Ownership Structure and Real Earning Management of Food and Beverage Companies in Vietnam: The Moderating Role of Auditing Quality

1. Introduction

The separation between ownership and control in joint stock companies creates agency conflicts that may motivate managers to engage in earnings management practices to pursue personal interests at the expense of shareholders. Among different forms of earnings manipulation, real earnings management (REM) has attracted increasing attention because it is conducted through actual business activities and is therefore more difficult for auditors and regulators to detect than accrual-based earnings management. Managers may manipulate earnings through abnormal production costs, discretionary expenses, or sales activities in order to meet earnings targets or market expectations. Such practices can adversely affect firm value and reduce the reliability of financial reporting.

Ownership structure is considered one of the most important corporate governance mechanisms influencing managerial behavior and financial reporting quality. Different ownership types, including state ownership, managerial ownership, foreign ownership, and institutional ownership, may influence managers’ incentives and monitoring effectiveness in different ways. On the one hand, foreign ownership or institutional ownership can strengthen monitoring and reduce opportunistic behavior. On the other hand, excessive managerial ownership or state control may create entrenchment effects that encourage managers to manipulate earnings for private benefits. Prior empirical studies have also documented mixed findings regarding the relationship between ownership structure and earnings management. Specifically, while many studies found that state ownership, managerial ownership and foreign ownership are negatively associated with firms’ earnings management (Ben-Nasr et al., 2015; Dong et al., 2020; Guo et al., 2015; Alzoubi, 2016; Al-Begali et al., 2024; Han et al. (2022), Gu et al., 2023; Saleh and Mansour, 2024; Warfield et al., 1995; Teshima and Shuto, 2008), several studies provide evidence to the contrary (Nguyen et al., 2020; Nguyen et al., 2021; Tran and Dang, 2021; Abubakar et al., 2020; Mellado and Saona, 2020). These findings suggest that the effectiveness of ownership mechanisms may depend on institutional contexts and other governance factors.

Audit quality is another important governance mechanism that can constrain managerial opportunism and improve the credibility of financial statements. High-quality auditors are expected to enhance monitoring effectiveness, strengthen compliance with accounting standards, and reduce managers’ ability to engage in earnings manipulation. Some empirical studies have reported that higher audit quality leads to lower levels of earnings management (Al-Duais et al., 2022; Yasmin et al., 2024; and Amara et al., 2025). Moreover, based on agency theory, audit quality can moderate the relationship between ownership structure and REM by either reinforcing or substituting the monitoring function of different ownership structures, depending on the institutional context and the relative strength of internal versus external governance mechanisms. Therefore, examining the moderating role of audit quality in the relationship between ownership structure and REM is necessary to provide a more comprehensive understanding of corporate governance effectiveness.

In a transitional economy such as Vietnam, earnings management is particularly important because the institutional and corporate governance systems are still developing, creating greater opportunities for managers to manipulate financial reporting. In addition, the rapid growth of the Vietnamese capital market, combined with relatively weak investor protection, information asymmetry, and inconsistent enforcement mechanisms, increases the risk of opportunistic managerial behavior. Since reliable financial information is essential for attracting investment and supporting sustainable market development, understanding earnings management in Vietnam is important for improving transparency, strengthening corporate governance, and enhancing investor confidence. Especially, the food and beverage (F&B) industry plays a significant role in the Vietnamese economy because of its contribution to economic growth, employment creation, and consumer demand. Companies in this sector often face intense market competition and pressure to maintain stable financial performance, which may increase incentives for real earnings manipulation. Despite the growing importance of this industry, empirical evidence on the effect of ownership structure on REM and the extent to which audit quality moderates this relationship within Vietnamese F&B companies remains limited.

This study is devoted to filling this gap in the literature by investigating the effect of ownership structure on REM and particularly examining whether audit quality moderates this relationship among F&B companies listed on the Vietnam stock market. The study contributes to the existing literature in several ways. First, while previous studies have mainly focused on accrual-based earnings management, this study focuses on REM, which has received relatively less attention in the context of Vietnam. By examining REM, the study provides deeper insights into managers’ operational manipulation activities that may not be fully captured by traditional accrual measures. Second, this study contributes to the existing literature by providing the first empirical evidence on the effect of ownership structure on REM in the F&B sector, a strategically important industry that plays a significant role in the Vietnamese economy. Compared with other industries, F&B companies are characterized by intense market competition, high consumer demand, and pressure to maintain stable financial performance, which may create stronger incentives for managers to engage in REM practices. Third, the study enriches the auditing and corporate governance literature by investigating the moderating role of audit quality in the relationship between ownership structure and REM. This approach provides additional evidence on whether high-quality auditing can strengthen or weaken the effectiveness of ownership monitoring mechanisms in constraining managerial opportunistic behavior. Finally, this study provides practical implications for regulators and investors. The findings may assist regulators in improving corporate governance policies and auditing regulations aimed at enhancing financial reporting quality in the Vietnamese stock market. In addition, investors may benefit from a better understanding of how ownership characteristics and audit quality influence firms’ earnings management behavior, thereby supporting more informed investment decisions.

