The effect of accounting standards on Earnings Management in U.S. and Europe

(1)

The effect of accounting standards on Earnings Management in

U.S. and Europe.

Student number : 11131241

Name : André Nguyen

Master Thesis date : 26 June 2017

Word count : 12,477

MSc Accountancy & Control : Specialization Accountancy Thesis Supervisor : Dr. G. Georgakopoulos Faculty of Economics and Business, University of Amsterdam

(2)

1

Abstract

Earnings management can be distinguished as either accruals-based earnings management and Real earnings management. In prior literature (Zang, 2012; Chunhui et al., 2014; Evans et al., 2015), IFRS accounting standards has a negative effect on Real earnings management and U.S. GAAP has a negative effect on accrual-based earnings management. A few researchers (Beest, 2009); (Zang, 2012); (Evans, 2015) describe that a combination of both types of earnings management are important. This study examines the effect of applying accounting standards on the level of earnings management. The observation is based on the countries France, Germany and United States and the selected time horizon is from 2005 to 2015. This sample have 11,436 observations. Accrual bases earnings is measured by the modified Jones model (Jones, 1991). The three proxies to measured Real earnings management are abnormal production, abnormal cash flow from operation and abnormal expenditures(Roychowdhury, 2006). The hypotheses that are examined in this study are that IFRS accounting standards have a negative effect on Real earnings management and the U.S. GAAP have a positive effect on accrual-based earnings management. This study concluded that there is one significant effect on the accounting standards on the level of earnings management.

Keywords: Earnings management; Real earnings management; Accrual based earnings

management Accounting standards; IFRS; U.S. GAAP; CEO; Agency theory; Rules based standards; Principle based standards

(3)

2

The general purpose and objective of financial reports is to provide financial information about the reporting firm. Information is useful to lenders and other creditors, potential and existing investors, and suppliers in making decisions about providing resources to the entity (IFRS, 2010b, §OB2). The information in the financial reports could be used as a reflection of the internal knowledge of the management.

The preparer of financial reports has more information regarding the entity’s performance than users of the financial reports (Scott, 2012). When this event occurs between the preparer and users of the financial reports this is called information asymmetry. This information asymmetry emerges when the preparer of financial reports is able to use earnings management to meet their objectives (Healy & Whalen, 1999). For the users of financial reports it is difficult to monitor the actual performance of management (Scott, 2012). Due to different motives the management apply earnings management (Watts & Zimmerman, 1986). In other empirical studies done over the years, the motives that Watts and Zimmerman (1986) developed. Researchers such as Eisenhardt (1989), Healy & Whalen (1999) and Scott (2012), supported the motives.

In the previous decade, there had been an increase in the number of accounting frauds, most notably the scandals at Lehman Brothers, Bernie Madoff and Enron. These caused tremendous harm to the capital markets (Hellstron, 2005). Often, financial statements are difficult to compare across borders, due differences in political, economic systems and regulation (Hellstron, 2005). An important condition for well-functioning capital markets and economy as a whole is high quality accounting information (Hellstron, 2005). “Value relevance” is one of the basic attributes for the quality of financial statements (Barth et al., 2008). The value relevance attributes to the usefulness of financial statement information for the focuses on equity valuation and stakeholders (Bart et al., 2006; Lin et al., 2012). Francis and Schipper (1999) defined this concept of the value relevance of accounting information as ‘the ability of accounting numbers to summarize the information underlying the stock prices’ (Francis & Schipper, 1999; Liu & Liu 2007). (p.56)

In 2005, all European listed firms were mandated to adopt International Financial Reporting Standards (IFRS). In the capital market, advocators of IFRS favoured the transition from domestic GAAP accounting standards to IFRS because it conveys new information (Barth et al., 2006). Merited by the increase of political, multinational firm’s movements and economic interactions between countries, the demand for common standards of financial reporting was evident (Hunton et al., 2004). This new regulation change would assist investors in making

(6)

5 informed decisions, aid in forecasting future cash flows thus leading to a higher accounting quality through transparency (Hunton et al., 2004).

Accruals are a way to influence reported earnings. The gaps between cash flows and reported earnings are accruals (Dechow, 1994). This difference includes accounting postings like provision, interest, depreciation and accounts receivable or payable (Dechow, 1994). Accruals are used to capture the financial transactions or events that occurs before the actual payment or receipt of cash. The full economic consequences of transactions are captured in a specific period when applying accrual accounting (Bart et al., 2008). When preparing the periodic financial reports, it is important to note the consequences of current events which may impede the expectations on future cash flow (Palepu et al., 2007). The financial data might be valuable for users since managers use their discretion to communicate their inside knowledge of the company (Palepu et al., 2007). Another way to use accruals is to smooth income and avoid volatility in earnings and to move towards an expected level of reported earnings (Beattie et al., 1994). Managers may smooth earnings to avoid the volatile reported earnings and give a better reflection on the firm’s current or future performance to investors (Dechow, 1994). Regarding a consistent earnings growth, managers may think this gives a positive signal to investors. These are among the incentives that managers may have to use their accounting discretion to influence the reported financial data (Dechow, 1994).

Real earnings management is another manner to influence reported earnings. The focus of Real earnings management is on manipulating real activities or the deviation from operational and normal business activities (Kim et al., 2010). Real earnings management has a direct effect on the cash flow (Dechow & Sloan, 1991). At the end of the fiscal year, the unmanaged earnings can be manipulated. Firms attempt to meet the earnings benchmark by the sales of assets or by declining the research and development expenditures (Kim et al., 2010). According to Dechow and Sloan (1991), on the near end of the tenure, managers reduce the research and developments expenditures to increase the earnings in the short term. The evidence in the study of Bamber et al., (1991) shows that firms reduce the research and development expenditure to meet earnings objectives from the previous years. It is not possible to apply real earnings activities when Accrual based earnings management are completely applied (Kim et al., 2010).

1.1 Motivation

The introduction of the SOX act in 2002 has had a negative impact on earnings management (Chang, 2008). Dechow et al. (1995) has shown that a strong corporate governance has a negative influence on committing financial fraud. The study of Libby et al. (2009) shows that the

(7)

6 interpretation in accounting standards influences the degree of earnings management. Prior studies (Heemskerk & Van der Tas, 2006; Jeanjean & Stolowy, 2008; Tendeloo & Vanstraelen, 2005) have mostly focused on Accrual based earnings management instead of Real earnings management. Since the implementation of the SOX in 2002, consideration on Real earnings management has increased (Evans et al., 2015). The studies (Heemskerk & Van der Tas, 2006; Tendeloo & Vanstraelen, 2005) have only measured Accrual based earnings management on level of IFRS standards. Real based earnings management was therefore not clearly measurable (Roychowdhury, 2006). Cohen et al. (2008), Cohen and Zarowin (2010) and Zang (2012) measured Real earnings management based on the method of Roychowdhury (2006) which is the most effective manner.

According to Habib (2004), the most valuable relevant researches assume that the financial information in the financial statements are either biased reporting by management or free of error. Stock prices and manipulating could possibly be detected by investors (Habib, 2002; Habib, 2004). This is an interesting topic for more research because in the past there has been inconsistent results regarding the differences and the declining value relevance of accounting information (Ball & Brown, 1968; Beaver, 1968; Lev, 1989; Collins, 1997).

