Real earnings management & mergers and acquisitions : real earnings management by acquiring firms prior to stock-for-stock mergers

(1)

Real Earnings Management

&

Mergers and Acquisitions

Real Earnings Management by Acquiring Firms prior to stock-for-stock mergers

Name: Lisa Beekhuis

Student number: 10335854

Date: 20-06-2017

Word Count: 10761

Supervisor: Mario Schabus MSc

MSc Accountancy & Control, specialization Accountancy Amsterdam Business School

(2)

2

Statement of Originality

This document is written by student Lisa Beekhuis who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Abstract

This study investigates how stock-for-stock mergers and acquisitions influence the use of income increasing real earnings management in the year prior to such a merger or acquisition. This examination is done by performing regressions of mergers and acquisitions on income increasing real earnings management. A sample of 385 M&A deals were selected for the period of 2000 up to 2012, of which 66 M&A deals are stock-for-stock mergers. Adding control firm-years, my entire dataset contains 5189 observations. Examining every type of M&A first, this study finds no evidence that acquiring firms engage in income increasing real earnings management in the year prior to a M&A. However, consistent with prior literature, this study finds that acquiring firms engage in income increasing real earnings management in the year prior to a stock-for-stock merger. This study contributes to existing literature by providing evidence on real earnings management that may be valuable for directors, shareholders and standard setters.

Key words: earnings management; real earnings management; stock-for-stock mergers; mergers; acquisitions; M&A

(4)

4

1. Introduction

Since the introduction of the Sarbanes-Oxley Act in 2002, the use of accrual-based earnings management has decreased (Aono and Guan, 2007; Zhou 2007). An important part of the Sarbanes-Oxley Act is Section 302, which obliges managers to be transparent in their reporting and provide a fair view of the economic situation, with the main purpose to recover investor confidence in financial reporting (SEC, 2002). Conversely, the use of real earnings management is significantly increased since the introduction of the Sarbanes-Oxley Act (Cohen et al., 2008). The recent shift from accrual-based earnings management to real earnings management provides a gap in the literature, and therefore it is important to do research on the use of real earnings management.

Previous research is extensively focused on accrual based earnings management. Louis (2004) examines the market efficiency and provides one explanation for the often seen underperformance after mergers and acquisitions. He tried to explain the before mentioned by measuring the impact of pre-merger abnormal accruals (accrual-based earnings management). Since accrual-based earnings management is only a matter of timing, the underperformance Louis (2004) finds, can be explained by the fact that those reversals are not fully anticipated by financial analysts. From the research of Louis (2004), and the evidence of more real-earnings management post-SOX, the following research question is derived: How do stock-for-stock mergers and acquisitions influence the use of Real Earnings Management? Due to existing literature I expect that acquiring firms engage in income increasing real earnings management in the year prior to a stock-for-stock merger or acquisition.

Real earnings management is important to study, because according to Cho and Chun (2016) real earnings management compared to accrual based earnings management can be more harmful for the firm. The reason for this is that real earnings management can adversely affect firm value by distorting normal operating activities. Consequently, relationship with stakeholders - such as customers, employees and communities – can be harmed. By manipulating normal operating, investing, and financing activities, real earnings management distorts optimal business decisions and therefore can be firm destroying (Cho and Chun, 2016). The impact real earnings management can have on firms makes the findings of this research are interesting to directors, shareholders, and standard setters.

(6)

6 This research contributes to the existing literature on real earnings management, because to the best of my knowledge there is no study that investigates the use of real earnings management in the year prior to stock-for-stock mergers and acquisitions. Cohen and Zarowin (2010) mentioned in their research: “Combined with the empirical evidence documented in Roychowdhury (2006) and Zang (2006, wp), our results suggest that future research on earnings management should focus on real activities manipulation as well as accrual-based manipulation.“

The sample of this empirical research consists of 385 deals of acquiring companies that are listed on the S&P500. Of these 385 deals, 66 deals are stock-for-stock mergers or acquisitions and 322 are cash purchases. In all of the firm years, the acquiring firms are audited a Big-4 company i.e. 100%, which results in the variable BIG4 being omitted because of lack of variation. Furthermore, in 66.7% of the firm-year observations, both the acquirer and the target operate in the same industry, and in 50.2% of the firm-year observations the CEO has also the position of chairman of the board.

The findings of this research are as follows. If I pool all types of M&As together, this study finds no evidence that acquiring firms engage in income increasing real earnings management in the year prior to a M&A. Consistent with prior literature, this study finds that acquiring firms engage in income increasing real earnings management in the year prior to a for-stock merger. This study finds that acquiring firms involved in stock-for-stock M&As engage more in income increasing real earnings management if the target does not operate in the same industry, compared to stock-for-stock M&As in which both the acquirer and the target operate in the same industry. Furthermore, the results indicate that firms that are involved in stock-for-stock M&As prior to the introduction of SOX engage in more real earnings management, compared to firms that are involved in stock-for-stock M&As after the introduction of SOX.

The remainder of this research is organized in the following manner. The second section provides at first some theoretical background, and gives an analysis of existing literature, about stock-for-stock mergers and both accrual-based- and real earnings management. Thereafter, the hypotheses are developed. The third section provides information about the methodology of this research, after which section four shows the results. Finally a conclusion is provided.

(7)

7

2. Literature Review

This section provides information of theories that are applicable to mergers and acquisitions. Theories that will be discussed are agency theory and theories that can describe the reason of engaging in a merger or acquisition. Thereafter existing literature provide insights about stock-for-stock mergers, both accrual-based earnings management and real earnings management, and the shift from accrual-based earnings management to real earnings management after the implementation of SOX. Finally, hypotheses will be developed.

2.1 Theoretical background

2.1.1 Agency theory

A well-known theory within economics is the agency theory. Jensen and Meckling (1976) were the first researchers that wrote about this phenomenon, which can be explained as follows: Within a company there is a principal and an agent (for example, shareholders and managers respectively). Both might have different preferences – shareholders for example want to maximize value, while managers might prefer their own wealth, which can cause problems in running a business. Jensen and Meckling (1976) state that it is impossible for the principle at zero cost to ensure that the agent make optimal decisions from the principals’ viewpoint. To come as close as possible, the principle and the agent have to incur positive monitor and bonding costs, which together are defined as the agency costs.

Linder and Foss (2015) explain agency theory as two self-interested individuals that can have potential gains from a trade, in which the principal delegates a task to the agent. The choices and actions of the agent determine the payoff of both parties. The different interests both parties can have, are explained by Linder and Foss (2015) using three examples. First, ‘conflict in outcome-type preferences’, which include for example the conflict of interest of the managers that want to build an empire, instead of maximizing shareholders’ returns (shareholders interest). Second, ‘conflict in risk preferences’, which can be the situation when managers want to engage in capital expenditures that have the highest chance of survival for the firm, while shareholders

(8)

8 prefer to maximize returns and to engage in more risky projects to diversify risks. Third, ‘conflict of time horizon’, explain the conflicts of interest resulting from managers that are planning not to remain at the company for a long time, and therefore make more short-term oriented decisions, while long-short-term oriented shareholders might have other preferences.