This paper is structured into five sections. Following the introduction, Section 2 provides the theoretical framework and a review of the empirical literature. Section 3 describes the data and methodology. Section 4 reports the findings and discussion, and Section 5 concludes the study.

2. Theoretical Background and Empirical Literature Review

2.1. Theoretical Background

Agency theory developed by Jensen and Meckling (1976) serves as a theoretical foundation for explaining the effect of ownership structure on REM of firms. Generally, agency theory explains the relationship between principals (shareholders) and agents (managers), emphasizing that conflicts arise when managers do not fully act in the best interests of shareholders. These conflicts stem from divergent interests, information asymmetry, and the difficulty of effectively monitoring managerial actions, which may lead to opportunistic behaviors such as inefficient decision-making or earnings manipulation. To mitigate these issues, firms implement governance mechanisms, such as managerial ownership, performance-based incentives, and external monitoring through boards and audits, to better align managerial interests with those of shareholders.

From an agency theory perspective, ownership structure plays a critical role in shaping REM among Vietnamese listed firms in general, and listed F&B companies in particular, where corporate governance mechanisms remain in the process of development. In this context, higher managerial ownership tends to reduce REM by aligning managers’ interests with long-term firm value, thereby discouraging opportunistic real activities manipulation. Similarly, foreign ownership is likely to be associated with lower REM, as foreign investors typically demand greater transparency, possess superior monitoring capabilities, and exert stronger governance pressure on management. In the context of Vietnam, the political and institutional characteristics of state ownership can weaken the monitoring role. Many listed firms in Vietnam originated as state-owned enterprises, and state representatives are often involved in the board of directors and the board of supervisors. The indirect and bureaucratic nature of state ownership, often exercised through government agencies, can dilute monitoring effectiveness, allowing managers greater discretion to engage in REM.

Additionally, on the basis of agency theory, audit quality generally constrains REM by enhancing external monitoring, reducing information asymmetry, and increasing the cost of managerial opportunism. Moreover, audit quality moderates the relationship between ownership and REM by either reinforcing or substituting the monitoring function of different ownership structures, depending on the institutional context and the relative strength of internal versus external governance mechanisms. On the one hand, high audit quality can strengthen the governance role of ownership structure. For example, in firms with foreign, institutional, or concentrated ownership, the presence of a high-quality auditor enhances transparency and complements internal monitoring, thereby further constraining REM. In this case, audit quality and ownership structure function as complementary governance mechanisms, jointly reducing agency problems and limiting managerial opportunism. On the other hand, agency theory also suggests a potential substitution effect. When audit quality is high, external monitoring becomes more effective, which may reduce the marginal monitoring role of certain ownership types. As a result, the incremental impact of these ownership structures on REM may weaken.

2.2. Empirical Literature Review

A growing body of empirical literature investigates the effect of ownership structure on real earnings management (REM), with evidence drawn from both developed and emerging markets. Ownership structure, typically captured through state ownership, foreign ownership, and managerial ownership, influences the intensity of monitoring, the alignment of interests, and ultimately managers’ incentives to engage in real activities manipulation. Empirical studies on earnings management have provided diverse evidence across different institutional settings, particularly regarding how internal governance structures and external monitoring interact to influence real activities manipulation.

• State ownership and REM

Empirical evidence on the effect of state ownership on REM of firms remains inconclusive. Several studies concluded that state ownership is positively associated with REM of firms. Using a sample of 350 privatized firms from 45 countries, Ben-Nasr et al. (2015) found that state ownership is negatively associated with earnings quality. This finding suggests that the state has stronger incentives to report lower earnings quality in order to conceal the potential tunneling of corporate resources for political purposes. In the context of Vietnam, Nguyen et al. (2020) provides evidence to confirm that state ownership is positively associated with earnings management among listed real estate companies in Vietnam. Similarly, using a sample of 489 companies listed on the Vietnamese stock market during the period from 2009 to 2018, Nguyen et al. (2021) documented that a higher level of state ownership is associated with greater earnings management. Contrary to the first category, some studies confirms that state ownership serves as a constraint on real earnings management. Cheng et al. (2015) examined earnings management of firms around the initial public offerings (IPO) in Chinese stock markets and found that state-owned enterprises engage in less earnings management than non-state-owned enterprises in the period surrounding IPOs. Additionally, Dong et al. (2020) reported that Chinese listed firms with state control are less likely to engage in REM. Similarly, evidence provided by Tran and Dang (2021) suggests that state ownership reduces earnings management among listed firms in Vietnam. Furthermore, Wang et al. (2023) emphasize that state ownership can significantly moderate the opportunistic behaviors of other shareholder groups, suggesting a complex governance role where the state serves as both a source of pressure and a stabilizing force. Based on the existing literature and the specific context of Vietnam, we propose the following hypothesis:

