The effect of mandatory IFRS adoption on earnings management in Europe,moderated by investor protection.

(1)

The effect of mandatory IFRS adoption

on earnings management in Europe,

moderated by investor protection.

Master Thesis

Author: Ilse M.W. Coenen, MSc (10543058)

Institution: University of Amsterdam, Faculty Economics and Business Program: Accountancy and Control

Supervisor: dr. A. (Alexandros) Sikalidis Second Supervisor:

Date: June, 22th₂₀₁₄ Concept: Final version

(2)

Key words: IFRS, investor protection, earnings management, agency theory Abstract

Prior studies suggest that mandatory IFRS adoption is associated with a decrease in earnings management because of the stronger regulations. In this study I examine whether mandatory IFRS adoption in Europe affect earnings management when this relation is affected by investor protection. Many studies are done regarding this topic, yet not the effect moderated by investor protection. With higher investor protection, earnings management would decrease because of the fewer incentives available to disguise the performance of the firm. This is the result of the strong investor protection that limits managers’ ability to acquire private control benefits (Leuz et al., 2003). As mandatory IFRS adoption is the results of many scandals, I would expect that the investor protection would be higher which results in less earnings

(3)

Acknowledgements

The goal of writing this master thesis is to obtain the Master of Science degree in Accountancy & Control at the Amsterdam Business School, a step further in the process of being a register accountant. I am most thankful to Alexandros Sikalidis, for the support during the writing of my master thesis. I appreciate the guidance and assistance he delivered during this process. His advice, suggestions and comments were very helpful.

I would also like to thank KPMG Amstelveen for giving me the opportunity to write my thesis at their office. Especially I would thank Marjolein Hoevenaars for her support and time.

(4)

1. Introduction

European listed companies are necessitated to report consolidated financial statements prepared according to IFRS and no longer to local GAAP or other standards, since the mandatory IFRS adoption in 2005 (Ball, 2006). This move forced over 7,000 public European companies to replace various domestic accounting standards with IFRS (Byard et al., 2011). Improving the accounting quality with the development of an internationally acceptable set of high quality financial reporting standards (Barth et al., 2007) in order to contribute to better functioning capital markets (Quigley, 2007) and to harmonize corporate accounting practice is the main goal of the International Accounting Standard Board (IASB) (Tendeloo & Vanstraelen, 2005). Therefore, mandatory IFRS adoption has the potential to facilitate cross-border comparability, increase reporting transparency, decrease information costs, reduce information asymmetry, and also increase the liquidity, increase the competitiveness and increase the efficiency of markets (Ball, 2006; Choi & Meek, 2005). The success of this change, in the sense of the benefits exceeding the costs, has economic consequences for firms within the EU, but also for firms worldwide, as IFRS increasingly will become the global standard.

The purpose of this paper is to help fill the void in the literature by examining the association between IFRS adoption and earnings management in Europe. Mandatory IFRS adoption and earnings management are often-discussed topics in the literature. Recent research focused on the importance of understanding how firms manage earnings through real earnings management in addition to accrual-based earnings management (Cohen and Zarowin, 2010; Cohen et al., 2008; Gunny, 2010; Roychowdhury, 2006; Zang, 2012). However, no research has pointed out the effect of IFRS adoption on earnings management, moderated by investor protection in Europe. As a result, this paper focuses on institutional factors that influence the relation between mandatory IFRS adoption and earnings management. This study seeks to contribute to the scientific literature by providing insight into the effect of the mandatory IFRS adoption on earnings management by European firms. Given the increasing popularity of earnings management within firms, it is socially relevant to determine if the mandatory IFRS adoption leads to earnings management, when investors are protected.

(6)

to engage in earnings management and affect the quality of reported earnings, particularly when this relation is influenced by investor protection. Prior research provides disagreements in the effect of mandatory IFRS-adoption on earnings management. For example, Doukakis (2013) did research into the effect of mandatory IFRS adoption on real-based and accrual-based earnings management activities. He used a sample of 15,206 observations from 22 European countries between 2000 and 2010. The results suggest that mandatory IFRS adoption had no significant impact on either real-based or accrual-based earnings management practices. In contrast, Jeanjean and Stolowy (2008) focused on Australia, France and the UK, three IFRS first-time adopter countries. After the introduction of IFRS, the level of earnings management didn’t decline in these countries, earnings management even increased in France. Jeanjean and Stolowy (2008) suggest that the standard setters and commissions should focus on modifying incentives and institutional factors instead of modifying current accounting standards. Also Capkun et al. (2012) find evidence that earnings management decreased after the adoption of IAS/IFRS for early adopters (pre-2004). The results also show that earnings management increased for late and mandatory adopters after their adoption of IFRS in 2005. These contradictions in the previous literature lead to the following research question:

“Does mandatory IFRS adoption lead to earnings management and what is the

effect of investor protection on this relation?

Following Healy and Wahlen (1999), earnings management is defined as “the alteration of firms’ reported economic performance by insiders either to mislead stakeholders, or to influence contractual outcomes”. According to Leuz et al., (2003) a conflict of interest is the main reason for having incentives to misrepresent firm performance through earnings management. At the expense of stakeholders (like investors), controlling owners or managers can use their control over the firm to benefit themselves (Leuz et al., 2003). As a result, some value is not shared with non-controlling outsiders, like investors and enjoyed only by inside managers. Non-controlling outsiders, like investors, can be protected for these actions through legal systems like investor protection (e.g., La Porta et al., 1998; Claessens et al., 2002). Legal systems that effectively protect outside investors reduce insiders’ need to conceal their activities. As a result, I propose that there is more earnings management in European countries where is less legal protection of outside investors, because as a result, in these

(7)

countries insiders (managers) enjoy greater private control benefits and hence have stronger incentives to modify firm performance (Leuz et al., 2003).

To test the above expectation, I use a quantitative research method, particularly empirical archival research. For my data I use the database of Compustat and the World Bank. The analysis is based on financial accounting data from 2001 to 2013 for 9807 firms from 30 European countries.

The results indicate that there is no correlation between mandatory IFRS adoption and earnings management. There is also no correlation between investor protection and earnings management. An increase in investor protection doesn’t lead to less earnings management. In the additional analysis, I examine whether investor protection is influenced by legal origin. There is a correlation between legal origin and investor protection.

The remainder of this paper is organized as follows. In section 2, I review related literature and develop my hypothesis. I discuss the sample and variables in section 3. Section 4 reports tests and results. Finally, I discuss the conclusions, limitations, and directions for future research in section 5-7. The end contains the references and appendix.

(8)

2. Literature review and hypotheses

2.1 Accounting Standards

Accounting standards are resulted from a complex interaction among numerous parties including agencies of the Federal government, state regulatory commissions, public accountants, quasi-public accounting standard-setting boards, and corporate managements. The main goal of these parties is to expend resources to influence the setting of accounting standards, in the past and they will continue doing this (Watts & Zimmerman, 1978, p. 112). Before the mandatory IFRS adoption in Europe in 2005, most companies used GAAP. In this paragraph, IFRS is described.