Two important features of agency theory are moral hazard and adverse selection, and these two often equates the terms hidden action and hidden information, respectively. Moral hazard arises, according to Darrough and Stoughton (1986), as “when the action undertaken by the agent is unobservable and has a differential value to the agent as compared to the principal.” Performance measures could help reduce moral hazard. Adverse selection is a problem of information asymmetry in which the agent has more information than the principal. Adverse selection can be reduced by using self-selection in contracting (Darrough and Stoughton, 1986).

Applying agency theory on corporate events, such as mergers and acquisitions, literature shows that discretion over accounting information can be used in an informative way: if companies show higher earnings, this results in a higher stock price, because higher earnings are a signal for positive economic forecasts. Schatt et al. (2016) states two hypotheses about the disclosure of information. First, the private information hypothesis states that disclosures may convey management’s private information. Second, the earnings management hypothesis states that managers may have incentives to not disclose reliable information. So, every disclosure gives information to shareholders, however, the shareholders have to determine whether the disclosed information is reliable and faithful.

2.1.2 Merger Motives

In literature there are several theories described that may be motives for mergers and acquisitions. Seven theories will be explained shortly, following Trautwein (1990): efficiency theory, monopoly theory, valuation theory, empire-building theory, process theory, raider theory and disturbance theory. The first theory is the efficiency theory, which says there are three types of synergies – financial synergies, operational synergies and managerial synergies - which are assumed to be the main goal of mergers and acquisitions. Financial synergies that lower the cost of capital, which can be achieved by

(9)

9 increasing the size or by decreasing the systematic risk of the company. Operational synergies to combine operations or to transfer knowledge, and managerial synergies, which occur when the acquirer thinks the managerial skills can benefit the target’s performance. Efficiency theory is assumed to be the best motive out of the seven (Trautwein, 1990).

Second, monopoly theory views that achieving market power is the main goal of mergers. Third, valuation theory assumes that a merger occurs when managers have better information about the target than the stock market has. Fourth, empire-building theory states that managers plan and execute mergers to increase their own utility instead of their shareholders’ value. Fifth, process theory assumes mergers are a result of strategic decision and processes within a company. Sixth, raider theory gives as a motive for mergers a person that transfers wealth, such as greenmail or excessive post-merger compensation. At last, disturbance theory is described by Gort (1969) and means that mergers are caused by economic disturbance, that result in changing individual expectations and increasing uncertainty. A very extended summary of above mentioned theories is provided by Trautwein (1990).

Above theories may explain motives for mergers, but do not give any motives for the use of real earnings management. For that reason, above theories will not be used or applied in the remaining of this paper.

2.2 Stock-for-stock mergers

Mergers and acquisitions take place in different forms. According to Erickson and Wang (1999) approximately 20-30% of the mergers that took place between 1975 and1999 are in form of parties exchanging stock. In the case of stock-for-stock mergers and acquisitions, the consideration the target receives is an X number of shares of the acquirers’ stock (Botsari and Meeks, 2008). The total number of shares the target receives is dependent of the exchange ratio, which is inversely related to the stock price around the agreement date. For that reason, acquirers might have an incentive to increase accounting earnings prior to the merger or acquisition, such that the cost of buying the target reduces (Botsari and Meeks, 2008).

(10)

10 Njah and Jarboui (2013) suggest from the research of Myers and Majluf (1984) that bidders that are overvalued in the financial market usually are the ones that are willing to use their stock to finance a merger or acquisition. Because stock payments are, as the word says, dependent on the value of the stock, this type of payments increase uncertainty (increase the information asymmetry) about the bid value. Overall, previous studies (Fishman, 1989; Sudarsaman, 1995; Henry, 2004) conclude that cash-based bids have the highest possibility for success of the merger, compared to stock payments and mixed payments.

Erickson and Wang (1999) state that in case of stock-for-stock mergers, the stock is almost always newly issued. For that reason, shareholders of both the acquirer and the target have to agree with the merger agreements, and generally they receive a substantial premium (above current market price).

According to Fu et al. (2013) previous research on stock-for-stock mergers find that firms that are overvalued (high market-to-book ratio) can create value for shareholders if they use stocks as the method of payment of a merger and the probability of engaging in stock-for-stock mergers is increasing for the level of equity overvaluation (Ang and Cheng, 2006; Dong et al., 2006; Rhodes-Kropf et al., 2005). Fu et al. (2013) study mergers of US firms that are announced and completed between 1985 and 2006. Fu et al. (2013) find that acquirer CEOs in overvaluation driven acquisitions face an increase in their compensation, that outweighs the relatively small decrease in the CEO’s equity holding in the firm. Besides that, they also find evidence that overvalued acquiring firms have weak governance structures prior to the merger. This finding is consistent with the hypothesis that overvaluation generates agency costs.

2.3 Earnings Management

2.3.1 Accrual-based Earnings Management

There are two types of earnings management: accrual-based earnings management and real earnings management. Zang (2012) studies whether managers use real earnings management and accrual-based earnings management as substitutes in managing earnings, using a sample of more than 6,500 earnings management suspect firm-years over the period 1987 – 2008. The empirical results show that accrual based earnings

(11)

11 management is constrained by the presence of high-quality auditors, heightened scrutiny of accounting practice after the passage of the Sarbanes-Oxley Act of 2002, and firms’ accounting flexibility. The empirical results in the research of Zang (2012) show that real earnings management is constrained by firms’ competitive status in the industry, financial health, scrutiny from institutional investors, and the immediate tax consequences of manipulation. Zang (2012) finds that managers trade off the two approaches according to the related costs and the timing, and concludes managers use the two approaches as substitutes.

Accrual-based earnings management may only have a short-term impact on reported earnings and no direct cash flow consequences, because “earnings are ‘borrowed’ from future accounting periods, or expenses are delayed” (Botsari and Meeks, 2008). However, Botsari and Meeks (2008) argue that accrual-based earnings management can have long-term irreversible consequences in the case of stock-for-stock mergers, because accrual-based earnings management can change the terms of the deal and whether the deal succeeds.

Prior research about accrual-based earnings management is very well developed. Erickson and Wang (1999) investigate whether the “benefits” of managing earnings exceed the costs and find that acquiring firms in the period prior to the stock-for-stock merger announcement manipulate total accruals and hence manage earnings upward. Beside that they find no evidence of pre-merger earnings management of completed cash-paid mergers Their sample consists of 55 mergers performed by US companies between 1985 and 1990.

An extended research on accrual-based earning management is the research of Louis (2004). Louis (2004) examines the markets efficiency and wants to provide an explanation for the often-seen underperformance after mergers and acquisitions. The sample used for this research consists of mergers of publicly traded US companies that were announced between 1992 and 2000. The method used in this research is an event study, using the time windows [-1,1] and [-21,1]. Louis (2004) has find strong evidence, in line with Erickson and Wang (1999), that earnings are overstated, especially in the quarter preceding a stock swap announcement. Besides that, Louis (2004) has find that a significant determinant of post-merger (short- and long term) performance is the reversal of pre-merger earnings management.

(12)

12 Other research about accrual management around economic events finds that accrual-based earnings management is related to lawsuits (DuCharme et al., 2004; Gong et al., 2008). Gong et al. (2008) find that pre-merger accruals have a positive strong relationship with post-merger lawsuits. Their evidence also suggests that lawsuits contribute to the post-merger underperformance. DuCharme et al. (2004) have the same findings, namely that around stock offers, accruals are unusually high, especially for firms that are subsequently sued.