H1: State ownership is positively associated with REM.

• Foreign ownership and REM of firms.

The monitoring role of foreign investors has been widely documented in the literature. Foreign investors are often considered sophisticated stakeholders with stronger monitoring capabilities, better access to information, and greater demand for transparency. Empirical evidence suggests that foreign ownership reduces earnings management practices by enhancing external oversight and corporate governance mechanisms. Specifically, Guo et al. (2015) investigated whether foreign investors significantly influence REM among Japanese firms. This study provides evidence that foreign ownership effectively curtails REM. Similarly, Ben-Nasr et al. (2015) found that foreign ownership reduces firms’ earnings management. The same evidence was also found by Alzoubi (2016) and Al-Begali et al. (2024) for companies listed on the Amman Stock Exchange. Abubakar et al. (2020) explored the effect of ownership structure and REM using a sample of 72 Nigerian firms over the period from 2014 to 2018. The findings reveal that foreign ownership constrains managers from manipulating the financial statements. In addition, Han et al. (2022) determined the effect of foreign ownership on earnings management among Chinese firms. The study identifies foreign investors as a key mechanism for reducing earnings management. In Malaysia, Al-Duais et al. (2022) provide evidence that foreign and institutional ownership significantly alleviate REM practices by demanding higher transparency. Moreover, using a sample of 13,860 firms across 41 economies from 2000 to 2017, Gu et al. (2023) found that foreign institutional investors is negatively associated with REM. In another analysis, Saleh and Mansour (2024) documented a negative association between foreign ownership and earnings management among Palestinian listed companies. In the context of Vietnam, Nguyen et al. (2021) reported that foreign ownership is negatively associated with earnings management. However, Tran and Dang (2021) documented that an increase in foreign ownership results in an increase in level of earnings management. Based on agency theory and the reviewed empirical evidence, we propose the second hypothesis as follows:

H2: An increase in foreign ownership results in a derease in level of REM.

• Management ownership and REM.

The effect of managerial ownership on earnings management is commonly examined within the agency theory framework, highlighting both alignment and entrenchment effects. Empirical studies generally support the alignment hypothesis, suggesting that higher managerial ownership is associated with lower levels of earnings management. When managers hold equity stakes, their interests are more closely aligned with those of shareholders, reducing incentives to manipulate earnings for short-term gains. As an early study in this field, Warfield et al. (1995) demonstrated that managerial ownership is positively associated with earnings quality. Additionally, Teshima and Shuto (2008) developed a theoretical model demonstrating the effect of managerial ownership on earnings management and concluded that higher managerial ownership is associated with lower levels of earnings management. Using a sample of Japanese firms, their empirical findings support the theoretical predictions. Similarly, analyzing a sample of 62 companies from the Amman Stock Exchange, Alzoubi (2016) found that managerial ownership effectively mitigate earnings management. Moreover, Dong et al. (2020) documented that managerial ownership plays a key role in constraining REM in Chinese firms. In Vietnam, Nguyen et al. (2021) also identified managerial ownership as a significant deterrent to earnings management. In contrast, some studies provided evidence supporting the entrenchment hypothesis, indicating that managerial ownership is positively associated with earnings management. Specifically, evidence from Abubakar et al. (2020) indicates that greater managerial ownership leads to higher levels of earnings management among Nigerian firms. Similarly, Mellado and Saona (2020) demonstrated that higher insider ownership encourages managers to engage more intensively in REM. In light of the theoretical framework and empirical findings discussed above, the following hypothesis is developed:

H3: Management ownership constrains REM.