2.1.1 IFRS

The International Accounting Standards Board (IASB) issues accounting standards, namely International Financial Reporting Standards (IFRS) (Ball, 2006). The IASB is an organization based in London, United Kingdom, designed for the use by profit-oriented entities and is fully independent. According to the IASB, IFRS is a set of rules that ideally would apply to financial reporting by all public companies worldwide. An entity claiming compliance with IFRS must comply with all standards and interpretations, including disclosure requirements (Holla, 2006). Before the presence of the IASB, from 1973 till 2000, the International Accounting Standards Committee (IASC) issued international standards. The IASC is established by professional accountancy bodies in Australia, Canada, France, Germany, Ireland, Japan, Mexico, the Netherlands, United Kingdom and the United States (Ball, 2006). Since the presence of the IASB, in April 2001, the rules-based International Accounting Standards (IAS) of the IASC are replaced by more principle-based standards. In 2001, as a non-profit corporation, the IASC Foundation was integrated in the United States. As a result, the IASB is the legal daughter of the IASC. Although the IASB continues to use some of the prior rules of the IASC, it describes the current rules under a new label, namely ‘International Financial Reporting Standards’. Overall, the new standard setter is better funded, better staffed and more independent than its predecessor, the IASC (Ball, 2006). But nevertheless, there are still a lot of agreements between the two kinds of Accounting Standards.1

(9)

2.2 Earnings Management

“Earnings management is the choice by a manager of accounting policies, or actions affecting earnings, so as to achieve some specific reported objective” (Scott, 2011, p.403). Following Healy and Wahlen (1999, p. 366), when judgment is used in structuring transactions and financial reporting, earnings management occurs. The main cause of the judgment is a modified financial report. There are two main reasons for modifying financial reports. First contractual outcomes that depend on this reported accounting practices can be influenced or second to mislead investors about the underlying economic performance of the company. The judgment is the result of imperfect auditing. Leuz et al. (2003, p. 506) argue that a conflict of interest between the managers and investors is the cause of incentives to misrepresent firm performance through earnings management. At the expense of investors, managers can use their power and control over the organization to benefit themselves. In the end, some value is enjoyed entirely by insiders and not shared with non-controlling outsiders, which is defined as earnings management.

Like Bartov (1993, p. 841) demonstrates, earnings management can be achieved by using accounting methods and estimates (i.e., accrual-based manipulation) or by undertaking transactions that make reported income closer to some target number than it would otherwise be, rather than maximize the firm’s discounted expected cash flow (i.e., real manipulation). In other words, accruals-based earnings management activities have no direct cash flows consequences and real activities manipulations do affect cash flows (Cohen & Zarowin, 2010). So, real earnings management activities are significantly different than accrual-based activities as they have direct cash flow effects.

Consistent with the literature, there is a distinction between earnings management activities using discretionary accrual activities and real operational activities (Healy and Wahlen, 1999; Fudenberg and Tirole, 1995; Dechow and Skinner, 2000). Cohen et al. (2008) did research into the changes in earnings management, both real-based and accrual-based, after and before the passage of Sarbanes-Oxley Act (SOX). They wanted to research if the passage of SOX influenced both ways of earnings management. In their paper they show, with graphical illustrations, that earnings management increased steadily over the sample period. But out of these results there is a distinction between accrual-based earnings management and based earnings management. After the passage of SOX, in 2002, real-based earnings management increased, while accrual-real-based earnings management declined.

(10)

As a conclusion, after the passage of SOX, firms shifted to real-based earnings management. Graham et al. (2005) also suggested this conjecture between accrual-based earnings management and real-based earnings management. They show that the combination of the Sarbanes–Oxley Act requirements and accounting scandals at WorldCom and Enron could have changed the preferences of managers for the mix between accrual-based earnings management and real-based earnings management.

2.4 The baseline relation between mandatory IFRS adoption and earnings management

Agency theory focuses on the conflict that arises from the separation between principals and agents (Walker, 2013). The main reason to believe that the agent (manager) will not always act in the best interest of the principle (investor) is because both parties in the relationship want to maximize their own utility (Jensen & Meckling, 1976, p. 308). The abnormal activities that negatively influence the principles can be limited by establishing appropriate incentives for the agent and by incurring monitoring costs. Through this, the principal could limit divergences from his interest (Walker, 2013). In some situations the agent receives money for expending bonding costs. This is a guarantee to the principles that the agent will not take certain actions that would harm them. Expending bonding costs can also be done to ensure that the principal will be compensated if the agent does take actions that harm the principles. It is generally impossible for both the agent as the principle to ensure that the agent will make perfect decisions that doesn’t negatively impact the principles, at zero cost. The agent and the principle will obtain positive bonding costs and monitoring. In addition there will be some divergence between the decisions of the agent and those decisions that would maximize the welfare of the principal. A cost of the agency relationship is the dollar equivalent of the principle’s reduced welfare experience, and this latter cost can be referred as the “residual loss” (Jensen & Meckling, 1978, p. 308). According to Walker (2013), agency theory emphasizes that firms operate under conditions of uncertainty and this leads to potential information asymmetries between the agent and the principal. There are two main themes according to information asymmetry: moral hazard problems and adverse selection problems. First, if principles are unable to observe the actions and choices that agents have made, a moral hazard problem arises. Second, an adverse selection problem arises if agents have private access to value- relevant information (Walker, 2013). In other words, there could be problems according to hidden action and hidden information (Jensen & Meckling, 1978).

(11)

To influence the agency conflict between outside investors (principles) and controlling insiders (agents), investor protection is necessary. In other words, the quality of financial information reported to investors can be improved by investor protection.

Figure 1. Conceptual model

Graham et al. (2005) show that the combination of the Sarbanes–Oxley Act requirements and accounting scandals at WorldCom and Enron could have changed the preferences of managers for the mix between accrual-based earnings management and real-based earnings management. In their results, Cohen et al. (2008, p. 785) show with graphs that after the passage of SOX real-based earnings management increased, while accrual based earnings management declined. Hence, with the mandatory adoption of IFRS managers have more flexibility according to financial statements. I acknowledge that this would result in more real earnings management and in less accrual based earnings management, but overall in less earnings management because of the fact that mandatory IFRS adoption is the consequence of many scandals. I propose the following hypothesis:

Hypothesis 1. Mandatory IFRS adoption leads to more real earnings management and

less accrual based earnings management.

The paper of Leuz et al. (2003) shows that there are systematic differences across 31 countries in the level of earnings management. They perform a couple of clusters to identify groupings of countries with almost similar institutional characteristics. According to these clusters, they describe the difference in earnings management between the groupings of these countries. According to this analysis they show that countries with relatively strong investor protection ensure lower levels of earnings management compared countries with weak investor protection. Taken together, this discussion leads to the second hypothesis, which concerns the interaction between investor protection and earnings management:

Mandatory IFRS adoption

Earnings management Investor protection

(12)

Hypothesis 2. High investor protection has a negative effect on both real earnings

management and accrual-based earnings management.

In countries where the investor protection is stronger, there will be less earnings management. The underlying reason is that incentives to mask firm performance will be reduced by strong investor protection. When the investor protection is weak, it appears to result in poor-quality reporting which results in more earnings management (Leuz et al., 2003, p. 309).

(13)

3. Research methodology

This chapter covers the methodological justification of the research surrounding the research approaches and choice of research strategy used. Until today there is still no information on the effect of mandatory IFRS adoption on earnings management in Europe when this relation is moderated by investor protection. To obtain information, data is collected from Compustat and the World Bank.

3.1 Research method

Worldwide, there is more focus on real earnings management instead of accrual-based earnings management after the implementation of SOX (Cohen et al., 2008). Nevertheless, during the research I focus on both real earnings management and accrual based earnings management before and after the in 2005 implemented mandatory adoption of IFRS in Europe.

