The effect of audit quality of the MNC on the financial reporting quality of the subsidiary

1

The effect of audit quality of the MNC on the

financial reporting quality of the subsidiary

Name: Gerrit Calkhoven Student number: 2573881

Master Accountancy, Rijksuniversiteit Groningen Supervisor: Mrs. Rusanescu

Date: June, 2019 Word count: 6.333

2

Abstract

This study examines the relation between the audit quality of the multinational corporation (MNC) and financial reporting quality of the subsidiary and the effect that a fully independent audit committee has on this relation. Studies have indicated that parent companies could affect the financial reporting quality of their subsidiaries by managing their earnings. A sample of 1.871 material subsidiaries and a sample period of 2015-2017 resulted in a regression of 5.605 firm-year observations. I found no significant relation between audit quality of the parent company and financial reporting quality of the subsidiary. Also audit committee independence appears to have no significant effect on this relation.

Key words: Multinational corporations, subsidiaries, audit quality, financial reporting quality, audit committee independence.

3

Table of content

4

1. Introduction

The 2018 report by the United Nations Conference of Trade and Development (UNCTAD) shows that the foreign operations of the top 100 global multinational corporations (MNCs) represented 9 percent of the world foreign assets, 17 percent of world foreign sales and 13 percent of foreign employment. These top 100 MNCs have 70% of their total assets invested abroad, and about half of the world gross domestic product (GDP) stems from foreign subsidiaries of MNCs, according to a report of the UNCTAD in 2014. This emphasizes the importance of foreign subsidiaries for big MNCs.

The aim of this study is to examine the relationship between the financial reporting quality (FRQ) of the subsidiary and audit quality of the MNC. This results in the following research question: ‘how does the subsidiary financial reporting quality relate to the MNC audit quality?

Central in this study is a dimension of financial reporting quality; namely the extent to which managers alter reported earnings for their own benefit (Beneish, 2001), or in other words, earnings management. While most studies with respect to earnings management focus on ways and incentives to manage earnings (Bergstresser & Philippon, 2006; Dechow & Skinner, 2000), this study takes the parent – subsidiary relation into account.

Bonacchi et al. (2018) find evidence that parent companies drive the earnings management of subsidiaries so that the parent can meet specific benchmarks. Beuselinck et al. (2018) show that the MNC headquarter can exert influence over the reporting practices of the foreign subsidiaries. Given the importance of financial reporting quality for firms and negative consequences of poor financial reporting quality for firms, MNCs have incentives to manage earnings through a reporting strategy across subsidiaries over which they exert significant influence. Furthermore, I examine whether the independence of the audit committee of the MNC affects the relation between audit quality at the MNC and the FRQ of the subsidiary.

This research focuses on a sample with 1.871 material subsidiaries of U.S. MNCs and covers the period 2015-2017. The lists of material subsidiaries are hand-collected from Exhibit 21 of the 10-K from the Edgar database. Necessary financial data of the subsidiary is obtained from the database Orbis, while data of the parent company is obtained from Compustat. I obtain data of the composition of the audit committees from the database MSCI (formerly KLD and GMI). In this study, earnings management is used as a measure of financial reporting quality. The level of earnings management is measured with the discretionary accruals, which I estimate using the modified Jones model (Dechow & Dichev, 2002). The measure of audit quality is based on the size of the auditor (Big 4 vs non-Big 4), which is one of the most popular measures in extant literature (DeAngelo, 1981; Lai, 2009). A Big 4 audit firm is expected to provide higher quality audit services, because they can spend more on training and other resources (DeAngelo, 1981), which increases the knowledge and competence of big auditors. Furthermore, Big 4 audit firms also have higher litigation risk and greater

reputational risk, so they also have more incentives to provide higher quality audit services. Finally, the audit committee is considered as independent when it fully consists of outside directors (Klein, 2002).

5

The findings of this study contribute to the existing literature on the relation between audit quality and financial reporting quality. To the best of my knowledge, the relation between subsidiary FRQ and the MNC audit quality has not been investigated before. Research has been done about the association between FRQ and audit quality (Francis et al., 1999; Chen et al., 2005; Becker et al., 1998), but never about a parent-subsidiary relation. The UNCTAD reports made clear that subsidiaries are crucial for big MNCs.

Furthermore, the findings of this study contribute to the literature on the financial reporting quality of MNCs. Beuselinck et al. (2018) and Bonacchi et al. (2018) also examined the financial reporting quality in MNCs, but this study takes the audit quality of the MNC into account.

The remaining of this paper is as followed: The next section discusses the theoretical

framework and hypotheses development. Section 3 describes the research methodology, while section 4 provides results of the empirical tests. Lastly, section 5 includes the conclusions and implications of this research.

