Abstract
Purpose
This study aims to see the moderating effect of board diversity on the relationship between ownership structure and real earnings management.
Design/methodology/approach
This study uses unbalanced panel data of 75 listed energy firms (346 firm-year observations) from three South Asian emerging economies (Bangladesh, India, and Pakistan) from 2015 to 2019. The two-step system GMM estimation is used for data analysis. This study also uses fixed effect regression to obtain robust findings.
Findings
The findings show that firms with a greater ownership concentration and managerial ownership significantly reduce real earnings management. In contrast, the data refute the idea that institutional and foreign ownership affect real earnings management. We also find that board diversity interacts significantly with ownership concentration and managerial ownership, meaning that board diversity moderates the negative link of the primary relationship that reduces real earnings management. On the other hand, board diversity has no interaction with institutional and foreign ownership, implying no moderating effect exists on the primary relationship.
Originality/value
To the best of the authors’ knowledge, this is unique research investigating how different ownership structures affect real earnings management in the emerging nations’ energy sector, which the earlier studies overlook. More specifically, this research focuses on how board diversity moderates the relationships between ownership structure and real earnings management, which could be helpful for future investors.
Keywords
Citation
Akter, A., Wan Yusoff, W.F. and Abdul-Hamid, M.A. (2024), "The moderating role of board diversity on the relationship between ownership structure and real earnings management", Asian Journal of Accounting Research, Vol. 9 No. 2, pp. 98-115. https://doi.org/10.1108/AJAR-10-2022-0307
Publisher
:Emerald Publishing Limited
Copyright © 2024, Aklima Akter, Wan Fadzilah Wan Yusoff and Mohamad Ali Abdul-Hamid
License
Published in Asian Journal of Accounting Research. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode
1. Introduction
Earnings management (EM) is a financial reporting method that projects a good financial and economic picture. But, when earnings management obscures investors’ fair judgments, the consequence is terrible since it may reduce the quality of profit information in financial statements. After several accounting scandals in the past, including Waste Management, Enron Corporation, WorldCom, Tyco International (Corporate Finance Institute, n.d.), etc., multiple revisions to the Corporate Governance Code have been made around the world throughout time, and the authorities have implemented numerous accounting regulations, policies, and laws that may have little or no effect. For instance, in recent years, there have been scandals involving Cronos Group Inc. (2019–2022), a cannabis business with headquarters in Toronto (US Securities and Exchange Commission, 2022); Luckin Coffee Shop in 2019 (BBC News, 2020); and Wirecard Company, based in Germany in 2020 (Browne, 2020), etc. all of the scandals and corporate failures are linked to earnings management and inadequate corporate governance standards that can erode the level of trust of investors and affect the economic condition. Nasir et al. (2019) argue that companies that are proven to have committed fraud from financial statements suffer several adverse effects, including backlash from the public, a loss of current and potential clients, falling stock prices, and strict regulatory scrutiny. Thus, earnings management has become a global problem, and it is pertinent nowadays.
Over the last two decades, firms have shifted their earning management practices from accrual earnings management (AEM) to Real earnings management (REM) because of changed accounting and tax legislation, the implementation of IFRS, higher accounting standards, and better audit quality (Cohen et al., 2008). Li et al. (2021) also find that the REM method is more practical in assessing whether firms engage in earnings management. Moreover, the majority of REM research has been carried out in developed nations like the USA and Europe (Cohen and Zarowin, 2010), while little research has been carried out in developing or emerging nations. Thus, there is a growing interest in researching emerging markets on REM.
Earlier literature finds that corporate governance mechanisms (CGMs) and their impact on reducing earnings management have been studied extensively (El Diri et al., 2020). Binaebi (2020) opines that ownership structure (OWNS) as a CGMs is crucial for corporate accounting behavior that substantially impacts EM. In this sense, emerging or developing nations play a crucial role because the majority of firms in these countries are governed mainly by shareholders with majority shareholdings (Masud et al., 2018). Also, the market environments of such countries are more involved in earnings management practices (Fatima et al., 2020). Furthermore, unlike other countries, for the high transparency of the standards, the listed firms need to disclose the ownership of the firms (Mishra, 2021); thus, these countries are the ideal place for monitoring the role of ownership structure in alleviating opportunistic managerial behavior that is detrimental for investors. However, the majority of previous research focused on the effects of ownership structure on AEM (Chen et al., 2020; Saona et al., 2020); only a few researchers have explored its connection to REM (AL-Duais et al., 2022). Thus, this research addresses that gap by examining ownership structure’s impacts on REM. So, the primary research question of this study is: What is the effect of ownership structure on REM?
Earlier literature indicates that the board of directors liaises between capital owners and managers to maximize shareholders’ wealth (Githaiga et al., 2022). According to the agency perspective, diversified boards with an acceptable proportion of female directors and board independence improve earnings quality (Githaiga et al., 2022). Furthermore, the participation of foreign directors can also aid in reducing earnings management by enhancing the board of directors’ independence (Kouaib and Almulhim, 2019). Following reading the preceding literature, we are inspired to investigate the second research question: How does board diversity moderate the impact of ownership structure on REM?
In answer to the research questions, the study aims to shed light on two specific research objectives: (1) To examine the effect of ownership structure on REM; (2) To examine moderating effect of board diversity on the relationship between ownership structure and REM. This study uses unbalanced panel data of 75 firms (346 firm-year observations) from three South Asian emerging economies (Bangladesh, India, and Pakistan) to achieve the research objectives.
This study contributes to empirical studies in several ways. Firstly, this is a unique study to look at a wide variety of ownership structures combining concentrated ownership, management, institutional, and foreign ownership with REM from energy sectors of South Asian emerging markets, whereas prior studies have focused on the impact of ownership structure on AEM in developed markets with other financial or non-financial sectors (Chen et al., 2020; Saona et al., 2020). Secondly, this paper examines whether board diversity moderates the relationship between ownership structure and REM, whereas other studies overlooked it. Finally, this work contributes by addressing unobservable heterogeneity and endogeneity resulting from the dynamic nature of earnings management by adopting a dynamic panel model with GMM, which is overlooked in the previous studies of earnings management research (Ashraf and Qian, 2021; Githaiga et al., 2022; Nguyen et al., 2020).
2. Literature review and hypotheses development
2.1 Ownership concentration and earnings management
Globalization has been coupled with market development and dynamism, as well as increased insecurity among large and well-known corporations, and recent worldwide financial scandals have raised concerns about the reliability of financial statements (Salehi et al., 2022a, b). Again, it shows that there are conflicts of interest among owners, creditors, and managers and that when management has easier access to information, it seems that they put their own interests ahead of maximizing shareholder value (Adeneye and Kammoun, 2022). However, the effect of corporate governance mechanisms in ensuring the reliability and integrity of financial statements is incredible. In line with agency theory (Jensen and Meckling, 1976), Saona et al. (2020) claim ownership concentration as a governance mechanism that has a tremendous incentive to settle agency conflict through direct business control, therefore protecting shareholder interests in investment and thus increasing the reliability of the financial statement. Asian countries have concentrated ownership, whereas developed countries (US and UK) have decentralized ownership (Nguyen et al., 2020). When capital is concentrated in the hands of a few large shareholders, they have the authority to supervise their investments and managers’ opportunistic behavior, and vice versa when money is concentrated in the hands of a few small owners (minority) (Nguyen et al., 2020). By following the agency theory, researchers have discovered negative connections between ownership concentration and earnings management, as high ownership concentration lowers EM (Saona et al., 2020). In line with the agency, this study suggests the following hypothesis:
Ownership concentration has a significant negative relation with REM.