3.1.1 Accrual-Based Earnings Management

As in DeFond & Subramanyam (1998) and Cohen et al. (2008), I use a cross-sectional version of the modified Jones model. Following Kim et al. (2012), the model is estimated for observation: 𝑇𝐴, 𝐴𝑠𝑠𝑒𝑡𝑠_, = 𝛼0 1 𝐴𝑠𝑠𝑒𝑡𝑠_, + 𝛼1 ∆𝑅𝐸𝑉, 𝐴𝑠𝑠𝑒𝑡𝑠_, + 𝛼 ∆𝑃𝑃𝐸, 𝐴𝑠𝑠𝑒𝑡𝑠_, + 𝜖, where:

TAit = total accruals for a firm i at year t, EBXIit - CFOit, where EBXI is the

earnings before extraordinary items and discontinued operations and CFO is the operation cash flows taken from the statement of cash flow;

ΔREVit = change in net revenues in year t from year t-1;

PPEit = gross property, plant, and equipment;

(14)

𝜖 _, = residual, proxy for discretionary accruals.

3.1.2 Real Activities Manipulation

I rely on prior studies to develop proxies for real earnings management. As in Roychowdhury (2006), I consider the following four measures to observe real activities manipulation:

1. Abnormal levels of operating cash flows (AB_CFO); 2. Abnormal production costs (AB_PROD);

3. Abnormal discretionary expenses (AB_DISCX); and

4. A combined measure of real activities manipulation (RM_PROXY).

Abnormal levels of the first three real activities manipulation measure will be measured as the residual from the relevant models estimated by year. The fourth measure is a combined measure of the three real earnings management measures. This measure is calculated as the residuals of AB_CFO – AB_PROD + AB_DISCX.

According to Cohen et al. (2008), I use three individual proxies as well as a combined proxy. The following model will be used to estimate the normal level of operating cash flows:

𝐶𝐹𝑂_, 𝐴𝑠𝑠𝑒𝑡𝑠_, =𝛼1 1 𝐴𝑠𝑠𝑒𝑡𝑠_, +𝛼2 𝑆𝑎𝑙𝑒𝑠 𝐴𝑠𝑠𝑒𝑡𝑠_, + 𝛼3 ∆𝑆𝑎𝑙𝑒𝑠_, 𝐴𝑠𝑠𝑒𝑡𝑠_, + 𝜖, where:

CFOt = cash flow from operations in year t;

Assets = total assets; ΔSales = Salest – Salest-1;

𝜖, = residual, proxy for combined measure of real activities manipulation.

For every firm-year, abnormal cash flow from operations (AB_CFO) is the residual from the corresponding industry-year model and the firm-year’s sales and lagged assets.

The second measure is abnormal production costs. According to Kim et al. (2012), I define production costs as the sum of COGS and change in inventory during the year, and they express expenses as a linear function of contemporaneous sales. I estimate the following model for normal COGS:

𝐶𝑂𝐺𝑆, 𝐴𝑠𝑠𝑒𝑡𝑠_, = 𝛼0 1 𝐴𝑠𝑠𝑒𝑡𝑠_, + 𝛼1 𝑆𝑎𝑙𝑒𝑠𝑖, 𝑡 𝐴𝑠𝑠𝑒𝑡𝑠_, + 𝜖,

(15)

Similarly, I estimate the model for normal inventory growth using the following equation: 𝛥𝐼𝑁𝑉_, 𝐴𝑠𝑠𝑒𝑡𝑠, = 𝛼0 1 𝐴𝑠𝑠𝑒𝑡𝑠, +𝛼1 ∆𝑆𝑎𝑙𝑒𝑠 𝐴𝑠𝑠𝑒𝑡𝑠, +𝛼1∆𝑆𝑎𝑙𝑒𝑠, 𝐴𝑠𝑠𝑒𝑡𝑠, + 𝜖, where ΔINVt is the change in inventory in year t.

Using the previous two equations, I estimate the normal level of production costs as: 𝑃𝑟𝑜𝑑_, 𝐴𝑠𝑠𝑒𝑡𝑠, = 𝛼0 1 𝐴𝑠𝑠𝑒𝑡𝑠, + 𝛼1 𝑆𝑎𝑙𝑒𝑠 𝐴𝑠𝑠𝑒𝑡𝑠, +𝛼2 ∆𝑆𝑎𝑙𝑒𝑠 𝐴𝑠𝑠𝑒𝑡𝑠, +𝛼3 ∆𝑆𝑎𝑙𝑒𝑠 𝐴𝑠𝑠𝑒𝑡𝑠, + 𝜖, Abnormal production cost (AB_PROD) is the residual from the model.

The third measure is abnormal discretionary expenses. Following Roychowdhury (2006), Kim et al. (2012), Cohen et al. (2008), Badertscher (2011), and Zang (2012), I estimate the normal level of discretionary expenses using the following equation:

,

= 𝛼0

,

+ 𝛼1

∆

,

+ 𝜖

,

In this equation, discretionary expense is modeled as a function of the current sales. This results in significantly lower residuals (𝜖_, ) when firms manage sales upwards. In this case, you have a wrong estimation of the residuals, which creates a problem. By creating a lagged sales function for estimating discretionary expenses, this issue can be addressed. The following model will be used:

𝐷𝐼𝑆𝐸𝑋𝑃_, 𝐴𝑠𝑠𝑒𝑡𝑠, = 𝛼0 1 𝐴𝑠𝑠𝑒𝑡𝑠, + 𝛼1 𝑆𝑎𝑙𝑒𝑠, 𝐴𝑠𝑠𝑒𝑡𝑠, + 𝜖,

where DISEXPi,t is the discretionary expenses in year t, defined as the sum of R&D and

SG&A expenses. For every firm-year, abnormal discretionary expenditure (AB_EXP) is the residual from the model.

3.2 Data sample

Most of the data of European companies are obtained from Compustat. Compustat comprises historical financial data from annual reports of companies around the whole world, for more than ten years. Data about the inflation and institutional factors to define investor protection are obtained from the World Bank. Excluded from my empirical analysis are banks and financial institutions. I restrict the sample to all non-financial firms with available data. Each firm-year observation must have the data necessary to calculate the proxies for accrual-based earnings management and real earnings management I employ in the analysis. As a result, when a firm-year observation doesn’t have all the variables needed to calculate the required

(16)

proxies, the firm will be excluded. I started with a sample of more than 40,000 firm-year observations. The final sample consists of 9807 firm-year observations, across 30 countries and 1313 non-financial firms for the fiscal years 2001 to 2013, 4 years before and 9 years after the mandatory IFRS adoption in Europe.

(17)

4. Tests and results

4.1 Descriptive statistics

Table 1 presents the number of firm-year observations per country, the frequency, as well as other descriptive statistics for the sample firms from European countries. There is a significant variation in the number of firm-year observations across countries due to differences in capital market development, country size, and the availability of complete financial data. The mean firm’s sales in US$ can be interpreted as relative, were Russia has the highest mean firm size in sales and Cyprus has the lowest value. Table 1 also shows average GDP per capita and inflation, where both numbers are collected from the World Bank.

