The effect of gender from top executives on earnings management

(1)

Master Thesis:

The effect of gender from top

executives on earnings management

Final version

Dennis Pals (10024778) 19th_{of June 2015} Supervisor: Dr. P. Kroos

MSc Accountancy & Control, variant Accountancy Amsterdam Business School

(2)

2 Statement of Originality

This document is written by Dennis Pals 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.

Abstract

This paper examines whether women in top executive positions engage less in earnings management than men in top executive positions. Using a sample of 18332 manager year observations over a period from 2007 to 2013 there is no evidence found that women in top executive positions engage less in earnings management than men in top executive positions. The study also answer calls from prior literature for research regarding the differences in gender effect between top executives. Despite of the findings which indicate that there is a significant difference in the level of earnings management between male and female CFOs, there is no evidence found supporting that the gender effect is stronger for CFO positions than for CEO positions.

(3)

3

Stakeholders use the information embedded in the annual report of companies for investment decisions. To appropriately assess a company’s current position and financial performance, they expect that reported earnings closely reflect the economic reality of the organization’s financial position and performance throughout the reporting period.

However, there can be a difference between the reported earnings and the actual economic performance of a firm. The degree to which the reported earnings capture the firm’s

economic reality is referred as the quality of earnings (Healy and Wahlen, 1999).

The difference between the reported earnings and the actual economic performance may be the result of earnings management. This principle of earnings management involves making use of the discretion inherent in financial reporting in such a way that the reported earnings are higher (or lower) than warranted by the economic reality. The company or the managers benefit from this higher level of reported earnings at the expense of external stakeholders.1_{Graham et al. (2005) perform a survey amongst more than 400 executives}

and concluded that earnings management is an ethical issue. In an organization there are certain factors that can motivate managers to engage in earnings management. The goal of the survey is to determine the factors that drive reported earnings and disclosure decisions. Prior research identified some of these conditions, for example the violation of debt

covenants, reputation with stakeholders,2_{performance based compensation or pending}

litigation. (Beaver and Engel, 1996; Graham et al 2005; Hall and Stammerjohan, 1997; Healy, 1985; Jones, 1991). On the other hand, regulators, investors and other users of financial statements are interested in the mechanisms that can prevent managers to engage in earnings management.

Prior research on gender and ethics showed that women display more ethical behavior than men (Beltramini, Peterson, & Kozmetsky, 1984; Ferrell & Skinner, 1988; Betz, O'Connell, & Shepard, 1989; Akaah, 1989; Ruegger & King, 1992; Nguyen, Basuray, Smith, Kopka, & McCulloh, 2008). As Bruns and merchant (1990) indicate that earnings

management also features an ethical dimension, prior research also examined the influence

1_{Here, the focus will be on the opportunistic management of earnings numbers.}

2_{By managing earnings, firms are able to enhance their reputation with stakeholders, such as suppliers,}

(5)

5 of gender on earnings management. Krishnan and Parsons (2008) find that gender diversity in senior management improves the quality of reported earnings. Shawver, Bancroft, and Senneti (2006) suggest that female accountants were less likely to perform earnings management. In general, prior research documented a positive relation between female employees and earnings quality.

Another stream of literature looked into the question which executives are

predominantly responsible for accrual-based earnings management. While most of the prior research established a relationship between CEO incentives and earnings management, Jiang et al. (2010) show that especially CFO incentives are associated with earnings

management. Furthermore, Feng et al. (2011) show that powerful CEOs pressure CFOs into earnings management. This paper combines both aforementioned streams of literature by looking at the gender of top executives and how this influence earnings management behavior.

1.2 Research question

The goal of this paper is to bring the research on gender and earnings management together and examine whether there is a difference in earnings management between male and female chief executive officers (CEOs) and chief financial officers (CFOs). To explore the difference in gender effect for CEOs and CFOs, first the study need to re-establishes the gender effect for top executives on earnings quality. So, the paper will first investigate the following research question:

“To what extent does the gender of top executives affect earnings management?”

Subsequently, the paper will perform an analysis of the differences in gender effect on earnings management between CEOs and CFOs. Because CFO’s primary responsibility is financial reporting, CFOs should play a stronger role than CEOs in earnings management (Jiang et al. 2010). Therefore, the gender effect can be different for CFOs than for CEOs. This investigation yield to the second research question of the paper:

(6)

6

1.3 Relevance

This study contributes to the existing literature by re-examining the relationships between gender and earnings management in a broad sample of U.S. firms over an extensive time range.

The results of the study contribute to the current knowledge about the advantages of women in senior management positions. Prior research showed that female employees engage less in earnings management (Krishnan and Parsons, 2008; Shawver, Bancroft, and Senneti, 2006). The main contribution follows from the fact that prior research has not covered the gender effect for CEOs and CFOs. Prior research on the gender effect of CEOs cannot unambiguously be extrapolated to CFOs given the distinctively different roles that both executive managers have within companies. So, the differences of the gender effect between these positions is not examined before and could give us new insights about how the gender effect varies across different executives management positions.

The research question also have great practical relevance given the societal discussions on gender diversity. There is discussion about the diversity aspect in board rooms. Especially gender diversity is a hot issue and is receiving considerable attention in the work place nowadays. In 2009 KPMG did a survey of 955 companies about diversity. The study reveals that more than 61% of these companies have a diversity policy in place

monitoring gender, age, race/ethnicity and disability (KPMG, 2009). This is not the only study about the diversity in companies in recent years. After the introduction of several corporate governance codes all over the world, the attendance on diversity in top management has increased significantly. So, the outcomes of this study may fuel the societal debate on specifically gender diversity in the board rooms.

1.4 Structure

The remainder of the paper will be structured as follows. Chapter 2 will discuss the relevant literature and the hypothesis. This section dives into the key topics of the paper: earnings management, gender characteristics and the relevant differences between CEOs and CFOs. In section 3 a description of the research methodology that will be used in the study is provided. The results of the research are presented in chapter 4. Finally, chapter 5 will point out the conclusions and the research limitations.