2.2 Managerial ownership and earnings management
Earnings management is a technique used by management and insiders to conceal confidential information from stakeholders (Ghorbani and Salehi, 2021). At the same time, higher management ownership is a suitable option for dealing with agency issues since it helps align the interest between owners and managers (alignment effect agency theory). As a result, it anticipates that growing managerial shareholding has a detrimental impact on earnings management, meaning that a ratio of ownership by management reduces earnings management (Piosik and Genge, 2020). Based on these motives and by following agency theory, this study proposes the following hypothesis:
Managerial ownership has a significant negative relation with REM.
2.3 Foreign ownership and earnings management
According to Jensen and Meckling (1976), if insiders’ and outsiders’ interests are well-aligned, firms will strive to maximize firm value by increasing financial openness, which will also help foreign investors. As an appealing avenue for mobilizing money, foreign investment can potentially reduce earnings management (Nguyen et al., 2020). Since Real earnings management fosters investment inefficiency and leads firms to overinvest, AL-Duais et al. (2022) argue that companies with a high level of foreign ownership are very effective in such a situation that monitoring management and limiting REM. Debnath et al. (2021) also provide similar evidence that sophisticated foreign ownership significantly affects REM restrictions. Based on the strong theoretical analysis described above, this study proposes the following hypothesis:
The presence of foreign ownership significantly reduces REM.
2.4 Institutional ownership and earnings management
Global economic competitiveness is fierce due to rapid environmental changes, and competitiveness, which promotes economic success across industries, has garnered attention (Salehi et al., 2022a, b). For sustained economic success, institutional ownership plays an important supervisory function in minimizing agency costs, controlling directors, and boosting present financial performance (Salehi et al., 2022a, b). In line with agency theory, Institutional investors help to supervise and control managers’ opportunistic activities. Several researchers investigated the effect of institutional ownership on EM and discovered that institutional owners’ supervisory function mitigates EM. For example, AL-Duais et al. (2022) found that institutional owners’ efficient monitoring results in an inverse link between EM and institutional share ownership, supported by Piosik and Genge (2020). Further, Saona et al. (2020) discovered that institutional investors’ active monitoring limits corporate managers’ opportunistic reporting behavior. From the above discussion, this study argues that institutional investors desiring stable shareholdings may not allow REM since it would lower the value of their investment. The current study thus proposes the following hypothesis:
The relevant shareholdings of institutional investors decrease REM.
2.5 The moderating effect of board diversity
Our first four hypotheses argue that different forms of ownership structures have a negative impact on earnings management. However, firms also function in setting contractual conflicts of interest between the contractual partners due to their wide range of incentives (Jensen and Meckling, 1976). Managers within the firms often seek their own goals and direct their management to make profit adjustments in opposition to the owner’s potential regarding maximizing their benefits in the company (Dang Ngoc and Tran Manh, 2020). This distinction between control and ownership leads to a possible divergence of interests between the owners and managers that increase agency costs in the firm (Jensen and Meckling, 1976). In this sense, Internal pressures that are typical in practice may additionally impact managers' propensity to engage in REM (Brink et al., 2020). Corporate governance (CG) mechanisms, thus, are allegedly effective in reducing conflict between managers and owners by reducing information asymmetries (Widagdo et al., 2023). In this situation, the board of directors is vital for CG and can effectively alleviate agency issues between owners and management (Fama and Jensen, 1983). In line with the agency theory, Ghaleb et al. (2021) claim that a diverse board of directors in the boardroom protects the interests of various owners by minimizing managers’ opportunistic conduct (for example, earnings management). Similarly, Githaiga et al. (2022) find evidence that board independence and gender diversity have a detrimental and significant impact on EM since board independence is an effective way of handling agency issues (Shafeeq Nimr Al-Maliki et al., 2023). Ashraf and Qian (2021) also noted in another study that foreign directors (i.e. national diversity) improve boards’ ability to supervise management, which reduces the need for corporate executives to manage earnings. Thus, the above discussion strengthens to propose board diversity as a moderator in the association between ownership structure and real earnings management. Accordingly, the following is the testable hypothesis:
Board diversity significantly moderates ownership concentration and REM.
Board diversity significantly moderates managerial ownership and REM.
Board diversity significantly moderates foreign ownership and REM.
Board diversity significantly moderates institutional ownership and REM.
3. Research method
3.1 Sample and data collection procedure
This study chose the listed non-financial firms of the three South Asian emerging economies (i.e. Bangladesh, India, Pakistan) from a total of eight South Asian countries (Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri Lanka). This study restricts its scope to three South Asian developing markets because these countries have a considerable economic impact on the rest of the continent. For example, according to the World Bank Database [1], these three countries account for 95.94% of the region’s GDP and 85.20% of its population. In addition, companies in these countries are governed mainly by shareholders with majority shareholdings (Masud et al., 2018). We purposely selected a sample of the 75 energy firms from the listed non-financial firms from these three countries (i.e. Bangladesh, India, Pakistan) to contribute to this specific area as the energy sector’s healthy development is essential strategically because it is a significant foundation for economic growth (AL-Duais et al., 2022).
The research period includes five years, from 2015 to 2019. The study excludes finance-related companies and other firms with different regulations, disclosure requirements, accounting systems, capital structures, and governance structures compared to non-financial firms (Latif, 2018) to maintain data homogeneity. Companies that lacked financial information, incomplete governance structure data, or annual reports were also excluded. Data for the study have been manually gathered from annual reports of listed energy firms in three SA emerging countries: Bangladesh (listed on the Dhaka Stock Exchange [2], i.e. DSE), India (Nifty 500 index companies [3] listed on the National Stock Exchange [4], i.e. NSE), and Pakistan (listed on the Pakistan Stock Exchange [5], i.e. PSX). The annual reports have been gathered from the websites of the indicated stock exchanges and sample firms. Table 1 shows the sample selection procedure of the energy sectors.
3.2 Measurement of variables
3.2.1 Dependent variable
3.2.1.1 Real earnings management (REM)
To estimate REM, this study employs three proxies, as defined by Roychowdhury’s (2006) models of REM, such as (1) abnormal cash flow from operations (ACFOit), (2) abnormal production costs (APCit), and (3) abnormal discretionary expenses (ADEXPit). The study calculates the proxies of REM using the cross-sectional ordinary least squares (OLS) regression by following Roychowdhury’s (2006) models of REM. The first model indicates the actual cash flow from operations, which is the combination of normal and abnormal cash flow from operations as follows:
The second model indicates the actual production costs, which are the combination of normal and abnormal production costs as follows:
The third model indicates the actual discretionary expenses, which is the combination of normal and abnormal discretionary expenses as follows:
In the three models, CFOit = cash flow from operations of firm i in year t; TASit = total assets of firm i in year t−1; Sit = total sales revenue of firm i in year t; ∆Sit = changes in total sales revenue of firm i in year t; ∆Sit−1 = changes in total sales revenue of firm i in year t−1; PCit = production costs (sum of the cost of goods sold, i.e. COGSit and change in inventories, i.e. ∆INVit. Thus, PCit = COGSit + ∆INVit); DEXPit = discretionary expenses (sum of selling, general, & administrative expenses, i.e. SGAEXPit, advertising expenses, i.e. ADEXPit, and research and development expenses, i.e. RDEXPit. Thus, DEXPit = SGAEXPit + ADEXPit + RDEXPit.); β0, β1, β2, β3, and β4 are the beta coefficients.