TABLE 1

Descriptive statistics of sample countries (2001-2013)

Country Frequency Percent Mean firm size in

Sales in US$ Average GDP per capita Inflation Austria 212 2,2 1351,76 35752 2,11 Belgium 282 2,9 1503,03 34290 2,17 Bulgaria 11 0,1 108,82 3798 4,91 Croatia 20 0,2 40958,08 9740 2,75 Cyprus 12 0,1 34,18 21740 1,93 Czech Republic 22 0,2 32871,97 12524 2,51 Denmark 324 3,3 8358,40 45058 2 Estonia 17 0,2 495,87 10173 3,99 Finland 482 4,9 2588,28 36059 1,82 France 948 9,7 6030,97 32276 1,67 Germany 1677 17,1 5089,88 32610 1,63 Greece 163 1,7 526,39 20588 2,73 Hungary 21 0,2 167082,76 9823 5,23 Ireland 137 1,4 1487,19 44033 2,36 Iceland 35 0,4 590,40 39725 5,79 Italy 347 3,5 6585,87 28839 2,16 Lithuania 16 0,2 204,74 7595 3,14 Luxembourg 61 0,6 12367,76 77087 2,32 Netherlands 354 3,6 7278,97 37637 2,09 Norway 213 2,2 34785,01 66482 1,92 Poland 92 0,9 6875,97 8017 3,03 Portugal 10 0,1 4889,59 17391 2,37 Romania 7 0,1 1233,10 4704 8,53 Russia 112 1,1 256614,89 6305 11,38 Slovakia 3 0 8985,72 11032 4,21

(18)

Panel A of Table 2 presents for each country in the sample the descriptive statistics on the institutional characteristics from 2000 till 2005. The institutional variables are drawn from La Porta et al. (1997, 1998). Legal Origin is presented in the second column and Legal Tradition assignment is presented in the third column. For Legal Tradition, there are two options; CD and CM. CD indicates a code-law country and CM indicates a common-law country. Outside Investor Rights describes how strongly the legal system favors minority shareholders against managers or dominant shareholders in the corporate decision-making process, including the voting process (La Porta et al., 1998, p.1127). There are five anti-director rights measures: (1) one share means one vote, (2) proxy by mail allowed, (3) shares not blocked before meeting, (4) cumulative voting/proportional representation, and (5) oppressed minority. A country gets on each measure a score of 1 if it protects minority shareholders according to this measure, and a score of 0 otherwise. An average score across the level of corruption, an index of the rule of law and an index of the legal system’s efficiency is the Legal Enforcement measure. This measure is calculated for each country separately. All three variables range from zero to ten. A strong system of Legal Enforcement could substitute for weak rules since active and well-functioning courts can step in and rescue investors abused by the management (La Porta et al., 1998, p. 1127). The Importance of Equity Markets is measured by an average rank across three variables. A higher score on the average rank indicates a greater importance of the stock market. The first variable is the ratio of the aggregate stock market held by minorities to gross national product. Second is the number of listed domestic stocks relative to the population. And third is the number of IPOs relative to the population. In the ten largest privately held non-financial firms, the median percentage of common shares owned by the largest three shareholders is called Ownership concentration (La Porta et al., 2008). According to La Porta et al. (2008), concentration of

Slovenia 22 0,2 18217,49 17525 3,14 Spain 137 1,4 3835,36 24434 2,7 Sweden 926 9,4 16230,14 38886 1,39 Switzerland 784 8 5240,01 57384 0,62 United Kingdom 2360 24,1 4709,65 35226 2,29 Total 9807 100 657132,27 Mean 326,9 3,33 21904,40 27557,8 3,16 Min 3 0 34,18 3798 0,62 Max 2360 24,1 256614,89 77087 11,38

(19)

extent to which investors are protected through disclosure of ownership and financial information. The index ranges from zero to ten, with higher values indicating more disclosure (World Bank). The countries with EU accession in 2004 or 2007 are excluded. These are Bulgaria, Croatia, Cyprus, Czech Republic, Estonia, Hungary, Lithuania, Poland, Romania, Slovakia and Slovenia. Also excluded are Iceland, Luxembourg and Russian Federation, because of the missing number on Important of Equity Market and Ownership Concentration. For the tests of the hypotheses, these countries will be excluded as well.

Panel B of Table 2 contains the descriptive statistics on the institutional characteristics from 2006 till 2013, the post mandatory adoption period, drawn from The World Bank Database2_{. The strength of legal protection of minority investors against misuse of corporate} assets by company directors for their personal gain is measured by the Protecting Investors indicator. The indicators distinguishes three dimensions of investor protection: (1) rules on the approval and disclosure of related-party transactions (extent of disclosure index), (2) liability of company executives for self-dealing (extent of director liability index) and (3) shareholders’ competence to access corporate information before and during litigation (ease of shareholder suits index). The standard case study assumes a related-party transaction between Company A (buyer) and Company B (seller) where one individual is the controlling shareholder of both the buyer and the seller and a member of both their boards of directors.

The ranking in the strength of investor protection index is the average of the percentile rankings in the extent of disclosure; extent of director liability and ease of shareholder suits indices. According to the World Bank, a higher ranking indicates that an economy’s regulations offer stronger investor protections. The indicator does not measure all aspects related to the protection of minority investors. This measure illustrates the distance of an economy to the country with the highest level of investor protection, which represents the best performance observed across all economies and years included since 2005. An economy’s distance to country with the highest level of investor protection is indicated on a scale from 0 to 100, where 0 represents the lowest performance and 100 the highest level of investor protection. The highest level of investor protection, 100, isn’t included in the table because it doesn’t contain a European sample country. Table 2, Panel C indicates the investor protection for each year, where the last column presents the average of years from 2006 till 2013.

(20)

Figure 3. Investor protection 3

TABLE 2 Panel A

Descriptive statistics of sample firms and countries Investor protection (2000-2005)

Country Legal _Origin Legal _Tradition

Outside investor rights Legal Enforcement Important of Equity Market Ownership concentration Disclosure index Austria German CD 2 9,4 7 0,51 5 Belgium French CD 0 9,4 11,3 0,62 8 Denmark Scandinavian CD 2 10 20 0,4 7 Finland Scandinavian CD 3 10 13,7 0,34 6 France French CD 3 8,7 9,3 0,24 10 Germany German CD 1 9,1 5 0,5 5 Greece French CD 2 6,8 11,5 0,68 7 Ireland English CM 4 7,5 17,3 0,36 10 Italy French CD 1 7,1 6,5 0,6 7 Netherlands French CD 2 10 19,3 0,31 3 Norway Scandinavian CD 4 10 20,3 0,31 7 Portugal French CD 3 7,2 11,8 0,59 6 Spain French CD 4 7,1 7,2 0,5 5 Sweden Scandinavian CD 3 10 16,7 0,28 2 Switzerland German CD 2 10 24,8 0,48 0 United Kingdom English CM 5 9,2 25 0,15 10 Mean 2,56 8,84 14,17 0,43 6,13

(21)

TABLE 2 Panel B

Descriptive statistics of sample firms and countries Institutional variables over the years (2006-2013)

Country Extent of disclosure index (0-10) Extent of director liability index (0-10) Ease of shareholder suits index (0-10) Strength of investor protection index (0-10) Austria 5 5 5 5.0 Belgium 8 6 7 7.0 Bulgaria 10 2 6 6.0 Croatia 1 5 4 3.3 Cyprus 8 4 7 6.3 Czech Republic 2 5 8 5.0 Denmark 7 5 7 6.3 Estonia 8 3 6 5.7 Finland 6 4 7 5.7 France 10 1 5 5.3 Germany 5 5 5 5.0 Greece 7 4 5 5.3 Hungary 2 4 7 4.3 Iceland 7 5 6 6.0 Ireland 10 6 9 8.3 Italy 7 4 7 6.0 Lithuania 7 4 6 5.7 Luxembourg 6 4 3 4.3 Netherlands 4 4 6 4.7 Norway 7 6 7 6.7 Poland 7 2 9 6.0 Portugal 6 5 7 6.0 Romania 9 5 4 6.0 Russian Federation 6 2 6 4.7 Slovak Republic 3 4 7 4.7 Slovenia 5 9 8 7.3 Spain 5 6 4 5.0 Sweden 8 4 7 6.3 Switzerland 0 5 4 3.0 United Kingdom 10 7 7 8.0