The incentives of management are crucial to examining the effect of earnings management on the value relevance of financial information. Managers have the discretion to make a decision that leads to negative or positive impacts on the value relevance of financial information. For standard setters and regulators, it is a difficult task to determine the space of discretion for the preparers of financial reports.

In the world, there are two predominant accounting standards IFRS and U.S. GAAP. The main difference is that IFRS is a more principle based, while U.S. GAAP is more rule based. In study is it interesting to examine whether there are differences between the accounting standards in the value relevance of the reported earnings and in the level of earnings management.

In the past, due to the accounting scandals the capital markets have been seriously harmed and firms have provided financial reports that the investors did not trust. It is important to get insight within this topic by comparing value relevance of reported earnings and the level of earnings management between the accounting standards IFRS and U.S. GAAP.

The research question for this study is formulated as:

‘What is the effect of Real earnings management for firms that are using IFRS accounting standards compared to firms that are using U.S. GAAP accounting standards?’

(8)

7 The rest of this study is structured as follows. Chapter two will describe the theoretical background of this study and a broad literature review will be given. Topics such as earnings management, agency theory and accounting standards are employed and the hypothesis development is discussed. The next chapter contains the research methodology used to test these hypotheses. The description is given of how the sample was selected and which models are used. In the fourth chapter, the research results of the effect of accounting standards on earnings management are presented. Finally, the last chapter contains the conclusion and limitations and the feasibility for future research.

(9)

8

2. Literature review and Hypothesis development

In this chapter the topics earnings management, accounting standards and agency theory will be discussed. The theoretical framework for this study will be explained in this paragraph. In the last paragraph, the hypothesis will be formed based on the results of theoretical framework.

2.1 Earnings Management

The subject of earnings management is well-researched in the literature. For example, researchers such as Healy and Wahlen (1999) and Schipper (1989) have written a definition regarding earnings management. Scot (2012) took their definition and rewrote it. That makes this topic a comprehensive and complex concept. For earnings management there are different definitions in the literature.

Firms communicate through financial reporting to provide stakeholders information regarding the performance of the firm. The performance reflects on earnings. Earnings management is defined by Leuz & Nanda (2003) as managers altering the reported financial performance to influence contractual performance or to mislead stakeholders. The two definitions of earnings management most commonly used from the papers are from Healy and Wahlen (1999) and Schipper (1989). Healy and Wahlen (1999) defines earnings management as: “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.” (p. 368)

According to this definition, misstating the financial reports could mislead stakeholders in their investment judgement. It doesn’t reflect the true underlying economic performance of a company. Maximizing the shareholder’s value by exploiting assets by debt and equity is the main objective of a firm. The only incentive for shareholders to invest would be if there is a positive future firm performance expectation. It is in the firm’s interest to report positive profits, sustain recent performance and meet analyst expectations (Degeorge et al., 1999). Other researchers defines their own interpretations of earnings management. Schipper (1989) defines earnings management as:

“Earnings management is really disclosure management in the sense of a purposeful intervention in the external financial reporting process, with the intent of obtaining some private gain as opposed to merely facilitating the neutral operation of the process.” (p. 91-92)

According to Scott (2009), earnings management is used to minimize or maximize income in order to meet a certain earnings objective. Managers could manipulate their earnings management downward or upward (Scott, 2009). A firm in financial stress already expects future

(10)

9 losses and might manage earnings downward to make the loss enormous (Scott, 2009). Scott (2009) distinguishes two types of information asymmetries. First, the adverse selection where one party has less information in a transaction than the other party. Second, moral hazard distinguished between control and ownership whereby information asymmetry occurs.

Baralexis (2004) mentions earnings management as creative accounting, preparers process the financial statement and intentionally violate the accounting rules for self-interest. Despite the negative views on earnings management, Beneish (2001) and Ronen and Yaari (2008) mention the positive sides. Managers use earnings management for their expectation of the future to inform the stakeholders through the financial statements (Beneish, 2001). Earnings management is divided into three alternative definitions by Ronen and Yaari (2008). The alternatives for Earnings Management are “white”, “grey” and “black”. The first definition, white: “Earnings management is taking advantage of the flexibility in the choice of accounting treatment to signal the manager’s private information on future cash flows.”(p. 25). As second alternative definition, grey: “Earnings management is choosing an accounting treatment that is either opportunistic (maximizing the utility of management only) or economically efficient” (p. 25). In the last alternative definition, black: “Earnings management is the practice of using tricks to mispresent or reduce transparency of the financial report” (p. 25). For this study, the earnings management definition of Scot (2009) will be used. Scott (2009) defines:“Earnings management is the choice by a manager of accounting policies, or orations affecting earnings, so as to achieve some specific reported earnings objective” (p. 403). Where other authors give a negative or positive opinion regarding earnings management, Scot (2009) describes in general and does not have a preference.

2.2 Two forms of earnings management

This study examines earnings management and whether managers manage more earnings in IFRS standards or in U.S. GAAP standards. Earnings management can be distinguished in two forms, (a) Real earnings management and (b) Accrual based earnings management.

2.2.1 Real earnings management

To meet financial targets earlier than estimated, managers use Real earnings management. Real earnings management are discrepancies from ordinary operational processes. Managers have intentionally misled stakeholders as to whether the financial targets have been met by earlier normal operation activity. This is attained by postponing actual expenditures or cutting in the research and development expenditures (Chunhui et al., 2014). Undoubtedly, these accounting practices do not add firm value. Real earnings management must be applied in the advanced fiscal year end (Evans et al., 2015; Roychowdhury, 2006). This often has an uncertain impact on

(11)

10 earnings (Evans et al., 2015). Cohen (2008), Roychowdhury (2006) and Zang (2012) examined Real earnings management due to increasing earnings by shrinking discretionary expenditures and reducing the cost of goods sold by overproducing the inventory. This disrupts cash flow of entities (Evans et al., 2015). Roychowdy (2006) measured real earnings manipulation activities through (a) the abnormal level of cash flow from operations, (b) the abnormal levels of discretionary expenses and (c) the abnormal level of production costs (Roychowdy, 2006). Cohen et al. (2008) and Zang (2012), measured Real earnings management in the same manner as Roychowdy (2006). Roychowdhury (2006) use these variables to measure the manipulation method and their effects: (a) sales manipulation, (b) reduced discretionary expenditures and (c) overproduction (Roychowdhury, 2006).

Roychowdhury(2006) defines sales manipulation as more lenient credit terms or offering temporarily discount to increase sales. The reduction of discretionary expenditures are expenditures that do not generate immediate income and revenue (Roychowdy, 2006). As last, overproduction is to manage overproduction more than is necessary to meet the production planning (Roychowdy, 2006).