(7)

7 2. Literature review and hypothesis

2.1 Financial reporting

Most of the public listed companies are owned by investors, for example pension funds, private investors, banks, etc. However, the companies are managed and controlled by managers on behalf of the investors. Agency theory tries to describe the relationship between the owners and the managers using the metaphor of a contract (Jensen and Meckling, 1976). The separation of ownership and control leads to information asymmetry between the owners and the managers of the firm because the managers have an

information advantage over the owners. This results into two specific agency problems. First, adverse selection refers to a situation where one of the parties has an information advantage in a business transaction, which he can exploit at the cost of the other party. This is referred to as ex ante asymmetry because the problem arises before signing the contract. Secondly, moral hazard refers to the problem of inducing agents (managers) to supply proper amounts of productive inputs in fulfillment of the contract while their actions cannot be observed by the principal (owners). This problem arises after signing the contract, hence this is ex post information asymmetry (Holmstrom, 1982; Janssen, 2015). The agency theory is concerned with resolving these agency problems. Which arises when the principle and the agent, both want to maximizes their own benefit (Eisenhardt, 1989).

To resolve the agency problem and to align the interest of the managers with the interest of the shareholders, a mechanism of incentives based payment is put in place. The incentives are based on the outcome of financial reporting. Financial reporting alleviates the problems associated with separation of ownership and control because the periodic

disclosure of information by managers decreases the information asymmetry between the owners and the managers.

According to Easterbrook and Fischel (1984), several public and private responses to the problem of information asymmetry establish powerful incentives for firm to disclose what investors want to know. However, firms may also have incentives to not disclose some information. To control the agency cost, mandatory disclosure specifies the minimum level of information that has to be communicated by corporate insiders (Mahoney, 1995). This does not mean that all the problems are solved. Given that the reported accounting information is based on accrual accounting, which requires the use of judgement and

(8)

8 estimates, the discretion in accounting can be used by managers for personal gain or

opportunistic behavior.

2.2 Earnings management

Earnings management is a problem which arises within the agency theory. As stated before, the players in the agency theory want to maximize their own benefits. Incentives are introduced as an attempt to align the interest of managers with those of shareholders. As long as the respective performance measure is congruent with firm value, the interest of the managers and the firm are perfectly aligned. However, if the targets may not be reached, managers can engage in earnings management and choose accounting estimates to influence contractual outcomes that depend on reported accounting numbers.

Healy and Wahlen (1999) pointed out that stakeholders expect that the annual report closely reflect the economic reality of the organization’s financial position throughout the reporting period. As stakeholders use the reported numbers for their decision making (Graham et al. 2005; Dichev et al. 2013). Earnings management is one of the dimensions of earnings quality. Earnings quality refers to the degree to which reported earnings capture a firm's economic reality.

If earnings are managed the reported numbers do not reflect firm’s economic reality anymore (Healy and Wahlen, 1999). Therefore, a high level of earnings management would imply a low level of earnings quality. This is supported by Dechow and Schrand (2004) who indicates that earnings management clearly deteriorates earnings quality. Healy and Wahlen (1999) comes up with the following definition of earnings management:

“Earnings management occurs when managers use judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers.”

There are several determinants of earnings management. Those can be divided into two different categories. On the one hand, Healy and Wahlen (1999), Dechow and Skinner (2000), Fields et al. (2001) and Graham et al. (2005), provides several incentives for why managers might exercise accounting discretion to achieve some specific earnings goals. A first motivation for managers to engage in earnings management are the bonuses they

(9)

9 receive for reaching a specific target. Secondly, a higher level of earnings would imply a higher stock price and as a lot of managers have shares of their company, they have an incentive to boost stock prices. Thirdly, Bowen et al. (1995) and Burgstahler and Dichev (1997) state that by managing earnings, companies are able to enhance their reputation with stakeholders, such as suppliers, creditors, and customers, and hence get better terms of trade. Another aspect that helps to explain the desire of managers to hit earnings targets are career concerns. According to Graham et al. (2005), managers feel that their inability to hit the earnings target is seen as a ‘‘managerial failure.’’ Finally, bond covenants may also motivate managers to engage in earnings management. Some research proposes that earnings might be managed to reduce the probability of violating a covenant, and hence the expected cost of debt (Watts and Zimmerman, 1990).

On the other hand, there are aspects which influences the opportunity to engage in earnings management. One well known example is the internal control environment. A weak internal control environment increase the possibility of earnings management. The level of audit quality is the other aspect which can make it easier for managers to manage earnings. There are several studies examining the relation between earnings management and audit quality. DeAngelo (1981) provides the following definition of audit quality:

“The quality of audit services is defined to be the market-assessed joint probability

that a given auditor will both (a) discover a breach in the client’s accounting system, and (b) report the breach”

Becker et al. (1998) and Francis et al. (1999) find that discretionary accruals in firms audited by Big Six auditors are less than those in firms audited by non-Big Six auditors. Becker et al. (1998) state that “Also, consistent with earnings management, we find that the mean and median of the absolute value of discretionary accruals are greater for firms with non-Big Six auditors. This result also indicates that lower audit quality is associated with more "accounting flexibility".”

The characteristics of the executives determine how an executive respond to the above discussed incentives and opportunities. This paper examines the relationship between one particular characteristic of top executives and its influence on earnings management, namely gender. There is much literature which indicates that characteristics and incentives of executives affect earnings management (see e.g. Cheng and Warfield,

(10)

10 2005; Davidson et al., 2007; Meek et al., 2007; Jiang et al., 2010; Matsunaga and Yeung, 2008).

2.3 General overview of gender studies

Several studies present findings, with respect to money and finance, about the differences in behavior between men and woman in organizations. Men are more focused on making money and getting ahead in organizations, while women are more concerned with helping others (Bernardi and Arnold, 1997; Betz et al., 1989). According to Barber and Odean (2001), women are expected to be more risk averse, which influence the type of investment decisions they make. Men are assumed to have more confidence with money matters. If women lack confidence with respect to financials and are less focused on

financial results that maximizes their own benefit, this may affect the way men and women are handling financial reporting (Krishnan and Parsons, 2008).

The overall consensus of prior research is that gender differences impact ethical behavior. Betz et al. (1989) mentioned that women are less likely than men to break the laws for their own benefit. Bernardi and Arnold (1997) argue that female executives may have higher moral standards than male executives. On the other hand, Owhoso (2002) indicate that male and female auditors with the same experience level are equally likely to detect fraud risks.

As a result of the differences in behavior between men and women with respect to money and finance, they vary in their success with investing money. This is supported by Barber and Odean (2001), who find that women hold securities in their portfolio for a longer period than men do. As a result, women have higher return on their investments. The difference is attributed to men’s overconfidence in their abilities to trade. In addition, a study of Bliss and Potter (2002) amongst mutual fund managers find that female managers outperform men in domestic funds. In managing international funds men and women perform equally.