The current study aggregates all the dimensions of REM (see in equation-4). Eng et al. (2019) noted that in capturing EM, the aggregate measure for REM is more effective than the individual measure for REM. When a firm engages in income-generating EM by manipulating sales, producing excessive quantities, or reducing discretionary expenses, it reports a low CFO, increased production costs, and low discretionary expenses. Thus, the standard residuals from discretionary expenses and the cash flow from operations are multiplied by −1 and added to the standard residuals of production costs to calculate REM, as shown in equation (4). The higher the value, the more REM is indicated.
3.2.2 Independent and control variables
The measurement of all independent and control variables are shown in Table 2.
3.2.3 Moderating variable
This study uses board diversity (BDIV) as a moderating variable. Gender diversity (GDIV), national diversity (NDIV), and board structural diversity (BSDIV) are used to create an index that measures board diversity, as shown in Table 3.
3.3 Data analysis techniques and model specification
This study employs a dynamic panel model to account for the dynamic nature of earnings management. For example, earnings management in the current year is affected by earnings management in the previous year (El Diri et al., 2020). In this situation, the dynamic panel model (using GMM) will give better results as it can handle endogeneity problems (Majumder et al., 2023; Uddin et al., 2021) arising from the dynamic nature of earning management. However, the ordinary least square (OLS) estimator will not solve the dynamic panel model as OLS ignores individual firms’ heterogeneity or the different endogeneity problems. Again, while the fixed effect can handle the unobservable heterogeneity problem, it does not solve the endogeneity as it follows the strict exogeneity of the variables. This study uses a two-step system GMM to control endogeneity. The dynamic panel model using GMM, as proposed by Blundell and Bond (1998), will give a trustworthy solution to both problems by introducing instrumental variables (El Diri et al., 2020). This study also uses the best panel regression technique (fixed effect) among Pooled OLS, fixed effect, and random effect regression (given in the part of tabulated regression results) for robust outcomes. Brahma et al. (2021) suggest that fixed effect regression is a useful complement to GMM for reliable findings. To test the hypotheses H1 to H4 (direct effect) and H5 (moderating effect), the following dynamic models are developed:
In models 5–6, i indicates firms, t indicates the time (in years), and k indicates countries. REMikt indicates real earnings management; REMikt−1 indicates REM for the year t−1. OWNCONikt = ownership concentration; FAMOWNikt = family ownership; MANOWNikt = managerial ownership; INSOWNikt = institutional ownership; FOROWNikt = foreign ownership; FSIZEikt = firm size; ROAikt = return on assets or firm profitability; LEVikt = firm leverage; FAGEikt = Firm age; TDM = Time dummies; CDM = Country dummies; BDIVikt = Board diversity; all the multiples terms are interactions terms indicating a moderating relationship between dimensions of ownership structure and REM. εikt is the error term, and β1 to β16 shows the beta coefficients.
4. Empirical results and discussion
4.1 Descriptive statistics
Table 4 presents the descriptive analysis of the sample used in this study. The average REM value for all sample enterprises is 0.03, with India exhibiting the highest average earnings manipulation at 0.04, in comparison to Pakistan (0.02) and Bangladesh (0.01). With respect to the independent variables, the average concentration of ownership among the sample firms is 41.78%. Among the sample firms, India exhibits a higher concentration of ownership (62.77%) compared to Bangladesh (25.79%) and Pakistan (30.08%). The data indicates that the mean reported managerial shareholdings of 15.25% suggest that the sample firms own a mere 15.25% ownership attributed to management. Notably, Bangladeshi firms exhibit the largest proportion of managerial ownership (18.99%) when compared to Indian firms (15.56%) and Pakistani firms (12.39%). The institutional ownership percentage in Bangladesh is greater (25.76%) compared to India (19.27%) and Pakistan (10.88%). The sample firms indicate an average institutional ownership of 17.81%. The mean proportion of foreign ownership among the selected enterprises is 4.90%. Pakistan has a higher proportion of foreign ownership (5.90%) compared to both India (4.39%) and Bangladesh (4.21%). In relation to the moderating variable, the average value of board diversity across various countries is 1.24. This value signifies that, on average, boards have experienced a limited level of diversification in terms of gender, nationality, and structure, with India exhibiting the highest mean value of diversified boards (1.39) compared to Pakistan (1.09) and Bangladesh (1.21). Details of descriptive statistics of the control variables are presented in Table 4.
4.2 Correlation analysis
Table 5 shows the Pearson correlation matrix. As a rule of thumb, a correlation coefficient of more than 0.80 indicates a problem with multicollinearity (Gujarati and Porter, 2009, p. 338). The table’s highest correlation coefficient value between variables is 0.581, less than 0.80, indicating no multicollinearity difficulties between variables and hence no threat to the estimated variables.
4.3 Empirical results
The results of the system GMM model and the fixed effect model are summarized in Table 6. Table 5 shows that the coefficient on the one-year lagged REM is statistically positive at 1%. This suggests that the past earnings management of all listed firms (Bangladesh, India, Pakistan) significantly affects the current one. This finding aligns with recent research of El Diri et al. (2020), implying that past earnings management should be considered a key variable in controlling for the dynamic nature of the ownership structure and REM connection. OWNCON is significant and negative at the 1% level in the system GMM and fixed effect model, supporting the hypothesis that ownership concentration balances managers’ and shareholders’ interests and reduces REM usage. Thus, current research recommends H1 in line with recent studies (Alhmood et al., 2023; Saona et al., 2020). The GMM and Fixed effect models reveal that a firm’s higher management ratio prevents manipulation of valid findings. Hence H2 is not rejected and is consistent with earlier research (Al-Haddad and Whittington, 2019; Piosik and Genge, 2020) and supports agency theory. Foreign and institutional ownership have insignificant impacts on REM under the GMM paradigm, thus disproving H3 and H4 as consistent with prior findings (Al-Haddad and Whittington, 2019; Lemma et al., 2018). In contrast, Institutional ownership has a negative link with REM but none with foreign ownership under the fixed effect model. Further, both models show that ROA negatively affects REM, indicating that high-performing firms are less likely to engage in earnings management. FSIZE and REM have a strong positive correlation under both models, indicating that larger firms can control real earnings more effectively than smaller firms. System GMM does not show a significant relationship between REM and LEV, but a fixed effect model shows a meaningful positive relationship. Fixed effect model results show that firms with high leverage levels use REM more frequently to avoid defaulting on debt arrangements. Both models found no correlation between the Firm age and REM.
