THE INFLUENCE OF BOARD INDEPENDENCE ON FINANCIAL REPORTING QUALITY: EVIDENCE FROM A MULTINATIONAL SETTING
Master thesis, MSc Accountancy and Controlling University of Groningen, Faculty of Economics and Business
June 22, 2020
Marie-Louise Erren Student number: 3853446 E-mail: m.erren@student.rug.nl
Supervisor S. Rusanescu, PhD
Word count: [10602]
ABSTRACT
This study investigates the relation between parent’s board independence and the financial reporting quality of its foreign subsidiaries. The parents of multinational corporations can exert influence over their subsidiaries and have more opportunities to manage their earnings because of their group structure. The subsidiaries’ financial reporting quality affects the financial reporting quality of the multinational corporation due to the consolidation process. Since the parent’s board monitors the financial reporting quality of the consolidated financial statements, the board is likely to be concerned about the financial reporting quality of the subsidiaries to protect the interests of the shareholders. Following the agency theory, board independence drives the monitoring effectiveness of the board of directors, because independent directors are argued to better protect the interests of the shareholder. Therefore, I predict that a higher proportion of independent directors in the parent’s board is positively associated with the financial reporting quality of its subsidiaries. Furthermore, I expect that a low institutional quality of the subsidiaries’ home country strengthens this positive relationship. Parent’s board independence might be incrementally beneficial in curbing the subsidiaries’ increased level of earnings management in absence of a strong external monitoring mechanism. This study is based on 8,205 firm-year observations from 394 unique U.S. multinationals and 2,186 unique European subsidiaries between the period 2011-2017. The results do not confirm my hypotheses, as they suggest that the board of multinational corporations’ parents is not an effective corporate governance mechanism to curb their subsidiaries’ level of earnings management and this is not influenced by the institutional quality of the subsidiaries’ home country. However, the results do show that a high institutional quality of the subsidiaries’ home country leads to lower levels of earnings management and thus a higher financial reporting quality.
Keywords: U.S. multinational corporations, European subsidiaries, financial reporting quality, earnings management, board independence, institutional quality, rule of law.
TABLE OF CONTENT
ABSTRACT ... 2
1. INTRODUCTION ... 4
2. THEORY & HYPOTHESES DEVELOPMENT ... 8
3. METHODOLOGY ... 14
4. RESULTS ... 23
5. CONCLUSION ... 30
REFERENCES ... 34
APPENDIX A: VARIABLE DEFINITIONS ... 39
APPENDIX B: RESULTS ADDITIONAL TEST ... 40
1. INTRODUCTION
This study examines the relationship between the board independence of multinational corporations’ parents and their subsidiaries’ financial reporting quality. Multinational corporations (MNCs) are organisations that have foreign subsidiaries and/or affiliates in different institutional environments through which they conduct their international activities (Durnev et al., 2017). According to a report by the United Nations Conference on Trade and Development (UNCTAD, 2018), 62 percent of the total assets of the world’s largest 100 MNCs were invested in foreign assets in 2017. In the same year, the subsidiaries of MNCs accounted for 30 percent of the share of foreign value added in total exports (UNCTAD, 2018). Due to the consolidation process, the earnings reported by the subsidiaries of the MNC-parent contribute to the consolidated earnings of the MNC. Therefore, the financial reporting quality of the subsidiaries is likely to influence the financial reporting quality of the MNC. Due to the economic importance, the financial reporting quality of the subsidiaries of MNCs deserves attention.
Financial reporting quality is important since the financial statements convey information about the firm’s underlying performance (Koo et al., 2017) which is used in contracts and by investors to value the firm and monitor its performance (Bradbury et al., 2006). Following the agency theory, managers will not necessarily act in the best interests of the shareholders (Bukit and Iskander, 2009), due to information asymmetry and conflicts of interests between the manager and the shareholders (Jensen and Meckling, 1979). This can create opportunities for managers to engage in earnings management to influence the firm’s earnings in their own favour (Healy and Wahlen, 1999). In the multinational context, MNC-parents can exert influence over the financial reporting process of their subsidiaries (Doz and Prahalad, 1981; Beuselinck et al.
2019) and have more opportunities (Beuselinck et al., 2019) to manage their earnings compared to stand-alone firms.
Corporate governance characteristics are argued to influence the firm’s financial reporting quality (Bushman and Piotroski, 2006). A key corporate governance mechanism is the board of directors (Alves, 2014). The proportion of independent directors on the board is an important factor that may affect the ability of the board to serve as an effective internal monitoring mechanism (Beasley, 1996; Dechow et al., 1996). A more independent board is expected to increase the financial reporting quality (Morck et al., 1989), since independent directors are argued to better protect the interests of the shareholders (Beasley, 1996; Dechow et al., 1996)
by mitigating the private incentives of the manager (Alves, 2014). Prior research argued that MNC-parents prefer to manage their earnings at the subsidiary-level (Beuselinck et al., 2019;
Durnev et al., 2017; Dyreng et al., 2012) through the reporting guidelines that the MNC-parents impose on their subsidiaries (Bonacchi et al., 2018; Prencipe, 2012). Due to the consolidation process, the subsidiaries’ financial reporting quality affects the financial reporting quality of the MNC. Since the MNC-parent’s board monitors the financial reporting quality of the consolidated financial statements, the board is likely to be concerned about the subsidiaries’
financial reporting quality to protect the interests of the shareholders. Therefore, I expect that a MNC-parent’s board comprised of a higher proportion of independent directors is better able to curb the level of earnings management in the consolidated financial statements by also limiting the subsidiaries’ level of earnings management, thereby protecting the interests of the shareholders of the MNC.
In addition, previous research has found a substantial cross-country difference in the financial reporting quality of firms around the world (Bushman and Piotroski, 2006; Leuz et al., 2003;
Dyreng et al., 2012). Since MNCs have subsidiaries in different institutional environments, they can use this cross-country variation to manipulate earnings away from enforcers (Ayers et al., 2011). Indeed, a MNC-parent is more inclined to manage earnings at the subsidiary-level when the institutional quality of the subsidiary’s home country is low compared to the institutional quality of the MNC-parent’s home country (Beuselinck et al., 2019; Durnev et al., 2017; Dyreng et al., 2012). The MNC-parents included in this study are located in the United States, which is argued to have one of the strictest legal environment (Durnev, 2017). This implicates that the MNCs of interest might have less opportunities to manage earnings at the parent-level and engage in earnings management at the subsidiary-level due to less stringent regulations, poor enforcement and low litigation risks (Beuselinck et al., 2019; Durnev et al., 2017; Dyreng et al., 2012). When the institutional quality of a firm’s home country is low, a firm can benefit from a more independent board of directors (Dahya et al., 2008), since the internal monitoring mechanism might work as a substitute for the weak external monitoring mechanism. The MNC- parent’s board independence might be incrementally beneficial in curbing the level of earnings management in the consolidated financial statements by limiting also the level of earnings management reported by its foreign subsidiaries, which is argued to be higher when the subsidiaries are located in a country with a low institutional quality. Therefore, I predict the positive relation between the MNC-parent’s board independence and the subsidiaries’ financial
reporting quality to be stronger when the subsidiaries are located in a country with a low institutional quality.