(22)

TABLE 2 Panel C

Descriptive statistics of sample firms and countries Investor protection per year (2006-2013)

Country 2006 2007 2008 2009 2010 2011 2012 2013 Mean Austria 50 50 50 50 50 50 50 50 50 Belgium 70 70 70 70 70 70 70 70 70 Bulgaria 60 60 60 60 60 60 60 60 60 Croatia 33,3 33,3 33,3 33,3 33,3 33,3 33,3 33,3 33,3 Cyprus 50 50 50 50 50 50 63,3 63,3 53,3 Czech Republic 50 50 50 50 50 50 50 50 50 Denmark 63,3 63,3 63,3 63,3 63,3 63,3 63,3 63,3 63,3 Estonia 56,7 56,7 56,7 56,7 56,7 56,7 56,7 56,7 56,7 Finland 56,7 56,7 56,7 56,7 56,7 56,7 56,7 56,7 56,7 France 53,3 53,3 53,3 53,3 53,3 53,3 53,3 53,3 53,3 Germany 50 50 50 50 50 50 50 50 50 Greece 30 30 30 33,3 33,3 33,3 33,3 46,7 33,7 Hungary 43,3 43,3 43,3 43,3 43,3 43,3 43,3 43,3 43,3 Iceland 50 50 53,3 53,3 53,3 53,3 60 60 54,2 Ireland 83,3 83,3 83,3 83,3 83,3 83,3 83,3 83,3 83,3 Italy 60 60 60 60 60 60 60 60 60 Lithuania 50 50 50 50 50 50 56,7 56,7 51,7 Luxembourg 43,3 43,3 43,3 43,3 43,3 43,3 43,3 43,3 43,3 Netherlands 43,3 43,3 43,3 43,3 43,3 43,3 43,3 46,7 43,7 Norway 66,7 66,7 66,7 66,7 66,7 66,7 66,7 66,7 66,7 Poland 56,7 60 60 60 60 60 60 60 59,6 Portugal 60 60 60 60 60 60 60 60 60 Romania 56,7 60 60 60 60 60 60 60 59,6 Russia 46,7 46,7 46,7 46,7 46,7 46,7 46,7 46,7 46,7 Slovakia 46,7 46,7 46,7 46,7 46,7 46,7 46,7 46,7 46,7 Slovenia 63,3 63,3 63,3 66,7 66,7 66,7 66,7 73,3 66,3 Spain 50 50 50 50 50 50 50 50 50 Sweden 43,3 56,7 56,7 56,7 56,7 63,3 63,3 63,3 57,5 Switzerland 30 30 30 30 30 30 30 30 30 United Kingdom 80 80 80 80 80 80 80 80 80 Mean 53,69 54,02 54,13 54,22 54,22 54,44 55,33 56,11 54,5 Min 30 30 30 30 30 30 30 30 30 Max 83,3 83,3 83,3 83,3 83,3 83,3 83,3 83,3 83,3

(23)

Table 3 provides summary statistics of the full sample. Sample firms have a 27 percent annual growth in sales. The operating cycle is, on average, approximately 445 days, suggesting that accruals are not likely to reverse in the subsequent year.

Total Accruals (total accruals deflated by prior-year total assets) is positive at 0,054

with a standard deviation of 0,134. In contrast, the average DA (discretionary accruals) is -0.00 with a standard deviation of 0,130. While the average DA is zero, I find that positive discretionary accruals (Positive_DA) is 0,00 with a standard deviation of 0,11.

The main variable of interest is the absolute value of discretionary accruals (ABS_DA). The average for ABS_DA is 0 with a standard deviation of 0,114.

Graham et al. (2005) mention in their survey that accrual based earnings management is less costly than real based earnings management. The last rows of Table 3 represent the real earnings management proxies. When comparing the percentiles of accrual based earnings management (DA) with the percentiles of the real based earnings management proxies (R_CFO, R_PROD, R_DISCX and RM_PROXY), accrual-based earnings management takes larges values.

(24)

TABLE 3 Descriptive Statistics

N Minimum Maximum Mean Median

Std.

Deviation Percentiles

Statistic Statistic Statistic Statistic Statistic Statistic 25th 75th

Total Assets 9807 0,069 12068139 15286,715 406,809 242395,907 83,697 2308,500 Sales 9807 0,001 4665631 9988,346 378,200 100757,097 82,642 2255,847 Growth of Sales 8502 -1 916 0,2785 0,054 10,516 -0,048 0,168 Total Accruals 8502 -1,44 4,36 0,054 0,046 0,134 0,010 0,088 Current Ratio 9807 0 30,76 0,6695 0,601 0,61795 0,424 0,799 Operating Cycle 9806 -1539,97 1122375 445,482 184,104 13126,986 138,034 252,392 DA 8502 -4,140 1,585 -0,000 0,006 0,130 -0,034 0,042 Positive_DA 1259 -0,393 1,292 0,000 -0,029 0,110 -0,049 0,010 ABS_DA 8502 -1,331 4,059 0,000 -0,022 0,114 -0,050 0,018 R_CFO 8502 -20,591 4,740 0,000 0,023 0,302 -0,035 0,074 R_PROD 8502 -2,407 14,239 0,000 0,021 0,273 -0,115 0,123 R_DISCX 8502 -2,780 6,629 0,000 -0,053 0,269 -0,149 0,092 RM_PROXY 8502 -6,045 34,577 -0,000 -0,018 0,541 -0,186 0,146 Variable Definitions:

Total Assets = annual Compustat data item 6; Sales = annual Compustat data item 12;

Growth of Sales = the change in sales divided by lagged sales;

Total Accruals = the difference between operating cash flow adjusted for income before extraordinary

items divided by lagged total assets;

Current Ratio = current assets divided by current liabilities; Operating Cycle =

( ) + ( );

DA = discretionary accruals using computed using the Modified Jones Model;

Positive_DA = the value of positive discretionary accruals computed using the Modified Jones Model; Negative_DA = the value of negative discretionary accruals computed using the Modified Jones Model;

ABS_DA = the absolute value of discretionary accruals computed using the Modified Jones Model; R_CFO = the level of abnormal cash flows from operations;

R_PROD = the level of abnormal production costs, where production costs are defined as the sum of

cost of goods sold and the change in inventories;

R_DISX = the level of abnormal discretionary expenses, where discretionary expenses are the sum of

R&D expenses and SG&A expenses;

RM_PROXY = The sum of standardized three real earnings management proxies, i.e., R_CFO, R_PROD

(25)

4.2 Test of hypothesis 1

Hypothesis 1. Mandatory IFRS adoption leads to more real earnings management and

less accrual based earnings management.