2.3.2 Real Earnings Management

Beside accrual-based earnings management, another way to manage earnings is due to real earnings management. Research on real earnings management is far less developed, and therefore less evidence is available. Roychowdhury (2006) defines real activities manipulation as “departures from normal operational practices, motivated by managers’ desire to mislead at least some stakeholders into believing certain financial reporting goals have been met in the normal course of operations”.

In his research, Roychowdhury (2006) has find evidence that real earnings management is used for different purposes. Roychowdhury (2006) find evidence that real earnings manipulation is used to temporarily increase sales (by means of price reductions), to lower cost of goods sold (by means of overproduction), to improve reported margins (by means of reduction of discretionary expenses), and less robust evidence to meet annual analyst forecasts.

Graham et al. (2005) did survey-based research to determine which factors drive reported earnings and disclosure decisions. They document a widespread use of real earnings management: about 80% of the CFO’s in their survey say that they would decrease R&D , advertising and maintenance expenses, in order to meet earnings targets. In their research they provide evidence that the following benchmarks are the most important to CFO’s: Previous quarter earnings per share (EPS), zero earnings, analyst consensus forecast, and same quarter last year.

Research on the effect of real earnings management shows different results. Taylor and Xu (2010) find, using a matched sample analysis, that on average, real earnings management does not have a significant negative effect on firms’ subsequent operating

(13)

13 performance. A regression analysis that controls for different factors confirms the results of the matched sample analysis.

Cohen and Zarowin (2010) examine both real and accrual-based earnings management activities around seasoned equity offerings (SEOs). They find that firms engage in real earnings management and that the decline in post-SEO performance due to real activities management is more severe than that due to accrual management. The importance of their findings is that underperformance is not only driven by accrual management. Cohen and Zarowin (2010) used 1511 US companies for their sample. Relying on previous research, they used three proxies for real earnings management: the abnormal levels of cash flow from operations, discretionary expenses, and production costs. A shortcoming of their research in my opinion is that they don’t use stock returns in their research, while stock returns are valuable measures for shareholder value.

Kothari et al. (2012) want to assess whether earnings management (both accrual-based and real earnings management) really mislead the market. Kothari et al. (2012) find that both real and accrual earnings management strategies result in negative future operating performance. Kothari et al. (2012) build further on the research of Cohen and Zarowin (2010), but Kothari et al. take post-SEO stock returns into account. Instead of Cohen and Zarowin (2010), Kothari et al. (2012) measure earnings management by focusing more on R&D expenses.

2.3.3 The shift from Accrual-Based to Real Earnings Management

After different highly publicized corporate failures and scandals such as Enron, Tyco and WorldCom, users of financial information had raising concerns about the integrity of accounting information provided to them, which resulted in a drop in investors’ confidence (Cohen et al., 2008; Jain and Rezee, 2006; Rezaee, 2004). As a reaction to those scandals, in 2002 the Security Exchange Commission (SEC) introduced to Sarbanes-Oxley Act (SOX) to restore investors trust such that results reflect underlying performance (McEnroe, 2007; SEC, 2002; Zhou, 2007).

Part of the Sarbanes-Oxley Act is Section 302. Section 302 requires the chief executive officer and the principal financial officer to certify in every report filed with the SEC that the financial statements and results of operations are a fair representation of the economic situation of a company. Another driver for more integrity is the personal

(14)

14 liability executives have according to Section 302: executives are personally responsible for deviations from accounting principles, which is in the case of the United States, the U.S. Generally Accepted Accounting Principles, US GAAP (McEnroe, 2007; SEC, 2002).

There is a lot of literature about the Sarbanes-Oxley Act, and a sub-category within this literature is the literature that focusses on the use of earnings management before and after introducing the Sarbanes-Oxley Act. Literature shows that accrual-based earnings management has decreased by the coming of the Sarbanes-Oxley Act (Aono and Guan, 2007; Zhou 2007). Conversely, real earnings management significantly increased in the post-SOX area, which suggests there has been a shift from accrual-based earnings management to real earnings management (Cohen et al., 2008).

2.4 Hypotheses development

Earnings management is a well-studied topic within academic research. The primary focus of earnings management research to date, has been on detecting whether and when earnings management takes place. The evidence from prior literature is consistent with firms managing earnings to window-dress financial statements prior to corporate events, such as initial public offerings and mergers and acquisitions (Healy and Wahlen, 1999). However, Erickson and Wang (1999) do not find any evidence to support the findings of Healy and Wahlen (1999) in the case of mergers and acquisitions financed with cash.

Following prior literature, I suggest that in the year prior to a corporate event, in this case a merger or acquisition, the level of real earnings management in acquiring firms is higher, compared to firm-years in which no merger or acquisition took place. Hence I test for the presence of real earnings management before any type of M&A. This is formulated in the first hypothesis:

H1: Prior to a stock-for-stock or cash-paid merger acquiring firms engage in more income increasing real earnings management, compared to years in which no merger or acquisition takes place.

Following, I focus specifically on stock-for-stock mergers (as opposed to all kind of mergers). Literature shows that overvalued acquirers are usually the ones that are willing to use their stock to finance a merger or acquisition, so that they can gain from an

(15)

15 artificially inflated valuation (Ang and Cheng, 2006; Dong et al., 2006; Fu et al, 2013; Rhodes-Kropf et al., 2005). This overvaluation is result from analysts, overinterpretation of upward distorting earnings numbers.

For those mergers in which a merger is financed with stock, the total numbers of shares issued by the acquiring firm is determined by a negotiated exchange ratio, agreed on by both parties. This exchange ratio may be set based on the acquiring firms’ stock value. Due to the uncertainty of the exchange ratio for a stock-for-stock merger, and therefore the transaction price, managers might have incentives to manage earnings (Erickson and Wang, 1999).

Following the literature I suggest that acquiring firms that engage in a stock-for-stock merger engage in higher levels of income increasing real earnings management in the year prior to the merger, as compared to years in which no stock-for-stock merger takes place. Hence, I form the following hypothesis:

H2: Prior to stock-for-stock mergers acquiring firms engage in more income increasing real earnings management, compared to years in which no stock-for-stock merger takes place.

3. Research Methodology

First, this section provides information on the data gathering process. Thereafter, the dependent- and independent variable are extensively discussed, as well as the control variables. Finally, this section provides the regression model that will be used in Section 4, the regression analysis.

3.1 Data

The sample of this research consist of a set of publically available data including both stock-for-stock mergers and cash-paid mergers, from companies of which the acquirer is listed on the S&P500 Index in the United States. Firstly, the sample is obtained from Zephyr. The search criteria used to get the data are the following:

(16)

16

Search Criteria Observations

Indices: S&P 500 (Acquirer) 25,497

Current deal status: Completed – confirmed1 _19,400

Method of payments: Cash, shares 8,736

Time period: on and after 01/01/2000 and up to and including 01/01/2012 (completed-confirmed)

5,659

From the sample of 5,659 the observations that are paid with both cash and shares, and the observations for which the Deal Type is a share buyback or capital increase are removed. This results in a sample of 3,253 observations.