2.2.2 Accrual based earnings management

Accruals are costs that are related to the operations of a firm without cash outflow. Managers have the opportunity to use Accrual based earnings management to consider whether they to take an accrual or not at the end of the fiscal year (Scot, 2009). Accrual accounting is defined as: “attempts to record the financial effects on an entity of transactions and other events and circumstances that have cash consequences for the entity in the periods in which those transactions, events, and circumstances occur rather than only in the periods in which cash is received or paid by the entity” [FASB 1985, SFAC No. 6, para. 139]. Preparers of financial statements use accrual accounting to help aim investors in regards to decisions of financial performance during the year. The reported earnings provide more understandable information than cash flows for the investors (Dechow & Skinner, 2000). Zang (2012) wrote that accrual-based earnings management can be achieved by predictions used in the financial statement or by changing the accounting methods (Zang, 2012). The likelihood is that the reported earning can be biased through revaluating the estimated provision for doubtful account receivables or use another depreciation method for fixed assets (Zang, 2012). Discretionary accruals were used as a proxy for Accrual based earnings management in prior literature (Jones, 1991). Discretionary accruals show the difference between the normal level of accruals and the actual accruals of a firm (Zang, 2012). The study of Dechow et al. (1995) examines accrual bases earnings management and which model is the best to measure Accrual

(12)

11 based earnings management. The models that were investigated are: (a) Healy Model, (b) the DeAngelo Model (c) the Jones Model, (d) the Modified Jones Model and (e) the Industry Model (Dechow et al., 1995). According to Dechow et al. (1995), the modified Jones model provided the most robust tests for earnings management. To estimate discretionary accruals, in the study of Zang (2012) and Cohen et al, (2008) the modified Jones (1991) model was used to measure discretionary accruals .

Model to measure accrual-based earnings management

In the research of Dechow et al. (1995) it was shown that accrual-based earnings management is measured in different methods. The Modified Jones Model (Jones, 1991), Healy Model (Healy, 1985) and DeAngelo model (DeAngelo, 1986) was used in the research of Dechow et al. (1995) to measure accrual-based earnings management. The assumption is that in the previous year Earnings management is not applied, therefore, the DeAngelo model (DeAngelo, 1986) and the Healy model (Healy, 1985) will be used. To measure the proxy discretionary accruals the value has to be absolute to measure the degree of Accrual based earnings management (Cohen, 2004); (Braam et al., 2015).

Healy model

The Healy model (Healy, 1985) use the average of total accruals (Taτ) divided by scaled total assets. The measurements of nondiscretionary accruals are from the estimated period between the years (Healy, 1985). The formula of the Healy model (Healy, 1985) for the nondiscretionary accruals in the current year is (Healy, 1985; Bartov et al., 2000):

NDAτ = 1/nΣτ (TAt/Aτ-1)

Where:

NDA = estimated non-discretionary accruals Aτ -1 = Total Asset year t -1

N = number of years in the estimation period TA = total accruals scaled by lagged total assets;

T = 1, 2, ... T is a year subscript for years included in the estimation period; and τ = a year subscript indicating a year in the event period”

DeAngelo Model

The DeAngelo model (DeAngelo, 1986) measures the nondiscretionary accruals (NDA_τ) by using the previous period’s total accruals scaled by lagged total assets. The discretionary ratio of accruals is the difference between total accruals in the current year t scaled by At-1 and NDAτ. The formula for the nondiscretionary accruals is (Bartov et al., 2000; Deangelo, 1986):

(13)

12 Where :

NDA = estimated non-discretionary accruals TA = total accruals scaled by lagged total assets; A = Total Asset; and

T = 1, 2, ... T is a year subscript for years included in the estimation period;

If a manager wants a higher profit in the financial statements, the managers will decrease or avoid accruals and release the provision in order to manipulate the financial profits (Graham, 2005). When the manager has incentive to artificially lower the financial results, the manager will use more provisions (Jones et al., 1991). Accrual based earnings management can defined as a manner of manipulation to the results on whether there is a revenue shift, or to take or not to take a provision.

2.3 Agency Theory

Jensen and Meckling (1976) describe the Agency theory as the principal-agent relationship that exists between the management that manages the firm and owners. The agency relationship is defined by Jensen and Meckling (1976) as “a contract under which one or more persons (the principal(s) engage(s) another person (the agent) to perform some service on behalf which involves delegating some decision making authority to the agent” (p. 308). Agency problems emerges when there is a conflict of interest between self-interest and duty. Moreover, the agent is assumed to be driven by self-interest (Dalton et al., 2007). The agent has incentives to engage in behaviour that is more beneficial for the agent rather than the principal (Jensen & Mecklin, 1976). The agency problems are distinguished in two types, namely moral hazard and adverse selection. Moral hazard arises when the agent behaves divergently as he has not been fully disclosed to the risks and does not need to worry about the consequences. Adverse selection emerges when one party has more informational advantage over another party (Scott, 2009). The agent representative in the name of the principal by managing the business on a daily basis, while the principal receives periodic updates (Jensen & Mecklin, 1976). It is difficult for the principal to control and monitor the agent when the agent has more access to information and more knowledge in general. The information asymmetry arise between the agent and the principal. In this situation gives the agent the opportunity to behave opportunistically (Jensen & Mecklin, 1976; Shapiro, 2005).

To reduce divergence and opportunistic actions by the agent, the principal can use a monitoring system or establish appropriate incentives (Shapiro, 2005). The principals align the interest of the agent with the firms which often offer a compensation plans to the agent. For example, through private contracting, incentives schemes or reward structures (Shapiro, 2005). Nonetheless, this compensation plans might have an adverse effect on the agent’s behaviour.

(14)

13 This might become incentive for the agent to manipulate earnings. To create firm value, management compensation contracts were designed to reduce the conflict of interest between agent and principal (Shapiro, 2005).

In the study of Jensen and Meckling (1976) they considered two types of monitoring which might increase firm value. The board of director is the important monitoring and the audit is a monitoring activity. For the firm, the monitoring system provides an ex post control system (Jensen & Meckling, 1976). The agent operates in the best interest of the stakeholders when the principals are able to obtain information about the agent’s actions in an adequate way. Less resources are needed to reduce the conflict of interest through incentives (Eisenhardt, 1989). The audit quality means that there is an assurance to detect omissions or material misstatements, hence an audit can reduce the level of information asymmetry (Wali, 2015).

Older people become more conservative in their actions (Peterson et al., 2001). In the study of Twenge and Campbell (2008) it is stated that older people have lower self-confidence levels than younger people and older people are more likely to show ethical behaviour. Shefrin (2005) stated that risk aversion increases until the age of 70. Over that age risk aversion decreases. Career and financial security are less important to younger executives than to older executives (Wiersema & Bantel, 1992). Younger executives seem to act less risk averse. In prior studies (Patterson, 2001; Peterson et al., 2001) it was concluded that there is a significant relationship between age and earnings management (Shefrin, 2005). Evidence showed that before a CEO goes into retirement the riskier his behaviour as he is more likely to engage in earnings management (Davidson et al., 2007).

Finkelstein and Hambrick (1990) found that tenure has a significant influence on organizational outcomes. According to Geiger and North (2006), in the first years of the CEO career there is a greater use of earnings management which reinforces the influence of CEO’s on earnings management. Executive management stays for a long tenure in companies, and change less in their business strategy (Geiger & North, 2006). If executives invest their time to staying at strategic level, the executives would gain more specific knowledge about the firm than average employees (Geiger & North, 2006; Finkelstein & Hambrick, 1990). This could imply that gaining by taking risks is smaller than the loss of long-tenure executives (Finkelstein & Hambrick, 1990). Zahra (2005) stated “The length of a CEO’s tenure is negatively associated with entrepreneurial risk taking, especially a family firm’s emphasis on innovation and venturing in domestic and international markets” (p. 35) Further in this study, evidence was found that the long tenure of CEO gives time due to the institutionalizing of systems and business processes (Zahra, 2005)

(15)

14

2.4 Accounting standards

The two predominant accounting standards are IFRS and U.S. GAAP. IFRS is used for publicly listed firms in most the countries around the world except for the United States. The United States uses the U.S. GAAP established by the Financial Accounting Standards Board (FASB). Developing a single set of enforceable, high quality, understandable, and globally accepted accounting standards is the main objective of the IASB (IASB, 2013). In the United States, there is still some resistance for the adoption of IFRS, albeit the Security and Exchange Commission (SEC) acknowledges the benefits of a global single set of accounting standards (Hunton et al., 2004).