Some studies found that gender diversity does not improve firm performance. Watson (2002) found that, after controlling for some variables, there was no difference in firm performance between firms controlled by males or by females. Carter et al. (2003) explored the effect of gender on firm performance. Unlike Watson (2002), they found that

(11)

11 gender diversity is positively related with firm performance. So, prior research of gender diversity on firm performance yields mixed results.

Prior research indicate that there are differences in behavior between men and women because they have different characteristics. Adams and Ferreira (2009) found that female directors exhibit greater diligence in monitoring. Adams, Gray and Nowland (2010) argue that female executives exhibit more independent thinking and improve the

monitoring process. Also, they show that investors value the addition of a female executive in the board because better monitoring and lower information asymmetry facilitate better earnings quality.

A vast body literature indicates that females are more conservative and risk averse than men (Johnson and Powell, 1994; Powell and Ansic, 1997; Jianakoplos and Bernasek, 1998; Sunden and Surette, 1998; Schubert, 2006). According to Srinidhni et al. (2011), prior studies on the attributes of women indicate that they might be less tolerant than men towards opportunistic behavior. Prior literature has shown that women exhibit lower tolerance to opportunism in their decision making (Robinson, Lewicki, and Donahue 2000; Schminke and Ambrose 2002; Ambrose and Schminke 2003; Thorne, Massey, and Magnan 2003; Bernardi and Arnold 1997, Krishnan and Parsons 2008). According to MacLeod Heminway (2007), females are more trustworthy than men and less likely to manipulate financials. This paper is build up on the previous literature of gender characteristics especially in conservatism, risk averseness and ethical behavior which may influence the level of earnings management.

2.4 Top executives and earnings management

In general, CEOs and CFOs are required to personally certify the accuracy and completeness of the financial information released by the company. CFOs have two main responsibilities within companies. On the one hand, as the CEO, they have as member of the board, important decision making responsibilities. They make decisions about taxes,

mergers and acquisitions, planning and budgeting, financing, dividends, etc. (Graham et al. 2005; Matsunaga and Yeung 2008). On the other hand, CFOs play a key role in financial reporting. Unlike CEOs, they are primarily responsible for the preparation and filing of the financial statements (Mian 2001).

(12)

12 Indjejikian and Matejka (2009) point out how CFO bonus plans are designed,

especially how these bonuses are connected to financial performance. They document how firms deemphasize the relation of bonuses with financial performance to motivate CFOs to focus more on their role in financial reporting to fairly represent the firm’s economic performance. So, when deciding about appropriate incentives to award CFOs, companies face a trade-off between their fiduciary responsibilities and their decision-making duties. Bonus incentives tied to accounting measures motivate CFOs to perform their decision-making responsibilities but also implies that CFOs are rewarded on performance measures they generate themselves. On the other hand, lower incentives benefit CFOs’ fiduciary

responsibilities but at the expense of properly incentivizing their decision-making duties.

CFOs are the top executives who have the ultimate responsibility over the preparation of financial statements (Mian, 2001). Prior research indicated that, relative to other executives, CFOs are in a unique position to engage in accounting manipulations (Feng, Ge, Lou, and Shevlin 2011). CFOs are viewed as watchdogs of the financial reporting quality because they monitor the process of preparing financial statements. This implies that they must safeguard the quality of the internal controls to reasonably assure that financial statements are

reliable and free of material mistakes.

Prior research finds that CEO equity incentives are related with accrual management (Bergstresser and Philippon, 2006). Commentators and policymakers have expressed their concerns that CFO equity-based compensation might especially contribute to earnings management. This is supported by Jiang, Petroni and Wang (2010), who found that especially CFO incentives are positively associated with accrual-based earnings

management. In addition, the survey of Graham, Harvey, and Rajgopal (2005) indicated that CFOs concedes that personal incentives embody their financial reporting decisions. The risk-taking incentives of CFOs are stronger related with earnings smoothing though accruals than the incentives of CEOs (Chava and Purnanandam, 2010). Whereas, this indicated that CFOs are motivated to engage in accrual-based earnings management to attain personal financial gain, it has also been argued that CFOs manipulate the financial reporting system due to pressure exerted by CEOs (Friedman 2014). Graham and Harvey (2001) stated that CFOs are the agents of the CEOs. Moreover, a CEO has the power to replace a CFO who does not follow the CEOs preferences (Mian, 2001; Fee and Hadlock, 2004). This is supported by the survey of Dichev, Graham, Harvey, and Rajgopal (2013), who reported that 91 percent of the

(13)

13 CFOs acknowledge that pressure from within the company is a reason to manipulate

economic performance in the financial statements. Given these results it may be the case that CFOs do not respond directly to their own incentives but only to the CEOs’ (Jiang, Petroni and Wang, 2010).

2.5 Hypothesis development

Financial reporting alleviates the problems associated with separation of ownership and control because the periodic disclosure of information by managers decreases the information asymmetry between the owners and the managers. Given that the reported accounting information is based on accrual accounting, which requires the use of judgement and estimates, the discretion in accounting can be used by managers for personal gain.

The determinants of earnings management can be divided into two different categories. Stakeholder motivations, stock price motivation, debt covenants, career concerns and bonuses represent incentives for why managers might exercise accounting discretion to achieve some specific earnings goals. However, prior literature also establish how the style of individual managers influence their reporting decisions. For example, Geiger and North (2006) find how the level of accruals in financial reporting changes when a different CFO takes office.

Specifically, the characteristics of the executives determine how an executive respond to the discussed incentives and opportunities. These characteristic differ

substantially between men and women. Betz et al. (1989) mentioned that women are less likely than men to break the laws for their own benefit. Bernardi and Arnold (1997) argue that female executives may have higher moral standards than male executives. In addition, a vast body literature indicates that females are more conservative and risk averse than men (Johnson and Powell, 1994; Powell and Ansic, 1997; Jianakoplos and Bernasek, 1998; Sunden and Surette, 1998; Schubert, 2006).