This research focuses on a sample of 8,205 firm-year observations corresponding to 2,186 unique subsidiaries of 394 unique non-financial listed U.S.-based MNCs between the period 2011-2017. The subsidiaries of interest are the material subsidiaries, which are subsidiaries that account for a significant proportion of the MNC’s consolidated earnings. The material subsidiaries are hand-collected from Exhibit 21.1 of the Edgar database. The financial reporting quality of the subsidiaries is measured through earnings management. I use the Modified Jones model (Dechow et al., 1995) to isolate the abnormal accruals, which is often used as a proxy for earnings management (Ebrahim, 2007). The financial data of the subsidiaries is obtained from the Orbis global database (Kalemli-Ozcan et al., 2015). The data of the composition of the board of directors of the MNC-parents is collected from the MSCI database while the financial information of the MNC-parents is obtained from Compustat. I use the rule of law as a proxy for the institutional quality of the subsidiaries’ home country. The rule of law derives from the Worldwide Governance Indicators (WGI) database, which previous studies have used to measure the institutional quality of a country (Dyreng et al., 2012; Beuselinck et al., 2019).
The results suggest that MNC-parent’s board independence is not an effective mechanism to curb the level of earnings management in foreign subsidiaries. A possible explanation might be that an independent audit committee may be more important in constraining the subsidiaries’
level of earnings management, since their primary role is to monitor the financial reporting process (Klein, 2002). Also, MNC-parents might focus on what happens at the parent-level and disregard the activities of its foreign subsidiaries, due to the geographical distance which leads to higher costs of gathering information (Beuselinck et al., 2011). Lastly, approximately half of the MNC firm-year observations show that the CEO is also the chairman of the board of directors. CEO-duality is argued to decrease the board oversight (Alabdullah, 2016), since it lowers the objectivity of the board (Fama, 1980; Fama and Jensen, 1983).
Also, the results suggest that a lower institutional quality does not strengthen the relationship between the MNC-parent’s board independence and its subsidiaries’ level of earnings management. The lack of evidence might be due to relatively small differences in the institutional quality among the European countries. However, this study does show that
subsidiaries located in countries with a high institutional quality engage less in earnings management and thus have a higher financial reporting quality.
This study makes two contributions to the literature. First, this research contributes to the scant literature on the influence of MNC-parent’s governance characteristics on the financial reporting quality of its subsidiaries. This strand of literature only documents that lower levels of family-ownership and higher analyst coverage of MNC-parents lead to higher levels of subsidiaries’ earnings management (Beuselinck et al., 2010). Prior research with respect to the influence of governance characteristics of stand-alone firms on their financial reporting quality does show that board independence is an effective internal monitoring mechanism to curb the level of earnings management (Alves, 2014; Beasley, 1996; Dechow et al., 1996; Jaggi et al., 2009; Klein, 2002; Liu et al., 2015; Peasnell et al., 2005; Xie et al., 2003). However, this has not been examined in a multinational setting yet. This implies that there is a lack of evidence regarding the effect of the MNC-parent’s board characteristics on the level of earnings management of its subsidiaries. To the best of my knowledge, this is the first study that investigates the influence of MNC-parent’s board characteristics on the financial reporting quality of its subsidiaries. The board of directors is seen as a key corporate governance mechanism (Alves, 2014) which protects the interests of the shareholders (Kim et al., 2005;
Alves, 2014) more effectively when the board is independent from the management (Beasley, 1996; Dechow et al., 1996). Therefore, it is important to investigate if the MNC-parent’s independent board of directors is able to protect the interests of the shareholders and curb the level of earnings management in the consolidated financial statements by limiting also the levels of earnings management reported by its material subsidiaries. My results show that board independence, as a corporate governance characteristic of the MNC-parent, does not effectively limit the level of earnings management of its subsidiaries.
Second, this research contributes to the literature on the effect of the institutional quality of the subsidiaries’ home country on their financial reporting quality. Dyreng et al. (2012) found that the consolidated earnings are managed to a lower extent when the MNC has a high concentration of subsidiaries in countries with a high institutional quality. However, they did not study whether the institutional quality of the home country of the subsidiaries affect their individual financial reporting quality, which contributes to the consolidated financial reporting quality. Beuselinck et al. (2019) did investigate this by arguing that MNC-level, as well as local factors, are likely to influence the financial reporting practices of its subsidiaries. Their results
show that subsidiaries located in countries with a high institutional quality have less tendency to manipulate earnings. The results of this study confirm this finding by showing additional evidence that subsidiaries located in countries with a low institutional quality tend to manage earnings more than subsidiaries located in countries with a high institutional quality.
The remainder of the paper is organized as follows. Section 2 will review the literature and develop the hypotheses. Section 3 describes the research methodology used in this study. In section 4, I report my results. Lastly, section 5 contains the conclusion of this research.
2. THEORY & HYPOTHESES DEVELOPMENT
According to agency theory, the separation between ownership and control can lead to conflicts of interests between the manager and the shareholders (Jensen and Meckling, 1979). It assumes that, as a result of the conflicts of interests, the manager will not act in the best interests of the shareholders (Bukit and Iskander, 2009), as they will pursue their private interests (Eisenhardt, 1989). Agency theory proposes that the alignment between the interests of the manager and the shareholders can be improved by monitoring mechanisms (Fama and Jensen, 1983; Jensen and Meckling, 1976). A classic distinction is made between external and internal monitoring mechanisms. The market for corporate control is the most outstanding external mechanism, whilst there are different internal mechanisms like ownership structure and the board of directors (Fama and Jensen, 1983; Kazemian and Sanusi, 2015; Zajac and Westphal, 1994).
The board of directors is considered to be a primary element of corporate governance (Alves, 2014), since the board is responsible for monitoring the managers and the quality of the financial reports to protect the interests of the shareholders (Fama and Jensen, 1983). Due to the self-interested incentives of managers to manage earnings, it is important that the board monitors the production of financial reports (Alves, 2014).
Managers manage earnings when they “use judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers” (Healy and Wahlen, 1999, p. 368). Researchers widely use earnings management as a measure to determine the financial reporting quality of firms. Biddle et al. (2009) define financial reporting quality as “the precision with which financial reporting conveys information about the firm’s operations, in particular its expected cash flows, that inform equity investors” (p. 113). So, financial reporting informs investors in making rational
investment decisions (FASB, 1987). However, when managers use earnings management in the financial reporting process, it may be intended to misinform investors (Watts and Zimmerman, 1978). Managers would only use earnings management when there is a potential for private benefits (Prior et al. 2008), which is in line with the agency theory. The more earnings management managers use, the more accruals will need to be revised in the future (Durnev et al., 2017), which reduces the earnings persistence and will eventually reduce the financial reporting quality (Dechow and Dichev, 2002).