Joos and Leung (2013) estimated the perception about the switch to IFRS of investors in the United States. The study of Joos and Leung (2013) scrutinized the stock market reaction to several events relating to IFRS adoption in the United States. According to Joos and Leung (2013) the findings suggest that investors react more positive to IFRS when IFRS is expected to lead to convergence benefits. Despite this, firms with higher litigation costs showed that their results were less positive regarding market reaction (Joos & Leung, 2013). Both accounting standards are principle-based, the main difference between the two standards is that U.S. GAAP is more rules-based set of standards, while IFRS is generally regarded as more principle-based (Joos & Leung, 2013). The interpretation that is deep-rooted in IFRS standards may be used to engage in earnings management. Nelson (2003) concludes that the level of flexibility within accounting standards increased the level of earnings management. In a principle-based environment, auditors allow more earnings management (Libby & Kiney, 2000). In the survey of Jamal and Tan (2010) they examined the influence of accounting standards on the level of aggressive reporting. Jamal and Tan (2010) discuss only that if the auditor’s attitude becomes more principle-oriented, than principle based standards reduce earnings management. Rule-based standards may be used for earnings management through transaction decisions (Nobes, 2005). According to Nobes (2005), rules-based accounting standards eliminate opportunities to engage earnings management through accounting decisions.

Researchers (Barth et al., 2012; Nelson, 2003) have begun to explore whether principles-based standards lead to different accounting outcomes when compared to rules-principles-based standards. Barth et al. (2012) examined whether the earnings of firms that reports in U.S. GAAP standards are comparable to firms that reports in IFRS standards. In the study of Bart et al. (2012) it is indicated that the financial reports under IFRS standards have a lower value relevance than financial reports under U.S. GAAP. Bart et al. (2012) also compared the other three dimensions of audit quality compared, which are timeless, income smoothing and accrual quality. The

(16)

15 findings of Bart et al. (2012) indicate that the differences are more mandatory for reporting firms under U.S. GAAP standards, but the overall accounting quality is lower for reporting firms under IFRS standards. In the literature it shows (Beest, 2009; Chunhui et al., 2014; Evans et al., 2015) that there is consensus that there is less flexibility inherent with U.S. GAAP standards than IFRS standards, because IFRS standards are more principle based. The flexibility may be used to engage in earnings management.

2.5 Principal-based & rules-based accounting standards

In this paragraph, the terms principles-based and rules-based accounting standards will be described and explained. Additionally, the terms principles-based and rules-based accounting standards will be compared.

2.5.1 Principles-based accounting standard

The preparer has the opportunity, in principle based accounting, to provide users with relevant information (Nobes, 2015) based on professional judgment. According to Nobes (2005) rules-based accounting standards are opposed to principle rules-based accounting standards. Whereas principle based accounting standards are not only focused on the content of the judicial definitions of financial information (Nobes, 2015). The users of financial statement desire relevant information and this can be attained through principle-based accounting standards (Nobes, 2015).

There is space for using earnings management in the domain of accounting standards. With managers managing the earnings easily in a certain direction under principle-based standards rather than under rules-based standards (Herz, 2003). There are a few ways to measure whether an adjustment in standards has a demotivation or motivation shock on the level of earning management (Healy, 1999). However, Libby et al. (2009) found that the degree of earnings management is influenced by the manager in the discretionary space of Principle-based Accounting standards. This should lead to more earning management (Libby et al., 2009). It was found by Benson et al. (2006) that principle-based accounting standards have more earnings management.

2.5.2 Rules-based accounting standard

Rules-based accounting standards has different interpretations. Rules-based standards provides detailed rules that others must comply to and this leads to less earnings management (Schipper, 2003). Financial statements under principle approach reduces the comparability (Schipper, 2003).

(17)

16 Preparing the financial statements is interpreted differently by regulators, accountants and managers. Under a principle based approach it is easier to manipulate earnings in the financial statements (Nelson, 2003). According to Benston et al. (2006), earnings management will be reduced by detailed regulation, all possibilities for Earnings management are mitigated.

Under rules based approach, the guideline provides a clear description regarding with the preparation of financial statements. It is impossible for the managers to be subjective and manage the earnings artificially. Nelson (2003) indicates that the rules have a restriction. According to Nelson (2003), it is complex for auditing when the amount of rules increases. Based on the results of the firm, forming an opinion for financial analysts and investors is challenging for auditors (Nelson, 2003).

Ewert & Wagenhofer (2005) found that under rules based approach the comparability of financial statements are enhanced. On the other hand, the question remains of whether it will reflect the true financial performance of the firm (Ewert & Wagenhofer, 2005). Rule based approach does not require specialized knowledge for economic and business underpinnings for the financial reporting of the firm (Ewert & Wagenhofer, 2005). The accountant focusses solely on the compliance of rules (Carmona & Trombetta, 2008). Managers should communicate the accounting choices effectively between stakeholders and auditors. Auditors have to checked whether rules are being complied rather than using their judgment on financial statements (Carmona & Trombetta, 2008). According to Schipper (2003), following an audit plan and checking the steps makes auditors blind for the business under rules based approach. The United States accounting regulation are too complex and exhaustive to monitor earnings management in the financial statements, as confirmed in the study of Herz (2003). According to findings of Segovia et al. (2003), earnings management cannot be detect if the financial statement complies with the accounting regulation. That idea is given by rules based approach (Segovia et al., 2003). Nobes (2003) mentions that principles could replace a vast part of the regulations. This will reduce the complexity of the financial statements (Nobes, 2003).

2.5.3 Difference between principle based and rules-based accounting standards

The system of principles-based accounting standards is with a reasonable degree of specificity and based on the four accounting fundamentals: (a) decision usefulness, (b) a true and fair view, (c) going concern and (d) substance over form (Schipper, 2003). There are fewer detailed guidelines about the implementations. Principle based required more professional judgment and there is more space on how to interpret (Beest, 2009).

(18)

17 Rules-based accounting system is based on comprehensive and detailed guidelines regarding the implementation. The precise details describe what is allowed and how financial statements should reported (Alexander & Jermakowicz, 2006). For auditors, rules-based accounting standards are a precise elaboration and comprehensive on what ensures that financial statements are more easy to compare. The information accuracy is easy to verify (Nelson, 2003; Schipper, 2003).

2.6 Hypothesis development

There are contradictions in several studies in the relationship between accounting standards and earnings management. The research of Nobes (2005) concludes that organizations report a decrease of earnings management under IFRS standards. Principles based standards provides space for professional judgment (Nobes, 2005).