6

2. Theoretical framework

Financial reporting quality is a complex concept, which is why various definitions have been used in previous research. For instance, Verdi (2006) defined financial reporting quality as ‘the precision with which financial reporting conveys information about the firm’s operations, in particular its expected cash flows, in order to inform equity investors.’ Jonas and Blanchet (2000) argue that quality financial reporting is ‘full and transparent financial information and not designed to mislead the users.’ These definitions are consistent with the Financial

Accounting Standards Board Statement of Financial Accounting Concepts No. 1 (1978), which states that one objective of financial reporting is to inform present and potential investors in making rational investment decisions and in assessing the expected firm cash flows.

Financial reporting quality is crucial for decision makers (McNichols & Stubben, 2008; Verdi, 2006), and research shows that poor FRQ has negative consequences for the firm. For instance, poor FRQ could have a negative effect on the investment efficiency (Biddle et al., 2009; Gomariz & Ballesta, 2014; Verdi, 2006). McNichols & Stubben (2008) find evidence that poor FRQ leads to overinvestment, because managers use distorted information to base their investment decisions on. Furthermore, FRQ has an effect on the corporate dividend policy of firms. Poor FRQ is associated with lower dividends and thus could have a

significant impact on a firms’ value (Koo et al., 2017). According to Sun et al. (2012), poor financial reporting quality has a negative impact on the value of corporate cash holdings. Based on these mentioned studies it could be said that poor financial reporting quality

exacerbates the level of information asymmetry between internal and external stakeholders of a firm.

One of the most popular ways of measuring financial reporting quality is earnings management. ‘Earnings management occurs when managers 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 & Wahlen, 1999). Earnings management can be divided into two categories: real earnings management and accruals earnings management (Lo, 2008). The difference between these two types of

earnings management is that real earnings management has consequences for the firms’ cash flows, while accruals earnings management does not affect cash flows (Roychowdhury, 2006). Obviously the level of earnings management is an inverse measure of financial reporting quality. The higher the level of earnings management, the lower the financial reporting quality, and vice versa.

Firms could have different incentives to engage in earnings management, and thus affect the financial reporting quality of the firm. These incentives show, according to Watts &

Zimmermann (1978), that managers make opportunistic choices to act in self-interest. A motive for earnings management is to satisfy the expectations of shareholders (Lo, 2008) and analysts (Degeorge et al., 1999). Managers could feel pressure of shareholders and analysts and therefore feel the need to report higher earnings. Another motive was

investigated by Bergstresser & Philippon (2006). They found evidence that managers whose overall compensation is more sensitive to the company’s share price, the higher the level of earnings management. Managers manipulate earnings to maximize their own compensation. The last incentive discussed in extant literature is the manipulation of earnings to avoid debt covenant violations (DeFond & Jiambalvo, 1994). They find that managers use earnings

7

management to increase earnings one year prior to debt covenant violations. This is called the ‘debt covenant hypothesis’ by Watts & Zimmerman (1978); the closer a firm comes to the violation of debt covenants, the more income-increasing accounting choices a firm will make to avoid these violations.

As discussed above, it is important for firms to report financial statements of high quality, due to all the negative consequences a low financial reporting quality could have. The external auditor plays a crucial role in monitoring this financial reporting quality (Cohen et al., 2004). Becker et al. (1998) argue that the external auditor acts as a constraint on the management’s choice of accounting procedures, whereby the effectiveness of this constraint is dependent of the audit quality. In existing literature, there is no universal recognized definition for audit quality. A definition that has been widely used in previous literature is the one from

DeAngelo (1981), which defines audit quality as ‘the market-assessed joint probability that a given auditor will both discover a breach in the client's accounting system, and report the breach.’

A number of studies have examined whether audit quality relates to financial reporting quality. Audit quality is typically proxied by the size of the auditor, where big auditors are expected to provide better quality audit services than smaller auditors. In this regard, Francis et al. (1999) found that firms with a big 6 auditor have lower discretionary accruals compared to firms with non-big 6 auditors, and thus have a higher financial reporting quality. Chen et al. (2005) investigated the relation between audit quality and earnings management for Taiwan IPO firms. They found that auditor size is associated with lower unexpected accruals and that high quality auditors constrain earnings management and provide more precise information. Becker et al. (1998) conclude that lower audit quality is associated with more accounting flexibility, because discretionary accruals are significantly higher for firms with non-big 6 auditors. Balsam et al. (2003) examined the effect of another dimension of audit quality, auditor industry specialization, on discretionary accruals of clients. They found evidence of a negative association between auditor industry specialization and absolute discretionary

accruals, which indicates that specialist auditors reduce earnings management by their clients. So, these studies have found that audit quality is positively related to financial reporting quality.

Earnings management can take place at the central management level of the parent company, but can also occur by choices of division or subsidiary managers (Dyreng et al., 2012).