Table 7 shows a substantial interaction between ownership concentration and board diversity, showing that board diversity moderates the association between ownership concentration and REM, thereby supporting H5a. The interaction between managerial ownership and board diversity is significant and unfavorable in both models, supporting (H5b) the negative link between managerial ownership and REM. As there is no interaction between foreign ownership and board diversity in the GMM system, H5c is not supported. Under the fixed-effect model, however, a significant association is found, which moderates the relationship between foreign ownership and REM. There is no moderating effect between institutional ownership and REM because its interaction does not significantly influence REM; thus, H5d is not supported.
4.4 Robustness checks
4.4.1 Alternative regression method (Fixed effect)
We ran a two-step system GMM since our study models include endogeneity issues, as previously described (Section 3.3). Prior researchers highlight the fixed effect model as the complement of GMM (Brahma et al., 2021; Piosik and Genge, 2020). Thus, the fixed-effect model is also used in this study to confirm the robustness of the findings. The findings are reported in Tables 6 and 7
4.4.2 Alternative measurement of REM
The idea that adding APC cost to ADEXP and ACFO results in duplication because these statistics are obtained from the same activities is the basis for the alternate measures of REM (Cohen and Zarowin, 2010). As a result, we construct REM by combining the three residuals by two: REM1 is the addition of ACFO and ADEXP, and REM2 equals the addition of ADEXP and APC (Cohen and Zarowin, 2010; Ghaleb et al., 2021). Thus, we re-investigated REM1 and REM2 using alternate system GMM and fixed-effect model estimations. The findings of REM1 and REM2 under system GMM and fixed effect models are shown in Table 8.
5. Conclusion
The study examined the relationship between ownership structure, board diversity, and real earnings management in South Asian emerging economies’ energy firms. Two-step GMM estimates demonstrate that the past year’s REM has significantly impacted the current year’s REM. Firms with a high degree of ownership concentration are exceptionally effective at monitoring management and restricting REM are consistent with prior findings (Alhmood et al., 2023; Saona et al., 2020). Managerial ownership, on the other hand, is also effective in restricting REM that is consistent with prior studies (Al-Haddad and Whittington, 2019; Piosik and Genge, 2020). The results, however, do not support the hypotheses H3 and H4, which are consistent with earlier findings (Al-Haddad and Whittington, 2019; Lemma et al., 2018), respectively, that foreign and institutional ownership reduces REM. The moderating effects of board diversity are also investigated. We discover a significant interaction of board diversity with ownership concentration and managerial ownership. However, there is no significant interaction between institutional and foreign ownership with board diversity, implying no moderating effect on the relationship between ownership structure and earnings management. Our findings are particularly robust under different REM measurements and regression models.
5.1 Implications
The findings have implications for policymakers, investors, management and other stakeholders. Our results are consistent with the idea that energy companies with concentrated ownership and management in developing South Asian emerging countries are less likely to manipulate earnings. Policymakers can consider the study’s findings, which show that firms with managerial ownership and concentrated ownership are more driven to self-monitor their operations and refrain from engaging in earnings manipulation. In line with the agency theory view, linking the manager’s interests to the company’s benefit decreases ownership and control. Thus, firms should increase manager-company alignment to reduce managerial self-interest and earnings management. Managers should be given preferential stock options or rewards to encourage company capital ownership. These results may also encourage investors to invest in firms with concentrated ownership and management ownership, as these firms generate earnings that accurately reflect the results of their actual business activities. The management of firms in South Asian emerging economies is recommended to develop and implement policies that will ensure the presence of a diverse board, particularly in terms of gender, nationality, and structural directors on the board because it is linked to a reduction in earnings management. This will make it easier for management and ownership-concentrated firms to direct earnings management measures in a way that will favor market value. This study will inform investors and other stakeholders of energy firms about the role that diverse boards play in decreasing agency conflict between owners and management in concentrated ownership firms by reducing information asymmetry, resulting in a reduction in earnings manipulation.
5.2 Limitation and future research direction
The current investigation exhibits specific constraints that could potentially offer avenues for future scholarly investigations. The current investigation refrains from classifying institutional ownership as either dedicated or transient due to insufficient data availability, which may lead to fluctuations in institutional ownership. If data becomes readily available, scholars in the future may engage in the endeavor of categorizing institutional ownership in less developed nations. Moreover, it is important to acknowledge that the topic of state ownership has not been discussed within this particular framework. Therefore, future research endeavors may focus on investigating this specific area. This study exclusively examines the phenomenon of real earnings management. Consequently, future researchers may consider exploring the realm of accruals earnings management. This study utilizes pre-Covid-19 data, and future research could provide a comparison analysis encompassing the periods before, during, and after the Covid-19 pandemic. Ultimately, the prospective researcher has the opportunity to do a comparative analysis across other nations by selecting samples from economies that share similar characteristics.
Sample selection procedure
| Sample selection procedure (2015–2019) | No. of firms |
|---|---|
| Total number of firms | |
| Bangladesh (listed on DSE) | 23 |
| India (collected from Nifty 500 index companies listed on NSE) | 28 |
| Pakistan (listed on PSX) | 33 |
| Total firms | 84 |
| Less: firms not fulfilled the criteria mentioned above (Bangladesh: 4; Pakistan: 5; India: 0) | 9 |
| Total final sample (Bangladesh: 19; Pakistan: 28; India: 28) | 75 |
| Total of observations in five years (75 × 05) (Bangladesh: 19*5; Pakistan: 28*5; India: 28*5) | 375 |
| Less: Data missing in firm year observations (Bangladesh: 10; Pakistan: 14; India: 5) | 29 |
| Final sample firm-year observations (Bangladesh: 85; Pakistan: 126; India: 135) | 346 |
Summary of variables
| Variables | Acronyms | Operationalization | References |
|---|---|---|---|
| Real earnings management | REM | Roychowdhury (2006) models | |