According to Bratton (2003), earnings management are reduced when the financial statements are reported under U.S. GAAP standards. The possibilities for earnings management are taken into account and minimized (Bratton, 2003). There are several ways to measure the adjustment in the rules. This has a demotivating or motivating impact on the size of earnings management (Healy & Wahlen, 1999). Under IFRS standards the degree of earnings management is affected by the discretion of the manager, this is shown in the studies of Benston et al. (2006) and Libby et al. (2009). Libby et al. (2009) found that IFRS standards lead to more earnings management. Earnings management appeared more in IFRS standard according to the study of Benston et al. 2006. There is evidence in prior studies (Chunhui et al., 2014; Graham et al., 2005; Nelson, 2003;) that non-U.S. firms reporting under IFRS arise more earnings management than in U.S. firms under U.S. GAAP (Barth et al., 2012; Evans et al. 2015). Under U.S. GAAP publicly listed firm adjust earnings upwards and prefer Real earnings management (Graham et al., 2005), while publicly listed firms under IFRS adjust earnings downwards and prefer Accrual based earnings management (Graham et al., 2005; Evans et al., 2015).

According to Nelson et al. (2003), the findings are that firms switching to a rules–based standard the use of Accrual based earnings management shift to Real earnings management. The earnings management remains the same (Nelson et al., 2003). In 2002, after the introduction of SOX, earnings management had still not decreased. The concerns of rules based regulatory had some influence on earnings management decisions (Cohen & Lys, 2008). Beest (2009) supports this finding, the research found that the nature of earnings management has changed and IFRS and U.S. GAAP lead to similar level of earnings management. The study of Beest (2009) showed that there is an increasing effect on Real earnings management occurred by rules based

(19)

18 accounting standard. In the study of Roychowdhury (2006) shows that the U.S. GAAP standard is significant to earnings management. According to Roychowdhury (2006), Real earnings management and accrual based management are significant. Zang (2012) supports Roychowdhury’s (2006) findings and mentions that both types of earnings management are significant in the U.S.. It is difficult to determine how earnings management is applied by the manager (Zang, 2012). According to Zhang (2012), managers always prefer a combination of both types of earnings management. Evans et al. (2015) mention that U.S. firms that report in U.S. GAAP standards prefer Real earnings management rather than Accrual based earnings management. However, there is no study that examined how different IFRS vs U.S. GAAP are used in different countries (Evans et al., 2015)

U.S. GAAP and IFRS accounting standards provide defined reporting boundaries and whether standards could have been violated with relatively objective determinations (Evans et al., 2015). According to Agoglia et al. (2011), IFRS are less precise accounting standards and allowing more reporting discretion than U.S. GAAP does. Detection in regulatory environment and relying on IFRS accountings standards are less effective than detection in environment relying on U.S. GAAP accounting standards (Evans et al., 2015).

The above studies show various claims concerning accounting standards in relation to earnings management. Several studies indicate that rules-based accounting standard has a significant relationship to Real earnings management based and principles-based has a significant relationship with accrual-based earnings management. In appendix A, Table A shows which papers are used for the development of the hypothesis. This has led to the following two hypotheses:

H1: Firms that report under IFRS standards are less likely to engage in Real earnings management than firms that report under U.S. GAAP.

H2: Firms that report under U.S. GAAP standards are less likely to engage in Accrual based earnings management than firms that report under IFRS.

(20)

19

3. Methodology

In this chapter, methodology and the structure of this study will be explained and described. As followed the data and sample description is explained. In the next three paragraphs the conceptual framework of the research will be discussed. In the following paragraph the operationalization of the variables will be explained. The fifth paragraph, the data and the method for gathering for data of the study will be explained. At last, the various steps of the study will be described.

3.1 Data and sample description

This study uses a quantitative research to test the hypotheses and to answer the research question. The population of three different countries are chosen and disturbing effects such as culture and political climate must be excluded. The two IFRS accounting standards population are Germany and France and one U.S. GAAP accounting standard population is the United States.

The initial sample includes all non-public and publicly listed from France, Germany and the United States for the fiscal years from 2005 to 2015. The period that was chosen is when the accounting standards were developed after several accounting scandals such as Enron and Ahold. The average time horizon is 20 years and in other similar studies such as Agoglia et al. (2011), Jamal & Tan (2010) and Zang (2012) they take a period from, respectively, 20 years, 20 years and 26 years. This period included the financial crisis. This study included ex ante financial crisis and post financial crisis (Agoglia et al., 2011; Jamal & Tan, 2010; Zang, 2012). In the period after the crisis, the economy is in a recovering phase and there is an opportunity for managers to apply more earnings management. This study included a time period of 10 years and compared it with other studies where this only a half of the average. The focusses is only on the post SOX and post IFRS adoption period for this study. For European companies, firms were selected based on the use of IFRS accounting standards. There is no distinguishing between publicly listed and non-listed firms. Data will be collected from Compustat Fundamentals database (Evans et al., 2015) and Bureau van Dijk. As in the studies of Roychowdhury (2006) and Zang (2012) the utility industry (SIC 4000-4900) and the financial industry (6000-6900) were excluded and countries were chosen randomly. In these industries, different regulations are applied which might affect the Real earnings management (Roychowdhury, 2006; Zang, 2012). The three datasets are obtain from 2 different databases and merged into one dataset.

The total amount of obervations without the selected criteria for the United States, France and Germany are 310,204 observations. In Europe, the sample contains 137,571 (France;

(21)

20 75,796 and Germany; 96,837) and in North America (United States) the sample contains 172,633. After eliminating accounting standards other than U.S. GAAP and IFRS, the initial observation of this study is 22,470 observations. In appendix B, Table B provide the sample observation. The three countries are divided in two accounting standards groups. The IFRS group contains France and Germany and the U.S. GAAP group contains United States. The respective observations are 12130 (France; 6134 and Germany; 5996) and 10,340. The total amount of firms that have a SIC code between 4400-4999 and 6000-6999 contains 10518 observations and are deleted from the total sample (Roychowdhury, 2006; Zang, 2012). In appendix B Table C divide the sample in industry codes.

3.2.1 The modified Jones Model

The modified Jones model is the most robust model to measure discretionary accruals (Bartov et al, 2000; Dechow et al., 1995). Bartov et al. (2000) stated “The modified Jones model is designed to eliminate the conjectured tendency of the Jones model to model discretionary accruals with error when discretion is exercised over revenues” (p. 9). The firm’s discretionary accruals are the differences between the normal accruals and the actual total accruals. Discretionary accruals remain for the model and need to be calculated from the dataset. The modified Jones model has the following formula to calculate discretionary accruals (Dechow et al., 1995):

TAt/At-1 = α1 NDAt + α3 DAt

NDAt = α1 (1/A_t-1) +α2 ((ΔREV_t- ΔREC_t)/ A_t-1) + α3 (PPE_t/ A_t-1) TAt/At-1= (ΔCAt - ΔCL t – Δ Cash t + Δ STD t – Dep t) / (At-1)

TAit = total accruals of firm x for year t. Calculated by the difference between income before extraordinary items and cash flow from operations

Ait-1 = total assets of firm x at the beginning of year t ΔREVit = Change in revenues of firm x for year t-1 to year t ΔRECit = Change in receivables of firms x for year t-1 to year t PPEit = Property, plant and equipment of firm x for year t

Eit = residual, representing discretionary accruals of firm x in year t

3.3 Model to measure Real earnings management

Three methods are illustrated and described as ‘abnormal levels of discretionary spending’, ‘abnormal levels of production expenses’ and ‘cash flow from operating’ for Real earnings management (Roychowdhury, 2006). To measure the degree of Real earnings management, Cohen et al. (2008), Cohen and Zarowin (2008) and Zang have shown that these methods measured the degree of Real earnings management effectively.