• The moderating effect of auditing quality on the relationship between ownership structure and REM

Prior studies have generally documented that audit quality plays an important monitoring role in constraining earnings management of firms. Specifically, Becker et al. (1998) examined the effect of auditing quality, proxied by a dummy variable distinguishing clients of Big Six auditors from clients of non-Big Six auditors, on earnings management. This study found that clients of Big Six auditors report lower income-decreasing discretionary accruals than clients of non-Big Six auditors. In the context of Malaysia, Al-Duais et al. (2022) documented that audit quality, measured by BIG4 auditors, is negatively associated with REM, suggesting that Big4 audit companies possess greater expertise and are more effective in constraining REM. Similarly, using a sample of 4,723 Indonesian firms from 2020 to 2023, Yasmin et al. (2024) found that firms audited by Big 4 auditors engage in less earnings management than the remaining firms. Additionally, Prayogi et al. (2022) demonstrated that superior audit quality curtails opportunistic reporting in Indonesian listed manufacturing firms. More recently, Amara et al. (2025) explored the effect of audit quality and the proportion of women on an audit committee on earnings management among UK listed companies. The empirical findings indicate that higher audit quality, measured by audit fees, leads to lower levels of earnings management.

Furthermore, several empirical studies provided evidence on the moderating role of audit quality in the relationship between ownership structure and earnings management. Most prior studies have documented that audit quality serves as a substitute for internal ownership monitoring, thereby weakening the marginal constraining effect of ownership structure on earnings management. Specifically, Gul et al. (2002) found that high quality auditing, proxied by Big 6 auditors, weakens the association between managerial ownership and discretionary accruals among Australian firms, suggesting that audit quality mitigates managers’ incentives to manipulate earnings. Similarly, Kouaib and Jarboui (2014) investigated the moderating role of auditing quality on the relationship between ownership structure and earnings management in Tunisian firms and asserted that the moderating effect is negatively associated with earnings management. This evidence implies that audit quality can substitute internal governance mechanisms in constraining earnings management. Recently, Naburgi et al. (2026) investigated the moderating role of audit quality in the relationship between board attributes and earnings management among Nigerian industrial goods firms. The study documented that higher audit quality is directly associated with lower levels of earnings management. However, the interaction effects between audit quality and board attributes on earnings management are significantly positive, suggesting that audit quality does not enhance the ability of corporate governance mechanisms to constrain earnings manipulation. In other words, the findings support the substitution perspective rather than the complementarity perspective between audit quality and corporate governance in constraining earnings management. From a different perspective in the Vietnamese context, Le (2025) argued that the effect of audit quality on REM is moderated by a firm’s financial health. Specifically, for firms with low financial leverage and high operating cash flows, audit quality is negatively associated with REM. Conversely, for firms under financial strain, reflected by high financial leverage and poor cash flow, higher audit quality results in lower levels of REM. Based on agency theory and the existing empirical literature, the following hypothesis is proposed for this study:

H4: The constraining effect of ownership structure on REM is weaker in firms with higher audit quality (substitution effect).

3. Data and Research Methodology

3.1. Data

This study utilizes a sample of 138 food and beverage (F&B) companies listed on Vietnam stock market during the period from 2022 to 2024. The study focuses on the 2022-2024 period because it represents the transition from the COVID-19 crisis to the post-pandemic recovery of Vietnam’s economy. During this period, as business operations and consumer demand gradually normalized, companies operated in a more stable environment than during the pandemic years. Consequently, the 2022-2024 period provides an appropriate setting to examine the governance role of ownership structure in shaping real earnings management among Vietnamese F&B companies. The data were obtained from the annual financial reports of the companies. All data were collected from the website of HOSE (www.hsx.vn) and HNX (www.hnx.vn).

3.2. Research Methodology

3.2.1. Measurement of REM

This study employs the model developed by Roychowdhury (2006) to measure REM, which has been widely adopted in the earnings management literature (Cohen and Zarowin, 2010; Dong et al., 2020; Al-Duais et al., 2022; Al-Begali et al., 2024; Le, 2025). According to Roychowdhury (2006), REM is proxied by three measures, abnormal discretionary expenses, abnormal cash flows from operations, and abnormal production costs, defined as the deviations of actual values from their normal levels that are measured by the residuals from regression models. Specifically, these proxies are specified as follows.

• Abnormal discretionary expenses (ADISEXP)

According to Roychowdhury (2006), discretionary expenses (DISEXP) are measured as the aggregate of advertising, research and development, and selling and administrative expenses. To estimate normal discretionary expenses, this study employs the following cross-sectional regression, estimated by year:

$${\displaystyle \frac{DISEX{P}_{it}}{T{A}_{i,t-1}}}={\beta }_0+{\beta }_1\left({\displaystyle \frac{1}{T{A}_{i,t-1}}}\right)+{\beta }_2\left({\displaystyle \frac{Sal{e}_{i,t-1}}{T{A}_{i,t-1}}}\right)+{\epsilon }_{it}$$

where:

• DISEXP_it: Discretionary expenses of company i in year t;
• TA_i,t-1: Total assets of company i at the end of year t–1;
• Sale_i,t-1: Net sales of company i at the end of year t–1.