| Ownership concentration | OWNCON | The percentage of shareholdings who own five percent or more in the firm by following prior literature | Dang Ngoc and Tran Manh (2020) |
| Managerial ownership | MANOWN | The proportion of the total shares held by executive directors to the total number of shares issued | Piosik and Genge (2020) |
| Institutional ownership | INSOWN | The number of shares held by the institutional shareholder as a percentage of the total number of shares outstanding | Alhadab et al. (2020) |
| Foreign ownership | FOROWN | The percentage of shares owned by foreign investors to the total number of shares issued | Dang Ngoc and Tran Manh (2020) |
| Firm size | FSIZE | The natural logarithm of the firm’s total assets | Uddin et al. (2021) |
| Return on asset | ROA | It is the percentage of net income to total assets | Majumder et al. (2023) |
| Leverage | LEV | The proportion of total debt scaled by the total asset | Saona et al. (2020) |
| Firm age | FAGE | The natural logarithm of the number of years since the establishment year of the firm | Dang et al. (2020) |
| Country dummies | CDM | Dummy variables for the countries Bangladesh, India, and Pakistan | By following studies (AlHares, 2020) |
| Year dummies | YDM | Dummy variables for the years 2015, 2016, 2017, 2018, and 2019 | By following studies (AlHares, 2020) |
Measurement of board diversity index
| Board diversity index | ||
|---|---|---|
| Dimensions | Operationalization | References |
| Gender diversity (GDIV) | is an indicator variable of 1 if one or more females are on a board and “0” otherwise | Ullah et al. (2020) |
| National diversity (NDIV) | coded as “1” if at least one of the board of directors from the sample company of the sample country is foreign nationals and “0” otherwise | Kouaib and Almulhim (2019) |
| Board structural diversity (BSDIV) | If the independent ratio of the board is greater than the sample median of companies in the same industry and year, it is entered as “1,” otherwise it is coded as “0.” Board’s Independent ratio is calculated as the percentage of independent directors on the board (Dang et al., 2020) | Author development |
Note(s): Board diversity index: The sum of the three (3) dimensions ranging from 0 to 3, the higher score indicates the greater board diversity and vice-versa
Source(s): Author development
Descriptive analysis of variables
| Variables | Combined sample | Bangladesh | India | Pakistan | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | Min | Max | Mean | SD | Min | Max | Mean | SD | Min | Max | Mean | SD | Min | Max | |
| REM | 0.03 | 0.44 | −4.44 | 8.80 | 0.01 | 0.14 | −2.02 | 2.76 | 0.04 | 0.58 | −4.44 | 8.80 | 0.02 | 0.40 | −1.01 | 3.04 |
| OWNCON (%) | 41.78 | 19.10 | 13.75 | 82.25 | 25.79 | 8.99 | 13.75 | 70.20 | 62.77 | 6.33 | 33.00 | 82.25 | 30.08 | 11.09 | 17.25 | 79.55 |
| MANOWN(%) | 15.25 | 7.36 | 0.00 | 55.00 | 18.99 | 8.44 | 0.00 | 55.00 | 15.56 | 7.88 | 0.00 | 49.00 | 12.39 | 4.05 | 0.60 | 25.90 |
| INSOWN(%) | 17.81 | 10.72 | 0.00 | 70.56 | 25.76 | 10.89 | 0.00 | 70.56 | 19.27 | 9.98 | 0.00 | 68.71 | 10.88 | 5.98 | 0.06 | 43.22 |
| FOROWN (%) | 4.90 | 2.16 | 0.00 | 14.03 | 4.21 | 1.20 | 0.00 | 6.15 | 4.39 | 1.56 | 0.00 | 8.00 | 5.90 | 2.77 | 0.00 | 14.03 |
| BDIV | 1.24 | 0.26 | 0.00 | 3.00 | 1.21 | 0.19 | 0.00 | 3.00 | 1.39 | 0.28 | 0.00 | 3.00 | 1.09 | 0.16 | 0.00 | 3.00 |
| FSIZE | 12.11 | 3.21 | 4.79 | 26.00 | 9.68 | 2.01 | 4.79 | 14.65 | 10.33 | 1.21 | 16.22 | 18.66 | 15.66 | 2.01 | 12.65 | 26.00 |
| ROA (%) | 9.07 | 11.58 | −17.00 | 87.00 | 5.33 | 6.80 | −16.21 | 25.40 | 9.80 | 13.59 | −0.30 | 72.00 | 10.80 | 11.23 | −17.00 | 87.00 |
| LEV | 0.34 | 0.23 | 0.01 | 0.89 | 0.45 | 0.24 | 0.02 | 0.89 | 0.17 | 0.03 | 0.01 | 0.87 | 0.44 | 0.25 | 0.02 | 0.84 |
| FAGE | 2.33 | 1.46 | 0.71 | 5.89 | 2.50 | 1.04 | 1.02 | 3.70 | 3.65 | 0.57 | 1.07 | 5.89 | 0.80 | 0.72 | 0.71 | 4.69 |
| Total firm year observations | 346 | 85 | 135 | 126 | ||||||||||||
Note(s): SD = Standard deviation; Min = Minimum value; Max = Maximum value; The definitions of all variables are presented in Table 2
Correlation matrix
| Variable | REM | OWNCON | MANOWN | INSOWN | FOROWN | BDIV | FSIZE | ROA | LEV | FAGE |
|---|---|---|---|---|---|---|---|---|---|---|
| REM | 1 | |||||||||
| OWNCON | −0.241*** | 1 | ||||||||
| MANOWN | −0.447*** | 0.412*** | 1 | |||||||
| INSOWN | −0.231*** | 0.316*** | 0.032 | 1 | ||||||
| FOROWN | −0.191 | 0.012** | −0.113** | 0.061 | 1 | |||||
| BDIV | −0.211*** | 0.462*** | −0.312** | 0.282*** | 0.252*** | 1 | ||||
| FSIZE | 0.301*** | 0.082** | 0.212*** | 0.073 | 0.031 | 0.212** | 1 | |||
| ROA | −0.271*** | 0.231*** | 0.112** | 0.131*** | 0.139*** | 0.137*** | 0.226*** | 1 | ||
| LEV | 0.323*** | 0.061 | 0.218*** | −0.042 | 0.321** | −0.332*** | 0.043 | 0.581*** | 1 | |
| FAGE | 0.414 | 0.212* | 0.091 | 0.181* | 0.141 | 0.123 | 0.403*** | 0.342** | 0.415*** | 1 |
Note(s): ***p < 0.01, **p < 0.05, *p < 0.1
Impact of ownership structure on REM
| Variable | Dependent variable: REM | |
|---|---|---|
| Fixed effect | System GMM | |
| REMikt−1 | 0.044*** (5.645) | |
| OWNCON | −0.084*** (−7.631) | −0.024*** (−4.665) |
| MANOWN | −0.037** (−2.572) | −0.031* (−2.201) |
| FOROWN | −0.022** (−2.301) | −0.031 (−1.008) |
| INSOWN | −0.082 (−1.002) | −0.031 (−1.941) |
| FSIZE | 0.021*** (5.636) | 0.014*** (3.616) |
| ROA | −0.057*** (−3.665) | −0.021** (−2.401) |
| LEV | 0.125** (2.261) | 0.064 (1.024) |
| FAGE | 0.022 (1.114) | 0.002 (1.071) |
| YDM | Yes | Yes |
| CDM | Yes | Yes |
| Constant | −0.148*** (−5.636) | −0.117*** (−4.438) |
| R2 | 0.434 | |
| F-statistic | 68.981*** | 55.004*** |
| Chow test (p-value) | 0.001 | |
| Breusch-Pagan LM test (p-value) | 0.000 | |
| Hausman F/R test (p-value) | 0.004 | |
| Durbin-Wu-Hausman test for endogeneity (p-value) | 0.000 | |
| Hansen J-statistic (p-value) | 0.523 | |
| AR(1) p-value | 0.023 | |
| AR(2) p-value | 0.412 | |
| No. of observations | 346 | 346 |
Note(s): Asterisks denotes significance level at 10% (*), 5% (**) and 1% (***). The numbers in parenthesis indicate t-statistics. In this study, we used the two-step system GMM as our baseline model. We conduct the Durbin-Wu-Hausman test for the endogeneity test of all regressors as a group. Endogeneity testing via Durbin-Wu-Hausman implies a rejection of the null hypothesis that the regressors are exogeneous; hence the system GMM necessarily be used. In addition, Hansen’s J-statistic results show that we cannot reject the null hypothesis that the instrumental variables used in the system GMM model are valid, which supports the validity of over-identifying restrictions and the arguments for using the system GMM. The F-statistic value is significant, indicating that the model has predictive power. AR(1) shows the presence of first-order autocorrelation of residuals, whereas AR(2) indicates that there is no second-order autocorrelation. The p-values of both Chow and Breusch-Pagan LM tests indicate that either fixed-effect or random-effect regression procedures should be used, based on the results. The Hausman fixed, or random (F/R) test findings confirm that the fixed effect is the optimal approach. Year dummies and country dummies are unreported
Moderating impacts of board diversity on the relationship between ownership structure and REM