Figures 4-7 provide graphical illustrations of the results from the appendix. Figure 4 indicates that the period before the mandatory IFRS adoption was, indeed, associated with a high level of accrual-based earnings management. After the IFRS adoption in 2005, there was a big decline in earnings management. In 2006, the level of accrual-based earnings management increased and during 2008 it even exceeded the level before the mandatory IFRS adoption. Figure 5 plots the trend in positive discretionary accruals. Positive discretionary accruals peaked before the mandatory IFRS adoption and the figure shows a decline after the mandatory IFRS adoption.

Among the real earnings management variable do not show an increasing trend over the pre mandatory IFRS adoption period, abnormal production costs increased in the post-mandatory IFRS adoption period, shown in Figure 6. Abnormal cash flows also increased, after one year of decline, after the mandatory IFRS adoption period. On the other hand, abnormal discretionary expenses were, after one year of increase, lower in the post-mandatory adoption period. Figure 7, which is the combined proxy for real earnings management shows a decreasing trend over time but shows a big increase after the mandatory IFRS adoption.

(26)

Absolute value of discretionary accruals over the 2001-2013 period, where the purple lines present the mandatory IFRS adoption period.

Positive value of discretionary accruals over the 2001-2013 period, where the purple lines present the mandatory IFRS adoption period.

0 0,02 0,04 0,06 0,08 0,1 0,12 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 ABS _D A

Absolute Value of Discretionary Accruals over Time, 2001-2013 FIGURE 4 -0,15 -0,1 -0,05 0 0,05 0,1 0,15 0,2 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 FIGURE 5

(27)

Abnormal cash flow from operations (R_CFO), abnormal production costs (R_PROD) and abnormal discretionary expenses (R_DISCX) over the 2001-2013 period, where the purple lines present the mandatory IFRS adoption period.

This figure plots the total real earnings management computed using the Modified Jones Model over the 2001-2013 sample period, where the purple lines present the mandatory IFRS adoption period.

-0,3 -0,2 -0,1 0 0,1 0,2 0,3 0,4 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 FIGURE 6

Abnormal cash flow from operations over Time, 2001-2013.

Reeks2 Reeks3 Reeks1 R_CFO R_PROD R_DISCX 0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4 0,45 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 FIGURE 7

(28)

Table 4 shows the correlation between the accrual-based earnings management proxy and the real earnings management proxies. I find a significant positive relation (,079) between accrual-based earnings management (ABS_DA) and real-based earnings management (RM_PROXY). I find a negative, not significant, relation (-,013) between discretionary accruals (ABS_DA) and production costs (R_PROD). There is a negative significant relation (-,379) between abnormal cash flows (R_CFO) and production costs (R_PROD) and discretionary expenses (R_DISCX), which indicates a substitution effect: while abnormal cash flows decreased, overall firms increased the use of production costs and discretionary expenses.

In summary, I found a positive significant correlation between accrual-based earnings management and abnormal cash flows, and a positive significant correlation between accrual based earnings management and discretionary expenses. This indicates that if accruals-based earnings management increases, abnormal cash flows and discretionary expenses increases as well. As a result, I have to reject my first hypothesis.

TABLE 4

Correlations (2001-2013)

R_CFO R_PROD R_DISCX ABS_DA RM_PROXY

R_CFO Correlation Coefficient 1,000 -,379** _-,057** _,101** _-,580** Sig. (2-tailed) . ,000 ,000 ,000 ,000 N 8502 8501 8502 8502 8501 R_PROD Correlation Coefficient -,379** _1,000 _-,694** _-,013 _,850** Sig. (2-tailed) ,000 . ,000 ,226 ,000 N 8501 8501 8501 8501 8501 R_DISCX Correlation Coefficient -,057** _-,694** _1,000 _,104** _-,329** Sig. (2-tailed) ,000 ,000 . ,000 ,000 N 8502 8501 8502 8502 8501 ABS_DA Correlation Coefficient ,101** _-,013 _,104** _1,000 _,079** Sig. (2-tailed) ,000 ,226 ,000 . ,000 N 8502 8501 8502 8502 8501 RM_PROXY Correlation Coefficient -,580** _,850** _-,329** _,079** _1,000 Sig. (2-tailed) ,000 ,000 ,000 ,000 . N 8501 8501 8501 8501 8501

(29)

4.3 Test of hypothesis 2

Table 5 provides descriptive statistics for the four individual earnings management measures as well as the aggregate earnings management score. The countries are sorted in descending order based on their aggregate score. EM1 is the accrual-based earnings management measure. EM2, EM3 and EM4 are the real-based earnings management measures where EM2 is the abnormal cash flow, EM3 is the production costs and EM4 is the discretionary expenses. The last columns of table 5 present the aggregate earnings management scores. This is the average rank of each country across the four earnings management measures (EM1-EM4), where the country with the highest amount of earnings management gets the highest rank. The aggregate earnings management measures show high ranks for countries as Czech Republic, Russia, Croatia and Denmark, and low ranks for countries such as Ireland, Switzerland and United Kingdom in both panels. The higher the scores on the aggregate earnings management measure, the more earnings management.

(30)

TABLE 5 Panel A

Country scores for earnings management measures for the period from 2001 till 2005

Kolom1 EM1 Rank EM2 Rank EM3 Rank EM4 Rank

Aggregate EM Czech Republic _{0,37410125341} _{20 0,00000007709633396030} _{19 0,000000032259802390863} _{19 0,000000014394715205} ₂₀ _19,5 Russia _{0,00000333257} _{16 0,00000000020403482717} _{15 0,000000000039582833839} _{14 0,000000000432779122} ₁₇ _15,5 Croatia _{0,00000000016} _{4 0,00000000155902479262} _{17 0,000000056160936127537} _{20 0,000000000920382048} ₁₈ _14,75 Luxembourg _{0,00000000037} _{6 0,00000000028042178415} _{16 0,000000002256247434992} _{18 0,000000002447078275} ₁₉ _14,75 Italy _{0,00004627485} _{19 0,00000000000000146835} _{1 0,000000001255815413996} _{16 0,000000000018667005} ₁₅ _12,75 Denmark _{0,00000000340} _{12 0,00000000001150205520} _{14 0,000000000000064624758} _{6 0,000000000091047054} ₁₆ ₁₂ Germany _{0,00000001000} _{14 0,00000000000135146670} _{9 0,000000000037883643390} _{13 0,000000000003655690} ₁₀ _11,5 Austria _{0,00001393554} _{18 0,00000000000031107391} _{5 0,000000000011178229945} _{12 0,000000000002456978} ₉ ₁₁ Netherlands _{0,00000000004} _{2 0,00000000000174982100} _{11 0,000000001409323978057} _{17 0,000000000015656116} ₁₃ _10,75 Norway _{0,00000479485} _{17 0,00000000000363620278} _{13 0,000000000000206969494} _{5 0,000000000001613550} ₈ _10,75 Estonia _{0,00000000257} _{11 0,00000000832270088281} _{18 0,000000000000794868535} _{2 0,000000000003755453} ₁₁ _10,5 France _{0,00000000035} _{5 0,00000000000339590815} _{12 0,000000000009522840528} _{11 0,000000000016371128} ₁₄ _10,5 Sweden _{0,00000000462} _{13 0,00000000000138911682} _{10 0,000000000000530868160} _{3 0,000000000008193122} ₁₂ _9,5 Greece _{0,00000033633} _{15 0,00000000000115464481} _{8 0,000000000001933431849} _{7 0,000000000000097653} ₄ _8,5 Finland _{0,00000000013} _{3 0,01151239596135220000} _{20 0,000000000000257961394} _{4 0,000000000000123612} ₅ ₈ United Kingdom _{0,00000000171} _{10 0,00000000000000415380} _{2 0,000000000081247022046} _{15 0,000000000000002174} ₁ ₇ Switzerland _{0,00000000078} _{7 0,00000000000003818663} _{3 0,000000000003847878483} _{9 0,000000000000323986} ₇ _6,5 Iceland _{0,00003579178} _{9 0,00000072450699997728} _{6 0,000001883719999873360} _{8 0,000000139924000053} ₂ _6,25 Ireland _{0,10205361013}- _{1 0,05440309769500000000} _{7 0,686223360425000000000} _{10 0,265358225272000000} ₃ _5,25 Belgium _{0,00000000088} _{8 0,00000000000016171738} _{4 0,000000000000008326380} _{1 0,000000000000271217} ₆ _4,75