These gender differences between men and women especially in conservatism and ethical behavior are expected to affect earnings management. Therefore, I expect female top executives to engage less in earnings management. This leads to the first hypothesis:

H1: “Women in top executive positions will engage less in earnings management than men

(14)

14 According to Mian (2001), CFOs are the executives who have the ultimate

responsibility over the preparation of financial statements. Prior research indicated that, relative to other executives, CFOs are in a unique position to engage in accounting manipulations (Feng, Ge, Lou, and Shevlin 2011). CFOs are viewed as watchdogs of the financial reporting quality because they monitor the process of preparing financial statements. This implies that they must safeguard the quality of the internal controls to reasonably assure that financial statements are reliable and free of material errors. Jiang, Petroni, & Wang (2010) come up with empirically evidence that CFOs have a higher influence on earnings management than CEOs. Friedman (2014) argued that CFOs

manipulate the financial reporting system due to pressure exerted by the CEO. Still, Prior research documents a change in financial reporting when another CFO enters the firm. Geiger and North (2006) document a change in the level of accruals when another CFO takes office. Dichev et al. (2013) find in their survey that larger accruals represent an indication of earnings management. Ge et al. (2010) document how individual CFOs exhibit different choices in terms of meeting benchmarks, the likelihood of accrual misstatements, and so forth.

So, I expect that the effect of gender on earnings management is higher for CFO positions than for CEO positions because CFOs have the ultimate responsibility over the preparation of financial statements and therefore their influence on the financial statement is more pronounced. This yield to the second hypothesis:

(15)

15 3. Research methodology

3.1 Sample Description

The sample of this paper is based on the U.S. market from 2007 to 2013. I start with data from the Standard & Poor´s ExecuComp database, which offers compensation statistics on the five best-paid executives for public U.S. companies. The ExecuComp database

roughly corresponds with the S&P 1500, in this sample there are for firm i in year t two observations (CEO and CFO). This yields an initial sample of 30.782 manager year observations. The variables for the modified jones (1991) model to estimate earnings management and the control variables, requires data from the Compustat database. After calculating the discretionary accruals according to the modified jones (1991) model and merging the ExecuComp sample3_{with Compustat sample,}4_{the sample is reduced to 22.422}

manager year observations. After dropping all observations with missing values the sample is reduced to 21.523 manager year observations. I eliminate firms in the financial (SIC codes 6000-6999), utility (SIC codes 4000-4999) and government (sic code 9900) industry because of their special accounting practices and executives might have different motivations to manage earnings in these heavily regulated industries (Burgstahler & Eames, 2003; Cheng & Warfield 2005). The exclusion of these SIC codes leads to a final sample of 18.332 manager year observations. Additionally, al continuous variables were winsorized at the 1st_{and 99}th

percent of the distribution to address any potential impact of outliers. Table 1 provides an overview of the sample selection process.

Table 1: Sample selection

Number of observations ExecuComp sample 30.782

Less: Observations lost by calculating discretionary 8.360 accruals and merging Less: observations from missing values 899

Less: Observations from specific SIC codes5_3.191

Final sample 18.332

3_{Variable downloaded from ExecuComp are: CEOANN, CFOANN and GENDER.}

4_{Variable downloaded from Compustat are: AT, PPEGT, RECT, LT, CEQ, SALE, NI, OANCF, IBC, MKVALT.}

5_{Observations are dropped from Financial Institutions (SIC codes6000-6999), Utility sectors (SIC codes}

(16)

16

3.2 Empirical models 3.2.1 Main model

Prior studies suggest that the unsigned magnitude of discretionary accruals is a potential “red flag” that companies are engaging in earnings management (Dechow & Schrand, 2004). Whether female top executives are related to less earnings management, is tested by means of the following empirical model:

(1)

EM

it = α0

+

α

₁

D_Fem

it

+ Controls + ε

it

Where EMit is the absolute value of the estimation residuals from Eq. (3), tested by the

modified jones (1991) model. D_Femit is a dummy variable which equals 1 for observations

with female top executives (CEO or CFO) and 0 for observations with male top executives. The coefficient α₁ gives the relationship between female executives and the unsigned discretionary accruals. A significantly negative α₁ indicates that female top executives are less likely to engage in earnings management. The control variables, which are included in the model, are described in section 3.3, measurement of control variables.

To examine the second hypothesis a second empirical model is required. The

following model is implemented to measure the difference in the gender effect for CEO and CFO positions:

(2)

EM

it = β0 + β1

D_Fem

it + β2

D_CFO

it + β3

D_Fem* D_CFO

it

+ Controls + ε

it

Where D_CFOit is a dummy variable that takes the value of 1 for CFO observations and takes

the value 0 for CEO observations. The interpretation of model 2 is summarized in table 2. The effect from male CEOs on earnings management is captured by β0. The coefficients β0 +

β1 provides the relationship between female CEOs and earnings management. β0 + β2

represent the relation between male CFOs and earnings management. The effect from female CFOs on earnings management is given by the coefficients β0 + β1 + β2 + β3. It is

expected by the second hypothesis that the gender effect is stronger for CFO positions than for CEO positions. As stated before, a negative value of a coefficient means a lower level of earnings management. Therefore, the gender effect for CFOs ((β0 + β1 + β2 + β3) – (β0 + β2))is

(17)

17 β3 should be smaller than zero. Hence, ((β0 + β1 + β2 + β3) – (β0 + β2)) < (β0 + β1 - β0) => β1 + β3

< β1 => β3 < 0. Table 2: Interactions model 2 CEO CFO Male β0 β0 + β2 Female β0 + β1 β0 + β1 + β2 + β3 3.2.2 Auxiliary model

Consistent with prior research, the modified Jones (1991) model is used to detect discretionary accruals. Dechow et al. (1995) and Guay et al. (1996) compared the Jones and modified Jones models with other models and conclude that the Jones models perform the best in detecting abnormal accruals. Jones (1991) is a model that relaxes the assumption that nondiscretionary accruals are constant. The model attempts to control for the effect of a firms’ economic circumstances on nondiscretionary accruals. The modified Jones (1991) model for total accruals is:

(3)

TA

it = λ0 + λ1(1/

A

t-1) + λ2(Δ

REV

it – Δ

REC

it) + λ3(

PPE

it) +

ε

it

Where

TAit = EARNit – CFOit;

EARNit = earnings before extraordinary items for firm i in year t scaled by total assets

at t-1;

CFOit = cash flow from operations for firm i in year t scaled by total assets at t-1;

At-1 = total assets at t-1;

ΔREVt = revenues in year t less the revenues in year t-1 scaled by total assets at t-1;

ΔRECt = net receivables in year t less the net receivables in year t-1 scaled by total

assets at t-1; and

PPEt = gross property plant and equipment in year t scaled by total assets at t-1.

The changes in revenues minus receivables and PPE are used to control for expected (i.e., economic-based) components in total accruals. The nondiscretionary accruals are captured in the first part of the model and the discretionary accruals are captured in the residuals. Similar to prior literature, the absolute value of residuals is used as a proxy for earnings

(18)

18 management (Dechow & Schrand, 2004). Larger absolute value of residuals indicates a higher level of earnings management.