| Variables | Dependent variable: REM | |
|---|---|---|
| Fixed effect | System GMM | |
| REMikt−1 | 0.021*** (4.952) | |
| OWNCON | −0.042*** (−5.315) | −0.033*** (−3.631) |
| MANOWN | −0.017* (−1.892) | −0.003* (−1.951) |
| FOROWN | −0.032** (−2.001)) | −0.005 (−1.208) |
| INSOWN | −0.075 (−1.045) | −0.021 (−1.012) |
| BDIV | −0.142*** (−6.315) | −0.047*** (−4.346) |
| OWNCON × BDIV | −0.282*** (−7.215) | −0.282*** (−4.235) |
| MANOWN × BDIV | −0.092** (−2.351) | −0.042* (−1.761) |
| FOROWN × BDIV | −0.057 (−1.037) | −0.037 (−1.091) |
| INSOWN × BDIV | −0.082*** (−3.255) | −0.098 (−0.029) |
| FSIZE | 0.035** (2.281) | 0.061*** (3.361) |
| ROA | −0.087*** (−6.684) | −0.007* (−1.674) |
| LEV | 0.029* (1.761) | 0.012 (1.124) |
| FAGE | 0.008 (1.214) | 0.002 (1.013) |
| YDM | Yes | Yes |
| CDM | Yes | Yes |
| Constant | −0.124*** (−3.438) | −0.161*** (−4.538) |
| R2 | 0.322 | |
| F-statistic | 59.092*** | 66.004*** |
| Chow test (p-value) | 0.000 | |
| Breusch-Pagan LM test (p-value) | 0.000 | |
| Hausman F/R test (p-value) | 0.009 | |
| Durbin-Wu-Hausman test for endogeneity (p-value) | 0.000 | |
| Hansen J-statistic (p-value) | 0.589 | |
| AR(1) p-value | 0.012 | |
| AR(2) p-value | 0.367 | |
| No. of observations | 346 | 346 |
Note(s): Asterisks denotes significance level at 10% (*), 5% (**) and 1% (***). The numbers in parenthesis indicate t-statistics. In this study, we used the two-step system GMM as our baseline model. We conduct the Durbin-Wu-Hausman test for the endogeneity test of all regressors as a group. Endogeneity testing via Durbin-Wu-Hausman implies a rejection of the null hypothesis that the regressors are exogeneous; hence the system GMM necessarily be used. In addition, Hansen’s J-statistic results show that we cannot reject the null hypothesis that the instrumental variables used in the system GMM model are valid, which supports the validity of over-identifying restrictions and the arguments for using the system GMM. The F-statistic value is significant, indicating that the model has predictive power. AR(1) shows the presence of first-order autocorrelation of residuals, whereas AR(2) indicates that there is no second-order autocorrelation. The p-values of both Chow and Breusch-Pagan LM tests indicate that either fixed-effect or random-effect regression procedures should be used, based on the results. The Hausman fixed, or random (F/R) test findings confirm that the fixed effect is the optimal approach. Year dummies and country dummies are unreported
Ownership structure and real earnings management (Alternative measurements of REM)
| Variables | Direct effects | Moderating effects | ||||||
|---|---|---|---|---|---|---|---|---|
| Fixed effect | System GMM | Fixed effect | System GMM | |||||
| REM1 | REM2 | REM1 | REM2 | REM1 | REM2 | REM1 | REM2 | |
| REMikt−1 | 0.034*** (4.221) | 0.014*** (3.244) | 0.027*** (5.902) | 0.015*** (3.062) | ||||
| OWNCON | −0.041*** (−5.433) | −0.031*** (−3.631) | −0.013*** (−5.532) | −0.020*** (−4.231) | −0.062*** (−3.115) | −0.032*** (−4.212) | −0.002** (−2.185) | −0.023*** (−5.311) |
| MANOWN | −0.017** (−2.022) | −0.021** (−2.213) | −0.007* (−1.522) | −0.027** (−2.022) | −0.015** (−2.315) | −0.047* (−1.793) | −0.089* (−1.893) | −0.043*** (−5.961) |
| FOROWN | −0.032 (−1.012) | −0.002 (−1.031) | −0.022 (−1.203) | −0.001 (−1.081) | −0.022 (−1.001) | −0.082** (−2.581)) | −0.084 (−1.091) | −0.025 (−1.358) |
| INSOWN | −0.052 (−1.211) | −0.022 (−1.071) | −0.031 (−1.004) | −0.031 (−1.321) | −0.042 (−1.221) | −0.028 (−1.301) | −0.019 (−1.024) | −0.042 (−1.235) |
| BDIV | −0.123*** (−3.015) | −0.112*** (−4.415) | −0.112*** (−3.115) | −0.034*** (−5.143) | ||||
| OWNCON × BDIV | −0.122*** (−4.315) | −0.145** (−2.091) | −0.102*** (−3.220) | −0.162** (−2.042) | ||||
| MANOWN × BDIV | −0.192** (−2.403) | −0.131* (−1.754) | −0.084* (−1.801) | −0.092* (−1.703) | ||||
| FOROWN × BDIV | −0.077 (−1.237) | −0.081 (−1.092) | −0.033 (−1.097) | −0.027 (−1.051) | ||||
| INSOWN × BDIV | −0.058 (−0.029) | −0.044* (−1.681) | −0.038 (−1.009) | −0.071 (−1.024) | ||||
| FSIZE | 0.011*** (4.436) | 0.025*** (5.186) | 0.001** (2.234) | 0.031** (2.656) | 0.045*** (4.221) | 0.033** (2.211) | 0.042** (2.281) | 0.048*** (3.581) |
| ROA | −0.043*** (−4.636) | −0.021*** (−3.802) | −0.001* (−1.783) | −0.031** (−2.331) | −0.035*** (−3.314) | −0.057*** (−4.124) | −0.087 (−1.057) | −0.103 (−1.214) |
| LEV | 0.521** (2.441) | 0.626** (2.452) | 0.032 (1.024) | 0.028* (1.866) | 0.023* (1.871) | 0.521** (2.591) | 0.041** (2.424) | 0.432** (2.532) |
| FAGE | 0.029* (1.831) | 0.029** (2.204) | 0.022 (1.133) | 0.008 (1.071) | 0.053* (1.034) | 0.095** (2.434) | 0.028 (1.034) | 0.019 (1.054) |
| YDM | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| CDM | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Constant | −0.124*** (−3.232) | −0.138*** (−3.006) | −0.131*** (−4.196) | −0.153*** (−3.623) | −0.115*** (−4.531) | −0.151*** (−5.038) | −0.126*** (−5.234) | −0.153*** (−4.033) |
| R2 | 0.381 | 0.452 | 0.421 | 0.324 | ||||
| F-statistic | 58.383*** | 54.125** | 63.002*** | 53.721*** | 49.719** | 51.221** | 55.291** | 59.201*** |
| Chow test (p-value) | 0.002 | 0.001 | 0.000 | 0.000 | ||||
| Breusch-Pagan LM test (p-value) | 0.000 | 0.000 | 0.000 | 0.000 | ||||
| Hausman F/R test (p-value) | 0.003 | 0.005 | 0.007 | 0.019 | ||||
| Durbin-Wu-Hausman test for endogeneity (p-value) | 0.000 | 0.000 | 0.000 | 0.000 | ||||
| Hansen J-statistic (p-value) | 0.321 | 0.432 | 0.382 | 0.431 | ||||
| AR(1) p-value | 0.012 | 0.042 | 0.054 | 0.035 | ||||
| AR(2) p-value | 0.321 | 0.442 | 0.445 | 0.383 | ||||
| No. of observations | 346 | 346 | 346 | 346 | 346 | 346 | 346 | 346 |
Note(s): Asterisks denotes significance level at 10% (*), 5% (**) and 1% (***). The numbers in parenthesis indicate t-statistics. In this study, we used the two-step system GMM as our baseline model. We conduct the Durbin-Wu-Hausman test for the endogeneity test of all regressors as a group. Endogeneity testing via Durbin-Wu-Hausman implies a rejection of the null hypothesis that the regressors are exogeneous; hence the system GMM necessarily be used. In addition, Hansen’s J-statistic results show that we cannot reject the null hypothesis that the instrumental variables used in the system GMM model are valid, which supports the validity of over-identifying restrictions and the arguments for using the system GMM. The F-statistic value is significant, indicating that the model has predictive power. AR(1) shows the presence of first-order autocorrelation of residuals, whereas AR(2) indicates that there is no second-order autocorrelation. The p-values of both Chow and Breusch-Pagan LM tests indicate that either fixed-effect or random-effect regression procedures should be used, based on the results. The Hausman fixed, or random (F/R) test findings confirm that the fixed effect is the optimal approach. Year dummies and country dummies are unreported
Notes
World Bank Data, 2019 (available at https://data.worldbank.org/country)
Top 500 companies in India based on full market capitalization
References
Adeneye, Y. and Kammoun, I. (2022), “Real earnings management and capital structure: does environmental, social and governance (ESG) performance matter?”, Cogent Business and Management, Vol. 9 No. 1, doi: 10.1080/23311975.2022.2130134.