Bonacchi et al. (2018) find evidence that parent companies drive the earnings management of subsidiaries so that the parent can meet specific benchmarks. Thus, the parent company could affect the financial reporting quality of their subsidiaries. A high audit quality of the parent company should mean that the auditor constraints the level of earnings management not only within the parent company, but also in the subsidiaries. This would lead to a higher financial reporting quality at the parent level and at the subsidiary level. Therefore, the following hypothesis is tested in this study:

H1: There is a positive relation between audit quality at the MNC level and FRQ at the subsidiary level.

Audit committees have been mandatory for companies listed on the U.S. stock exchanges since 1978 (Vanasco, 1994). The primary role of the audit committee is to help ensure high quality financial reporting (Bedard et al., 2004). The audit committee meets regularly with the firm’s outside directors and internal financial managers to review the firm’s financial

8

Sarbanes-Oxley Act was introduced with the goal to improve the financial reporting quality. It was a reaction to some major accounting scandals, like Enron and WorldCom. The Sarbanes-Oxley Act mandates that, among other things, at least one member of the audit committee has to be a financial expert. The Sarbanes-Oxley Act ensured that the role of the audit committee became more important (Holm & Laursen, 2007).

Various researchers have shown that audit committee characteristics influence financial reporting quality. For instance, Klein (2002) found a negative relation between audit

committee independence and abnormal accruals, which implies that a higher audit committee independence is associated with lower earnings management, and thus a higher financial reporting quality. This is consistent with the findings of Bedard et al. (2004). They found that a fully independent audit committee significantly reduce the likelihood of aggressive earnings management. Lin et al. (2006) found a negative relation between audit committee size and the occurrence of earnings restatement, but no significant relation between audit committee independence and earnings quality.

According to Abbott et al. (2004), the most important reason why an independent audit committee is associated with a higher financial reporting quality is that independent directors do not have psychological or economic ties that constrain their task to question management. An independent audit committee can thus better perform on their job to monitor financial reporting quality not only at the parent level, but also at the subsidiary level. This leads to the following hypothesis:

H2: The positive relation between audit quality at the MNC level and FRQ at the subsidiary level is stronger when the internal audit committee of the parent company is fully

9

3. Research methodology

Sample

This study is based on a sample of 1871 material subsidiaries of US multinational

corporations and covers the period 2015-2017. The first step in the sample selection is to use Compustat to identify non-financial listed MNCs located in the US during the period

analyzed. For each MNC and year, I hand-collect the list of material subsidiaries from Exhibit 21 of the 10-K (Securities Exchange Commission’s (SEC) Edgar database). Then I select only the subsidiaries that are non-financial private companies located in Austria, Belgium,

Bulgaria, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Luxembourg, Netherlands, Norway, Poland, Portugal, Romania, Spain, Sweden, or United Kingdom. I obtain the necessary financial data of the subsidiaries from the database Orbis and that of the parent company from the database Compustat. The data about the composition of the audit committees is obtained from the database MSCI (formerly KLD and GMI).

Variables FRQ

The dependent variable in this research is the subsidiary FRQ, proxied by the level of earnings management. The most popular model used to estimate the level of earnings management is the original Jones (1991) model. This model is able to decompose total accruals into discretionary and non-discretionary accruals (Islam et al., 2011). Dechow et al. (1995) proposed the modified Jones model, where changes in revenues are adjusted for the change in receivables. The reason for this change is that it is easier to manage earnings by exercising discretion over the recognition on revenue on credit sales than it is to manage earnings by exercising discretion over the recognition on revenue on cash sales, assuming that all changes in credit sales in the event period result from earnings management (Dechow et al., 1995). In this research, the modified Jones model is used to estimate discretionary

accruals, as this model is more powerful than the original Jones (1991) model (Dechow et al., 1995).

I will first calculate the total accruals as follows:

𝑇𝐴𝐶𝐶_𝑡 = ∆𝐶𝐴_𝑡− ∆𝐶𝑎𝑠ℎ_𝑡− ∆𝐶𝐿_𝑡+ ∆𝐷𝐶𝐿_𝑡− 𝐷𝐸𝑃_𝑡 (1)

Where

𝑇𝐴𝐶𝐶𝑡 = the total accruals in year t

∆𝐶𝐴_𝑡 = the change in current assets in year t

∆𝐶𝑎𝑠ℎ_𝑡 = the change in cash and cash equivalents in year t ∆𝐶𝐿_𝑡 = the change in current liabilities in year t

∆𝐷𝐶𝐿_𝑡 = the change in short term debt in current liabilities in year t 𝐷𝐸𝑃_𝑡 = the depreciation and amortization expense in year t

Then, I estimate the discretionary accruals (DiscAcc) with the Dechow et al. (1995) model.