3.3 Measurement of control variables

Several control variables will be included in the empirical models to control for differences that are associated with accruals. Watts and Zimmerman (1990) has suggested a negative relation between the level of earnings management and company size. As larger firms are under more scrutiny of the outside world, there are less opportunities for executives to engage in earnings management. To control for the size of a company the natural logarithm of total assets (Ln(ASSET)it) will be included. According to Smith and Watts

(1992), growth options increase monitoring difficulties which creates more opportunities for executives to engage in earnings management. The market to book ratio, which is defined as the market value of equity divided by the book value of equity, will be added to control for growth (GROWTHit). Third, to control for differences in performance the yearly return on

assets (ROAit) will be included (Dechow et al. 1995). Total debt divided by total assets is

include as the fourth control variable to control for leverage (LEVERAGEit), as financial

leverage is positively associated with earnings management (Dechow et al. 1995). Finally, industry and year control variable dummies are included to improve the robustness of the findings.

(19)

19 4. Results

4.1 Descriptive statistics

Table 3 shows the percentage of women representation in top management

positions during the years 2007-2013. Women represent on average 3.3 percent of the CEO positions, which is higher than found by prior research in the period before 2006 (Mohan and Ruggiero, 2007; Wolfers, 2006).6_{This increase is acknowledged by table 3 as it indicates}

an increase in female CEOs from three towards four percent over the sample years. The increase in female executives during the years emphasizes the importance of controlling for year effect in the models. The descriptive statistics also reveal that women occupy about 8.6 percent of the CFO positions. 7

Table 3:

Top executive gender across years

Year Observations Firms with female

CEO # (%)

Firms with female CFO # (%) 2007 2822 40 (2.84) 119 (8.43) 2008 2759 41 (2.97) 114 (8.27) 2009 2727 42 (3.08) 112 (8.22) 2010 2610 46 (3.52) 109 (8.36) 2011 2544 43 (3.37) 112 (8.83) 2012 2482 46 (3.70) 116 (9.35) 2013 2388 48 (4.01) 109 (9.14) Total 18332 306 (3.33) 791 (8.64)

Table 4 shows the percentage of female top executives across industries during the sample period, based on a classification by two-digit SIC codes. There are more female CEOs and CFOs in the wholesale and retail industries and less in the mining, construction, industry and manufacturing sectors. The gender differences in top executives across industries show that it is important to control for industry effect in the analysis.

6_{Mohan and Ruggiero (2007) suggest that less than 1% of the Fortune 1000 firms have female chief executives}

in 2002. Wolfers (2006) indicate that 1.3% of the S&P 1500 companies have female CEOs over the period 1992-2004.

7_{Firms in the U.S. have a relative low percentage of women in top executive positions. Ye et al. (2010)}

documents that 4% percent of the Chinese firms have Female CEOs and that 25% of the companies have female CFOs during the period 2001-2006.

(20)

20 Table 4:

Top executive gender across industries Two digit

Sic code

Industry Observations Firms with

female CEO # (%)

Firms with female CFO # (%) 00-09 Agriculture, forestry & fishing 52 0 (-) 0 (-)

10-19 Mining & construction 1420 0 (-) 45 (6.33)

20-39 Industry and manufacturing 10397 154 (2.96) 409 (7.87) 50-59 Wholesale & retail trade 2627 89 (6.78) 151 (11.5)

70-89 Services 3836 63 (3.28) 186 (9.72)

Total 18332 306 (3.34) 791 (8.64)

Table 5 reports descriptive statistics for earnings management and the control variables for firms with male and female CEOs and CFOs. On average, the firms in the sample exhibit growth options, the ROA is about 4% and firms finance about half of their assets through debt. Table 5 also tests for differences in means between males and females. There is a significant difference in the level of earnings management between male and female CFOs. The difference indicates that when a company has a female CFO the company has lower discretionary accruals compared to firms with a male CFO. In percentage it suggests that companies with a women as CFO has 7.70% (0.006/0.078) lower discretionary accruals than firms with a male CFO. Shortly, this suggests that a female as CFO has a decreasing effect on the level of earnings management.

Table 5:

Descriptive statistics: Means

Variables Total Gender of CEO Gender of CFO

Male Female Difference Male Female Difference

EM 0.077 0.077 0.080 -0.003 0.078 0.072 0.006**

Ln(Assets) 7.427 7.426 7.474 -0.048 7.427 7.433 -0.006

GROWTH 3.147 3.146 3.177 -0.031 3.229 2.299 0.93

ROA 0.036 0.036 0.032 0.004 0.035 0.054 -0.019***

LEVERAGE 0.510 0.509 0.554 -0.045*** 0.509 0.526 -0.017**

Note: EMit is the absolute value of the estimation residuals from Eq. (3), the proxy for earnings management.

Ln(Assets)it is the natural logarithm of total assets; GROWTHit is the market value of equity divided by the book

value of equity; ROAit is the firms return on assets; LEVERAGEit is the total debt divided by the total assets. All

(21)

21 Table 6 provides Spearman correlation coefficients between the main variables. EMit

is negatively correlated with Ln(Assets)it, GROWTHit and ROAit. As firms are larger they

receive more attention. Therefore, it becomes more difficult to engage in earnings

management. On the other hand, if companies are growing and have a high return on assets ratio, there is less need for managers to engage in earnings management because the incentives to engage in earnings management (e.g., meeting benchmarks) are lower. Leverage is positively correlated with accrual-based earnings management. Finally, the magnitude of the correlations provide no concerns about multicollinearity.

Table 6:

Spearman correlation matrix

EM Ln(Assets) GROWTH ROA LEVERAGE

EM 1.0000

Ln(Assets) -0.1706* 1.0000

GROWTH -0.0679* 0.0767* 1.0000

ROA -0.2693* 0.0948* 0.4922* 1.0000

LEVERAGE 0.0172* 0.4111* 0.0415* -0.2184* 1.0000

(*) Correlations are significant at the 5% significance level.

4.2 Main analysis

4.2.1 Female top executives and earnings management

Table 7 provides data on regressions with earnings management as dependent variable and the dummy female as independent variable. The table reports for different models estimated by Ordinary Least Squares (OLS) where I do not include control variables, include control variables, additionally control for year-effects, control for year- and industry-effects, respectively. The fifth and final model also includes the regular control variables and control for year- and industry-effects, but now the empirical model is estimated by robust regression instead of OLS. Robust regression are less sensitive to outliers as it eliminates observations with cooks D > 1 and subsequently puts less weight on those observations that would otherwise strongly influence the final regressions results.