AL-Duais, S.D., Malek, M., Abdul Hamid, M.A. and Almasawa, A.M. (2022), “Ownership structure and real earnings management: evidence from an emerging market”, Journal of Accounting in Emerging Economies, Vol. 12 No. 2, pp. 380-404, doi: 10.1108/JAEE-01-2021-0008.
Alhadab, M., Abdullatif, M. and Mansour, I. (2020), “Related party transactions and earnings management in Jordan: the role of ownership structure”, Journal of Financial Reporting and Accounting, Vol. 18 No. 3, pp. 505-531, doi: 10.1108/JFRA-01-2019-0014.
Al-Haddad, L. and Whittington, M. (2019), “The impact of corporate governance mechanisms on real and accrual earnings management practices: evidence from Jordan”, Corporate Governance: The International Journal of Business in Society, Vol. 19 No. 6, pp. 1167-1186, doi: 10.1108/CG-05-2018-0183.
Alhmood, M., Shaari, H., Al-Dhamari, R. and Sani, A.A. (2023), “Does ownership concentration matter for the relationship between CEO characteristics and real earnings management: evidence from Jordan”, EuroMed Journal of Business, Vol. ahead-of-print No. ahead-of-print, doi: 10.1108/EMJB-09-2022-0155.
Ashraf, B.N. and Qian, N. (2021), “The impact of board internationalization on real earnings management: evidence from China”, Sage Open, Vol. 11 No. 3, doi: 10.1177/21582440211032640.
BBC News (2020), “China's Luckin coffee slumps on ‘fake’ data news”, 3 April.
Binaebi, B. (2020), “The effect of corporate governance and ownership structure on earnings management of manufacturing companies in Nigeria”, Journal of Accounting and Financial Management, Vol. 6 No. 2, pp. 39-49.
Blundell, R. and Bond, S. (1998), “Initial conditions and moment restrictions in dynamic panel data models”, Journal of Econometrics, Vol. 87 No. 1, pp. 115-143, doi: 10.1016/S0304-4076(98)00009-8.
Brahma, S., Nwafor, C. and Boateng, A. (2021), “Board gender diversity and firm performance: the UK evidence”, International Journal of Finance and Economics, Vol. 26 No. 4, pp. 5704-5719, doi: 10.1002/ijfe.2089.
Brink, A.G., Gouldman, A., Rose, J.M. and Rotaru, K. (2020), “Effects of superiors' compensation structures on psychophysiological responses and real earnings management decisions of subordinate managers”, Management Accounting Research, Vol. 48 No. May 2018, 100691, doi: 10.1016/j.mar.2020.100691.
Browne, R. (2020), The Enron of Germany: Wirecard Scandal Casts a Shadow on Corporate Governance, CNBC, June, available at: https://www.cnbc.com/2020/06/29/enron-of-germany-wirecard-scandal-casts-a-shadow-on-governance.html
Chen, H., Jory, S. and Ngo, T. (2020), “Earnings management under different ownership and corporate governance structure: a natural experiment with master limited partnerships”, The Quarterly Review of Economics and Finance, Vol. 76, pp. 139-156, doi: 10.1016/j.qref.2019.05.005.
Cohen, D.A. and Zarowin, P. (2010), “Accrual-based and real earnings management activities around seasoned equity offerings”, Journal of Accounting and Economics, Vol. 50 No. 1, pp. 2-19, doi: 10.1016/j.jacceco.2010.01.002.
Cohen, D.A., Dey, A. and Lys, T.Z. (2008), “Real and accrual-based earnings management in the pre-and post-Sarbanes-Oxley periods”, The Accounting Review, Vol. 83 No. 3, pp. 757-787, doi: 10.2308/accr.2008.83.3.757.
Corporate Finance Institute (n.d.), “Top accounting scandals”, available at: https://corporatefinanceinstitute.com/resources/knowledge/other/top-accounting-scandals/
Dang Ngoc, H. and Tran Manh, D. (2020), “The impact of ownership structure, financial distress on earnings management: the case of Vietnam”, Advance. doi: 10.31124/advance.12721907.v1.
Debnath, N.C., Chowdhury, S.P. and thai, S. (2021), “Ownership structure and real earnings management: an empirical study on emerging economy”, Corporate Ownership and Control, Vol. 18 No. 2, pp. 74-89, doi: 10.22495/cocv18i2art6.
El Diri, M., Lambrinoudakis, C. and Alhadab, M. (2020), “Corporate governance and earnings management in concentrated markets”, Journal of Business Research, Vol. 108, pp. 291-306, doi: 10.1016/j.jbusres.2019.11.013.
Eng, L.L., Fang, H., Tian, X., Yu, T.R. and Zhang, H. (2019), “Financial crisis and real earnings management in family firms: a comparison between China and the United States”, Journal of International Financial Markets, Institutions and Money, Vol. 59, pp. 184-201, doi: 10.1016/j.intfin.2018.12.008.
Fama, E.F. and Jensen, M.C. (1983), “Separation of ownership and control”, The Journal of Law and Economics, Vol. 26 No. 2, pp. 301-325, doi: 10.1086/467037.