𝑇𝐴𝐶𝐶𝑡 𝑇𝐴𝑡−1 = 𝛼1 1 𝑇𝐴𝑡−1+ 𝛼2 ∆𝑅𝐸𝑉𝑡−∆𝑅𝐸𝐶𝑡 𝑇𝐴𝑡−1 + 𝛼3 𝑃𝑃𝐸𝑡 𝑇𝐴𝑡−1+ 𝜀𝑡 (2) Where

𝑇𝐴𝐶𝐶_𝑡 = total accruals in year t

10 ∆𝑅𝐸𝐶_𝑡 = the change in receivables in year t

𝑃𝑃𝐸_𝑡 = the property, plant and equipment in year t 𝑇𝐴𝑡−1 = the total assets in year t – 1

𝜀_𝑡 = the estimated discretionary accruals in year t.

Equation (2) is estimated for each year and industry combination. In this study I use the absolute value of the discretionary accruals as a proxy for earnings management for two reasons. First, several influential studies use the absolute value of discretionary accruals to measure the level of earnings management (Dechow & Dichev, 2002; Klein, 2002;

Bergstresser & Philippon, 2006). Second, the sign of the value of discretionary accruals is deemed not to be relevant, since all deviations from underlying earnings reduce financial reporting quality, regardless the direction (Baxter & Cotter, 2009).

Audit quality

The independent variable is the audit quality of the U.S. MNCs. To measure the audit quality, I follow prior research (Choi et al., 2012; DeAngelo, 1981; Lai, 2009) and use the size of the auditor hired by the MNC. Particularly, Big4 is a dummy variable equal to 1 if the MNC is audited by one of the big 4 auditors (Deloitte, Ernst & Young, KPMG and

PriceWaterhouseCoopers) and 0 is the MNC is audited by a non-big 4 audit firm. Audit committee independence

Audit committee independence is the moderating variable in this study. The Audit Committee is considered as independent, when all of the members are independent outside directors (Klein, 2002; Piot & Janin, 2007). Therefore, I define ACind as a dummy variable equal to 1 if the committee is fully composed of outsiders, and 0 otherwise.

Empirical model

The following model is estimated to investigate the relation between subsidiary FRQ and the MNCs audit quality:

𝐷𝑖𝑠𝑐𝐴𝑐𝑐_𝑖𝑡 = 𝛽₀ + 𝛽₁Big4 + 𝛽₂Sub_Size + 𝛽₃Par_Size + 𝛽₄Sub_ROA + 𝛽₅Par_ROA + 𝛽₆Sub_Loss + 𝛽₇Par_Loss + 𝛽₈Sub_Lev + 𝛽₉Par_Lev + 𝛽₁₀Year dummies + 𝛽₁₁Industry dummies + 𝛽₁₂Country dummies + ɛ

Control variables

Prior research shows that larger firms tend to have lower levels of discretionary accruals compared to smaller firms, because larger firms with more subsidiaries could have better opportunities to diversify their earnings management across their subsidiaries (Beuselinck et al., 2018). Therefore, I include Par_Size, which is defined as the logarithm of the total assets of the parent company. I also include Sub_Size, which is defined as the logarithm of the total assets of the subsidiary. I also control for firm performance, as discretionary accruals are related to firm performance (Dechow et al., 1995). Poorly performing firms have more incentives to engage in earnings management, which leads to higher discretionary accruals (Dechow et al., 2010). Therefore, I include return on assets as a proxy for firm performance for the parent company and the subsidiary. Par_ROA is calculated as net income of the parent company divided by the total assets of the parent company, while Sub_ROA is calculated as net income of the subsidiary divided by the total assets of the subsidiary. Furthermore, I take leverage into account, since there is evidence that firms with more debt have greater

incentives to manage earnings, due to debt covenant constraints (DeFond & Jiambalvo, 1994; Piot & Janin, 2007). I include leverage for the parent company and the subsidiary. Par_Lev is

11

calculated as total liabilities of the parent company divided by shareholders’ equity of the parent company. Sub_Lev is calculated as total liabilities of the subsidiary divided by shareholders’ equity of the subsidiary. I also add a dummy variable to control for loss of the parent company and the subsidiary. Par_Loss equals 1 if the earnings of the parent company are negative and 0 if the earnings of the parent company are positive, while Sub_Loss equals 1 if the earnings of the subsidiary are negative and 0 if the earnings of the subsidiary are

positive. Lastly, I control for year, industry and country effects using dummy variables. For the industry classification of the subsidiaries I used the 12 categories of Fama & French. After excluding the categories with no or very little data, 8 categories remain: (1) Consumer, non-durables, (2) Consumer, non-durables, (3) Manufacturing, (4) Chemicals, (5) Business equipment, (6) Shops, wholesalers, retailers and some services, (7) Healthcare, medical equipment and drugs stores, and (8) Others.