The variable of interest is D_Femit. The findings are not consistent with my first

hypothesis. That is, the coefficient on D_Femit is not significant in all the regressions. So, on

(22)

22 Table 7:

Regression results Eq. (1)

EM (1) (2) (3) (4) (5) intercept 0.082 (90.27***) 0.13 (48.94***) 0.12 (39.52***) 0.09 (9.09***) 0.076 (11.66***) D_Fem -0.004 (-1.29) -0.003 (-1.44) -0.003 (-1.29) -0.002 (-0.87) -0.001 (-0.38) ROA -0.142 (-45.0***) -0.138 (-43.47***) -0.135 (-43.01***) -0.24 (-95.45***) Ln(Assets) -0.009 (-25.19***) -0.009 (-24.34***) -0.010 (-27.10***) -0.004 (-15.70***) LEVERAGE 0.043 (86.48***) 0.043 (87.80***) 0.044 (89.47*** 0.002 (5.79***) GROWTH -0.000 (-1.72*) -0.000 (-1.47) -0.000 (-1.08) 0.000 (1.39)

Year effects NO NO YES YES YES

Industry effects NO NO NO YES YES

Model significance 0,20 0.000 0.000 0.000 0.000

Table 7 reports regression results from EMit on D_Femit. Column one to four are estimated by OLS. Column 5 is

estimated by robust regression. Test-statistics are reported in parentheses. The coefficients that are significant on the 0.10, 0.05 and 0.01 level are denoted with *, ** and *** respectively. The panel data set consist of 18332 S&P 1500 manager year observations over the period 2007-2013.

With regard to the control variables, the results consistently show that firm size and accounting performance are negatively associated with accrual-based earnings

management, while leverage is positively associated with earnings management. The latter two findings indicate that when performance decreases and leverage increases, the

motivations to engage in earnings management become greater. Finally, most models are significant. In sum, I conclude that there is insufficient evidence to accept the hypothesis that women in top executive positions will engage less in earnings management than men in top executive positions.

(23)

23

4.2.2 Gender effect for CEOs and CFOs

The data on regressions of the second equation, with earnings management as dependent variable and the female dummy, CFO dummy and the female-CFO dummy as independent variables, are provided in table 8. The second hypothesis tests whether the effect of gender on earnings management is higher for CFO positions than for CEO positions. The variable of interest in the second equation is D_Fem*D_CFOit. In order for the second hypothesis to not be rejected the coefficient on β3 needs to be significantly smaller than zero.

Table 8: Regression results Eq. (2)

EM (1) (2) (3) (4) (5) intercept 0.082 (64.49***) 0.13 (47.77***) 0.12 (38.20***) 0.093 (9.05***) 0.076 (11.64***) D_Fem 0.001 (0.23) 0.000 (0.05) 0.001 (0.21) 0.004 (0.86) 0.000 (0.08) D_CFO 0.001 (0.37) 0.001 (0.67) 0.001 (0.67) 0.001 (0.71) 0.000 (0.16) D_Fem*D_CFO -0.009 (-1.10) -0.005 (-1.01) -0.005 (-1.10) -0.008 (-1.60) -0.001 (-0.33) ROA -0.143 (-44.98***) -0.138 (-43.46***) -0.135 (-42.98***) -0.236 (-95.43***) Ln(Assets) -0.009 (-25.20***) -0.009 (-24.35***) -0.010 (-27.11***) -0.004 (-15.70***) LEVERAGE 0.043 (86.48***) 0.043 (87.81***) 0.044 (89.49***) 0.002 (5.76***) GROWTH -0.000 (-1.72*) -0.000 (-1.47) -0.000 (-1.09) 0.000 (1.39)

Year effects NO NO YES YES YES

Industry effects NO NO NO YES YES

Model significance 0,40 0.000 0.000 0.000 0.000

Table 8 reports regression results from EMit on D_Femit, D_CFOit and D_CFO*D_Femit. Column one to four are

estimated by OLS. Column 5 is estimated by robust regression. Test-statistics are reported in parentheses. The coefficients that are significant on the 0.10, 0.05 and 0.01 level are denoted with *, ** and *** respectively. The panel data set consist of 18332 S&P 1500 manager year observations over the period 2007-2013.

(24)

24 The findings are not consistent with the second hypothesis. That is, the coefficient on D_Fem*D_CFOit is not significant smaller than zero in all the regressions. Therefore, I will

reject the second hypothesis. With respect to the control variables, the results consistently indicate that firm size and accounting performance are negatively associated with earnings management, while leverage is positively associated with earnings management. These two results show that when size and performance decreases and leverage increases, the

motivations to engage in accrual-based earnings management become greater. Finally, most models are significant. Altogether, I conclude that there is insufficient evidence to accept the second hypothesis that the effect of gender on earnings management is higher for CFO positions that for CEO positions.

(25)

25 5. Conclusion

This study brings the research on gender an earnings management together and examines whether women in top executive positions engage less in earnings management than men in top executive positions. Based on prior research it is argued that women have different characteristics, especially in conservatism and ethical behavior. Due to these gender differences, women are expected to be less sensitive to the incentives for engaging in earnings management. The first findings indicate that there is a significant difference found in the level of earnings management between male and female CFOs. However the results of the first regression suggest that the results in this paper are consistent with the projections of prior literature. Using a sample of 18332 manager year observations over a period from 2007 to 2013 there is no evidence found that women in top executive positions engage less in earnings management than men in top executive positions. This fits into the widely dispersed prior literature.

This paper answer calls from prior literature for research regarding the differences in gender effect between top executives. CFOs are the top executives who have the ultimate responsibility over the preparation of financial statements. Therefore, their influence on the financial statement is more pronounced and are they in a unique position to engage in accounting manipulations. So, the gender effect is expected to be stronger for CFO positions than for CEO positions. As far as known to the author, this study is first to examine whether the gender effect is stronger for CFO positions than for CEO positions. Thereby, extending prior research on diversification among top executives. Despite of the first findings, which indicate that there is a significant difference in the level of earnings management between male and female CFOs, there is no evidence found supporting that the gender effect is stronger for CFO positions than for CEO positions.

The results of this study are subjected to the following limitation. The data for the regressions are obtained for large publicly listed American firms (S&P 1500). On that account, generalization of findings is limited, for small and mid-sized companies, none American firms and privately held firms. It is up to further research to examine whether the results are the same in different settings. Other recommendations for further research are to investigate to what extent other heterogeneity characteristics of top executives, such as functional background, age, tenure and education affect earnings management.