(22)

21 Firms trying to avoid losses by engaging to exceed the budgeted production plans to decline the cost of goods sold, offering lower sales prices to increase sales temporarily and reducing or postponing the discretionary expenditures to improve earnings (Chunhui et al., 2014; Roychowdhury, 2006). The study of Roychowdhury (2006) used the following formulas to examine the level on abnormal production (PROD, discretionary expenditures (DIX) and abnormal cash flows (CFO) (Chunhui et al., 2014; Cohen, 2007; Gunny, 2005).

PROD

According to Roychowdhury (2006), managers can override the production plans to increase earnings. By producing more units, fixed costs per unit can be declined due to the fact that fixed overhead costs are divided over a larger quantity of units. With a lower cost of goods sold than originally planned in the production plans, the company is able to report higher gross margins (Roychowdhury, 2006). The abnormal levels of production is estimated by the regression model of Roychowdhury (2006): COGS Assets t − 1= β0 + β1 ( 1 Assets t − 1) + 𝛽2 ( Sales t Assets t − 1) + εt ∆ INVT t Assets t−1 = β0 ( 1 Assets t − 1) β1 ( ( ∆ Sales t Assets t − 1) + β2 ( ∆ Sales t − 1 Assets t − 1) + εt PROD t Assets t − 1 = β0 + β1 ( 1 Assets t − 1) + β2 ( Sales t Assets t − 1) + β3 ( ∆ Sales t Assets t − 1) + β4 (∆ Sales t − 1 Assets t − 1) + εt Where:

PRODit = Sum of the cost of sales in year T plus the change in the stock related to the year T-1. (COGS + ΔINV)

Assets t-1 = Total assets in year T-1

Sales t = Net sales of company x in year T

Δ Sales t = Change net sales of company x in year T Δ Sales t-1 = Change net sales of company x in year T-1

Eit = St. Error

DIX

Postponing or cutting in selling, general & administrative and research and development expenses will decrease the entity’s costs and increases earnings for the short term. These expenses are settled in cash, this will affect the next period that the cash flow increase (Roychowdhury, 2006). Dechow et al. (1998) developed a model that used to generate the normal levels of

(23)

22 discretionary expenses (Dechowet al., 1998). Roychowdhury (2006) also implemented this model is his study. The Abnormal levels of discretionary expenditure formula of Roychowdhury (2006):

𝐷𝐼𝑆𝑋 t 𝐴𝑠𝑠𝑒𝑡𝑠 t − 1 = 𝛽0 + 𝛽1 ( 1 𝐴𝑠𝑠𝑒𝑡𝑠 t − 1) + 𝛽2 ( 𝑆𝑎𝑙𝑒𝑠 t − 1 𝐴𝑠𝑠𝑒𝑡𝑠 t − 1) + 𝜀𝑡 Where:

DIXit = Discretionary spending including the sum of R&D, advertising, sales, general and administrative costs of company in year T

Assets t-1 = Total assets of company x in year T-1 Sales t-1 = Net sales of company x in year T-1

𝜀𝑡 = St. Error

CFO

Sales temporarily increases on short terms when the firm is using lenient credit terms and price discounts. Once the firm reverses the normal prices, this sales boost will disappear. This will affect the cash flow as lower in the current period, but the additional sales will boost the current earnings (Roychowdhury, 2006). The following are equations according to Roychowdhury (2006) to estimate the normal cash flow from operation (Roychowdhury, 2006):

CashFO t Assets t₋₁ = β0 + β1 ( 1 Assets t₁) + β2 ( Sales t Assets t₁) + β3 ∆Sales t Assets t₋₁+ εt Where:

CFOt = operational cash flow from year t; At-1 = total assets in year t–1;

Sales t = sales in year t

Δ Sales t-1 = mutations in net seals in year t relative to t-1

Et = st. error

3.4 Dependent variables

In previous studies (Kim et al., 2012, Roychowdhury, 2006; Zang, 2012) to detect Real earnings management there are three proxies. Proxy abnormal production costs (AB_PROD) is the first proxy in the study of Kim et al. (2012), Roychowdhury (2006) and Zang, (2012). Abnormal operating cash flows (AB_CFO) is the second one in their study. The last proxy is abnormal discretionary expenses (AB_DIX). This study will use the three proxies of Roychowdhury (2006) to measure Real earnings management. The three proxies are into three separate formulas of Roychowdhury (2006) . This study use has the three following formulas (Roychowdhury, 2006): PRODit /Ait-11 = ß0+ ß 1(1/Ait-1) + ß2 (Sit/Ait-1) + ß3 (ΔSit/Ait-1) + ß4 (Δ

Sit-1/Ait-1) + εt

CFOit / Ait-1 = β0 + β1 (1/Ait-1) + β2 (Sit/Ait-1) + β3 (∆Sit/Ait-1) + εt DISXit /Ait-1 = ß 0 + ß 1(1/Ait-1) + ß2 (Sit-1/Ait-1) + εt

(24)

23 The combined REM model to calculate Real earnings management for this study is;

Combined REM = PRODit – CFOit - DISXit

This study use the modified Jones model (Jones, 1991) to detect Accrual based earnings management (ABS_DA). According to Dechow et al. (1995) the modified jones (Jones, 1991) is the most powerful model to recognizing discretionary accruals. The formula for the modified jones model (Jones, 1991) as follow:

TAt/At-1 = α0 + α1 (1/At-1) +α2 (ΔREVt - ΔRECt / At-1) + α3 (PPEt/ At-1) + εt

3.5 Control variables

This research uses different control variables which are explained in this paragraph. This study consists the variable accounting standard. Accounting standard (AS) is added as a dummy variable. When IFRS standard is used, the research variable will have dummy 1. The U.S. GAAP standard gets a dummy variable with value 0. In the dataset, countries are a variable. Due to dividing the countries in accounting standard groups, this will cover the variable countries to run a regression.

Other factors that might have an influence on Accrual based earnings management and Real earnings management. For this reason, other control variables are needed in this study. The size of a firm is an important control variable that may have an enhancing effect on earnings management (SIZE 𝜀𝑡) (Klein 2002). The entity size is measured in the total assets of a company (Klein 2002). Besides, the included control variables in other studies are the cash flow from operation (CFO) or the cash flow from ordinary operations (Roychowdhury, 2006; Zang, 2012). Ibrabim & Loyd (2011) used the control variable cash flows including in operating activities on earnings management which may have an increasing effect.

In studies of Roychowdhury (2006) and Zang (2012), a firm’s profitability is measured by net income/total assets. Roychowdhury (2006) and Zang (2012) use the return on asset (ROA 𝜀𝑡) as a control variable. The control variables to manage earnings are the firm’s leverage ratio (LEV 𝜀𝑡) and debt to equity ratio (De 𝜀𝑡) are included for leverage-related incentives (Sweeney, 1994; Teoh et al, 1998). The LEV 𝜀𝑡 is measured by total debt divided by total equity and DE 𝜀𝑡 is measured by long term debt divide by total equity.