Table 1: Definition of the variables Variable Definition

ACInd Dummy variable equaling 1 if the audit committee of the parent committee fully consists of independent directors, 0 otherwise.

Big4 Dummy variable equaling 1 if the auditor of the parent company is a big 4 audit firm, 0 otherwise.

DiscAcc The absolute value of discretionary accruals of the subsidiary estimated using the modified Jones model (Dechow et al., 1995).

Par_Lev Total liabilities of the parent company divided by shareholders’ equity of the parent company.

Par_Loss Dummy variable equaling 1 if net income of the parent company is negative, 0 if net income of the parent company is positive.

Par_ROA Return on Assets (ROA), calculated as the ratio of net income of the parent company divided by total assets of the parent company.

Par_Size Logarithm of total assets of the parent company.

Sub_Lev Total liabilities of the subsidiary divided by shareholders’ equity of the parent company.

Sub_Loss Dummy variable equaling 1 if net income of the subsidiary is negative, 0 if net income of the subsidiary is positive.

Sub_ROA Return on Assets (ROA), calculated as the ratio of net income of the subsidiary divided by total assets of the subsidiary.

12

4. Results

Descriptive statistics

Table 1 contains the descriptive statistics of all the variable used in this study. All continuous variables are winsorized at the 5% and 95% level to eliminate the effect of outliers. The absolute value of discretionary accruals has a mean of 0,132 and a median of 0,09. What stands out is the relatively high mean of Big4 and ACInd, with a value of respectively 0,968 and 0,885. This means that 96,8 % of the total sample has a big 4 auditor and 88,5 % of the companies of the final sample has a fully independent audit committee.

Obviously, the mean and median of Par_Size are bigger than the mean and median of

Sub_Size, as the parent should have more total assets than the subsidiary. Furthermore, based on the mean of Sub_Loss and Par_Loss, subsidiaries did report a loss more often than the parent company; 18,1 % of the subsidiaries reported a loss, while only 13,8 % of the parent companies reported a loss. When we compare the mean of Sub_ROA and Par_ROA, we can conclude that the subsidiaries are less profitable than the parent companies in this sample. The mean of Sub_lev is higher than the mean of Par_lev, with a value of respectively 1,938 and 1,328.

Table 2: Descriptive statistics of the variables

Variable n Mean Median Std.

Dev. Min 25% 75% Max

DiscAcc 5.605 0,128 0,090 0,115 0,000 0,040 0,190 0,497 Big4 5.605 0,968 1,000 0,177 0,000 1,000 1,000 1,000 ACInd 5.605 0,885 1,000 0,319 0,000 1,000 1,000 1,000 Sub_Lev 5.605 1,825 0,880 3,057 -2,850 0,320 2,220 11,142 Par_Lev 5.605 1,980 1,590 1,413 0,303 1,000 2,480 5,755 Sub_Loss 5.605 0,181 0,000 0,385 0,000 0,000 0,000 1,000 Par_Loss 5.605 0,138 0,000 0,345 0,000 0,000 0,000 1,000 Sub_ROA 5.605 0,053 0,050 0,085 -0,146 0,010 0,100 0,231 Par_ROA 5.605 0,055 0,050 0,046 -0,029 0,020 0,080 0,151 Sub_Size 5.605 10,079 10,060 1,845 6,190 8,800 11,320 14,156 Par_Size 5.605 15,848 15,740 1,392 12,953 14,890 16,620 18,874

The variable definitions are provided in Table 1.

For the industry classification of the subsidiaries I used the 12 categories of Fama & French. After excluding the categories with no or very little data, 8 categories remain: (1) Consumer, non-durables, (2) Consumer, durables, (3) Manufacturing, (4) Chemicals, (5) Business equipment, (6) Shops, wholesalers, retailers and some services, (7) Healthcare, medical equipment and drugs stores, and (8) Others.

Panel A shows that the most represented industries in my sample are manufacturing with practically 30%, and business equipment with 22,20 %.

13

Table 3: Sample distribution

Panel A: Distribution by industry group

Industry group Freq. Percent

(1) Consumer non-durables 510 9,10 (2) Consumer durables 273 4,87 (3) Manufacturing 1.659 29,60 (4) Chemicals 483 8,62 (5) Business Equipment 1.215 21,68 (6) Shops 346 6,17 (7) Healthcare 555 9,90 (8) Others 564 10,06 Total 5.605 100,00

Panel B shows the distribution by year. It shows that the years are almost equally distributed with 1.871 observations from 2015 and 2016, and 1.863 from 2017.

Panel B: Distribution by year

Panel C shows the distribution by country. The most represented countries in my sample are the United Kingdom with 24,19 %, Italy with 15,27 %, and France with 12,90 % of the total sample.