Agency theory states that independence from the management drives the ability of the board of directors to serve as an effective internal monitoring mechanism (Beasley, 1996; Dechow et al., 1996) to protect the interests of the shareholders (Kim et al., 2005; Alves, 2014). In this respect, researchers distinguish dependent directors from independent directors. Whereas dependent directors are also managers, independent directors are entirely independent of the management (Davidson et al., 2005). Despite the fact that dependent directors will acquire more information to monitor, they are more likely to collaborate with the management (Fama, 1980; Fama and Jensen, 1983). This is because the interests of dependent directors are aligned with those of the management (Kim et al., 2005). On the contrary, independent directors are independent of the management (Fama, 1980; Fama and Jensen, 1983) and are therefore “not subject to the same potential conflicts of interests that are likely to affect the judgement of inside directors”
(Rhoades et al., 2000, p. 78). Independent directors are believed to protect the interests of the shareholders more effectively compared to dependent directors, since the value of their human capital depends to a certain extent on the effectiveness of their monitoring performance (Fama and Jensen, 1983). Specifically, when they fail to satisfactorily monitor the management, their human capital value might decrease (Ebrahim, 2007) which decreases their reputations on the labour market (Davidson et al., 2005). Therefore, the proportion of independent directors on the board plays an important role in determining the board’s monitoring effectiveness.
Indeed, prior research finds monitoring to be stronger when the board is comprised of a higher proportion of independent directors (Morck et al., 1989). Hence, independent directors may play an important role in monitoring the production of financial reports by managers (Alves, 2014) and constraining opportunistic earnings management (Alves, 2014; Beasley, 1996;
Dechow et al., 1996; Jaggi et al., 2009; Klein, 2002; Liu et al., 2015; Peasnell et al., 2005; Xie et al., 2003). These studies show that firms with a higher proportion of independent directors on the board have a higher financial reporting quality. Following this reasoning, it is likely that
independent directors improve the financial reporting quality by reducing the level of earnings management (Jaggi et al., 2009). However, this view is challenged by Monks and Minow (1995), who argue that independent directors need proper incentives to become effective monitors. This view suggests that an increase in independent directors does not automatically lead to a decrease in the level of earnings management, all else held constant. This is consistent with the finding that all independent directors may aim to restrain earnings management, but only those that have financial expertise are actually able to do so (Chtourou et al., 2001) and that only independent directors from financial intermediaries are able to reduce earnings management (Park and Shin, 2004). In contrast, Bradbury et al. (2006) do not find statistically significant evidence of a relation between board independence and earnings management. They attribute this to the fact that in their sample of firms from Malaysia and Singapore the proportion of independent directors on the audit committee is related to a lower level of earnings management instead of to the proportion of independent directors on the board.
Board independence and financial reporting quality
A MNC is structured as a single economic entity consisting of a parent firm and subsidiaries that are legally independent with the same source of control (La Porta et al., 1999). Previous research suggests that a MNC-parent has significant power and can exert extensive influence over its subsidiaries (Beuselinck et al., 2019). This is in line with the management and business literature stating that a MNC-parent is likely to control the behaviour of its subsidiaries (Doz and Prahalad, 1981), to ensure that the overall corporate strategy will be achieved (Andersson and Forsgren, 1996). Therefore, Dyreng et al. (2012) assume that a MNC-parent can also exert influence over the financial reporting decisions of its subsidiaries. Indeed, since the MNC- parent carries the responsibility to prepare and publish the consolidated financial reports, the MNC-parent requires its subsidiaries to report in accordance with specific reporting guidelines (Prencipe, 2012). The earnings that the subsidiaries report contribute to the consolidated earnings of the MNC (Beuselinck et al., 2010; Beuselinck et al., 2019). Hence, the MNC-parent can manage its consolidated earnings at the subsidiary-level through the reporting guidelines (Bonacchi et al., 2018; Prencipe, 2012) and thus influencing the financial reporting quality of its subsidiaries.
The incentives of MNCs to engage in earnings management are mostly the same as those of stand-alone firms (Beuselink et al., 2019). However, a MNC has more opportunities to engage in earnings management than stand-alone firms because of its group structure (Beuselinck et
al., 2019). This is related to the differences in (1) the levels of complexity of the financial reporting, (2) the corporate governance structures and (3) the institutional quality of countries.
The increased level of complexity of the MNC’s financial reporting is due to the difficulty to track cash flows within the MNC. The intercompany transactions within MNCs make the transfer of cash flows less transparent and more difficult to track than those of stand-alone firms (Beuselinck et al., 2019; Kao and Chen, 2004; Prencipe, 2012). Therefore, MNCs might manage the earnings at the subsidiary-level in order to avoid getting caught (Dyreng et al., 2012), since it is more difficult for the auditors of the MNC-parents to monitor the financial reporting quality of the subsidiaries (Stewart and Kinney, 2013). When getting caught, MNCs might incur lower legal and reputational sanctions if the earnings are managed at the subsidiary- level compared to the parent-level, because the MNC-parents could argue that they were not aware of the misreporting of their subsidiaries (Dearborn, 2009).
Corporate governance characteristics of firms are proven to affect their financial reporting quality (Bushman and Piotroski, 2006). Kao and Chen (2004) examine the influence of board independence as a governance characteristic on the level of earnings management. They hypothesized and found that Taiwanese firms will be less likely to engage in earnings management when they have a greater proportion of independent directors and that this relationship is more significant in group affiliation firms than in non-group affiliation firms. As already mentioned, a MNC has foreign subsidiaries and/or affiliates (Durnev et al., 2017).
Depending on the level of ownership, a firm is classified as a subsidiary or an affiliate of the MNC-parent. By definition, when the parent has a minority (majority) stake in the firm, it is considered to be an affiliate (a subsidiary). Whereas the parent can exercise majority control over its subsidiaries, the parent cannot over its affiliates due to the differences in ownership.
Since Kao and Chen (2004) found a relationship between a governance characteristic and earnings management in group affiliated firms, it is likely that the MNC-parent’s governance characteristics affect the level of earnings management of its subsidiaries, because the parent can exercise majority control over its subsidiaries. Indeed, Beuselinck et al. (2010) found evidence that corporate governance characteristics of a MNC-parent play a crucial role in affecting the extent of earnings management of its subsidiaries. Specifically, their results show that lower levels of family-ownership and higher analyst coverage of MNC-parents leads to higher levels of earnings management of their subsidiaries.
Summarizing, a more independent board is expected to be better able to constrain earnings management through more effective monitoring according to the agency theory. This suggests that the financial reporting quality of firms is likely to be higher with a higher proportion of independent directors. This relationship is even more important in a multinational setting, because the MNC-parent can exert influence over the financial reporting decisions of its subsidiaries and has more opportunities to engage in earnings management at the subsidiary- level compared to stand-alone firms. Due to the consolidation process, the subsidiaries’
financial reporting quality affects the financial reporting quality of the MNC. Since the MNC- parent’s board monitors the financial reporting quality of the consolidated financial statements, the board is likely to be concerned about the subsidiaries’ financial reporting quality to protect the interests of the shareholders. Therefore, I predict that when the board of directors of the MNC-parent is comprised of a higher proportion of independent directors, this will lead to less earnings management at the subsidiary-level. This leads to the following hypothesis:
H1: The board independence of the MNC-parent has a positive relation with the financial reporting quality of its subsidiaries.