(26)

26 References

Adams, R. B., & Ferreira, D. (2009). Women in the boardroom and their impact on governance and performance. Journal of Financial Economics, 94, 291–309.

Adams, R., S. Gray, and J. Nowland. (2010). Is there a business case for female directors? Evidence from the market reaction to all new director appointments. Working paper, City University of Hong Kong

Ambrose, M. L., & Schminke, M. (2003). Organization structure as a moderator of the relationship between procedural justice, interactional justice, perceived

organizational support, and supervisory trust. Journal of Applied Psychology, 88, 295–305.

Akaah, I. P. (1989). Differences in research ethics judgments between male and female marketing professionals. Journal of Business Ethics, 8(5), 375−381.

Barber, B. M. and T. Odean: (2001), 'Boys WiU Be Boys: Gender, Overconfidence, and Common Stock Investment', The Quarterly Journal of Economics 116(1), 261-292.

Becker, C.L., DeFond, M.L., Jiambalvo, J., Subramanyam, K.R., (1998). The effect of audit quality on earnings management. Contemporary Accounting Research 15, 1–24.

Beltramini, R. F., Peterson, R. A., & Kozmetsky, G. (1984). Concerns of college students regarding business ethics. Journal of Business Ethics, 3, 195−200.

Bergstress Der. and Philippon, T. (2006). CEO incentives and earnings management. Journal

of financial economics, 8: 511-529.

Bernardi, R. A. and D. F. Arnold: (1997). 'An Examina tion of Moral Development within Public Accounting by Gender, Staff Level, and Firm', Contemporary Accounting

Research, 14(4): 653-668.

Betz, M., O'Connell, L., & Shepard, J. M. (1989). Gender differences in proclivity for unethical behavior. Journal of Business Ethics, 8(5), 321−324.

Bliss, R. T. and M. E. Potter. (2002), 'Mutual Fund Managers: Does Gender Matter?', Journal

(27)

27 Bowen, R., DuCharme, L., Shores, D., (1995). Stakeholders' implicit claims and accounting

method choice. Journal of Accounting and Economics 20 (3), 255-295.

Chava, S., Purnanandam, A., (2010) .CEOs vs. CFOs: incentives and corporate policies.

Journal of Financial Economics, 97, 263–278.

Beaver, W. and E. Engel: 1996, 'Discretionary Behavior with Respect to Allowances for Loan Losses and the Behavior of Security Prices', Journal of Accounting and Economics 22(1-3), 177-206.

Bruns, W. J., & Merchant, K. A. (1990). The dangerous morality of managing earnings. Management Accounting, 72(2), 22−25.

Burgstahler, D., Dichev, I., 1997. Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics 24, 99–126.

Burgstahler, D., Eames, M., (2003). Earnings management to avoid losses and earnings decrease: are analysts fooled? Contemporary Accounting Research, 20, 253–294.

Carter, D., Simkins, B. and Simpson, W. (2003), ‘‘Corporate governance, board diversity, and firm value’’, The Financial Review, Vol. 38, pp. 33-53.

Cheng, Q. and Warfield, T. (2005), ‘‘Equity incentives and earnings management’’, The Accounting Review, Vol. 80, pp. 441-76.

Davidson, W., Xie, B., Xu, W. and Ning, Y. (2007), ‘‘The influence of executive age, career horizon and incentives on pre-turnover earnings management’’, Journal of Management & Governance, Vol. 11, pp. 45-60.

DeAngelo, L. H. Auditor Size and Audit Quality. Journal of Accounting and Economics. (December 1981), 183-199

Dechow P. M., R. Sloan, and A. P. Sweeney. Detecting Eamings Management. The Accounting Review 70 (April 1995), 193-225.

Dechow, P. M., & Schrand, C. M. (2004). Earnings Quality. : Research Foundation of CFA Institute.

(28)

28 Dechow, P., Skinner, D., 2000. Earnings management: reconciling the views of accounting

academics, practitioners, and regulators. Accounting Horizons 14 (2), 235–250.

Dichev, I. D., J. R. Graham; C. R. Harvey; & S. Rajgopal. (2013). Earnings Quality: Evidence from the Field. Journal of Accounting and Economics, 56, 1–33.

Easterbrook, F. H., & Fischel, D. R. (1984). Fifty Years of Federal Securities Regulation: Symposium on Contemporary Problems in Securities Regulation. Virginia Law

Review, 70(4), 669-715.

Eisenhardt, K. M. (1989). Agency theory: An assessment and review, Academy of

Management Review, 14, 57-74.

Fee, C., Hadlock, C. (2004). Management turnover across the corporate hierarchy. Journal of

Accounting and Economics, 37, 3–38.

Feng, M., Ge, W., Luo, S., and Shevlin, T. (2011). Why do CFOs become involved in material accounting manipulations? Journal of Accounting and Economics, 51(1), 21–36.

Ferrell, O. C., & Skinner, S. J. (1988). Ethical behavior and bureaucratic structure in market research organizations. Journal of Marketing Research, 25(1), 103−109.

Fields, T., Lys, T., Vincent, L. (2001). Empirical research on accounting choice. Journal of

Accounting & Economics, 31, 255–307.

Friedman, H. (2014). Implications of power: When the CEO can pressure the CFO to bias reports. Journal of Accounting and Economics, 58(1), 117-141.

Francis, J., E. L. Maydew, and H. C. Sparks. (1999). The role of Big 6 auditors in the credible reporting of accruals. Auditing: A Journal of Practice & Theory, 18: 17-34.

Ge, W., Matsumoto, D., Zhang, J.L. (2010). Do CFOs have styles? An empirical investigation of the effect of individual CFOs on accounting practices. Working paper, University of

Washington and UBC.

Geiger, M., North, D. (2006). “Does hiring a new CFO change things?” An investigation of changes in discretionary accruals. The Accounting Review, 81, 781–809.

(29)

29 Graham, J., Harvey, C. (2001). The theory and practice of corporate finance: evidence from

the field. Journal of Financial Economics, 60, 187–243.

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

Guay, W. A., Kothari, S. P., and Watts, R. L. (1996). A market-based evaluation of discretionary accrual models. Journal of Accounting Research, 34, 83-105.

Hall, S. C. and W. W. Stammerjohan. (1997). Damage Awards and Earnings Management in the Oil Indus try. The Accounting Review, 72(1), 47-65.