Another sample is obtained from Compustat Annual Fundamentals. For this sample data is requested for the period 01/01/1998 – 01/11/2015, so that I have firm year observations for three years before and three years after the merger. This resulted in 217,860 observations. Again missing observations are omitted from the sample to get a usable dataset. This resulted in 21,095 observations.

A third sample is obtained from Execucomp, to get the variables BONUS and CEODuality. For this sample data is requested also for the period 1998 – 2015. This sample includes 39,466 observations.

To create a usable dataset for performing the regressions, all samples are merged and observations with missing data are removed. This resulted in a total amount of 5,189 observations, which includes multiple years per event (merger or acquisition). This number of observations contains 385 mergers, of which 66 are stock-for-stock mergers.

1_{The deal is included as completed – confirmed if, and only if the following is fulfilled: the deal was} successfully completed, the bidder is (or was at the time of acquisition) listed on S&P500, the bidder is a non-financial company (SIC Codes 60-69) and the transaction is either pure cash or pure shares.

(17)

17

3.2 Variable Measurement

3.2.1 Real Earnings Management

For this research I rely on the model of Roychowdhury (2006), which is often used by other researchers and is proved to be valid (Cohen et al., 2008; Cohen and Zarowin, 2010; Kothari et al., 2012; Taylor and Xu, 2010; Zang, 2012). Roychowdhury (2006) mentioned three ways of measuring real earnings management, using the models developed by Dechow et al. (1998): Sales manipulation, discretionary expenditures and production costs. To determine the use of real earnings management, the abnormal Cash Flow from Operations (Abn_CFO), the abnormal level of discretionary expenses (Abn_DISEXP), and the abnormal level of production costs (Abn_PROD) will be calculated. The abnormal levels (the proxy for real earnings management) are determined by subtracting the normal levels of the variables from the actual levels of the variables. This means that the error term captures the abnormal level. The models given below calculate the normal levels of the variables.

Firstly, I will look at sales manipulation, which can be measured by the cash flow from operations (Roychowdhury, 2006):

CFOt/At-1 = ∝

₀

+ ∝

₁

(

1 𝐴_𝑡−1

) + ∝

2

(

𝑆𝑎𝑙𝑒𝑠 𝐴_𝑡−1

) + ∝

3

(

Δ𝑆𝑎𝑙𝑒𝑠 𝐴_𝑡−1

) + 𝜀

𝑡 In which:

CFOt = Cash Flow from Operations in period t

At-1 = Total Assets at t-1

Sales = Sales at t

∆Sales = Change in sales between t and t-1 ԑt = Residual of the regression at time t

An example of managing cash flows is offering price discounts, which leads to lower cash flows in period t. So, a low value of Abn_CFO indicates a high level of real earnings management (Roychowdhury, 2006).

(18)

18 The second way Roychowdhury (2006) mentioned to manipulate earnings is by reducing discretionary expenses. Discretionary expenses are the sum of advertising, R&D and selling, general and administrative expenses.

DISEXPt/At-1 = ∝

₀

+ ∝

₁

(

1

𝐴𝑡−1

) + ∝

2

(

𝑆𝑎𝑙𝑒𝑠𝑡−1

𝐴𝑡−1

) + 𝜀

𝑡

In which:

DISEXPt = Discretionary expenses in period t

Salest-1 = Sales at t-1

If managers want to increase earnings, they can reduce discretionary expenses to boost them. So, a low value of Abn_DISEXP means a high level of real earnings management (Roychowdhury, 2006).

The final measure that will be used are the production costs (Roychowdhury, 2006). The production costs are the sum of costs of goods sold and the change in inventory:

𝑃𝑅𝑂𝐷𝑡 𝐴𝑡−1

= ∝

0

+ ∝

1

(

1 𝐴𝑡−1

) + ∝

2

(

𝑆𝑎𝑙𝑒𝑠𝑡 𝐴𝑡−1

) + ∝

3

(

Δ𝑆𝑎𝑙𝑒𝑠𝑡 𝐴𝑡−1

) +∝

4

(

Δ𝑆𝑎𝑙𝑒𝑠𝑡−1 𝐴𝑡−1

) + 𝜀

𝑡 In which:

PRODt = Production costs in period t

∆Salest-1 = Change in sales between t-1 and t-2

If managers want to increase earnings, they can produce more, such that they can spread the costs of goods sold over more units. This results in a decrease in cost of goods sold, and an increase in earnings. For above mentioned reason, then a high value of Abn_PROD, indicates a high level of real earnings management (Roychowdhury, 2006).

Afterwards, the three proxies will be accumulated to see the overall impact of real earnings management, which will be captured in the variable REM_proxy. The abnormal level of cash flow from operations and the abnormal level of discretionary expenses will be multiplied by -1 in the accumulation, because for that proxies, the lower the amount,

(19)

19 the higher the use of income increasing real earnings management. So by multiplying those proxies with -1, the interpretation of real earnings management is more straightforward. Now, higher values of REM_proxy indicate more income increasing real earnings management.

3.2.2 Mergers

To measure the use of real earnings management prior to a stock-for-stock merger, I need independent variables that measure mergers and acquisitions. Therefore, I build two dummy variables.

The first variable, ACQ, is a dummy variable which denotes 1 for the years t that company i made one of the two type of acquisitions, cash paid or stock-for-stock, and 0 otherwise. The dummy variable is included, because the sample period includes years where a firm does not do any acquisition. The coefficient of the variable ACQ provides an answer whether any kind of acquisition affects real earnings management, and therefore, to answer the first hypothesis. A positive coefficient of the variable ACQsupports the first hypothesis developed in section 2.4, because that means that if a firm does any kind of a merger or acquisition, it increases the use of real earnings management (if coefficient of variable ACQ increases, REM_proxy increases). Based on my H1, I expect a positive relation between ACQ and real earnings management.

The second variable, SFSA, is a dummy variable which denotes 1 for the years t that company i made a stock-for-stock merger or acquisitions, and 0 otherwise. The dummy variable is included to measure the use of real earnings management during the year prior the stock-for-stock merger or acquisition, and therefore, to answer the second hypothesis. A positive coefficient of the variable SFSA supports the second hypothesis developed in section 2.4, because that means that if a firm does a stock-for-stock merger or acquisition it increases the use of real earnings management (if the coefficient of the variable SFSA increases, REM_proxy increases). Based on my H2, I expect a positive relation between SFSA and real earnings management.

(20)

20 3.2.3 Control Variables

Based on existing literature, some firm-specific control variables are included. This study uses the following control variables: SHARES, BIG 4, BONUS, SIZE, LEVERAGE, ROA, MB, INDUSTRY and CEODuality.

Following Cohen and Zarowin (2010), I include a variable called SHARES, which is the natural logarithm of the total outstanding shares. Cohen and Zarowin (2010) argue that a greater number of shares outstanding requires more earnings management to achieve the target. Since there is no clear explanation whether the share effect would induce earnings management, no directional prediction is made about this variable.

A variable called BIG 4 is included following the argumentation of Cohen et al. (2008) who states that less earnings management is not necessary the result of higher quality monitoring. In the choice for an audit firm, firms can do some self-selection. Besides, Chi et al. (2011) include the variable BIG 4 because they argue that (opposed to other literature) clients of higher quality auditors are likely to resort to more real activities manipulation. The variable BIG 4 will denote one if a company is audited by a Big 4 firm, and zero if not. Since there is no clear explanation whether a BIG 4 auditor would induce earnings management, no directional prediction is made about this variable.