Institutional quality and financial reporting quality
Institutions can be defined as “formal and informal rules of the game, and their enforcement characteristics” (Kunčič, 2014, p. 137). Previous research has found an inverse relationship between a country’s institutional quality and the firm’s level of earnings management (Ball et al., 2000; Bushman and Piotroski, 2006; Cohen and Zarowin, 2010; Leuz et al., 2003). They consider the institutional quality of a firm’s home country as one of the drivers of the financial reporting quality, since firms located in countries with a high institutional quality are more conservative by recognizing bad news earlier.
As already mentioned, the differences in the institutional quality of countries provide MNCs with opportunities to engage in earnings management at the subsidiary-level due to their group structure (Beuselinck et al., 2019). Prior studies have examined the influence of the institutional quality on the financial reporting quality in a multinational setting (Beuselinck et al., 2019;
Durnev et al., 2017; Dyreng et al., 2012). Indeed, a MNC is found to be more inclined to manage its earnings through its subsidiaries which are located in countries where the institutional quality is low (e.g. less stringent regulations, poor enforcement mechanisms and low litigation risks) when the MNC-parent’s home country has a high institutional quality (Beuselinck et al., 2019;
Durnev et al., 2017; Dyreng et al., 2012). This is because the MNC faces a higher potential costs of earnings management detection (e.g. legal and reputational sanctions) when the earnings are managed at the MNC-parent (Beuselinck et al., 2019). Logically, subsidiaries are often located in countries with a lower institutional quality than the institutional quality of the MNC-parent’s home country (Beuselinck et al., 2019). Despite the fact that MNC-parents are responsible for the consolidated earnings, they consider the benefits to outweigh the costs of managing earnings through subsidiaries in countries with a low institutional quality (Beuselinck et al., 2019).
Thus, a low institutional quality of the subsidiaries’ home country can create opportunities to manage the earnings at the subsidiary-level, especially when the institutional quality of the MNC-parent’s home country is high. The MNC-parents of interest in this research are located in the United States, which is considered to have one of the highest institutional quality, in terms of legal, political and economic institutions (Kunčič, 2014; Durnev, 2017). Specifically, the United States is argued to have a strict legal environment (Durnev, 2017), strong reporting enforcement and high exposure to scrutiny (Beuselinck et al., 2019). This indicates that the MNC-parents located in the United States might manage their earnings at the subsidiary-level when their subsidiaries are located in countries with a lower institutional quality, which reduces the subsidiaries’ financial reporting quality.
Prior research indicates that governance characteristics affect the financial reporting quality of firms (Bushman and Smith, 2001), especially in countries with a low institutional quality (Bushman and Piotroski, 2006). Specifically, firms located in a country with a low institutional quality could benefit more from an independent board of directors (Dahya et al., 2008), due to the absence of a strong external monitoring mechanism. Also, shareholders might demand strong firm-level governance in countries without a strong country-level governance (Kim et al., 2007). This implies the need for a strong internal monitoring mechanism when the country’s institutional quality is low. Since the board of directors is seen as a key corporate governance mechanism (Alves, 2014), it is considered to act as a strong internal monitoring mechanism.
When subsidiaries are located in countries with a low institutional quality, they are more likely to engage in earnings management (Beuselinck et al., 2019; Durnev et al., 2017; Dyreng et al., 2012). Since the MNC-parent’s board monitors the financial reporting quality of the consolidated financial statements, and due to the consolidation process, is concerned about the financial reporting quality of its subsidiaries, MNC-parent’s board independence becomes an
even more important internal monitoring mechanism in the absence of a strong external monitoring mechanism. Conversely, subsidiaries that are located in countries with a high institutional quality have less tendency to manipulate earnings (Beuselinck et al., 2019), because the stronger legal environment serves as an external monitoring mechanism (Fama and Jensen, 1983). The strong external monitoring mechanism might work as a substitute for high levels of board independence.
Summarizing, I expect that the level of earnings management is higher when the institutional quality of the home country of the subsidiary is lower. Prior studies have argued that there is an inverse relation between the institutional quality of the home country and the level of earnings management (e.g. Ball et al., 2003; Leuz et al., 2003, Beuselinck et al., 2019) and the incremental need for a strong internal monitoring mechanism in the absence of a strong external monitoring mechanism (Dahya et al., 2008; Kim et al., 2007). MNC-parent’s board independence might be incrementally beneficial in curbing the level of earnings management in the consolidated financial statements by limiting the level of earnings management reported by its foreign subsidiaries, which are likely to be higher when the institutional quality of the subsidiaries’ home country is low. Therefore, I predict that the positive relation between the MNC-parents’ board independence and their subsidiaries’ financial reporting quality is stronger when the institutional quality of the home country of the subsidiaries is lower. This reasoning leads to the following hypothesis:
H2: The positive relation between the board independence of the MNC-parent and the financial reporting quality of its subsidiaries is stronger when the institutional quality of the subsidiaries’
home country is lower.
3. METHODOLOGY Sample selection
The sample selection starts with identifying all listed firms from the U.S. using Compustat.
Next, I discard firms that do not have the necessary financial information to carry out the research. Furthermore, I exclude MNC-parents with missing board related information. The financial information of the parents is obtained from Compustat, while the information about the parents’ board is gathered from the MSCI (former KLD and GMI) database. Then, for the listed non-financial MNC-parents located in the U.S., I use Exhibit 21.1 or 21 of the 10-K filings from the Edgar database to hand-collect the name and jurisdiction of all material subsidiaries
for each parent during the period 2011-2017. Next, I match the name and the jurisdiction of the material subsidiaries with those of the firms covered by the Orbis database. Given the coverage of Orbis, I select only the privately held, non-financial subsidiaries located in 30 European countries (EU, Norway and Switzerland). Then, I delete the subsidiaries operating in the financial industry, because it is difficult to estimate the discretionary accruals for financial firms (Klein, 2002). The financial information of the subsidiaries is collected from the Orbis database.
I discard the subsidiaries with unavailable information. I use the Worldwide Governance Indicators (WGI) database to obtain the rule of law, as a proxy for the institutional quality of the subsidiaries’ home country. The final sample consists of 8,205 firm-year observations corresponding to 394 unique MNCs and 2,186 unique subsidiaries over the period 2011-2017.
Table 1 presents the distribution of the subsidiary firm-year observations in the final sample over the different years, countries and industries. Table 1, Panel A presents the distribution for the period 2011-2017. It shows that the firm-year observations are increasing over the years.
This is due to the availability of the financial information in Orbis.
TABLE 1 Sample distribution Panel A: Distribution by year
Year Freq. Percent
2011 854 10.41
2012 995 12.13
2013 1,073 13.08
2014 1,167 14.22
2015 1,241 15.12
2016 1,403 17.10
2017 1,472 17.94
Total 8,205 100.00
Table 1, Panel B shows the distribution of the subsidiary firm-year observations across the different countries. The subsidiaries in the final sample are located in 21 European countries.
France is the most represented country (21.62 percent), followed by Italy (16.45 percent), Germany (15.04 percent), Belgium (8.86) and Spain (7.20 percent).