ADISEXP is defined as the residual obtained from the regression model. A negative ADISEXP indicates that managers reduce discretionary spending below normal levels, which increases current-period earnings.

-

Abnormal cash-flow from operations (ACFO)

Roychowdhury (2006) argues that managers can manage earnings by accelerating sales (e.g., price discounts, lenient credit terms). Such actions increase current sales but reduce operating cash flows, leading to abnormally low cash-flow from operations (CFO) relative to sales. Normal CFO is estimated using the following cross-sectional regression:

$${\displaystyle \frac{CF{O}_{it}}{T{A}_{i,t-1}}}={\beta }_0+{\beta }_1\left({\displaystyle \frac{1}{T{A}_{i,t-1}}}\right)+{\beta }_2\left({\displaystyle \frac{Sal{e}_{i,t-1}}{T{A}_{i,t-1}}}\right)+{\beta }_3\left({\displaystyle \frac{\Delta Sal{e}_{it}}{T{A}_{i,t-1}}}\right)+{\epsilon }_{it}$$

where:

• CFO_it: Cash-flow from operations of company i in year t;
• TA_i,t-1: Total assets of company i at the end of year t–1;
• Sale_it: Net sales of company i at the end of year t;
• Sale_i,t-1: Net sales of company i at the end of year t–1;

$\Delta Sal{e}_{it}=Sal{e}_{it}-Sal{e}_{i,t-1}$: Change in net sales from year t-1 to year t.

ACFO is defined as the residual derived from the regression model. A negative ACFO indicates abnormally low operating cash flows relative to sales, consistent with income-increasing REM through sales manipulation.

-

Abnormal production costs (APROC)

According to Roychowdhury (2006), managers may strategically overproduce to reduce reported cost of goods sold per unit by distributing fixed overhead across a higher level of production. While this strategy increases inventory and total production costs, it artificially inflates current-period earnings. As a result, firms that overproduce exhibit abnormally high production costs relative to sales. Using the model proposed by Roychowdhury (2006), normal production costs are estimated using the following cross-sectional regression:

$${\displaystyle \frac{PRO{C}_{it}}{T{A}_{i,t-1}}}={\beta }_0+{\beta }_1\left({\displaystyle \frac{1}{T{A}_{i,t-1}}}\right)+{\beta }_2\left({\displaystyle \frac{Sal{e}_{i,t-1}}{T{A}_{i,t-1}}}\right)+{\beta }_3\left({\displaystyle \frac{\Delta Sal{e}_{it}}{T{A}_{i,t-1}}}\right)+{\beta }_4\left({\displaystyle \frac{\Delta Sal{e}_{i,t-1}}{T{A}_{i,t-1}}}\right)+{\epsilon }_{it}$$

where:

• PROC_it: Production costs of company i in year t;
• TA_i,t-1: Total assets of company i at the end of year t–1;

$\Delta Sal{e}_{it}=Sal{e}_{it}-Sal{e}_{i,t-1}$: Change in net sales from year t-1 to year t;

$\Delta Sal{e}_{i,t-1}=Sal{e}_{i,t-1}-Sal{e}_{i,t-2}$: Change in net sales from year t-2 to year t-1;

APROC is measured as the residual from the regression model. A positive APROC indicates overproduction, reflecting income-increasing real earnings management.

Following Roychowdhury (2006), a composite measure of income-increasing REM is constructed by combining three individual REM proxies that capture distinct managerial actions. Because income-increasing REM is associated with lower ACFO and ADISEXP, but higher APROD, the signs of ACFO and ADISEXP are reversed. Specifically, the composite indicator of income-increasing REM is calculated as follows:

$$RE{M}_{it}=APRO{C}_{it}-ADISEX{P}_{it}-ACF{O}_{it}$$

Higher values of the composite REM measure indicate a greater degree of income-increasing REM. By capturing multiple REM channels simultaneously, the index provides a more comprehensive measure than any single proxy.