Auditors that execute higher quality audits are more likely to discover earnings management (Evans et al., 2015). In a study of Cohen et al. (2008) there is less Accrual based earnings management possible and audits of big 4 firm are of higher quality. Badersche et al. (2010) concluded similar results with prior research (Cohen et al., 2008), firms with a higher financial reporting quality are audited by big 4 auditors and engage less in earnings management.

(25)

24 As control variable (AUD), firms audited by non-big 4 firms gets a dummy variable 0 and firms that are audited by big 4 firm have a dummy variable 1.

Estes and Hosseini (2001) reported that men generally have a higher level of confidence in making business decisions. When the level of self-assurance is high due to a high level of confidence, opportunities and certain risks are taken by an executive (Estes & Hosseini, 2001). Therefore, it is expected that a man applies more Real earnings management than a woman. The value of the dummy variable for a woman is 0 and the value for a man is 1. Therefore, in this study the control variable GEND will be used to measure.

Changes in revenues (GROWTH) is the last control variable to measure the percentage change of sales (Barth et al., 2008). In the study of Barth et al. (2008) this control variable was also used for economic effects and companies have a greater incentive to use earnings management (Rusmin, 2014).

3.6 Empirical model

This study has resulted in two hypotheses to answer on the research question: ‘What is the effect of Real earnings management for companies that are using IFRS accounting standards than for companies that are using U.S. GAAP accounting standards?’ This study expects a negative relation between Real earnings management and IFRS accounting standards. The observations are random and independent from each other. This study contains four different dependent variables DA, CFO, DIX and PROD. A multiple linear regression model will be used in this study. The two hypotheses that will be tested with the regression model are:

The first hypothesis that will be tested:

The regression model to test the first hypothesis is:

H1: Firms that report under IFRS standards are less likely to engage in Real earnings management than firms that report under U.S. GAAP.

1R_ AB_PROD = β0 + β1AS 𝜀𝑡 + β2SIZE 𝜀𝑡 +β3CFO 𝜀𝑡 + β4ROA 𝜀𝑡 +β5LEV 𝜀𝑡 +β6DE 𝜀𝑡 + β7AUD 𝜀𝑡 + β8GEND 𝜀𝑡 + β9Growth 𝜀𝑡 + 𝜀𝑡

2R_ AB_EXP = β0 + β1AS 𝜀𝑡 + β2SIZE 𝜀𝑡 +β3CFO 𝜀𝑡 + β4ROA 𝜀𝑡 +β5LEV 𝜀𝑡 +β6DE 𝜀𝑡 + β7AUD 𝜀𝑡 + β8GEND 𝜀𝑡 + β9Growth 𝜀𝑡 + 𝜀𝑡

3R_AB_CFO = β0 + β1AS 𝜀𝑡 + β2SIZE 𝜀𝑡 +β3CFO 𝜀𝑡 + β4ROA 𝜀𝑡 +β5LEV 𝜀𝑡

+β6DE 𝜀𝑡 + β7AUD 𝜀𝑡 + β8GEND 𝜀𝑡 + β9Growth 𝜀𝑡 + 𝜀𝑡

(26)

25 The second hypothesis that will be tested:

H2: Firms that report under U.S. GAAP standards are less likely to engage in Accrual based earnings management than firms that report under IFRS.

Regression model to test the second hypothesis:

5 R_ABS_DA = β0 + β1AS 𝜀𝑡 + β2SIZE 𝜀𝑡 +β3CFO 𝜀𝑡 + β4ROA 𝜀𝑡 +β5LEV 𝜀𝑡 +β6DE 𝜀𝑡 + β7AUD 𝜀𝑡 + β8GEND 𝜀𝑡 + β9Growthit + 𝜀𝑡

(27)

26

4. Results

The results of this study are discussed in this chapter. In the first paragraph, the descriptive statistics of the sample are evaluated. Then, the Pearson’s correlation matrix is analysed and explained. Lastly, the results of the main analysis are discussed about what effect IFRS and U.S. GAAP standards have on the level of earnings management that are given in the coefficients tables.

4.1 Descriptive statistics

The sample that is used in this study will be illustrated by the descriptive statistics. This description contains the sample from United States and Europe. In the samples from Europe, France and Germany, various factors such as cultural differences and political climate have to be eliminated. Firms who operate in financial industries with (SIC codes between 6000 and 6500) or in regulated industries (SIC codes between 4400 and 4999) are eliminated from this study (Roychowdhury, 2006) ; (Zang, 2012).

The sample for the dependent variables AB_CFO and DA_ABS are respective 9353 and 9208. In comparison with the sample of Zang (2012) which 7592 is corresponding. Zang (2012) use a period from 1987 – 2008 and this study contains a period of 10 years (2005 - 2015. The dependent variable AB_PROD have sample of 7223 observations.

Outliers are removed in this dataset and are not able to influence the outcome. In the next table gives an overview of the sample. The manner that outliers are removed is through winsorizing at 1% or 99%. Table 1 shows the mean, standard deviation, median, minimum and maximum for each variables with accounting standards as factor.

(28)

27

Table 1 Descriptive statistics

Table 1 shows the mean for U.S. GAAP is higher on dependent variable DA_ABS and AB_REM than IFRS. In this study can make an assumption that in U.S. GAAP accounting standards applies more earnings management for Accrual based earnings management and Real earnings management. This study determines the effect of Earnings management through a conscious choice to select firms of various sizes as in the studies of Roychowdhury (2006) and Zang (2012). There are no particular abnormalities found in the other variables.

4.2 Multicollinearity of variables

This study has executed a multicollinearity test by using a Pearson’s correlation test to see there is a high degree of correlation between the variables. The Pearson correlation test is in Table 2 shows whether there is a bivariate relationship is between the ratio of variables. The correlation between the mulitcollinearity and independent variables is to determine, among other things that

Variables Acc. Standards Mean Median Std. Deviation Min. Max. 0 US GAAP 0.109 0.053 0.164 0.001 0.888 1 IFRS 0.032 0.024 0.040 0.001 0.888 0 US GAAP 0.556 0.416 0.582 0.000 3.429 1 IFRS 0.579 0.487 0.457 0.000 3.429 0 US GAAP 0.114 0.058 0.206 0.000 2.051 1 IFRS 0.436 0.356 0.315 0.000 2.051 0 US GAAP 0.044 0.073 0.246 -1.215 0.531 1 IFRS 0.056 0.072 0.139 -1.215 0.531 0 US GAAP 0.395 0.290 0.642 -1.346 3.244 1 IFRS 0.092 0.068 0.556 -1.346 3.244 0 US GAAP 6,118.796 404.916 15,841.947 2.044 84,017.000 1 IFRS 4,637.702 260.848 13,141.467 2.044 84,017.000 0 US GAAP 513.553 23.054 1,316.763 -112.836 6,961.372 1 IFRS 342.088 16.502 1,042.355 -112.836 6,961.372 0 US GAAP -0.034 0.031 0.285 -1.660 0.310 1 IFRS 0.006 0.032 0.134 -1.660 0.310 0 US GAAP 121.105 1.575 417.377 -107.149 2,234.179 1 IFRS 3.665 1.314 54.890 -107.149 2,234.179 0 US GAAP 0.300 0.006 0.915 -1.962 7.037 1 IFRS 0.464 0.235 0.909 -1.962 7.037 0 US GAAP 0.230 0.090 0.651 -0.740 3.968 1 IFRS 0.105 0.045 0.460 -0.740 3.968 0 US GAAP 0.802 1.000 0.399 0.000 1.000 1 IFRS 0.383 0.000 0.486 0.000 1.000 0 US GAAP 0.974 1.000 0.160 0.000 1.000 1 IFRS 0.759 1.000 0.428 0.000 1.000 GROWTH_W Auditor Gender SIZE_W CFO_W ROA_W LEV_W DE_W DA_ABS_W AB_PROD_W AB_DIX_W AB_CFO_W AB_REM_W

(29)

28 the independent variables is not too high. All dependent and control variables from Table 2 in this paragraph are explained as are the relations between the variables.