TABLE 1 (continued)
Panel B: Distribution by the home country of the subsidiaries
Country Freq. Percent
Austria 186 2.27
Belgium 727 8.86
Bulgaria 74 0.90
Croatia 52 0.63
Czech Republic 420 5.12
Denmark 40 0.49
Finland 115 1.40
France 1,774 21.62
Germany 1,234 15.04
Greece 85 1.04
Hungary 106 1.29
Italy 1,350 16.45
Luxembourg 26 0.32
Netherlands 58 0.71
Norway 402 4.90
Poland 377 4.59
Portugal 157 1.91
Romania 138 1.68
Slovenia 28 0.34
Spain 591 7.20
Sweden 265 3.23
Total 8,205 100.00
Table 1, Panel C presents the distribution of the subsidiary firm-year observations across the industries. The subsidiaries of the final sample operate in 11 different industries according to the two-digit NACE Rev 2. industry classification of European economic activities (European Commission, 2008, p. 57). The most represented industry is the wholesale and retail industry (42.57 percent), followed by the manufacturing industry (26.81 percent).
TABLE 1 (continued)
Panel C: Distribution by industry in which the subsidiaries operate
NACE Rev 2. industry classification Freq. Percent
C: Manufacturing 2,200 26.81
G: Wholesale and retail trade; repair of motor vehicles and motorcycles 3,493 42.57
H: Transportation and storage 35 0.43
I: Accommodation and food service activities 98 1.19
J: Information and communication 799 9.74
L: Real estate activities 79 0.96
M: Professional, scientific and technical activities 947 11.54 N: Administrative and support service activities 440 5.36
Q: Human health and social work activities 10 0.12
R: Arts, entertainment and recreation 53 0.65
S: Other service activities 51 0.62
Total 8,205 100.00
Financial reporting quality
The dependent variable is the financial reporting quality of the subsidiaries of the MNCs.
Financial reporting quality is widely measured through earnings management (Beuselinck et al., 2019; Dyreng et al., 2012; Durnev et al., 2017). Researchers often use discretionary accruals to measure the level of earnings management (Dechow et al., 1995; Bonacchi et al., 2018;
Durnev et al., 2017; Kao and Chen, 2004; Ebrahim, 2007; Xie et al., 2003) and commonly use the original Jones model (Jones, 1991) or the modified Jones model (Dechow et al., 1995) to estimate them. It is argued that the original Jones model does not capture earnings management if earnings are managed through discretionary revenues. Therefore, Dechow et al. (1995) propose the modified Jones model to “eliminate the conjectured tendency of the Jones model
to measure discretionary accruals with error when discretion is exercised over revenues”
(Dechow et al., 1995, p. 199). In this study, I use the modified Jones model since this model is proven to be more powerful in detecting earnings management than the original Jones model (Dechow et al., 1995). First, I calculate the total accruals using the following equation:
𝑇𝐴𝐶𝐶 = 𝛥𝐶𝐴 − 𝛥𝐶𝑎𝑠ℎ − 𝛥𝐶𝐿 + 𝛥𝐷𝐶𝐿 − 𝐷𝐸𝑃 Where
TACCt = total accruals in year t
ΔCAt = change in current assets in year t
ΔCasht = change in cash and cash equivalents in year t ΔCLt = change in current liabilities in year t
ΔDCLt = change in short term debt in current liabilities in year t DEPt = depreciation and amortization expense in year t
Second, I estimate the discretionary accruals of the subsidiaries using the modified Jones model as proposed by Dechow et al. (1995). The discretionary accruals are used as a proxy for earnings management and are estimated for each industry and year combination. The discretionary accruals of the subsidiaries (DACCSUB) are defined as absolute values, since this research is only interested in the extent of earnings management, regardless of its direction (Kim and Yi, 2006; Beuselinck et al., 2019).
𝑇𝐴𝐶𝐶
𝑇𝐴 = α 1
𝑇𝐴 + α (𝛥𝑅𝐸𝑉 − 𝛥𝑅𝐸𝐶 )
𝑇𝐴 + α 𝛥𝑃𝑃𝐸
𝑇𝐴 + ϵ
Where
TACCt = total accruals in year t TAt-1 = total assets in year t - 1 ΔREVt = change in revenues in year t ΔRECt = change in receivables in year t
PPEt = property, plant and equipment in year t ϵt = estimated discretionary accruals
Board independence
The board independence of the MNC-parent is the main independent variable in this study.
Consistent with previous research (Klein, 2002), I categorize directors as dependent, independent and affiliated. Whereas dependent directors are current employees, independent directors have no ties with the MNC-parent. Affiliated directors are past employees, have family ties with the CEO or have business relationships (Klein, 2002). The board independence of the MNC-parents (BOD_INDMNC) is measured as the number of independent directors divided by the total number of board members (Peasnell et al., 2005; Alves, 2014).
Institutional quality
Institutional quality is the moderating variable in this study. Prior research (Beuselinck, 2019;
Dyreng et al., 2012) used the Worldwide Governance Indicators (WGI) database created by the World Bank to measure the institutional quality of a country. This database captures six indicators of institutional quality, including the rule of law index. The rule of law index is “a summary indicator of the extent of compliance with laws and regulations that, by affecting investor protection, extent of self-dealing, and audit quality, influence managerial reporting behaviour” (Beuselinck et al., 2019, p. 64). The rule of law is computed annually and ranges from weak to strong corporate governance within the values -2.5 till 2.5. In line with previous studies that examined the effect of the institutional quality of the subsidiaries’ home country (Beuselinck, 2019; Dyreng et al., 2012), I use the RULE_OF_LAWSUB as a proxy for the institutional quality of the subsidiaries’ home country. The RULE_OF_LAWSUB is measured as the annual rule of law of the subsidiaries’ home country (Beuselinck et al., 2019). When the MNCs’ foreign subsidiaries are located in countries with a strong (weak) rule of law, i.e. a high (low) institutional quality, the RULE_OF_LAWSUB will be high (low).
Control variables
Prior research has found several variables that are associated with financial reporting quality for which I control in my empirical models. There is mixed evidence of the influence of firm size on the level of earnings management. On the on hand, larger firms are associated with lower levels of earnings management, since they face higher political costs due to higher scrutiny (Jiang et al., 2008; Peasnell et al., 2000). On the other hand, larger firms may use aggressive accounting policies to avoid scrutiny by banks and analysts (Chen et al., 2007). The mixed evidence suggests that the size of the subsidiary can be positively and negatively associated with earnings management. Furthermore, the parents of larger MNCs are
hypothesized to be better able to push down earnings management to their subsidiaries, thereby reducing the level of earnings management in their individual financial statements (Beuselinck et al., 2019). This implies that larger MNCs are more likely to manage earnings through their subsidiaries. Hence, I control for MNC (SIZEMNC) and subsidiary (SIZESUB) size, which are measured as the natural logarithm of the book value of their total assets (Beuselinck et al., 2019).
Leverage can also have a positive or negative influence on the level of earnings management.
Highly leveraged firms may have incentives to engage in earnings management due to debt covenant violations (Davidson, 2005; DeFond and Jiambalvo, 1994; Jiang et al., 2008; Klein, 2002), However, highly leveraged firms can also be hesitant to use earnings management because “they are under close scrutiny by lenders” (Alves, 2014, p. 30). Moreover, highly leveraged MNCs are more likely to use their subsidiaries to engage in earnings management.