Fatima, H., Haque, A. and Usman, M. (2020), “Is there any association between real earnings management and crash risk of stock price during uncertainty? An evidence from family-owned firms in an emerging economy”, Future Business Journal, Vol. 6 No. 1, p. 36, doi: 10.1186/s43093-020-00038-5.
Ghaleb, B.A.A., Qaderi, S.A., Almashaqbeh, A. and Qasem, A. (2021), “Corporate social responsibility, board gender diversity and real earnings management: the case of Jordan”, Cogent Business and Management, Vol. 8 No. 1, 1883222, doi: 10.1080/23311975.2021.1883222.
Ghorbani, A. and Salehi, M. (2021), “Earnings management and the informational and disciplining role of debt: evidence from Iran”, Journal of Asia Business Studies, Vol. 15 No. 1, pp. 72-87, doi: 10.1108/JABS-11-2019-0336.
Githaiga, P.N., Muturi Kabete, P. and Caroline Bonareri, T. (2022), “Board characteristics and earnings management. Does firm size matter?”, Cogent Business and Management, Vol. 9 No. 1, 2088573, doi: 10.1080/23311975.2022.2088573.
Gujarati, D.N. and Porter, D.C. (2009), Basic Econometrics, McGraw-Hill Education, New York.
Jensen, C. and Meckling, H. (1976), “Theory of the firm: managerial behavior, agency costs and ownership structure”, Vol. 3 No. No. 4, pp. 305-360, doi: 10.1016/0304-405X(76)90026-X.
Kouaib, A. and Almulhim, A. (2019), “Earnings manipulations and board's diversity: the moderating role of audit”, The Journal of High Technology Management Research, Vol. 30 No. 2, 100356, doi: 10.1016/j.hitech.2019.100356.
Latif, K. (2018), “Interactions between corporate governance, earnings quality attributes and value of firm: empirical analysis from non-financial sector of Pakistan”, [Doctoral Dissertation, International Islamic University, Pakistan], Pakistan Research Repository.
Lemma, T.T., Negash, M., Mlilo, M. and Lulseged, A. (2018), “Institutional ownership, product market competition, and earnings management: some evidence from international data”, Journal of Business Research, Vol. 90, pp. 151-163, doi: 10.1016/j.jbusres.2018.04.035.
Li, X., Than, E.T., Ahmed, R., Ishaque, M. and Huynh, T.L.D. (2021), “Gender diversity of boards and executives on real earnings management in the bull or bear period: empirical evidence from China”, International Journal of Finance and Economics, Vol. 28 No. 3, pp. 1-19, doi: 10.1002/ijfe.2562.
Majumder, M.T.H., Ruma, I.J. and Akter, A. (2023), “Does intellectual capital affect bank performance? Evidence from Bangladesh”, LBS Journal of Management and Research, Vol. 21 No. 2, pp. 171-185, doi: 10.1108/LBSJMR-05-2022-0016.
Masud, M., Kaium, A., Nurunnabi, M. and Bae, S.M. (2018), “The effects of corporate governance on environmental sustainability reporting: empirical evidence from South Asian countries”, Asian Journal of Sustainability and Social Responsibility, Vol. 3 No. 1, pp. 1-26, doi: 10.1186/s41180-018-0019-x.
Mishra, R.K. (2021), “Corporate governance: trends and challenges”, Indian Journal of Corporate Governance, Vol. 12 No. 1, p. 1, doi: 10.1177/0974686219854811.
Nasir, N.A.B.M., Ali, M.J. and Ahmed, K. (2019), “Corporate governance, board ethnicity and financial statement fraud: evidence from Malaysia”, Accounting Research Journal, Vol. 32 No. 3, pp. 514-531, doi: 10.1108/ARJ-02-2018-0024.
Nguyen, A.H., Nguyen, L.H. and Doan, D.T. (2020), “Ownership structure and earnings management: empirical evidence from vietnam real estate sector”, Real Estate Management and Valuation, Vol. 28 No. 2, pp. 37-51, doi: 10.1515/remav-2020-0014.
Piosik, A. and Genge, E. (2020), “The influence of a company's ownership structure on upward real earnings management”, Sustainability, Vol. 12 No. 1, p. 152, doi: 10.3390/su12010152.
Roychowdhury, S. (2006), “Earnings management through real activities manipulation”, Journal of Accounting and Economics, Vol. 42 No. 3, pp. 335-370, doi: 10.1016/j.jacceco.2006.01.002.
Salehi, M., Zimon, G., Arianpoor, A. and Gholezoo, F.E. (2022a), “The impact of investment efficiency on firm value and moderating role of institutional ownership and board independence”, Journal of Risk and Financial Management, Vol. 15, p. 4, doi: 10.3390/jrfm15040170.
Salehi, M., Zimon, G., Tarighi, H. and Gholamzadeh, J. (2022b), “The effect of mandatory audit firm rotation on earnings management and audit fees: evidence from Iran”, Journal of Risk and Financial Management, Vol. 15, p. 3, doi: 10.3390/jrfm15030102.
Saona, P., Muro, L. and Alvarado, M. (2020), “How do the ownership structure and board of directors' features impact earnings management? The Spanish case”, Journal of International Financial Management and Accounting, Vol. 31 No. 1, pp. 98-133, doi: 10.1111/jifm.12114.
Shafeeq Nimr Al-Maliki, H., Salehi, M. and Kardan, B. (2023), “The relationship between board characteristics and social responsibility with firm innovation”, European Journal of Management and Business Economics, Vol. 32 No. 1, pp. 113-129, doi: 10.1108/EJMBE-04-2020-0094.
Uddin, M.J., Majumder, M.T.H., Akter, A. and Zaman, R. (2021), “Do the diversification of income and assets spur bank profitability in Bangladesh? A dynamic panel data analysis”, Vilakshan - XIMB Journal of Management, Vol. 19 No. 2, pp. 177-194, doi: 10.1108/XJM-01-2021-0023.
Ullah, I., Fang, H. and Jebran, K. (2020), “Do gender diversity and CEO gender enhance firm’s value? Evidence from an emerging economy”, Corporate Governance: The International Journal of Business in Society, Vol. 20 No. 1, pp. 44-66, doi: 10.1108/CG-03-2019-0085.
U.S. Securities and Exchange Commission (2022), “SEC charges Canadian Cannabis company and former senior executive with accounting fraud”, 24 October, Washington D.C.
Widagdo, A.K., Arifah, S., Goestjahjanti, F.S., Goestjahjanti, F.S. and Kiswanto (2023), “The impact of ownership characteristics and gender on earnings management: Indonesian companies”, Journal of Risk and Financial Management, Vol. 16 No. 1, p. 17, doi: 10.3390/jrfm16010017.
Further reading
Majumder, M.T.H., Akter, A. and Li, X. (2017), “Corporate governance and corporate social disclosures: a meta-analytical review”, International Journal of Accounting and Information Management, Vol. 25 No. 4, pp. 434-458, doi: 10.1108/IJAIM-01-2017-0005.
Majumder, M.T.H., Li, X., Akter, A. and Begum, M. (2019), “Corporate attributes and corporate social disclosures: a meta-analytical review”, International Journal of Law and Management, Vol. 61 No. 1, pp. 45-72, doi: 10.1108/IJLMA-02-2018-0023.