The variable called BONUS is included, because according to Cohen and Zarowin (2010) recent research suggests that excess compensation is associated with earnings management. So the higher the bonus, the higher the level of earnings management. The variable BONUS is the bonus as a percentage of total compensation.

The remaining control variables – SIZE, LEVERAGE, ROA, MB, S_INDUSTRY and CEODuality – are some general firm-specific variables, that are included following recent literature (Chi et al., 2011; Cohen and Zarowin, 2010; Healy and Wahlen, 1999; Zang, 2012). A specification of these variables can be found in Appendix A.

(21)

21

3.3 Regression Model

The regression model follows Louis (2004). The dependent variable, REM_proxy, is a proxy for the use of income increasing real earnings management in de year prior to a merger or acquisition. This variable is a summation of the following proxies: Abnormal Cash Flows From Operations (Abn_CFO), Abnormal Discretionary Expenses (Abn_DISEXP), and Abnormal production costs (Abn_PROD). The dependent variable is defined as mentioned above to give a clear answer on the research question. The measurement of the dependent variable is described in Section 3.2.1.

𝑅𝐸𝑀_𝑝𝑟𝑜𝑥𝑦_𝑖,𝑡 = 𝛼 + 𝛽₁𝑆𝐹𝑆𝐴_𝑖,𝑡+ 𝛽₂𝐴𝐶𝑄_𝑖,𝑡 + 𝛽₃𝑆𝐻𝐴𝑅𝐸𝑆 + 𝛽₄𝐵𝐼𝐺4_𝑖,𝑡

+ 𝛽₅𝐵𝑂𝑁𝑈𝑆_𝑖,𝑡 + 𝛽₆𝑆𝐼𝑍𝐸_𝑖,𝑡+ 𝛽₇𝐿𝐸𝑉𝐸𝑅𝐴𝐺𝐸_𝑖,𝑡+ 𝛽₈𝑅𝑂𝐴_𝑖,𝑡+ 𝛽₉𝑀𝐵_𝑖,𝑡 + 𝛽₁₀𝐼𝑁𝐷𝑈𝑆𝑇𝑅𝑌_𝑖,𝑡+ 𝛽₁₁𝐶𝐸𝑂𝐷𝑢𝑎𝑙𝑖𝑡𝑦_𝑖,𝑡+ 𝜀_𝑖

4. Results

First, this section provides some descriptive statistics, some clarifying details about the dataset, about the fiscal years and industry, and Pearson correlations. Thereafter, I provide the analysis and discussion on the multivariate analysis. Finally, this section provides some sensitivity analyses, to examine the validity of the results.

4.1 Descriptive Statistics

The descriptive statistics of the sample used in this study are shown in Table 1. All variables are winsorized on 1st_{and 99}th_{percentile before using them, to mitigate the}

effect of outliers. Table 1 shows that on average the real earnings management proxy (REM_proxy) is 0.013. The mean of the variable SFSA is 0.185, which means that 18.5% of the firm-year observations include information of acquiring firms that engage in stock-for-stock mergers. The mean of the variable ACQ is 0.074, which means that 7.4% of the firm-years consist of a merger, which can be either a stock-for-stock merger or a merger paid in cash. The minimum and maximum value of the variable BIG4 is 1, which means that 100% of the firms are audited by a big-4 firm, which seems plausible, since the

(22)

22 sample consist of acquiring firms listed on the S&P 500. The variable BIG4 is omitted for the regressions, because of the lack of variation (i.e., BIG4 cannot explain any variation in the dependent variable). Both the return on assets (ROA) and the market-to-book ratio (MB) show a positive mean of 8.8% and 470.7%, respectively. The mean of the variable S_INDUSTRY is 0.667, which means that in 66.7% of the firm-year observations, the acquirer as well as the target is categorized in the same industry, according to US SIC codes. Lastly, the mean of the variable CEODuality, 0.502, means that in 50.2% of the cases the CEO is also chairman of the board.

Table 1 - Descriptive Statistics

Variable N Mean Std. Dev. Min. Max. P25% P50% P75%

REM_proxy 5189 0.013 0.324 -0.996 0.679 -0.1770 0.0608 0.2223 SFSA 5189 0.185 0.388 0 1 0 0 0 ACQ 5189 0.074 0.262 0 1 0 0 0 SHARES 5189 6.800 1.454 3.998 9.095 5.5070 6.8017 8.0209 BONUS 5189 0.087 0.145 0 0.671 0 0 0.1506 SIZE 5189 9.817 1.470 6.720 12.294 8.4927 9.8205 11.1721 LEVERAGE 5189 0.170 0.129 0 0.661 0.0704 0.1596 0.2391 ROA 5189 0.088 0.075 -0.327 0.242 0.0618 0.0918 0.1278 MB 5189 4.707 5.126 -13.117 37.395 2.5444 3.7833 5.4935 S_INDUSTRY 5189 0.667 0.471 0 1 0 1 1 CEODuality 5189 0.502 0.500 0 1 0 1 1

This table provides the descriptive statistics that are used the regression. It shows the number of observations (N), the mean, the standard deviation, the minimum value, the maximum value, and the percentiles 25%, 50% and 75%. See Appendix A for variable definitions.

Table 2 provides some details about in which sector the acquirer operates, using US SIC codes. The sectors Manufacturing and Services are well presented in the sample, with 216 and 130 deals respectively, while no acquiring company operates in the Construction, Finance, Insurance, Real Estate, and Public Administration sector.

The deals that are financed with shares (stock-for-stock mergers) are spread over three sectors: Manufacturing, Retail Trade and Services, with 46, 1 and 19 deals respectively. Of all the deals, in 256 cases both the acquirer and the target are in the same industry.

(23)

23

Table 2 - Sectors acquiring companies

US SIC Code – Sector SFSA Cash N Same Industry?

0100-0900: Agriculture, Forestry, Fishing 0 5 5 2

1000-1499: Mining 0 3 3 2

1500-1799: Construction 0 0 0 0

2000-3999: Manufacturing 46 170 216 131

4000-4999: Transportation & Public Utilities 0 6 6 4

5000-5199: Wholesale Trade 0 3 3 1

5200-5999: Retail Trade 1 21 22 14

6000-6799: Finance, Insurance, Real Estate 0 0 0 0

7000-8999: Services 19 111 130 102

9100-9999: Public Administration 0 0 0 0

Total 66 319 385 256

Table 3 provides the sample distribution by year. In the year 2000, the most stock-for-stock mergers took place, namely, 27 of the 66. The amount of stock-for-stock-for-stock-for-stock mergers decreased after 2002, which indicates that the incentive to participate in a stock swap is decreased by the introduction of the Sarbanes-Oxley Act, Section 203. This pattern cannot be seen for cash purchases. Cash purchases are distributed very equal across the years, with a peak in 2006.