Healy, P. M. (1985). The Effect of Bonus Schemes on Accounting Decisions. Journal of

Accounting and Economics. 7, 85-107.

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

Holmstrom, B. (1982). Moral hazard in teams. The bell journal of economics. 13, 324-340

Indjejikian, R., Matejka, M. (2009). CFO Fiduciary Responsibilities and Annual Bonus Incentives. Journal of Accounting Research. 47(4), 1061–1093.

Janssen, W. (2015). What are the economic roles for financial reporting? [PowerPoint slides]. Retrieved from http://blackboard.uva.nl/bbcswebdav/pid-5253730-dt content-rid-6886772_1/courses/2414E001.6314M0038Y.S12.1.2014/Seminar1_

small.pdf

Jensen, M., & Meckling, W. (1976). Theory of the firm: Man-agerial behavior, agency costs, and ownership structure. Journal of Financial Economics, 3, 305-360.

Jianakoplos, N. and Bernasek, A. (1998), ‘‘Are women more risk averse?’’, Economic Inquiry, 36, 620-30.

Jiang, J., Petroni, K., Wang, I. (2010). CFOs and CEOs: who have the most influence on earnings management? Journal of Financial Economics, 96, 513–526

Johnson, J. and Powell, P. (1994), ‘‘Decision making, risk and gender: are managers different?’’ British Journal of Management, 5, 123-38.

(30)

30 Jones, J. J. (1991). Earnings Management during Import Relief Investigations, Journal of

Accounting Research, 29, 193-228.

KPMG: (2009). ‘CSR and Diversity Threshold Identified’, February 24, Available at kpmg.com/KPMG UK News=.mht.

Krishnan, G. V., & Parsons, L. M. (2008). Getting to the bottom line: an exploration of gender and earnings quality. Journal of Business Ethics, 78, 65−76.

MacLeod Heminway, J. (2007), Sex, trust, and corporate boards. Hastings Women’s Law

Journal, 18, 173-93.

Mahoney, P. G. (1995). Mandatory disclosure as a solution to agency problems. The

University of Chicago Law Review, 6(2). 1047-1112.

Matsunaga, S.R. and Yeung, P.E. (2008). Evidence on the impact of a CEO’s financial experience on quality of the firm’s financial reports and disclosures. Proceedings of

the 2008 AAA Financial Accounting and Reporting Section Confe`rence.

Meek, G., Rao, R. and Skousen, C. (2007). Evidence on factors affecting the relationship between CEO stock option compensation and earnings management, Review of

Accounting and Finance, 6, 304-23.

Mian, S., (2001). On the choice and replacement of chief financial officers. Journal of

Financial Economics, 60, 143–175.

Mohan, N., & Ruggiero, J. (2007). Influence of firm performance and gender on CEO compensation. Applied Economics, 39, 1107−1113.

Nguyen, N. T., Basuray, M. T., Smith, W. P., Kopka, D., & McCulloh, D. (2008). Moral issues and gender differences in ethical judgment using Reidenbach and Robin's (1990) multidimensional ethics scale: implications in teaching of business ethics. Journal of

Business Ethics, 77, 417−430.

Owhoso, V. (2002). Mitigating Gender-specific Superior Ethical Sensitivity When Assessing Likelihood of Fraud Risk. Journal of Managerial Issues, 14(3), 360-374.

(31)

31 Powell, M. and Ansic, D. (1997). Gender differences in risk behaviour in financial

decisionmaking: an experimental analysis. Journal of Economic Psychology, 18, 605 628.

Robinson, R., Lewicki, R. J., & Donahue, E. (2000). Extending and testing a five factor model of ethical and unethical bargaining tactics: Introducing the SINS scale. Journal of

Organizational Behavior, 21, 649-664.

Ruegger, D., & King, E. W. (1992). A study of the effect of age and gender upon student business ethics. Journal of Business Ethics, 11(3), 179−186.

Shawver, T. J., Bancroft, P. C., & Senneti, J. (2006). Can the ‘clan effect’ reduce the gender sensitivity to fraud? The case of the IPO environment. Journal of Forensic

Accounting, 7, 185−208.

Schminke, M., Ambrose, M. L., & Cropanzano, R. (2000). The effect of organizational structure on perceptions of procedural fairness. Journal of Applied Psychology, 85, 294–304.

Schubert, R. (2006). Analyzing and managing risks – on the importance of gender difference in risk attitudes. Managerial Finance, 32, 706-15.

Smith, Clifford W., Jr. and Ross L. Watts. (1992). The investment opportunity set and corporate financing, dividend, and compensation policies. Journal of Financial

Economics, 32, 263-292.

Srinidhi, B., Gul, F. A., Tsui, J. (2011). Female directors and earnings quality. Contemporary

Accounting Research in press.

Sunden, A. and Surette, B. (1998). Gender differences in the allocation of assets in retirement savings plans. American Economic Review, 88, 207-11.

Thorne, L., D. W. Massey, and M. Magnan. (2003). Institutional context and auditors’ moral reason-ing: A Canada-U.S. comparison. Journal of Business Ethics, 43(4), 305–21.

Watson, J., (2002). Comparing the performance of male- and female-controlled businesses: relating outputs to inputs. Entrepreneurship: Theory and Practice, 26(3), 91–100.

(32)

32 Watts, R., Zimmerman, J. (1990). Positive accounting theory: a ten year perspective. The

Accounting Review, 65, 131–156.

Wolfers, J. (2006). Diagnosing discrimination: Stock returns and CEO gender. Journal of the

European Economic Association, 4(2–3), 531−541.

Ye, K., Zhang, R., Rezaee, Z. (2010). Does top executive gender diversity affect earnings quality? A large sample analysis of Chinese listed firms. Advances in Accounting,

(33)

33 Appendix

Variable definitions

Variable Definition

EM The absolute value of the discretionary accruals estimated by the modified jones (1991) model

D_Fem A dummy that equals 1 if the CEO or CFO is a women D_CFO A dummy that equals 1 if the observation is a CFO

D_Fem*D_CFO A dummy that equals 1 if the observations is a female CFO ROA Return on Assets, NI – Net Income divided by AT – Total Assets Ln(Assets) The natural logarithm of AT – Total Assets

LEVERAGE LT – Total liabilities divided by AT – Total Assets

GROWTH MKVALT – Market Value of Total Equity divided by CEQ – Common/Ordinary Equity

The effect of gender from top executives on earnings management

Master Thesis: