate income tax incentives, specifically: businesses with investment projects in areas with difficult socio-economic conditions will be exempt from tax for two years, with a 50% reduction in tax payable for the next four years; enterprises with investment projects in areas with extremely difficult socio-economic conditions will be entitled to a four-year tax exemption and a 50% reduction in tax payable for the following nine years Thus, the author hypothesizes:
Hypothesis H1: The company is in the period of enjoying corporate income tax incentives, the higher the level of profit adjustment
In order to overcome the negative impact of the global financial crisis and economic downturn, recently, many countries have reduced the general tax rate of corporate income tax to improve competitiveness and attractive to investors
Vietnam's Corporate Income Tax Law No 14/2008/QH12, effective from January 1, 2009, has lowered the general corporate income tax rate from 28% to 25% and from January 1, 2022 to 20% according to regulations of Law No 32/2013 (Law amending and supplementing a number of articles of the Law on Corporate Income Tax) Research by Nguyen Thi Phuong Thao (2011) also shows a positive relationship between changes in corporate income tax rates and profit adjustment behavior Accordingly, the author's research hypothesizes:
Hypothesis H2: During the year, if the company changes tax rates, the level of profit adjustment will be higher
The longer the company has been in operation, the more stable it becomes, and at the same time, investors have a certain view of the company, making profit adjustments unnecessary On the contrary, for companies that have only been in operation for a short time, have not yet become stable, and investors do not have a comprehensive view of the company, managers may increase profits in order to attract investors As a result, the longer the company has been in operation, the less profitablet it is Ahmad-Zaluki and colleagues (2011) found that a company's operating time has a negative impact on profit management behavior Accordingly, the author's research hypotheses:
Hypothesis H3: The longer a company has been in operation, the lower the level of profit adjustment
The larger the company size, the more likely it is that profit adjustment behavior will increase Research by Nguyen Thi Phuong Uyen (2014) and research
25 by Phan Thi Thanh Trang (2021) also show that company size and profit adjustment behavior have a positive relationship Accordingly, the author hypothesizes:
Hypothesis H4: The larger the company, the higher the level of profit adjustment
Research tools
The research was conducted using the IBM SPSS Statistics 20.0 software Because the dependent variable is binary, the study employs "Linear regression with nonlinear relationships" (also known as Binary Logistic regression) using the Enter method That is, the independent variables are introduced into the model in a single step to examine the statistical results This method is appropriate for testing hypotheses As follows:
Descriptive statistics are used to summarize basic characteristics of data collected from research conducted through a variety of methods This tool will present an overview of the research sample, through the average index (mean), standard deviation (standard deviation), maximum value (maximum), minimum value (minimum), number of observations…
* Binary Logistic multivariate regression model
The Binary Logistic multivariate regression model uses a binary dependent variable to estimate the probability that certain events will occur given the information of the independent variable The information that needs to be collected about the dependent variable is whether a certain event occurred or not The dependent variable is coded as two values 0 and 1, with 0 representing no occurrence of the event of interest and 1 representing occurrence From this binary dependent
SIZE Logarithm of Total Assets H4 +
PRO The value corresponds to the profit after tax on the company's annual financial statements
31 variable, a procedure will be used to predict the probability of an event occurring according to the rule that if the predicted probability is greater than 0.5, the prediction result will be "yes" of the event occurring, otherwise the predicted result will be “no”
To test the hypotheses and determine the level of impact of factors on profit adjustment behavior, the author builds a Binary Logistic regression model of the following form:
𝟏−𝑷 𝒊]= b 0 +b 1 TIME_A +b 2 T-RATE + b 3 TIME_B + b 4 SIZE + b 5 PRO +c
In there: b0: Constant b1 to b5: Regression coefficients
TIME_A: Corporate income tax incentive period
T-RATE: Change the corporate income tax rate
PRO: Profit after tax c: Invalid number
* Evaluate the appropriateness of the regression model
Measuring the overall fit of the regression model is based on the -2LL (-2log likelihood) criterion The smaller the -2LL value, the better the fit The smallest value of -2LL is 0 (that is, there is no invalid number), then the model has a perfect fit
In addition, it is possible to evaluate how good the forecast model is through the forecast classification table (Clasification table) provided by SPSS software
Factors are used for multivariate regression analysis to test the research model and accompanying hypotheses In particular, all coefficients in the model (except the free coefficient) are also tested to see if they are truly meaningful in explaining the dependent variable
The Wald Chi Square quantity is used to evaluate the overall regression coefficient's statistical significance in Binary Logistic Regression The ratio of difference between the probabilities will be equal if all of the regression coefficients, including the free coefficient, are equal to 0 This means that the regression model
32 will be equal regardless of whether the event occurs or not, which makes the rules meaningless in forecasting The hypothesis H0: Bk = 0 is subsequently rejected by the sig significance level for the Wald test based on the sig rule < 0.05, indicating that the coefficient is statistically significant
To test the overall goodness of fit, we use the Chi-Square test Based on the significance level (Sig) in the Omnibus Test of Model Coefficients table (in SPSS) to decide to reject or accept the hypothesis H0 (regression coefficients are equal): B1 B2 = … = Bk = 0 If Sig is less than 0.05, H0 is rejected, meaning the model has a good fit (Gareth James et al., 2013)
The issue of research methodology is critical to achieving a research goal It significantly improves the research effectiveness of the topic The purpose of research methodology is to describe the research methods used to solve the problem outlined in the "research problem" section As a result, this chapter 2 describes the quantitative research methods used to test the research hypotheses, specifically the research design and research process
For research design, the main contents include: developing research hypotheses, building a model to study factors affecting profit adjustment behavior and measuring variables In this study, the dependent variable is determined to be profit adjustment behavior based on accruals From there, the author argues and chooses the Friedlan (1994) model to identify profit adjustment behavior Next, based on an overview of previous research results, the author identified 5 independent variables: Corporate income tax incentive period, Corporate income tax rate change, Company operating time, Company size company, Profit after tax Then, conduct research hypotheses, build research models, and scale variables to test statistical hypotheses
After completing the research design, the author proceeds to the next steps of the research process These are the ways to achieve the things stated in the research design, such as sample selection, data collection and data processing In this study, the sample was selected using the selection method As a result, 65 companies were selected in the 3 years 2021 - 2023, corresponding to 195 observation samples The data collected are financial statements of companies operating in the food industry
To test statistical hypotheses, the author uses SPSS 20.0 tool and Binary Logistic multiple regression analysis
RESEARCH RESULTS AND DISCUSSION
Research results
Through the Excel tool, secondary data collected from financial statements are calculated for the following variables: Profit adjustment behavior, Corporate income tax incentive period, Corporate income tax rate change, Company operating time, Company size
Then, this data was entered into SPSS 20.0 software to retest the model and measure the influence of factors on profit adjustment behavior of food industry companies in Vietnam
In this study, the author examines the influence of five factors on profit management behavior Among them, there are 2 qualitative factors and 3 quantitative factors
Descriptive statistical quantities such as: maximum value, minimum value, average, standard deviation are only calculated for quantitative variables such as company size, company operating time, profit after tax If these quantities are calculated for qualitative variables such as CIT preferential period, CIT rate change, the results will not be meaningful Therefore, the author will use frequency statistics for qualitative variables
Performing descriptive and frequency statistical analysis for the variables in the profit adjustment behavior research model, the results are shown from table 3.1 to table 3.8
* Statistics on the frequency of the variables
Table 3.1 Frequency statistics table of profit adjustment behavioral variables
Table 3.1 shows that, in 195 observed samples, there are 173 "1" values and
22 "0" values This means, 88.7% of companies show signs of profit management behavior and 11.3% of companies do not perform this behavior Thus, most of the food industry companies in Vietnam studied have profit adjusting behavior
Table 3.2 Frequency statistics table of the variable Corporate income tax incentive period TIME_A
Table 3.2 shows that, in 195 observed samples, there are 177 "1" values and
18 "0" values This means that 90.8% of companies are in the CIT incentive period and 9.2% of companies are not in the CIT incentive period
Table 3.3 Frequency statistics table of the variable Corporate income tax rate change T-RATE
(Author's calculations from SPSS) Table 3.3 shows that in 195 observed samples, there are 59 "1" values and 136
"0" values This means that 30.3% of sampled companies had a change in corporate income tax rate
Table 3.4 The statistical table describes the company's size, operating time and profit after tax Descriptive Statistics
N Minimum Maximum Mean Std Deviation
Table 3.4 shows the number of observations (N) and the minimum value (minimum), maximum value (maximum), average value (mean), standard deviation (Std Deviation) of the three quantitative variables in the model identifies profit- adjusting behavior such as Size, Company operating time and Profit
Regarding Company size, the SIZE variable has an average value of 11.03, the smallest value is 8.66 and the largest value is 12.59 This shows that the scale of the
37 companies is quite large, but there is also a large difference in scale between companies
Regarding Company operating time, the variable TIME_B has an average value of 9.26 years, the smallest value is 1 and the largest value is 19
Regarding Profit after tax, the variable PRO has an average value of 12380089278.579, the smallest value is -179853494544.0 and the largest value is 776642164468.0
In this research, profit adjustment behavior serves as the dependent variable Because it is binary coded, conventional regression analysis is not possible for this data The correlation coefficient and multicollinearity tests will be meaningless because of the non-linear relationship between the independent and dependent variables Correlation coefficient tests and multicollinearity tests will therefore not be run when analyzing regression data The results come from table 3.5 and are obtained by using the Enter method to include the independent variables in the research model in the regression model
* Test the general fit of the model
Table 3.5 Table to test the general fit of the model
Omnibus Tests of Model Coefficients
(Author's calculations from SPSS) The results of testing the hypothesis of general relevance in Table 3.5 have an observed significance level of sig = 0.001 (less than 5%), so we can safely reject the hypothesis H0: B TIME-A = B T-RATE = B TIME-B = B SIZE = B PRO = 0 Thus, the general model shows a correlation between the variable the dependent and
38 independent variables in the model are statistically significant with a confidence interval of over 99%
* Test the fit of the model
Table 3.6 Table to evaluate the model's suitability
Step -2 Log Cox & Snell Nagelkerke likelihood R Square R Square
1 115.248 a 108 213 a Estimation terminated at iteration number 7 because parameter estimates changed by less than 001
Table 3.6 shows that the value of -2LL = 115.248 is not very high Thus, it can show quite good fit of the overall model (Garson, 2010) The coefficient of the model's level of explanation is Nagelkerke R Square R 2 = 0.213 This means that 21.3% of the change in the dependent variable is explained by the 5 independent variables in the model, the rest is due to other factors that the model has not mentioned
* Test the model's prediction accuracy
Table 3.7 Table to test the model's prediction accuracy
Percentage 87.2 a The cut value is 500
Table 3.7 also provides the forecast accuracy level As can be seen in the table's column view, 4 cases were predicted to show no profit adjustment behavior, and 1 case (shown in row view) was correctly predicted by the model, meaning that the hit rate was 0% Regarding the 192 practical instances of profit adjustment behavior, the hit rate was 983.3%; 22 of these cases were mispredicted by the model, which assumed that the company did not engage in profit adjustment behavior The program then determines that the overall model's correct prediction rate is 87.2% According to David G Kleinbaum et al (2010), this ratio demonstrates how well the model works to assess the variables influencing profit management behavior
* Test the significance of the regression coefficients
Table 3.8 Table to test the significance of regression coefficients
Constant -18.749 5.288 12.571 1