The dependent variable Accruals-based earnings management (DA_ABS) is stronger negatively correlated with the control variables. All control variables are significantly correlated with the dependent variable. Control variables that are negatively significantly correlated are the variables Accounting standards (p < 0.01), SIZE(p < 0.01), CFO(p < 0.01), ROA(p < 0.01), LEV(p < 0.01), and DE(p < 0.01). Compared with the study of Zang (2012) the variable ROA and SIZE are positively correlated. In the study of Zang (2012), the variable LOGSIZE is the strongest positive correlation with the other control variables. This study shows that the ROA (-.336) is the strongest negative correlation. To interpret this value of SIZE, it seems that the firm size is not as important a factor compared to the other variables. As the findings can explain, larger entities are not more likely to invest in project or assets than smaller entities. The coefficient of variable accounting standards (-.331) can be interpreted as the accounting standards between IFRS and U.S. GAAP do not have a high effect on Accrual based earnings management. The study of Evans et al. (2015) concluded these results. Evans et al. (2015) stated accounting standards are not a panacea for earnings management.

The correlation between the first dependent variable PROD of Real earnings management is more positive correlated with the control variables. In the study of Roychowdhury (2006) the variables Growth, CFO and ROA are negatively significantly correlated. This study showed that the control variable ROA (p < 0.01) and Growth (p < 0.01) have the opposite results and CFO (p > 0.1) is not significant. The SIZE is negative correlated and not significant. This means that the overproduction is not dependable on the firm’s size and the accounting standards do not override the production plans. Net income plays an important factor in overproduction The cost of goods sold is through overproduction lower. This effect that the gross margin gains.

The second dependent variable CFO (AB_CFO) of Real earnings management has a negatively stronger correlation with the control variables. All the control variables Accounting Standards (p < 0.05), SIZE (p < 0.01), CFO (p < 0.01) and ROA (p < 0.01) are positively correlated and significant as in study Roychowdhury (2006). The control Variables are negative significant LEV (p < 0.01), Growth (p < 0.01) and DE (p > 0.1) show the opposite results compared to the study of Zang (2012). DE (p > 0.1) is not significant.

The last dependent variable of Real earnings management is the abnormal expenses (AB_DIX). This variable is less significant compared to dependent variable AB_CFO and AB_PROD. The dependent variable AB_DIX has a positively stronger correlation to the control

(30)

29 variables. The variables Accounting Standards (p < 0.01), ROA (p < 0.01) and GROWTH (p < 0.01) are positive correlated. This study shows that variable SIZE (p > 0.1), CFO (p > 0.1), LEV (p > 0.1) and DE (p > 0.1) have no significant correlation to AB_DIX regarding to the studies of Roychowdhury (2006) and Zang (2012). The accounting standards constrains the firms to reducing or postponing in the operational expenditure or cutting in the research and development expenses.

The combined REM variable is more positively correlated to the control variables. The variable Accounting Standards (p < 0.01), SIZE (p < 0.01), ROA (p < 0.01), LEV (p < 0.01) and GROWTH (p < 0.01) are significant. The accounting standards cannot reduce Real earnings management. Therefore is the interpretation the coefficient (0.033) in which accounting standards is more Real earnings management.

(31)

30

Tabel 2 Pearson’s correlation

1 2 3 4 5 6 7 8 9 10 11 12 13 14 Pearson Correlation 1 Sig. (2-tailed) Pearson Correlation .175** 1 Sig. (2-tailed) 0.000 Pearson Correlation .031* 0.009 1 Sig. (2-tailed) 0.022 0.500 Pearson Correlation -.345** -.033* -.170** 1 Sig. (2-tailed) 0.000 0.014 0.000 Pearson Correlation .228** .827** -.416** -.263** 1 Sig. (2-tailed) 0.000 0.000 0.000 0.000 Pearson Correlation -.331** 0.022 .494** .033* -.243** 1 Sig. (2-tailed) 0.000 0.106 0.000 0.016 0.000 Pearson Correlation .033* 0.020 -.245** .104** .108** -.411** 1 Sig. (2-tailed) 0.015 0.130 0.000 0.000 0.000 0.000 Pearson Correlation .080** -.052** -.169** 0.011 .042** -.288** .102** 1 Sig. (2-tailed) 0.000 0.000 0.000 0.404 0.002 0.000 0.000 Pearson Correlation -.113** -0.017 -.158** .069** .046** -.051** .173** 0.018 1 Sig. (2-tailed) 0.000 0.218 0.000 0.000 0.001 0.000 0.000 0.173 Pearson Correlation -.112** -0.018 -.148** .130** 0.021 -.072** .198** 0.022 .913** 1 Sig. (2-tailed) 0.000 0.190 0.000 0.000 0.119 0.000 0.000 0.108 0.000 Pearson Correlation -.336** .050** -.125** .733** -.135** .095** .064** -0.005 .079** .115** 1 Sig. (2-tailed) 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.688 0.000 0.000 Pearson Correlation .104** -0.008 -.087** -.076** .061** -.212** 0.024 .053** -.062** -.061** -.108** 1 Sig. (2-tailed) 0.000 0.537 0.000 0.000 0.000 0.000 0.075 0.000 0.000 0.000 0.000 Pearson Correlation -.084** -0.017 -.067** 0.000 0.018 .087** .048** -0.017 .113** .062** 0.017 .033* 1 Sig. (2-tailed) 0.000 0.217 0.000 0.982 0.174 0.000 0.000 0.197 0.000 0.000 0.201 0.014 Pearson Correlation .283** .153** .064** -.078** .099** -.112** .058** 0.015 -.054** -.043** -0.007 0.015 -0.006 1 Sig. (2-tailed) 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.257 0.000 0.002 0.607 0.253 0.638 14. GROWTH_W

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed). c. Listwise N=5508 9. SIZE_W 10. CFO_W 11. ROA_W 12. LEV_W 13. DE_W 4. AB_CFO_W 5. AB_REM_W 6. Accounting Standards 7. Auditor 8. Gender 1. DA_ABS_W 2. AB_PROD_W 3. AB_DIX_W

The effect of accounting standards on Earnings Management in U.S. and Europe

The effect of accounting standards on Earnings Management in

U.S. and Europe.

Table of Contents