3.2.2. Research Models

To examine the impact of ownership structure on earnings management of food and beverage (F&B) companies listed on the Vietnamese stock market, this study first employs both fixed-effects (FEM) and random-effects (REM) models. The Hausman test results are then used to identify the most appropriate model for the study. It is important to note that the estimated models may be subject to heteroskedasticity and autocorrelation. When these issues are detected, the feasible generalized least squares (FGLS) estimator is employed as the final step to correct for such violations. Accordingly, the regression models are specified as follows:

$$\begin{array}{l}R{M}_{it}=\alpha +{\beta }_{1}S{O}_{i,t}+{\beta }_{2}F{O}_{i,t}+{\beta }_{3}M{O}_{i,t}+{\beta }_{4}BIG{4}_{it}+{\beta }_{5}S{O}_{it}*BIG{4}_{it}+{\beta }_{6}SIZ{E}_{i,t}+\\ {\beta }_{7}LEV{}_{it}+{\beta }_{8}RO{A}_{i,t}+{\beta }_{9}R{G}_{i,t}+{\epsilon }_{it}\end{array}$$

(1)

$$\begin{array}{l}R{M}_{it}=\alpha +{\beta }_{1}S{O}_{i,t}+{\beta }_{2}F{O}_{i,t}+{\beta }_{3}M{O}_{i,t}+{\beta }_{4}BIG{4}_{it}+{\beta }_{5}F{O}_{it}*BIG{4}_{it}+{\beta }_{6}SIZ{E}_{i,t}+\\ {\beta }_{7}LEV{}_{it}+{\beta }_{8}RO{A}_{i,t}+{\beta }_{9}R{G}_{i,t}+{\epsilon }_{it}\end{array}$$

(2)

$$\begin{array}{l}R{M}_{it}=\alpha +{\beta }_{1}S{O}_{i,t}+{\beta }_{2}F{O}_{i,t}+{\beta }_{3}M{O}_{i,t}+{\beta }_{4}BIG{4}_{it}+{\beta }_{5}M{O}_{it}*BIG{4}_{it}+{\beta }_{6}SIZ{E}_{i,t}+\\ {\beta }_{7}LEV{}_{it}+{\beta }_{8}RO{A}_{i,t}+{\beta }_{9}R{G}_{i,t}+{\epsilon }_{it}\end{array}$$

(3)

where:

• RM is real earnings management of a company;
• Ownership structure (SO, FO, MO) and other control variables (BIG4, SIZE, LEV, ROA, RG) are defined and presented in Table 1.

4. Empirical Results

4.1. Descriptive Statistics of the Sample

Based on the data of 138 F&B listed companies during the period from 2022 to 2024 (on 414 firm-year observations), the descriptive statistics of variables employed in the models are computed and reported in Table 2.

It is shown that the mean of real earnings management (RM) is -0.08 indicating that, on average, firms in the sample tend to engage in income-decreasing real earnings management. In other words, instead of inflating reported earnings, firms slightly reduce their reported performance through real activities manipulation. However, the wide range from -1.28 to 1.44 and a standard deviation of 0.32 suggest substantial variation in earnings management practices across firms.

Regarding ownership structure, Table 2 shows that State ownership (SO) has a relatively high mean of 20.32%, ranging from 0% to 97.78%. These statistics indicate that some firms remain heavily state-controlled while others have no state participation. In addition, Foreign ownership (FO) has a relatively low mean of 4.99%, indicating that foreign investors hold only a small proportion of shares in F&B listed companies. Despite the low average, the wide range of FO (from 0% to 95.56%) indicates that while many companies have little or no foreign participation, a few firms attract significant interest from international investors. Moreover, Table 2 reports that the mean of management ownership (MO) is 3.03%, indicating that executives and directors hold a relatively small equity stake in F&B listed companies. This low level of ownership suggests a separation between ownership and control, where managers are primarily hired professionals rather than significant shareholders.

In terms of firm characteristics, Table 2 reveals that firm size (SIZE), measured as the natural logarithm of total assets, has a mean of 27.41 (equivalent to 4,414.37 billion VND) with a moderate standard deviation of 1.76 (14,191.86 billion VND). Additionally, the average financial leverage (LEV) of companies in the sample is 0.44, indicating a moderate use of leverage. The wide range of LEV, spanning from 0.02 to 0.98, suggests considerable variation across firms. Moreover, statistics presented in Table 2 reveals that ROA has considerable variation across firms in the sample, with a mean of 3.72% but a wide range from -36.87% to 60.47%. Finally, revenue growth (RG) has an average of 5.43%, but also displays substantial dispersion, ranging from -99.88% to 149.41%, highlighting the volatility in sales growth among firms in the industry.

4.2. Regression Results

As presented above, this study initially performs both REM and FEM. The results of these models are reported in Table 3 and Table 4. The results of Hausman test confirm that the REM is more appropriate than FEM for the study. In addition, the results of Wooldridge test for autocorrelation and the ARCH test for heteroscedasticity presented in Table 5 indicate that heteroskedasticity and autocorrelation among the residuals exist in the model.