Table 3 – Sample distribution by year

Year SFSA Cash Total

2000 27 16 43 2001 17 25 42 2002 6 23 29 2003 4 25 29 2004 2 31 33 2005 5 29 34 2006 2 35 37 2007 2 32 34 2008 1 26 27 2009 0 19 19 2010 0 37 37 2011 0 21 21 2012 0 0 0 Total 66 319 385

(24)

24 Table 4 provides the Pearson correlations of the variables. The real earnings management proxy is significantly correlated at the 5% level with all the variables. The real earnings management proxy is significantly positively correlated with the variable SFSA, which indicates that a stock-for-stock merger increases the engagement of income increasing real earnings management in the year prior to a stock-for-stock merger. The real earnings management proxy is significantly negatively correlated with the variable ACQ, which indicates that in general, a merger or acquisition, lowers the engagement in income increasing real earnings management in the year prior to a merger or acquisition. The variable ACQ includes all types of mergers, in which 18.5% of the sample is represented by information on acquiring firms engaging in stock-for-stock mergers. The rest of the sample, 81.5%, includes information on acquiring firms engaging in mergers financed with cash.

(25)

Table 4 - Pearson Correlation Matrix Variables 1 2 3 4 5 6 7 8 9 10 11 1. REM_proxy 1 2. SFSA 0.0889** 1 3. ACQ -0.0333** -0.0099 1 4. SHARES -0.0334** 0.2514** -0.0212 1 5. BONUS -0.0989** 0.0727** 0.0528** 0.1410** 1 6. SIZE 0.1719** 0.1177** -0.0482** 0.8460** 0.0354** 1 7. LEVERAGE 0.3245** 0.0247* -0.0313** -0.0766** -0.0836** 0.1054** 1 8. ROA -0.4069** -0.0768** -0.0106 0.1590** 0.1294** 0.0993** -0.2075** 1 9. MB -0.2545** -0.0485** 0.0486** -0.0248* 0.0698** -0.0751** 0.0693** 0.1237** 1 10. INDUSTRY -0.0617** -0.1023** -0.0011 -0.0722** 0.0085 -0.0379** -0.0099 0.0348** -0.0259* 1 11. DUALITY 0.1036** -0.0098 -0.0024 -0.0604** 0.1026** 0.0946** 0.1732** -0.0143 0.0218 -0.0022 1

The symbols ** and * denote statistical significance at the 5% and 10% levels, respectively. The numbers in the first row refer to the numbers in the first column.

(26)

4.2 Multivariate Analysis

Any Type of Mergers and Acquisitions

For the first hypothesis the explanatory variable of interest is ACQ, which includes any type of mergers and acquisitions (i.e., stock-for-stock or cash paid M&As). This variable includes for 18.5% firm-year observations on firms that engage in stock-for-stock mergers, and for 81.5% firm-year observations on firms that engage in cash purchases. Firstly, I performed regressions on the separate proxies of real earnings management, which are the variables Abn_CFO, Abn_DISEXP and Abn_PROD. The results of those regressions are provided in Appendix B. The coefficients for the variable ACQ for those regressions are 0.0032, 0.0018 and -0.0007 respectively, and none of them is significant.

To draw overall conclusions on the engagement in real earnings management in the year prior to any type of merger or acquisition, I use the variable REM_proxy. The results of the regression analysis of mergers and acquisitions (SFSA and ACQ) on the proxy of engagement in real earnings management (REM_proxy) are provided in Table 5. The coefficient of the variable ACQ is negative (-0.0050), but not significant for the real earnings management proxy (REM_proxy). This provides no evidence of a statistical significant correlation, and therefore the findings do not support the first hypothesis. Contrary to the findings of Healy and Wahlen (1999) that a corporate event such as a merger or acquisition gives incentives to window-shell financial statements, this study provides no evidence to support this. However, results are consistent with the study of Erickson and Wang (1999), who also did not find any evidence for earnings management in the year prior to a merger or acquisition financed with cash.

The contrary relationship I find can be explained by the introduction of the Sarbanes-Oxley Act, introduced in 2002. This act requires management to give a true and fair view of the economic situation of the company, and executives are personally responsible for deviations from accounting principles (SEC, 2002). This indicates that executives are not willing to take the risk to deviate from the accounting principles, without a significant incentive.

(27)

27

Table 5 – OLS Regression on REM_proxy

Panel A: Variable Statistics

Variable Coefficients Constant -0.4462** (0.1990) SFSA 0.0830*** (0.0274) ACQ -0.0050 (0.0139) SHARES -0.1061*** (0.0245) BONUS 0.0020 (0.0832) SIZE 0.1235*** (0.0329) LEVERAGE 0.4496*** (0.1476) ROA -1.4849*** (0.2924) MB -0.0112*** (0.0031) S_INDUSTRY -0.0388 (0.0315) CEODuality -0.0085 (0.0413) Panel B: Model Statistics

Observations 5189

R-squared 0.3771

Prob. F 0.0000

The dependent variable in this regression is REM_proxy. Coefficients are the estimated coefficients. Standard errors are in parentheses. The symbols *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively. See Appendix A for variable definitions. Year fixed-effects are included. Standard errors are clustered on a firm level.

(28)

28 Stock-for-stock mergers and acquisitions

For the second hypothesis the explanatory variable of interest is SFSA, capturing stock-for-stock mergers. The coefficients for the variable SFSA for the regressions on Abn_CFO, Abn_DISEXP and Abn_PROD (i.e. proxies of real earnings management), provided in Appendix B, are -0.0062, -0.0445 and 0.0332 respectively. The coefficient of the variable Abn_CFO is not significant, but the coefficients of the variables Abn_DISEXP and Abn_PROD are significant, at the 5% and 1% level respectively. This indicates that in the year prior to a stock-for-stock merger, acquiring firms engage in income increasing real earnings management by having abnormal discretionary expenses and abnormal production costs.

To draw overall conclusions on the engagement in real earnings management in the year prior to a stock-for-stock merger or acquisition, again I use the variable REM_proxy. The coefficient of the variable SFSA, provided in Table 5, is 0.0830 and is significant at the 1% level. So, the coefficient of this proxy of real earnings management (REM_proxy) provides evidence that acquiring firms engage in income increasing real earnings management in the year prior to a stock for stock merger, and therefore supports the second hypothesis. These findings are in line with prior literature (Cohen and Zarowin, 2010; Graham et al., 2005; Roychowdhury, 2006).

Control Variables

With regard to the control variables, the results in Table 5 suggest that firms that have a higher leverage and are large in size, are more likely to engage in income increasing real earnings management. Besides that, the results suggest that CEO’s that have a higher bonus, are more likely to engage in income increasing real earnings management, but this result is not significant. Lastly, the results suggest that firms with a higher amount of outstanding shares, are less likely to engage in real earnings management. All of these results are in line with prior literature (Cohen and Zarowin, 2010; Healy and Wahlen, 1999).

(29)

29 Change in real earnings management

Afterwards, additional regressions are performed on the change in the real earnings management proxy (REM_proxy), which means the difference in engagement in income increasing real earnings management in the year prior to a merger or acquisition, and the engagement in income increasing real earnings management in the year of the merger or acquisition. This is captured in the variable dREM. The results of this analysis are provided in Table 6.