The financial constraints associated with high leverage can cause the MNC-parent to impose specific reporting guidelines on its subsidiaries to manage the earnings upward (Beuselinck et al., 2019). Hence, I control for MNC (LEVMNC) and subsidiary (LEVSUB) leverage, which are both measured as their total debt divided by the book value of total assets (Klein, 2002).
According to Dechow et al. (2010), earnings management is related to firm performance. Firms that perform poorly are more inclined to engage in earnings management (Defond and Park, 1997). When the MNC performs poorly, the MNC-parent can use its subsidiaries to manage the earnings upward (Beuselinck et al., 2019). Thus, I control for firm performance for the MNC (ROAMNC) and the subsidiary (ROASUB). I use the return on assets as a proxy for firm performance, which is the net income before extraordinary items scaled by the book value of total assets (Beuselinck et al., 2019).
Dechow et al. (2010) also argue that high growth firms, measured in terms of sales growth, are associated with lower earnings persistence due to more earnings management opportunities.
This also implies that high growth MNCs have more opportunities to engage in earnings management at the subsidiary-level. Therefore, I control for firm growth of both the MNC (GROWTHMNC) and the subsidiary (GROWTHSUB), which is calculated as the annual change in sales (Beuselinck et al., 2019). Following Beuselinck et al. (2019), I also control for losses of the MNC (LOSSMNC) and the subsidiary (LOSSSUB). I define these controls as a dummy variables equal to one if the net income is negative, and zero otherwise.
Duality of the chairman is found to increase the prevalence of earnings management (Dechow et al., 1996), because duality strengthens the power of the CEO and reduces the board oversight (Alabdullah, 2016). Indeed, Fama (1980) and Fama and Jensen (1983) argue that when the chairman of the board is also the CEO of the firm, it will lower the objectivity of the board of directors. Following the agency theory, this implies less effective board monitoring. I control for the duality of the chairman of the board of directors of the MNC-parent (CEO_DUALITYMNC), which is a dummy variable equal to one if the chairman of the board is also the CEO, and zero otherwise.
Also, MNCs are more likely to engage in more earnings management when they have subsidiaries located in tax havens, because “tax havens allow earnings management in a tax free manner, which implies that taxes are costly for the firm” (Dyreng et al., 2012, p. 658). In my sample, Austria, Belgium, Cyprus, Ireland, Luxembourg, Malta, the Netherlands and Switzerland are considered to be a tax haven according to Menkhoff and Miethe (2019). Hence, I add a dummy variable to control for the subsidiaries that are located in tax havens (TAX_HAVENSUB), with a value equal to one when the subsidiary is located in a tax haven, and zero otherwise.
Lastly, I control for year (YEARDUM), country (COUNTRYDUM) and industry (INDUSTRYDUM) effects by using dummy variables. To control for the industry, I use the two-digit NACE Rev 2. industry classification (European Commission, 2008). I winsorize all continuous variables at the 5th and 95th percentiles of their distribution.
Empirical model
Since the financial reporting quality of the subsidiaries is a continuous variable, I test both hypotheses by using ordinary least squares (OLS) regression models. The equation of model (3) examines the influence of the independence of the MNC-parent’s board of directors (BOD_INDMNC) on the financial reporting quality of its subsidiaries (DACCSUB). Given my prediction that the financial reporting quality (level of earnings management) of the subsidiary is higher (lower) when the board of the MNC-parent has a higher proportion of independent directors, I expect the coefficient of BOD_INDMNC to be negative.
𝐷𝐴𝐶𝐶 = 𝛽 + 𝛽 𝐵𝑂𝐷_𝐼𝑁𝐷 + 𝛽 𝑆𝐼𝑍𝐸 + 𝛽 𝑆𝐼𝑍𝐸 + 𝛽 𝐿𝐸𝑉 + 𝛽 𝐿𝐸𝑉 + 𝛽 𝑅𝑂𝐴 + 𝛽 𝑅𝑂𝐴 + 𝛽 𝐺𝑅𝑂𝑊𝑇𝐻
+ 𝛽 𝐺𝑅𝑂𝑊𝑇𝐻 + 𝛽 𝐿𝑂𝑆𝑆 + 𝛽 𝐿𝑂𝑆𝑆 + 𝛽 𝐷𝑈𝐴𝐿𝐼𝑇𝑌
+ 𝛽 𝑇𝐴𝑋_𝐻𝐴𝑉𝐸𝑁 + 𝛽 𝑌𝐸𝐴𝑅 + 𝛽 𝐶𝑂𝑈𝑁𝑇𝑅𝑌 + 𝛽 𝐼𝑁𝐷𝑈𝑆𝑇𝑅𝑌 + 𝜖
To measure the moderating effect of the institutional quality on the relationship between the board independence of the MNC-parent and the financial reporting quality of its subsidiaries, I use the model of equation (4). This model is an extension of model (3), which I obtained by adding the rule of law (RULE_OF_LAWSUB) as a proxy of the institutional quality and the interaction term (BOD_INDMNC *RULE_OF_LAWSUB). Given my prediction that a lower institutional quality strengthens the positive relationship between the board independence of the MNC-parent and the financial reporting quality of the subsidiaries, I expect the interaction term (BOD_INDMNC *RULE_OF_LAWSUB) to be positive.
𝐷𝐴𝐶𝐶 = 𝛽 + 𝛽 𝐵𝑂𝐷_𝐼𝑁𝐷 + 𝛽 𝑅𝑈𝐿𝐸_𝑂𝐹_𝐿𝐴𝑊 + 𝛽 𝐵𝑂𝐷_𝐼𝑁𝐷
∗ 𝑅𝑈𝐿𝐸_𝑂𝐹_𝐿𝐴𝑊 + 𝛽 𝑆𝐼𝑍𝐸 + 𝛽 𝑆𝐼𝑍𝐸 + 𝛽 𝐿𝐸𝑉 + 𝛽 𝐿𝐸𝑉 + 𝛽 𝑅𝑂𝐴 + 𝛽 𝑅𝑂𝐴 + 𝛽 𝐺𝑅𝑂𝑊𝑇𝐻
+ 𝛽 𝐺𝑅𝑂𝑊𝑇𝐻 + 𝛽 𝐿𝑂𝑆𝑆 + 𝛽 𝐿𝑂𝑆𝑆 + 𝛽 𝐷𝑈𝐴𝐿𝐼𝑇𝑌
+ 𝛽 𝑇𝐴𝑋_𝐻𝐴𝑉𝐸𝑁 + 𝛽 𝑌𝐸𝐴𝑅 + 𝛽 𝐶𝑂𝑈𝑁𝑇𝑅𝑌 + 𝛽 𝐼𝑁𝐷𝑈𝑆𝑇𝑅𝑌 + 𝜖
4. RESULTS Descriptive statistics
Table 2 provides the descriptive statistics of the final sample. The average of the absolute value of the discretionary accruals of the subsidiaries (DACCSUB), the dependent variable, is 0.164.
The main independent variable, which is the board independence of the MNC-parents (BOD_INDMNC), has an average of 0.778; meaning that 77.8 percent of the directors of the boards of MNC-parents are independent directors. The mean (median) value of the rule of law of the subsidiaries’ home country (RULE_OF_LAWSUB) is 1.211 (1.413). The minimum is -0.112, which is quite high considering that the value of the rule of law ranges between -2.5 and 2.5 worldwide. This is considered to be high especially since prior studies show that the subsidiaries are likely to be located in countries with a low institutional quality due to the opportunities of MNCs to manage the earnings abroad (Beuselinck et al., 2019; Durnev et al., 2017; Dyreng et al., 2012). However, the relatively high mean and minimum are not surprising since the final sample consists of subsidiaries located in European countries only. As expected, the average size of the MNCs (SIZEMNC) is higher than the average size of their subsidiaries (SIZESUB), which are 8.724 and 2.863 respectively. Furthermore, the average leverage of MNCs (LEVMNC) is 0.234 while the average of their subsidiaries (LEVSUB) is 0.035. This means that the total debt of the MNCs (subsidiaries) constitutes 23.4 percent (3.5 percent) of the value of the total assets. The average return on assets of MNCs (ROAMNC) is slightly higher than the average of the subsidiaries (ROASUB), with a value of respectively 0.061 and 0.050. This means that the MNCs as well as their subsidiaries have a positive return on their assets. The sales of the MNCs (GROWTHMNC) grow 4.0 percent while their subsidiaries’ sales (GROWTHSUB) grow 2.1 percent per year on average. Based on the reported losses from MNCs (LOSSMNC) and their subsidiaries (LOSSSUB), the subsidiaries are on average more likely to report a loss; 20.4 percent of the subsidiaries firm-year observations reported a loss compared to 10.1 percent of the MNCs. On average, in 49.8 percent of the MNC firm-year observations, the CEO of the MNC- parent is also the chairman of the MNC-parent’s board of directors (CEO_DUALITYMNC).
Lastly, 12.2 percent of the subsidiaries firm-year observations (TAX_HAVENSUB) are located in a tax haven.
TABLE 2 Descriptive statistics
Variables N Mean Median Std. Dev. Min 25% 75% Max
DACCSUB 8,205 0.164 0.098 0.177 0.007 0.039 0.219 0.658 BOD_INDMNC 8,205 0.778 0.818 0.149 0.000 0.714 0.900 1.000 RULE_OF_LAWSUB 8,205 1.211 1.413 0.562 -0.112 0.802 1.616 2.100 SIZEMNC 8,205 8.724 8.544 1.494 6.339 7.615 9.672 11.780 SIZESUB 8,205 2.863 2.817 1.744 -0.233 1.581 4.144 6.157 LEVMNC 8,205 0.234 0.230 0.139 0.000 0.134 0.329 0.498 LEVSUB 8,205 0.035 0.000 0.098 0.000 0.000 0.000 0.385 ROAMNC 8,205 0.061 0.057 0.048 -0.030 0.030 0.092 0.156 ROASUB 8,205 0.050 0.046 0.093 -0.171 0.008 0.098 0.244 GROWTHMNC 8,205 0.040 0.035 0.094 -0.140 -0.016 0.092 0.245 GROWTHSUB 8,205 0.021 0.032 0.200 -0.477 -0.064 0.127 0.409 LOSSMNC 8,205 0.101 0.000 0.301 0.000 0.000 0.000 1.000 LOSSSUB 8,205 0.204 0.000 0.403 0.000 0.000 0.000 1.000 CEO_DUALITYMNC 8,205 0.498 0.000 0.500 0.000 0.000 1.000 1.000 TAX_HAVENSUB 8,205 0.122 0.000 0.327 0.000 0.000 0.000 1.000
This table presents the descriptive statistics of all the variables used in this study based on the final sample.
All variables are defined in Appendix A.
Correlations
Table 3 contains the correlations between all the variables. In line with Klein (2002), I use Spearman correlations. The coefficient of the correlation between the discretionary accruals of the subsidiaries (DACCSUB) and the board independence of the MNC-parents (BOD_INDMNC) is negative and significant at the 10 percent level. This indicates that the subsidiaries’ level of earnings management is lower when the board independence of the MNC-parents is higher. The rule of law of the countries where the subsidiaries are located (RULE_OF_LAWSUB) is also negatively correlated with the discretionary accruals of the subsidiaries (DACCSUB) at the 10 percent significance level. This suggests that when the institutional quality of the subsidiaries’
home country is higher, subsidiaries engage less in earnings management and so their financial
reporting quality is higher. The rule of law of the subsidiaries (RULE_OF_LAWSUB) is also negatively correlated with the MNC-parents’ board independence (BOD_INDMNC) at the 1 percent significance level. This implies that MNC-parents with a less independent board have subsidiaries located in countries with a higher institutional quality. This is in line with the reasoning that the external monitoring mechanism might work as a substitute for high levels of board independence.
Furthermore, the size of the subsidiary (SIZESUB) is negatively correlated with the subsidiaries’
discretionary accruals (DACCSUB) at the 1 percent significance level. This implicates that the subsidiaries’ level of earnings management is lower, when the subsidiaries are larger. The MNCs’ return on assets (ROAMNC) is also negatively correlated with the subsidiaries’
discretionary accruals (DACCSUB) at the 1 percent significance level. Next, the subsidiaries’
return on assets (ROASUB), the level of growth of both the MNC (GROWTHMNC) and the subsidiary (GROWTHSUB) and the likelihood that the MNC reports a loss (LOSSMNC) as well as the of the subsidiaries (LOSSSUB) are positively correlated with the discretionary accruals of the subsidiaries (DACCSUB) at the 1 percent significance level. However, the subsidiaries’
discretionary accruals (DACCSUB) are not related to the size of the MNCs (SIZEMNC), the leverage of both the MNCs (LEVMNC) and the subsidiaries (LEVSUB), the CEO duality of the MNC-parents (CEO_DUALITYMNC) and to the likelihood of having subsidiaries located in tax havens (TAX_HAVENSUB). Finally, the correlations between each the variables do not exceed the threshold of 0.7, which indicates that there are no multicollinearity issues between the variables in my model.
TABLE 3 Correlation matrix
***, **, * Indicate statistical significance at the 1 percent, 5 percent and 10 percent levels, respectively.
This table presents the matrix of Spearman correlation coefficients of all the variables used in this study based on the final sample.
All variables are defined in Appendix A.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 DACCSUB 1.000
2 BOD_INDMNC -0.022* 1.000
3 RULE_OF_LAWSUB -0.020* -0.054*** 1.000
4 SIZEMNC -0.015 0.335*** -0.045*** 1.000
5 SIZESUB -0.234*** 0.173*** 0.150*** 0.380*** 1.000
6 LEVMNC -0.014 0.010*** -0.040*** 0.194*** 0.074*** 1.000
7 LEVSUB -0.014 -0.011 0.134*** 0.005 0.202*** 0.004 1.000
8 ROAMNC -0.035** 0.174*** -0.019* 0.150*** 0.081*** -0.138*** 0.016 1.000
9 ROASUB 0.033*** 0.048*** 0.055*** 0.007 -0.016 0.081*** -0.095*** 0.104*** 1.000
10 GROWTHMNC 0.053*** -0.154*** 0.020* -0.150*** -0.147*** 0.014 0.005 0.065*** 0.064*** 1.000
11 GROWTHSUB 0.080*** -0.022* -0.069*** -0.052*** -0.040*** -0.027** -0.020* 0.041*** 0.151*** 0.177*** 1.000
12 LOSSMNC 0.046*** -0.131*** 0.008 -0.180*** -0.090*** 0.071*** -0.021* -0.512*** -0.027** -0.029*** -0.012 1.000
13 LOSSSUB 0.090*** -0.068*** 0.005 -0.062*** -0.091*** -0.056*** 0.086*** -0.081*** -0.698*** -0.023** -0.104*** 0.018* 1.000
14 CEO_DUALITYMNC -0.008 0.154*** -0.039*** 0.265*** 0.135*** -0.006 -0.001 0.116*** -0.066*** -0.029 -0.029*** -0.116*** -0.009 1.000 15 TAX_HAVENSUB -0.001 0.011 0.182*** 0.041*** 0.053*** -0.004 0.126*** 0.001 -0.039*** -0.011 -0.016 0.007 -0.012 0.035*** 1.000
Regression analysis
Table 4 shows the results of the regression analysis based on my two models. For both models, I used the ordinary least squares (OLS) regression model for panel data. Column (1) reports the results when testing the first hypothesis according to model (3). This model includes the independent variable (BOD_INDMNC), the dependent variable (DACCSUB) and all control variables. The coefficient of BOD_INDMNC is not significant, suggesting that there is no relation between the MNC-parent’s board independence and the subsidiaries’ discretionary accruals.
Therefore, the results do not support hypothesis 1 which predicted that a higher proportion of independent directors on the MNC-parents’ boards would lead to lower levels of subsidiaries’
earnings management, and thus a higher financial reporting quality.
Column (2) reports the results when testing the second hypothesis according to model (4). This model adds the proxy for the institutional quality (RULE_OF_LAWSUB) and the interaction term (BOD_INDMNC *RULE_OF_LAWSUB) to the previous model. The interaction term measures the incremental effect of the institutional quality of the subsidiaries’ home country on the relation between MNC-parents’ board independence and the financial reporting quality of their subsidiaries. The coefficient of the interaction term is not statistically significant, suggesting that the institutional quality does not have an effect on the relation between the MNC-parent’s board independence and the subsidiaries’ level of earnings management. This means that the results do not support hypothesis 2 which predicted that a low institutional quality of the subsidiaries’ home country would strengthen the relation between the MNC-parents’ board independence and the subsidiaries’ level of earnings management. Nonetheless, this estimation does show that the institutional quality (RULE_OF_LAWSUB) has a negative coefficient which is significant at the 5 percent level. This indicates that a higher institutional quality in the subsidiaries’ home country leads to lower levels of earnings management, and thus a higher financial reporting quality.
TABLE 4 Regression analysis
Variables (1) (2)
BOD_INDMNC 0.016 -0.004
(0.016) (0.030)
RULE_OF_LAWSUB -0.066**
(0.033)
BOD_INDMNC *RULE_OF_LAWSUB 0.018
(0.023)
SIZEMNC 0.010*** 0.010***
(0.002) (0.002)
SIZESUB -0.027*** -0.027***
(0.002) (0.002)
LEVMNC -0.010 -0.010
(0.019) (0.019)
LEVSUB 0.080*** 0.080***
(0.025) (0.025)
ROAMNC -0.073 -0.075
(0.057) (0.057)
ROASUB 0.124*** 0.125***
(0.031) (0.031)
GROWTHMNC 0.041* 0.042*
(0.023) (0.023)
GROWTHSUB 0.062*** 0.061***
(0.009) (0.009)
LOSSMNC 0.006 0.006
(0.008) (0.008)
LOSSSUB 0.051*** 0.051***
(0.007) (0.007)
CEO_DUALITYMNC 0.005 0.005
(0.005) (0.005)
TAX_HAVENSUB 0.068 0.111**
(0.049) (0.054)
Constant 0.068 0.137**
(0.051) (0.062)
Year effect yes yes
Country effect yes yes
Industry effects yes yes
R2 0.102 0.103
Observations 8,205 8,205
***, **, * Indicate statistical significance at the 1 percent, 5 percent and 10 percent levels, respectively.
This table presents the OLS coefficients and in parentheses the standard errors of all the variables used in this study based on the final sample.
All variables are defined in Appendix A.
Additional test
To address a possible endogeneity problem between the variables in my two models, I perform the Hausman test. The results of the Hausman test (p=0.00 for both models) indicate the existence of endogeneity (Ayers et al., 2011). Therefore, I estimate the two models using firm fixed effects which control for within-year invariant factors at the subsidiary-level. This robustness test accounts for potential omitted variables related to invariant factors that might affect my results. The results are reported in Appendix B. The results show omitted variables for TAX_HAVENSUB, country effects and industry effects. However, in line with the regression results of table 4, the results of the robustness test also do not show a significant coefficient for the MNC-parent’s board independence and the interaction term.
5. CONCLUSION
This study examined the relationship between the MNC-parent’s board independence and the financial reporting quality of its foreign subsidiaries. A MNC-parent can exert influence over its subsidiaries (Doz and Prahalad, 1981; Beuselinck et al. 2019) and has more opportunities to manage their earnings compared to stand-alone firms (Beuselinck et al., 2019). Therefore, it is important to investigate if the MNC-parent’s board of directors is able to protect the interests of the shareholders and curb the level of earnings management in the consolidated financial statements by limiting also the levels of earnings management reported by its subsidiaries.
Following the agency theory, independent directors are believed to be better monitors and thus better able to constrain earnings management (e.g. Alves, 2014; Beasley, 1996; Fama and Jensen, 1983; Ebrahim, 2007). Since the consolidated financial statements of the MNCs are obtained from the financial statements of their subsidiaries, the financial reporting quality of the material subsidiaries is likely to influence the financial reporting quality of the consolidated earnings. Specifically, the MNC-parent’s board is likely to be concerned about the financial reporting quality of its subsidiaries as they try to enhance the financial reporting quality of the MNC’s consolidated financial statements. So, a higher proportion of independent directors on the MNC-parent’s board of directors is an important governance mechanism to curb the level of earnings management and thus increase the financial reporting quality of its subsidiaries.
Therefore, I expected a positive relation between the board independence of the MNC-parents and the financial reporting quality of their subsidiaries. However, my results do not support the first hypothesis as they suggest that MNC-parent’s board independence is not an effective mechanism to curb the subsidiaries’ level of earnings management. This can have different possible explanations. First, according to Bradbury et al. (2006), the independence of the directors on the audit committee, rather than the board of directors, is important to constrain the level of earnings management. This might be due to the different roles; the audit committee primarily monitors the financial reporting process (Klein, 2002), whereas the board of directors has a much broader role of protecting the interests of the shareholder (Fama and Jensen, 1983).
Second, the geographical distance between the MNC-parent and its subsidiaries can lead to high costs for the MNC-parent when gathering information about its subsidiaries (Beuselinck et al., 2011). The higher costs can be a result of more difficult interaction (Beuselinck et al., 2011) due to the more complex corporate organizational structure of MNCs compared to stand-alone firms (Chin et al., 2009). This could implicate that the MNC-parent’s board might focus on what happens at the parent-level and disregard the activities of the foreign subsidiaries. Third,