To address the heteroskedasticity and autocorrelation that exist in the model, the Feasible Generalized Least Squares (FGLS) approach is finally used in this study. The estimated results of the FGLS approach are reported in Table 5. As expected, the estimated results presented in Table 5 confirm that foreign ownership (FO) is negatively associated with real earnings management (REM) of listed F&B companies at the 1 percent level of significance. These findings suggest that foreign investors play an effective monitoring and governance role that constrains managerial opportunism. This evidence aligns with a substantial body of literature (e.g., Guo et al., 2015; Ben-Nasr et al., 2015; Alzoubi, 2016; Abubakar et al., 2020; Han et al., 2022; Al-Duais et al., 2022; Gu et al., 2023; Nguyen et al., 2021; Al-Begali et al., 2024), but contradicts the evidence provided by Tran and Dang (2021). The mitigating effect of foreign ownership on REM observed in this study can be attributed to several reasons as follows. First, Foreign shareholders are often more experienced and sophisticated, with stronger incentives to protect their investments and reputations. As a result, they tend to demand higher standards of transparency and financial reporting quality, which discourages managers from engaging in REM practices such as overproduction, excessive price discounts, or cuts in discretionary expenditures to manipulate earnings. Second, foreign investors typically bring better corporate governance practices, including stricter oversight, improved internal controls, and greater alignment between management and shareholder interests. Their presence can enhance board effectiveness and increase scrutiny over managerial decisions, thereby limiting the scope for real activities manipulation. Finally, firms with higher foreign ownership are often subject to greater attention from international markets and analysts, which increases the likelihood of detection and the potential reputational costs associated with earnings management.

Additionally, the coefficient of MO in Models 1 and 3 is negative and statistically significant at the 1 percent level, indicating that higher managerial ownership reduces REM in listed F&B companies. This finding is in line with previous findings of Warfield et al. (1995), Teshima and Shuto (2008), Alzoubi (2016), Dong et al. (2020), and Nguyen et al. (2021) and supports the alignment-of-interests view in corporate governance which argues that when managers own a larger proportion of the firm’s equity, their interests become more closely aligned with those of shareholders. As a result, managers are less likely to engage in REM practices, such as cutting discretionary expenses, overproduction, or sales manipulation, because these actions may harm the firm’s long-term value and, consequently, their own wealth. Especially, in the context of Vietnam, where legal enforcement, disclosure quality, and investor protection have improved but remain not yet fully robust, increasing managerial stakes can reduce opportunistic behavior because self-monitoring becomes more effective than relying solely on external oversight, thereby improving the quality of financial reporting. Moreover, the effect of MO on REM can be even more pronounced in the context of listed F&B companies. These firms often rely heavily on brand reputation, product quality, and stable customer relationships. Engaging in REM, such as cutting marketing or quality-related expenses, could damage brand equity and future sales. Therefore, manager-owners, with a stronger commitment to the firm’s long-term success, are less likely to compromise these fundamentals for short-term reporting gains.

Moreover, the results presented in Table 5 indicate that audit quality, as proxied by BIG 4 auditors, is negatively and significantly associated with REM, suggesting that firms audited by the BIG4 (Deloitte, PricewaterhouseCoopers, Ernst & Young, and KPMG) are less likely to engage in opportunistic real activities manipulation. This evidence is consistent with empirical findings of Becker et al. (1998), Al-Duais et al. (2022), Prayogi et al. (2022), Yasmin et al. (2024), and supports the view that high-quality auditors serve as an effective external governance mechanism by enhancing monitoring and constraining managerial discretion. Although REM is implemented through operational decisions such as overproduction or reductions in discretionary expenses, these actions generate abnormal patterns in financial statements that are more likely to be detected by auditors with superior expertise and rigorous audit procedures (Kim and Park, 2014, Choi et al., 2018). Besides, BIG 4 auditors possess substantial reputational capital and face greater litigation risk, which incentivizes them to enforce stricter financial reporting standards and challenge aggressive managerial behavior (Craswell et al., 1995; Choi et al., 2008). The presence of such auditors also creates a deterrence effect, discouraging managers from engaging in REM prior to its occurrence (Francis and Yu, 2009). This role is particularly important in the context of Vietnam, where institutional enforcement and investor protection mechanisms, while improving, remain relatively underdeveloped.

Especially, the interaction coefficient between FO and BIG4 in Model (2) is positive and significant at the 1% level, indicating that audit quality moderates the monitoring role of foreign investors in reducing real earnings management. According to agency theory (Jensen and Meckling, 1976), multiple governance mechanisms may either complement or substitute for one another in mitigating managerial opportunism. In this study, while both FO and BIG4 individually constrain REM, the positive interaction suggests that the joint monitoring effect of foreign investors and high-quality auditors is weaker than the sum of their individual effects, indicating a substitution relationship. This finding aligns with the evidence reported by Gul et al. (2002), Kouaib and Jarboui (2014), and Naburgi et al. (2026). While foreign investors are generally associated with effective monitoring and reduced REM, the marginal effectiveness appears to decline in the presence of Big 4 auditors who already provide rigorous external oversight. Additionally, foreign shareholders may rely more on the credibility of Big 4 and engage less in direct monitoring, further weakening their influence on REM.

Regarding firm-specific control variables, the results derived from the FGLS model reveal that SIZE, financial leverage (LEV) and returns on total assets (ROA) are determinants of REM of the listed F&B companies. Specifically, the findings indicate that SIZE is significantly and positively associated with REM, suggesting that larger firms are more likely to engage in real activities manipulation. This finding can be attributed to several reasons as follows. First, large firms possess greater operational flexibility, allowing managers to adjust production levels, sales timing, and discretionary expenditures more easily. At the same time, they face stronger pressure to meet market expectations given their higher visibility to investors and analysts, which increases incentives to manage earnings. In addition, the complexity of large organizations can obscure such practices, making REM more difficult for external stakeholders to detect. Moreover, large firms typically have more resources to absorb the potential costs associated with real activities manipulation. In contrast, LEV exhibits a negative relationship with REM, suggesting that firms with higher levels of debt tend to engage in lower levels of real earnings manipulation. Highly leveraged firms are often subject to stricter supervision by lenders, who closely monitor financial performance and impose restrictive debt covenants. This enhanced scrutiny can limit managerial opportunistic behavior and reduce managers’ ability to manipulate real operating activities. From this perspective, leverage acts as a governance mechanism that disciplines managers and lowers the level of REM. Similarly, ROA is negatively associated with REM, aligning with the findings of Abad et al. (2018) and Al-Duais et al. (2022), and Al-Begali et al. (2024). This finding indicates that firms with stronger financial performance have less incentive to manipulate real activities, as they are better able to meet earnings targets without resorting to opportunistic behavior.

5. Conclusions

This study is devoted to investigate the effects of ownership structure on earning management of listed F&B companies in Vietnam, with a particular focus on the moderating role of audit quality. Using FGLS approach on panel data from 138 listed F&B companies during the period from 2022 to 2024, the findings indicate that foreign ownership and managerial ownership significantly constrain REM practices, whereas state ownership does not exhibit a statistically significant effect. In addition, firms audited by Big 4 auditors tend to engage less in REM, confirming the important governance role of high-quality external auditing. Especially, the positive moderating effect of BIG 4 on the relationship between FO and REM indicates that the joint monitoring effect of foreign investors and high-quality auditors is weaker than the sum of their individual effects. This evidence supports the view that governance mechanisms can act as substitutes rather than complements in constraining REM, contributing to the broader literature on corporate governance in emerging markets.

These findings offer several practical implications for regulators, managers, and investors in Vietnam. First, for regulators, given that foreign ownership significantly reduces REM, policies should continue to encourage foreign institutional participation to enhance market transparency and corporate accountability. Second, for managers, Second, increasing managerial ownership can be an effective internal governance mechanism. When managers hold equity stakes in the firm, their interests become more aligned with long-term firm value rather than short-term financial reporting targets. Companies should design compensation policies that encourage equity-based incentives, thereby reducing the likelihood of value-destroying earnings manipulation. Additionally, the role of audit quality highlights the importance of engaging reputable external auditors. Firms audited by Big 4 auditors exhibit lower levels of REM, suggesting that high-quality audits enhance financial reporting credibility. Therefore, managers should prioritize auditor reputation and independence when selecting audit firms, rather than focusing solely on cost considerations. Finally, for investors, the study provides useful insights for evaluating the financial reporting quality of listed F&B companies. Investors should pay greater attention to ownership characteristics and audit quality when making investment decisions, as these factors may influence the likelihood of earnings management. In particular, companies with higher foreign ownership, higher managerial ownership, and engagement with Big 4 auditors may be perceived as having more reliable financial reporting and lower risks of opportunistic earnings manipulation.

Although this study has enriched the literature on the effect of ownership structure on REM of firms, particularly by highlighting the moderating role of audit quality, it still has several limitations that should be addressed in future studies. First, although the study adopts the widely used Roychowdhury (2006) model, REM may still suffer from measurement error and may not perfectly reflect all forms of real earnings manipulation. Second, while the FGLS approach helps address heteroskedasticity and autocorrelation, potential endogeneity issues may still persist in the model. These limitations could serve as valuable directions for future research.