In the first model, the coefficient of the variable SFSA is 0.0010, but not significant. In the second model, the coefficient of the variable SFSA is 0.0152, and it is significant at the 10% level. The insignificance of the variable SFSA in the first model is caused by including the control variables SHARES, BONUS, SIZE, ROA, and MB. This research provides evidence that acquiring companies engage more in income increasing real earnings management prior to a stock-for-stock merger, compared to years in which no merger or acquisition takes place. This is consistent with prior literature, that states that in the year prior to a stock-for-stock merger or acquisition, managers may have incentives to manage earnings upwards, such that the costs for buying the target reduces (Botsari and Meeks, 2008).

(30)

30

Table 6 – OLS Regression on dREM

Panel A: Variable Statistics

Variable Coefficients (1) Coefficients (2) Constant 0.0264 (0.0174) 0.1086 (0.0822) SFSA 0.0010 (0.0057) 0.0152* (0.0090) ACQ -0.0187** (0.0082) -0.0208** (0.0102) SHARES 0.0180*** (0.0035) BONUS -0.1750*** (0.0172) SIZE -0.0015 (0.0034) LEVERAGE 0.0699*** (0.0178) 0.0633 (0.0460) ROA -0.1066*** (0.0317) MB -0.0005* (0.0003) S_INDUSTRY 0.0081* (0.0043) 0.0043 (0.0071) CEODuality 0.0240*** (0.0045) 0.0178 (0.0141) Panel B: Model Statistics

Observations 4801 4801

R-squared 0.1129 0.0714

Prob. F 0.0000 0.0000

The dependent variable in this regression is dREM. Coefficients are the estimated coefficients. Standard errors are in parentheses. The symbols *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively. See Appendix A for variable definitions. Year fixed-effects are included. Standard errors are clustered on a firm level.

(31)

31

4.3 Sensitivity analysis

4.3.1 Same industry versus different industry

In prior literature about mergers and acquisitions related to earnings management, to the best of my knowledge, nothing is written about the impact industry can have on the engagement in real earnings management. Therefore, I perform a supplementary test to examine the effects of company characteristics on real earnings management. To do this additional analysis, two sub-samples are used: one that consists of mergers and acquisitions in which both the acquirer and the target operate in the same industry, and one that consists of mergers and acquisitions in which the acquirer and the target do not operate in the same industry. The results of the additional analysis are provided in Table 7.

For both subsamples “not same industry” and “same industry”, the coefficients of the variable SFSA are positive and significant at the 1% level. The coefficients of the variable SFSA are 0.1400 and 0.0432, for not in the same industry and in the same industry, respectively. The results are in line with the results of the whole sample on REM_proxy, which are provided in Table 5. By comparing the coefficients of the subsamples, the p-value is 0.0000, which means that the coefficients of the variable SFSA are significantly different from each other at the 1% level.

This study provides evidence that both acquiring firms of which the target operates in the same industry and acquiring firms of which the target does not operate in the same industry, participating in stock-for-stock mergers, exhibit income increasing real earnings management. Furthermore, this research provides evidence that when the acquirer and target involved in the M&A operate in a different industry, acquiring firms engage more in income increasing earnings management, compared to M&As where de acquirer and the target do not operate in the same industry (i.e. 0.1400>0.0432).

4.3.2 Prior-SOX versus Post-SOX

Cohen et al. (2008) find that engagement in income increasing real earnings management significantly increased after the introduction of SOX in 2002. Therefore, I perform another supplementary test to examine the effects of deal characteristics on real earnings management. To do this additional analysis, two sub-samples are used: one which

(32)

32 consists of mergers and acquisitions announced before the introduction of SOX, and one which consists of mergers and acquisitions announced after the introduction of SOX. The results of this additional analysis are also provided in Table 7.

For both the subsamples, prior and post SOX, the coefficients of the variable SFSA are positive and significant at the 1% level. The coefficients of the variable SFSA are 0.0983 and 0.0493, for prior-SOX and post-SOX respectively. By comparing the coefficients of the subsamples, the p-value is 0.0496, which means that the coefficients of the variable SFSA are significantly different from each other at the 5% level.

The results indicate that, both prior SOX and post SOX, firms participating in stock-for-stock mergers exhibit income increasing real earnings management. Furthermore, the results indicate that firms that are involved in stock-for-stock mergers and acquisitions prior to the introduction of SOX engage in more real earnings management, compared to firms that are involved in stock-for-stock mergers after the introduction of SOX (i.e. 0.0983>0.0493). This is not consistent with prior literature, which states that the use of real earnings management increased after the introduction of SOX (Cohen et al., 2008). The contrary results I find can be explained by the personal responsibility managers have since the introduction of SOX, which may reduce the incentive to engage in real earnings management. Besides, all acquiring firms in the sample of this research are listed on the S&P500 index, which includes the largest companies of the United States, and therefore are under scrutiny quite often, which also may reduce the incentive to engage in income increasing real earnings management.

(33)

33

Table 7 – OLS regression on REM_proxy - subsamples

Panel A: Variable Statistics

Variable Not same

industry

Same industry Prior-SOX Post-SOX

Coefficient Coefficient Coefficient Coefficient

Constant -0.3496*** (0.0510) -0.5032*** (0.0365) -0.4948*** (0.0542) -0.4444*** (0.0371) SFSA 0.1400*** (0.0145) 0.0432*** (0.0131) 0.0983*** (0.0131) 0.0493*** (0.0191) ACQ -0.0240 (0.0220) 0.0007 (0.0172) -0.0459* (0.0269) -0.0026 (0.0172) SHARES -0.1287*** (0.0099) -0.0922*** (0.0074) -0.0978*** (0.0100) -0.1102*** (0.0074) BONUS -0.0035 (0.0495) -0.0063 (0.0372) 0.0103 (0.0422) -0.0027 (0.0411) SIZE 0.1433*** (0.0101) 0.1074*** (0.0073) 0.1230*** (0.0104) 0.1254**** (0.0071) LEVERAGE 0.0557 (0.0451) 0.6853*** (0.0402) 0.6982*** (0.0545) 0.3723*** (0.0367) ROA -1.5687*** (0.0897) -1.4051*** (0.0664) -1.4284*** (0.0896) -1.4813*** (0.0663) MB -0.0115*** (0.0010) -0.0109*** (0.0010) -0.0106*** (0.0014) -0.0111*** (0.0009) S_INDUSTRY -0.1163*** (0.0120) -0.0003 (0.0097) CEODuality 0.0488*** (0.0127) -0.0201** (0.0101) -0.0105 (0.0128) -0.0068 (0.0010) Panel B: Model Statistics

Observations 1729 3460 1681 3508

R-squared 0.4384 0.3760 0.4487 0.3643

Prob. F 0.0000 0.0000 0.0000 0.0000

The dependent variable in the regressions is REM_proxy. Coefficients are the estimated coefficients. Standard errors are in parentheses. The symbols *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively. See Appendix A for variable definitions. Year fixed-effects are included.

(34)

34

5. Conclusion

This study investigates how stock-for-stock mergers and acquisitions influence the use of income increasing real earnings management in the year prior to such a merger or acquisition. This examination is done by performing regressions of mergers and acquisitions - of which the acquirer is listed on the S&P500 index - that were completed between 2000 – 2012, on the engagement in real earnings management.

There are two types earnings management: accrual-based earnings management and real earnings management. Accrual-based earnings management is described by Botsari and Meeks (2008) as: “Earnings are ‘borrowed’ from the future accounting periods, or expenses are delayed”. This means accrual-based earnings management only has a short-term, temporary effect. Real earnings management is described by Roychowdhury (2006) as “departures from normal operational practices, motivated by managers’ desire to mislead at least some stakeholders into believing certain financial reporting goals have been met in the normal course of operations”. This means real earnings management may have destroying consequences, since there is no reversal of the manipulation.

Literature shows that accrual-based earnings management has decreased by the coming of the Sarbanes-Oxley Act in 2002 (Aono and Guan, 2007; Zhou 2007). Conversely, real earnings management significantly increased in the post-SOX area, which suggests there has been a shift from accrual-based earnings management to real earnings management (Cohen et al., 2008). Real earnings management is important to study, because compared to accrual based earnings management it can be more harmful for the firm, i.e. it can adversely affect firm value by distorting normal operating activities (Cho and Chun, 2016). To the best of my knowledge, there is no other study that investigates the engagement in income increasing real earnings management prior to stock-for-stock mergers and acquisitions.

Botsari and Meeks (2008) suggest that participating in a stock-for-stock merger gives managers incentives to manage earnings, such that the acquirer pays less for the target. This incentive is confirmed by other research, who provide evidence of engagement in income increasing real earnings management, prior to a corporate event such as a merger or acquisition (Cohen and Zarowin, 2010; Graham et al., 2005; Kothari et al., 2012; Roychowdhury, 2006).

(35)

35 First, this study investigates whether any type of mergers and acquisitions affects the engagement in income increasing real earnings management. The results show a negative relationship between any type of M&A’s and the proxy for real earnings management, and the relationship is not significant. Therefore, the first hypothesis is rejected. These results are not consistent with prior literature (Healy and Wahlen, 1999), which can possibly be explained by the introduction of the Sarbanes-Oxley Act in 2002. The Sarbanes-Oxley Act requires managers to give a true and fair view of financial statements, and if they deviate from this requirement they are personally responsible (SEC, 2002). This may lower the incentive to participate in real earnings management.

Secondly, this study investigates whether participation in stock-for-stock mergers and acquisitions affects the engagement in income increasing real earnings management. The results show a positive relationship between stock-for-stock M&A’s and the proxy of real earnings management, and the relationship is significant at the 1% level. Therefore, the second hypothesis hold, and is consistent with prior literature.

Furthermore, this study provides two additional analysis, to examine the effects of firm characteristics and deal characteristics on real earnings management. The first sensitivity analysis finds that acquiring firms involved in stock-for-stock M&As engage more in income increasing real earnings management if the target does not operate in the same industry, compared to stock-for-stock M&As in which both the acquirer and the target operate in the same industry. Furthermore, the results of the second sensitivity analysis indicate that firms that are involved in stock-for-stock M&As prior to the introduction of SOX engage in more real earnings management, compared to firms that are involved in stock-for-stock M&As after the introduction of SOX.

There are some limitations to this study. First, this study uses a sample of only S&P500 listed firms, so it is not possible to apply the results for real earnings management to other world regions, such as Europe or Asia. Second, this study investigated 5189 firm-year observations of 385 M&A deals, of which 66 are stock-for-stock mergers. This may be too small to be a representative population.

The findings add to prior literature, because this is the first study which investigates the engagement in income increasing real earnings management prior to stock-for-stock mergers. Findings are consistent with prior literature about real earnings management prior to corporate events. Besides, real earnings management may be

(36)

36 harmful for the firm by distorting normal operating activities. Therefore, the results of this study are interesting to directors, shareholders and standard setters.

As a suggestion for future research, I think it is important to always investigate real earnings management together with accrual-based earnings management, since Zang (2012) shows that the two approaches are used as substitutes. Besides, I think it is important to do some additional research on this topic for other world regions, since differences can occur due to for example regulations and governance.

(37)

37

References

Ang, J., Cheng, Y., (2006). Direct evidence on the market-driven acquisition theory. Journal of Financial Research, Vol. 29, pp. 199 – 216.

Aono, J.Y., Guan, L., (2007). The impact of Sarbanes-Oxley Act on cosmetic earnings management. Research in Accounting Regulation, Vol. 20, pp. 205 – 215.

Botsari, A., Meeks, G., (2008). Do Acquirers Manage Earnings Prior to a Share for Share Bid? Journal of Business Finance & Accounting, Vol. 35(5/6), pp. 633-670

Chi, W., Lisic, L.L., Pevzner, M., (2011). Is Enhanced Audit Quality Associated with Greater Real Earnings Management? Accounting Horizons, Vol. 25(2), pp. 315 – 335. Cho, E., Chun, S., (2016) Corporate social responsibility, real activities earnings management, and corporate governance: evidence from Korea. Asia-Pacific Journal of Accounting & Economics, Vol. 23(4), pp. 400 – 431.

Cohen, D.A., Dey, A., 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). Accural-based and real earnings management activities around seasoned equity offerings. Journal of Accounting and Economics, Vol. 50, pp. 2 - 19

Darrough, M.N., Stoughton, N.M., (1986). Moral Hazard and Adverse Selection: The Question of Financial Structure. The Journal of Finance, Vol. 41(2), pp. 501 – 513 Dechow, P.M., Kothari, S.P., Watts, R.L., (1998). The relation between earnings and cash flows. Journal of Accounting and Economics, Vol. 25, pp. 133 – 168

(38)

38 Dong, M., Hirshleifer, D., Richardson, S., Teoh, S., (2006). Does investor misevaluation drive the takeover market? Journal of Finance, Vol. 61, pp. 725 – 762.

DuCharme, L.L., Malatesta, P.H., Sefcik, S.E., (2004). Earnings management, stock issues, and shareholder lawsuits. Journal of Financial Economics, Vol. 71, pp. 27 – 49 Erickson, M., Wang, S. (1999), Earnings Management by Acquiring Firms in Stock for

Stock Mergers. Journal of Accounting and Economics, Vol. 27(2), pp. 149-176 Fishman, M.J., (1989). Preemptive Bidding and the Role of Medium of Exchange Acquisitions. The Journal of Finance, Vol. 44(1), pp.41 - 57

Fu, F., Lin, L., Officer, M.S., (2013). Acquisitions driven by stock overvaluation: Are they good deals? Journal of Financial Economics, Vol. 109, pp. 24 – 39.

Gong, G., Louis, H., Sun, A.X., (2008). Earnings management, lawsuits, and stock-for-stock acquirers’ market performance. Journal of Accounting and Economics, Vol. 46, pp. 62 – 77

Gort, M. (1969). An economic disturbance theory of mergers. Quarterly Journal of Economics, Vol. 83, pp. 624-642

Graham, J.R., Harvey, C.R., Rajgopal, S., (2005). The economic implications of corporate financial reporting. Journal of Accounting and Economics, Vol. 40, pp. 3 - 73

Healy, P.M., Wahlen, J.M., (1999). A Review of the Earnings Management Literature and Its Implications for Standard Setting. Accounting Horizons, Vol. 13(4), pp. 363 –

383.

Henry, D., (2004). Corporate Governance and ownership structure of target companies and the outcome of takeovers. Pacific-Basin Finance Journal, Vol. 12, pp. 419 –

Real earnings management & mergers and acquisitions : real earnings management by acquiring firms prior to stock-for-stock mergers