Year Freq. Percent 2015 1.871 33,38 2016 1.871 33,38 2017 1.863 33,24 Total 5.605 100,00

14

Panel C: Distribution by country

Country Freq. Percent

Austria 84 1,50 Belgium 440 7,85 Bulgaria 51 0,91 Croatia 27 0,48 Czech Republic 231 4,12 Denmark 51 0,91 Estonia 12 0,21 Finland 119 2,12 France 723 12,90 Germany 144 2,57 Greece 42 0,75 Hungary 93 1,66 Ireland 144 2,57 Italy 856 15,27 Luxembourg 3 0,05 Netherlands 27 0,48 Norway 225 4,01 Poland 219 3,91 Portugal 149 2,66 Romania 120 2,14 Spain 450 8,03 Sweden 39 0,70 United Kingdom 1.356 24,19 Total 5.605 100 Pearson Correlations

Table 3 shows the Pearson correlations for the research variables. First, no significant relation is found between DiscAcc and Big4, which means that the audit quality of the parent company is not correlated with the financial reporting quality of the subsidiaries. Furthermore, no significant relation is found between ACInd and DiscAcc, so a fully independent audit committee at the parent company has no effect on the financial reporting quality of the

subsidiary. Sub_Lev is found to be negatively and significantly related to DiscAcc, so a higher leverage at the subsidiary level is related to lower discretionary accruals, and thus a higher financial reporting quality. This is not what I predicted, because previous literature showed that highly leveraged firms had more incentives to engage in earnings management due to debt covenant constraints, what would lead to a lower financial reporting quality (DeFond & Jiambalvo, 1994; Piot & Janin, 2007). Par_Loss and Sub_Loss are both positively related to DiscAcc and statistically significant at the 1% level. It means that the financial reporting quality of the subsidiary is lower, when the subsidiary or the parent company reports a loss, which is what I expected according to extant literature. Furthermore, the table shows that Sub_Size is negatively related to DiscAcc and statistically significant at the 1% level. This

15

means that bigger firms have lower discretionary accruals, and thus a higher financial

reporting quality. No significant relation is found between Par_Size and DiscAcc, so the size of the parent company does not affect the financial reporting quality of the subsidiary. Finally, there are no variables highly correlated to each other, so multicollinearity is not an issue.

16 Table 4: Pearson correlations

DiscAcc Big4 ACInd Sub_Lev Par_Lev Sub_Loss Par_Loss Sub_ROA Par_ROA Sub_Size Par_Size DiscAcc 1 Big4 0,003 1 ACInd 0,012 -0,009 1 Sub_Lev -0,043*** -0,007 -0,033** 1 Par_Lev -0,004 0,046*** -0,057*** 0,007 1 Sub_Loss 0,097*** -0,021 -0,004 -0,038*** 0,061*** 1 Par_Loss 0,049*** -0,072*** 0,035*** 0,012 0,176*** 0,057*** 1 Sub_ROA 0,012 0,029** 0,028** -0,075*** -0,055*** -0,681*** -0,040*** 1 Par_ROA -0,034** 0,082*** -0,013 -0,005 -0,227*** -0,099*** -0,638*** 0,102*** 1 Sub_Size -0,191*** 0,073*** -0,048*** 0,038*** 0,076*** -0,050*** -0,085*** -0,019 0,091*** 1 Par_Size -0,019 0,183*** -0,046*** 0,039*** 0,141*** -0,055*** -0,168** -0,033** 0,151*** 0,360*** 1

The variable definitions are provided in Table 1. *= significant at the 10% level

**= significant at the 5% level ***= significant at the 1% level

17 Regression analysis

Table 4 contains the results of the regression analysis. First, I conduct a regression where I estimate the baseline model to test H1. Column (1) includes the results of the regression with the baseline model, which includes the dependent variable (DiscAcc), the main independent variable (Big4), and all the control variables. I find no significant relation between Big4 and DiscAcc, which means that these results cannot support H1, since I predicted a positive relation. To test H2, I use an interaction variable, called Big4*ACInd, which is calculated as Big4 multiplied by ACInd. This interaction variable measures the effect that a fully

independent audit committee at the parent company has on the relation between Big4 and DiscAcc. I find that Big4*ACInd does not have a significant effect on the relation between Big4 and DiscAcc. This means that the outcomes of this study cannot support H2.

18 Table 5: Regression analysis

(1) (2)

VARIABLES DiscAcc DiscAcc

Big4*ACInd -0,004 (-0,005) Big4 0,004 0,008 (-0,009) (-0,01) Sub_Lev 0,000 0,000 (0,000) (0,000) Par_Lev -0,001 -0,001 (-0,001) (-0,001) Sub_Loss 0,052*** 0,052*** (-0,005) (-0,005) Par_Loss 0,013** 0,013** (-0,006) (-0,006) Sub_ROA 0,170*** 0,171*** (-0,024) (-0,024) Par_ROA 0,068 0,068 (-0,043) (-0,043) Sub_Size -0,014*** -0,014*** (-0,001) (-0,001) Par_Size 0,006*** 0,006*** (-0,001) (-0,001) Constant 0,164*** 0,164*** (-0,021) (-0,021)

Year effects Yes Yes Industry effects Yes Yes Country effects Yes Yes

Observations 5.605 5.605 R-squared 0,085 0,085

Note: standard errors in parenthesis.