(31)

TABLE 5 Panel B

Country scores for earnings management measures for the period from 2006 till 2013

Country EM1 Rank EM2 Rank EM3 Rank EM4 Rank

Aggregate EM Russia _{0,00001113356773} ₁₈ _{0,00003450716328617} ₂₀ _{0,00012786840970} _{20 0,0001035949616880} ₂₀ _19,5 Czech Republic _{0,00005578960173} ₁₉ _{0,00000000000640545} ₁₇ _{0,00000000049484} _{18 0,0000000000875603} ₁₇ _17,75 Croatia _{0,00000004132774} ₁₆ _{0,00000000001289839} ₁₈ _{0,00000000048014} _{17 0,0000000046104463} ₁₉ _17,5 Denmark _{0,00000003731578} ₁₅ _{0,00000000003956594} ₁₉ _{0,00000000000751} _{14 0,0000000001415299} ₁₈ _16,5 Austria _{0,00068313277529} ₂₀ _{0,00000000000001486} ₃ _{0,00000000000611} _{13 0,0000000000053576} ₁₃ _12,25 Iceland _{0,00000000097767} ₈ _{0,00000000000499683} ₁₄ _{0,00000000000055} _{8 0,0000000000084595} ₁₅ _11,25 Sweden _{0,00000000083343} ₇ _{0,00000000000563340} ₁₆ _{0,00000000000010} _{5 0,0000000000133908} ₁₆ ₁₁ Norway _{0,00000001070586} ₁₃ _{0,00000000000496305} ₁₃ _{0,00000000000008} _{3 0,0000000000055671} ₁₄ _10,75 Netherlands _{0,00000000656136} ₁₂ _{0,00000000000022901} ₆ _{0,00000000000128} _{11 0,0000000000052183} ₁₂ _10,25 Italy _{0,00000001679029} ₁₄ _{0,00000000000037508} ₈ _{0,00000000000155} _{12 0,0000000000000966} ₃ _9,25 Belgium _{0,00000000324515} ₁₁ _{0,00000000000151305} ₁₁ _{0,00000000000029} _{6 0,0000000000005081} ₈ ₉ Finland _{0,00000000039381} ₅ _{0,00000000000071605} ₁₀ _{0,00000004217857} _{19 0,0000000000000734} ₂ ₉ Germany _{0,00000000121143} ₉ _{0,00000000000543942} ₁₅ _{0,00000000000010} _{4 0,0000000000003840} ₇ _8,75 Estonia _{0,00000000000075} ₂ _{0,00000000000221875} ₁₂ _{0,00000000000092} _{9 0,0000000000045921} ₁₁ _8,5 United Kingdom _{0,00000000021661} ₄ _{0,00000000000023552} ₇ _{0,00000000004043} _{16 0,0000000000003789} ₆ _8,25 France _{0,00000000003693} ₃ _{0,00000000000039163} ₉ _{0,00000000002736} _{15 0,0000000000002167} ₅ ₈ Luxembourg _{0,00000005422979} ₁₇ _{0,00000000000015558} ₅ _{0,00000000000008} _{2 0,0000000000000013} ₁ _6,25 Switzerland _{0,00000000250097} ₁₀ _{0,00000000000013895} ₄ _{0,00000000000001} _{1 0,0000000000006446} ₉ ₆ Greece _{0,00000000000026} ₁ _{0,00000000000000182} ₂ _{0,00000000000092} _{10 0,0000000000015894} ₁₀ _5,75 Ireland _{0,00000000066413} ₆ _{0,00000000000000021} ₁ _{0,00000000000046} _{7 0,0000000000001785} ₄ _4,5

(32)

Table 6, Panel A. Correlations (2001-2005) Outside investor rights Legal enforce ment Important equity market Ownership concentration Disclosure index Aggregate EM Correlation Coefficient -,302 -,005 -,376 ,088 -,243 Sig. (2-tailed) ,293 ,988 ,185 ,765 ,403 NA 14 14 14 14 14

Outside investor rights

Correlation Coefficient ,145 ,595* _-,783** _,321 Sig. (2-tailed) ,621 ,025 ,001 ,263 N 14 14 14 14 Legal enforcement Correlation Coefficient ,531 -,376 -,562* Sig. (2-tailed) ,051 ,185 ,036 N 14 14 14

Important equity market

Correlation Coefficient -,513 ,004 Sig. (2-tailed) ,061 ,988 N 14 14 Ownership concentration Correlation Coefficient -,126 Sig. (2-tailed) ,667 N 14 Disclosure index Correlation Coefficient Sig. (2-tailed) N

*. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed).

(33)

Table 6, Panel B. Correlations (2006-2013) Director Liability Ease of Shareholder Suits Investor Protection Aggregate Earnings Management Disclosure Correlation Coefficient ,043 ,445* _,821** _-,127 Sig. (2-tailed) ,858 ,049 ,000 ,593 Director Liability Correlation Coefficient ,350 ,440 ,048 Sig. (2-tailed) ,130 ,052 ,839 Ease of Shareholder Suits Correlation Coefficient ,766** _-,014 Sig. (2-tailed) ,000 ,952 Investor protection Correlation Coefficient -,045 Sig. (2-tailed) ,851

*. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed).

(34)

Based on the above results, I can test the second hypothesis:

Hypothesis 2. High investor protection has a negative effect on both real earnings

management and accrual-based earnings management.

Table 6, Panel A and Panel B, present correlations among the aggregate earnings management score from Table 5 (aggregate EM) and institutional variables. Panel A presents the correlations for the years before the mandatory IFRS adoption, 2001-2005. Panel B presents the correlations for post mandatory IFRS adoption period, 2006-2013.

Before the mandatory IFRS adoption (2001-2005), ‘outside investor rights’ has an income effect (,595) on ‘important equity market’. When there is an increase in ‘outside investor rights’, ‘important equity market’ will increase as well. There is a substitute effect (-,783) on ‘outside investor rights’ and ‘ownership concentration’. This indicates that when an increase in ‘outside investor rights’ arise, the ‘ownership concentration’ decreases. There is also a substitute effect (-,562) between ‘legal enforcement’ and ‘disclosure index’.

After the mandatory IFRS adoption (2006-2013), there is an income effect (,445) between ‘disclosure’ and ‘ease of shareholder suits’ and ‘investor protection’. An income effect (,821) arise also between ‘ease of shareholder suits’ and ‘investor protection’.

Inconsistent with the hypothesis, there is no correlation among the institutional variables and the aggregate earnings management scores for both periods. As a result, investor protection has no significant effect on earnings management in Europe, so I have to reject the second hypothesis.

(35)

5. Additional Analysis

TABLE 7, Panel A. Correlations (2001-2005) Outside Investor Rights Legal Enforcement Important Equity Market Ownership Concentration Disclosure Index Legal Origin Correlation Coefficient ,651** _,171 _,570* _-,545* _,516* Sig. (2-tailed) ,006 ,527 ,021 ,029 ,041 N 16 16 16 16 16 Outside Investor Rights Correlation Coefficient 1,000 -,019 ,470 -,625** _,241 Sig. (2-tailed) . ,943 ,066 ,010 ,368 N 16 16 16 16 16 Legal Enforcement Correlation Coefficient -,019 1,000 ,572* _-,513* _-,368 Sig. (2-tailed) ,943 . ,021 ,042 ,161 N 16 16 16 16 16 Important Equity Market Correlation Coefficient ,470 ,572* _1,000 _-,561* _,055 Sig. (2-tailed) ,066 ,021 . ,024 ,839 N 16 16 16 16 16 Ownership Concentration Correlation Coefficient -,625** _-,513* _-,561* _1,000 _-,086 Sig. (2-tailed) ,010 ,042 ,024 . ,752 N 16 16 16 16 16 Disclosure Index Correlation Coefficient ,241 -,368 ,055 -,086 1,000 Sig. (2-tailed) ,368 ,161 ,839 ,752 . N 16 16 16 16 16

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

(36)

TABLE 7, Panel B. Correlations (2006-2013)

Legal Origin Investor Protection Legal Origin Correlation Coefficient 1,000 ,735** Sig. (2-tailed) . ,001 N 16 16 Investor Protection Correlation Coefficient ,735** _1,000 Sig. (2-tailed) ,001 . N 16 16

**. Correlation is significant at the 0.01 level (2-tailed).

In the sample, there are four legal origins: German, French, Scandinavian and English. To identify the correlation between legal origin and investor protection, I made dummies for the legal origin were German is 1, French is 2, Scandinavian is 3 and English is 4. The correlations are shown in Table 7, Panel A and B. ‘Legal origin’ has a positive significant effect on ‘outside investor rights’ (,651), ‘important equity market’ (,570) and ‘disclosure index’ (,516) during the period 2001-2005. ‘Legal Origin’ has a negative significant effect (-,545) on ‘ownership concentration’ during the period 2001-2005. During the period 2006-2013, ‘legal origin’ has a positive significant effect (,735) on investor protection. This result indicates that when legal origin is higher, the investor protection is higher as well. As a result, I can say that countries with an English legal origin have more investor protection than countries with a German legal origin. In Table 2, Panel C there are the legal origins for each country. As a result, Ireland and the United Kingdom have a higher investor protection than Austria, Germany and Switzerland. But as mentioned before, this relation has no impact on the aggregate earnings management measure.

(37)

6. Conclusion

This paper documents differences in the level of earnings management across the period before and after the mandatory IFRS adoption in 2005, and the impact of investor protection on this relation.

I first hypothesize that mandatory IFRS-adoption leads to less earnings management. The figures of chapter 4 represent changes in accrual-based earnings management and real-based earnings management over a time period from 2001 till 2013, to show the changes around the mandatory IFRS adoption of 2005. The results indicate there is no correlation between mandatory IFRS adoption and earnings management. For both accrual-based earnings management as real-based earnings management, there is a sufficient decrease after the mandatory IFRS adoption in 2005, but after 2006 this decrease sits in for a big increase for both types of earnings management.

I second hypothesize that investor protection leads to less earnings management, both real-based and accrual-based. With higher investor protection, earnings management would decrease because of the fewer incentives available to disguise the performance of the firm. This is the result of the strong investor protection that limits managers’ ability to acquire private control benefits (Leuz et al., 2003). Inconsistent with the hypothesis, there’s no correlation between investor protection and earnings management.

Overall, the findings are inconsistent with the hypotheses, which indicates that investor protection doesn’t lead to less earnings management. The findings should be interpreted cautiously as earnings management is difficult to measure.

(38)

7. Limitations

First, most of European countries legally adopt IFRS before 2005. In this situation, countries voluntary adopted IFRS. During this research, I didn’t distract these companies. Based on the data of the World Bank, most of these voluntary adoption countries adopt IFRS around 2002, after the Sarbanes-Oxley Act. For this reason you should see a difference in earnings management around 2002 until 2013.

Second, normally discretionary expenses are defined as the sum of Research and Development expenses (R&D), advertising expenses and Selling, General and Administrative expenses (SG&A). In this paper advertising expenses are excluded in this calculation, because these expenses are included in Compustat. For this reason I calculated discretionary expenses as the sum of R&D expenses and SGA expenses. This could have a small influence on the results.

And last, the findings are subject to several caveats. Earnings management is difficult to measure because it manifest itself in different forms. In this case, we have real-based earnings management and accrual-based earnings management. These two forms of earnings management have together four forms. To address this issue, I computed four proxies for earnings management and calculated the aggregate earnings management score by ranks, where every form of earnings management has an equal weight. Because there are not strict numbers available for earnings management, this way of measuring could also influence the results.

(39)

References

Badertscher, B. A. (2011). Overvaluation and the Choice of Alternative Earnings Management Mechanisms. Accounting Review. Vol. 86 Issue 5, p1491-1518.

Ball, R. (2006). International Financial Accounting Standards (IFRS): Pros and cons for investors. Accounting and Business Research (Special Issue): 5–27.

Barth, M. E., W. R. Landsman, M. Lang and C. Williams (2007), “Accounting Quality: International Accounting Standards and US GAAP,” Working Paper, University of North Carolina, pp. 1-56.

Bartov, E. (1993). The Timing of Asset Sales and Earnings Manipulation. The Accounting

Review, Vol. 68, pp. 840-855.

Byard, D., Li, Y., Yu, Y. (2011). The Effect of Mandatory IFRS Adoption on Financial Analysts’ Information Environment. Journal of Accounting Research. Vol. 49.

Capkun, V., Collins, D.W., Jeanjean, T. (2012). Does Adoption of IAS/IFRS Deter Earnings Management?

Choi, F. D. S. & Meek, G. (2005). International accounting, 5th ed. Upper Saddle River, NJ: Prentice Hall.

Claessens, S., Djankov, S., Fan, J., Lang, L., (2002). Disentangling the incentive and entrenchment effects of large shareholdings. Journal of Finance 57, pp. 2741–2772. Cohen, D.A., Dey, A. and Lys, T.Z. (2008). Real and Accrual-based Earnings Management in

the Pre- and Post-Sarbanes Oxley Periods. The Accounting Review. Vol. 83, pp. 757-787.

Cohen, D.A. & Zarowin, P. (2010). Accrual-Based and Real Earnings Management Activities around Seasoned Equity Offerings. Journal of Accounting and Economics. Vol. 50, pp. 1-19.

Dechow, P.M. & Skinner, D.J. (2000). Earnings Management: Reconciling the Views of Accounting Academics, Practitioners, and Regulators. Accounting Horizons, Vol. 14, pp. 235-250.

DeFond, M.L. & Subramanyam, K.R. (1998). Auditor changes and discretionary accruals.

Journal of Accounting and Economics. Vol. 25, pp. 35-67.

Doukakis, L.C., (2013). The Effect of Mandatory IFRS Adoption on Real and Accrual-Based Earnings Management Activities. Journal of Accounting and Public Policy.

The effect of mandatory IFRS adoption on earnings management in Europe,moderated by investor protection.