The variable definitions are provided in Table 1. *= significant at the 10% level

**= significant at the 5% level ***= significant at the 1% level

19

5. Conclusion

This study investigates the relation between audit quality of the MNC and financial reporting quality of the subsidiary. Based on previous literature I expected a positive relation between these variables. A higher audit quality of the parent company would mean that the auditor constraints the level of earnings management within the parent company and the subsidiary, and thus would lead to a higher financial reporting quality. The following hypothesis was formulated to answer my research question:

H1: There is a positive relation between audit quality at the MNC level and FRQ at the subsidiary level.

The findings of a linear regression showed that the audit quality of the parent company is not associated with the financial reporting quality of the subsidiary, which means that H1 can not be supported. Furthermore, I examined what the effect of a fully independent audit committee is on the relationship between the audit quality of the MNC and the financial reporting quality of the subsidiary. I predicted that a fully independent audit committee at the parent company would strengthen the positive relation between audit quality of the parent company and FRQ of the subsidiary. A fully independent audit committee would better perform their job to monitor the financial reporting quality, which would lead a higher financial reporting quality. This resulted in the following hypothesis that was tested:

H2: The positive relation between audit quality at the MNC level and FRQ at the subsidiary level is stronger when the internal audit committee of the parent company is fully

independent.

The results of a linear regression showed that a fully independent audit committee has no effect on the relation between audit quality of the MNC and financial reporting quality of the subsidiary and thus, H2 cannot be supported.

A limitation of this study is the proxy used to measure audit quality of the parent company. In particular, I only use auditor size (Big 4 vs non-Big 4). However, auditor quality is

multidimensional and inherently unobservable, so there is no single auditor characteristic that can be used to proxy for it (Balsam et al., 2003). In addition, further research may look into real earnings management instead of accrual based earnings. The reason for this is that this form of earnings management is more often used since the enactment of the SOX act, and accrual-based earnings has declined since then (Cohen et al., 2008). There is evidence in a recent survey of executives that managers prefer real earnings management activities

compared with accrual-based earnings management (Graham et al., 2005). Due to these facts, the results might change when focusing on real earnings management instead of accrual-based earnings management.

20

References

Abbott, L. J., Parker, S., & Peters, G. F. (2004). Audit committee characteristics and restatements. Auditing: A Journal of Practice & Theory, 23(1), 69-87.

Balsam, S., Krishnan, J., & Yang, J. S. (2003). Auditor industry specialization and earnings quality. Auditing: A journal of practice & Theory, 22(2), 71-97.

Baxter, P., & Cotter, J. (2009). Audit committees and earnings quality. Accounting &

Finance, 49(2), 267-290.

Becker, C. L., DeFond, M. L., Jiambalvo, J., & Subramanyam, K. R. (1998). The effect of audit quality on earnings management. Contemporary accounting research, 15(1), 1-24. Bédard, J., Chtourou, S. M., & Courteau, L. (2004). The effect of audit committee expertise, independence, and activity on aggressive earnings management. Auditing: A Journal of

Practice & Theory, 23(2), 13-35.

Beneish, M. D. (2001). Earnings management: A perspective. Managerial Finance, 27(12), 3-17.

Bergstresser, D., & Philippon, T. (2006). CEO incentives and earnings management. Journal

of financial economics, 80(3), 511-529.

Beuselinck, C., Cascino, S., Deloof, M., & Vanstraelen, A. (2018). Earnings management within multinational corporations. The Accounting Review.

Biddle, G. C., Hilary, G., & Verdi, R. S. (2009). How does financial reporting quality relate to investment efficiency?. Journal of accounting and economics, 48(2-3), 112-131.

Bonacchi, M., Cipollini, F., & Zarowin, P. (2018). Parents’ use of subsidiaries to “push down” earnings management: Evidence from Italy. Contemporary Accounting Research,

35(3), 1332-1362.

Chen, K. Y., Lin, K. L., & Zhou, J. (2005). Audit quality and earnings management for Taiwan IPO firms. Managerial Auditing Journal, 20(1), 86-104.

Choi, J. H., Kim, J. B., Qiu, A. A., & Zang, Y. (2012). Geographic proximity between auditor and client: How does it impact audit quality?. Auditing: A Journal of Practice & Theory,

31(2), 43-72.

Cohen, D., Dey, A., & Lys, T. (2008). Real and Accrual-Based Earnings Management in the Pre- and Post-Sarbanes-Oxley Periods. The Accounting Review, 83(3), 757-787.

Cohen, J. R., Krishnamoorthy, G., & Wright, A. (2004). The corporate governance mosaic and financial reporting quality. Journal of accounting literature, 87-152.

DeAngelo, L. E. (1981). Auditor size and audit quality. Journal of accounting and economics,

21

Dechow, P. M., & Dichev, I. D. (2002). The quality of accruals and earnings: The role of accrual estimation errors. The accounting review, 77(s-1), 35-59.

Dechow, P., Ge, W., & Schrand, C. (2010). Understanding earnings quality: A review of the proxies, their determinants and their consequences. Journal of accounting and economics,

50(2-3), 344-401.

Dechow, P. M., & Skinner, D. J. (2000). Earnings management: Reconciling the views of accounting academics, practitioners, and regulators. Accounting horizons, 14(2), 235-250. Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1995). Detecting earnings management.

Accounting review, 193-225.

DeFond, M. L., & Jiambalvo, J. (1994). Debt covenant violation and manipulation of accruals. Journal of accounting and economics, 17(1-2), 145-176.

Degeorge, F., Patel, J., & Zeckhauser, R. (1999). Earnings management to exceed thresholds. The Journal of Business, 72(1), 1-33.

Dyreng, S., Hanlon, M., & Maydew, E. (2012). Where do firms manage earnings? Review of Accounting Studies, 17(3), 649-687.

FASB, 1978. Statement of Financial Accounting Concepts No. 1. Objectives of Financial Reporting by Business Enterprises.

Francis, J. R., Maydew, E. L., & Sparks, H. C. (1999). The role of Big 6 auditors in the credible reporting of accruals. Auditing: a Journal of Practice & theory, 18(2), 17-34. Gomariz, M. F. C., & Ballesta, J. P. S. (2014). Financial reporting quality, debt maturity and investment efficiency. Journal of Banking & Finance, 40, 494-506.

Graham, J. R., C. R. Harvey, & S. Rajgopal. 2005. The Economic Implications of Corporate Financial Reporting. Journal of Accounting and Economics 40: 3-73.

Healy, P. M., & Wahlen, J. M. (1999). A review of the earnings management literature and its implications for standard setting. Accounting horizons, 13(4), 365-383.

Holm, C., & Laursen, P. B. (2007). Risk and control developments in corporate governance: changing the role of the external auditor?. Corporate Governance: An International Review,

15(2), 322-333.

Islam, M. A., Ali, R., & Ahmad, Z. (2011). Is modified Jones model effective in detecting earnings management? Evidence from a developing economy. International Journal of Economics and Finance, 3(2), 116-125.

Jonas, G.J., Blanchet, J. (2000). Assessing Quality of Financial Reporting. Accounting Horizons, 14(3), 353-363.

Jones, J. J. (1991). Earnings management during import relief investigations. Journal of

22

Klein, A. (2002). Audit committee, board of director characteristics, and earnings management. Journal of accounting and economics, 33(3), 375-400.

Koo, D. S., Ramalingegowda, S., & Yu, Y. (2017). The effect of financial reporting quality on corporate dividend policy. Review of Accounting Studies, 22(2), 753-790.

Lai, K. W. (2009). Does audit quality matter more for firms with high investment opportunities?. Journal of Accounting and Public Policy, 28(1), 33-50.

Lin, J. W., Li, J. F., & Yang, J. S. (2006). The effect of audit committee performance on earnings quality. Managerial Auditing Journal, 21(9), 921-933.

Lo, K. (2008). Earnings management and earnings quality. Journal of accounting and

economics, 45(2-3), 350-357.

McNichols, M. F., & Stubben, S. R. (2008). Does earnings management affect firms’ investment decisions?. The accounting review, 83(6), 1571-1603.

Piot, C., & Janin, R. (2007). External auditors, audit committees and earnings management in France. European accounting review, 16(2), 429-454.

Roychowdhury, S. (2006). Earnings management through real activities manipulation. Journal of accounting and economics, 42(3), 335-370.

Sun, Q., Yung, K., & Rahman, H. (2012). Earnings quality and corporate cash holdings.

Accounting & Finance, 52(2), 543-571.

United Nations Conference on Trade and Development (UNCTAD). (2014). World

Investment Report.

United Nations Conference on Trade and Development (UNCTAD). (2018). World

Investment Report.

Vanasco, R. R. (1994). The audit committee: an international perspective. Managerial

Auditing Journal, 9(8), 18-42.

Verdi, R. S. (2006). Financial reporting quality and investment efficiency. Available at SSRN

930922.

Watts, R. L., & Zimmerman, J. L. (1978). Towards a positive theory of the determination of accounting standards. Accounting review, 112-134.