The financial crisis, earnings management and CEO compensation

(1)

Accountancy & Control

The financial crisis, earnings

management and CEO

compensation

Student: Afram Kara Student number: 10867864 Student number: 10867864

Date: 22-06-2015

Education: MSc Accountancy and Control, variant Control Supervisor: P. Kroos

Word count: 11019

Amsterdam Business School

(2)

2

Statement of originality

This document is written by student Afram Kara 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 paper examines the relationship between CEOs incentive compensation and earnings management, and how the financial crisis moderates this relationship. I

hypothesize that managers engage more in earnings management to increase their total compensation, if there is more attention to incentive compensation in their contracts. Furthermore, due to big bath accounting I hypothesize that the financial crisis

moderates this relationship in a positive way. Using compensation and accruals data over the 2004-2010 time period, I document that managers with less incentive

compensation engage more in earnings management. This finding is consistent with the argument that managers have the incentives to time the release of good and bad news. Moreover, I also do not find evidence that the financial crisis influences the relationship in any way which is in contradiction with the predictions. Collectively, this study doesn’t provide evidence that incentive compensation motivates managers to manage their earnings and this doesn’t differ during the financial crisis.

Key words: financial crisis; bonus incentives; equity incentive;, incentive compensation;

earnings management; accrual;, meeting or beating analysts’ forecasts.

Data availability: data used in this study are available from the sources identified in the

(4)

4

1. Introduction

1.1 Background

In public companies there is a separation between ownership and control. Because of this separation of ownership and control, there is a potential conflict of interest between executives who manage the firm on a daily basis and the owners (shareholders) of the firm. Managers will have personal interests, which they could find more important than the interests of the shareholders. It could have bad consequences for the rest of the organization, when managers will pay more attention to their personal interests instead of the interests of the shareholders. Examples of these consequences are short-termism

due to a limited employment horizon and accepting excessively risky projects. To deal with these conflicts and to align the interest of shareholders and

managers, organizations make use of incentive compensation. This compensation could be in form of base salary, bonus and equity-based incentives. As managers own more shares, they are more likely to act according the interests of the shareholders (Cheng & Warfield, 2005). Incentive compensation becomes more and more important for

companies. This is reflected by comparing compensation of the managers over different periods. For example, Murphy (1999) shows that the total realized pay, salary and bonus increased for the years 1976 to 1996. Also the median exposure of CEO wealth to firm stock prices tripled between 1980 and 1994, and doubled again between 1994 and 2000 (Hall & Liebman, 1998). This indicates that companies find it more important to align the interests of the shareholders and managers.

However, the use of (equity-based) incentive compensation may also represent an incentive for managers to misrepresent the reported performance of the company. This misrepresentation of a company’s performance is called earnings management. Bergstresser and Philippon (2006) point out that large option packages increase the incentives for managers to manipulate their firms’ reported earnings. In this way the reported accounts performance differs with what is warranted by the economic reality. The misrepresentation of performance could influence the total compensation of the managers in a positive way. This is reflected by Healy (1985), who states that managers have an economic incentive to influence earnings in order to increase their cash

compensation. In his paper, Healy (1985) presents evidence that the accruals policies of managers are related to the nonlinear incentives inherent in their bonus contracts. Also Burgstahler and Dichev (1997) found evidence that firms manage earnings to avoid earnings decreases and losses. In both studies, the reported performances of companies

differ with the economic reality of the financial position of these companies. The main reason for earnings management is personal interest, as a result of the

compensation of CEOs (Bauman & Shaw, 2006; Bergstresser & Philippon, 2006). Besides that managers directly want to influence their own compensation, Graham et al. (2005) distinguishes alternative reasons for executives to manage earnings numbers. The most important reason to manage earnings is meeting or beating benchmarks. On the overall, managers have four reasons to meet or beat benchmarks: (i) build credibility with the capital market; (ii) maintain or increase stock price; (iii) improve the external

(6)

6 reputation of the management team; and (iv) convey future growth prospects (Graham et al., 2005). Research illustrates that some managers are prepared to participate in value-destroying actions (Graham et al. 2005) or accrual management (Barua et al.

2006) in order to meet/beat earnings benchmarks. However, meeting or beating benchmarks is more difficult during the financial

crisis because of the distressed environment. A financial crisis is a disruption to financial markets in which adverse selection and moral hazard problems become much worse, so that financial markets are unable to efficiently channel funds to those who have the most productive investment opportunities (Mishkin, 1992). The financial crisis is

characterized by a decline in GDP, reduction of industrial output, less spending from the common public as well as lack of liquidity for companies and individuals (Iatridis & Dimitras, 2013). Due to this environment demand and supply is lower in this period. The combination of less spending by customers and greater difficulties to raise capital by loans (Ivashina & Sharfstein, 2010), will make it more difficult for managers to accept projects and invest in new opportunities. Another characteristic of the financial crisis is that there are many companies which are in financial distress. If a firm is in financial distress, the managers of these companies can expect to have their bonuses cut, be replaced and suffer loss of reputation (Liberty & Zimmerman, 1986; Gilson, 1989). You expect than managers will manage earnings to compensate their loss of compensation in this way. However, it is unknown how the earnings management behavior of managers will react to these events when their company and also other companies are in a

financial distress, during the financial crisis of 2007.

1.2 Research question

The focus of this research will be on the effect of the financial crisis on the relationship between CEO compensation and earnings management. To study this area, the following research question will be used:

How does the financial crisis influence the relationship between CEO incentive compensation and earnings management?

1.3 Relevance of the question

Much literature has focused on the explanation of earnings management (Graham et al., 2005; Healy & Wahlen, 1999). Less research has specifically focused on the relation between CEO compensation and earnings management. This will be the focus of my thesis. Furthermore, prior studies predominantly looked at the uniform relationship between CEO compensation and earnings management. This study contributes by examining how the financial crisis has influenced the relation between CEO

compensation and earnings management. This question is especially interesting as the impact of the financial crisis is not a priori clear. For example, CEOs can attribute the poor performance of their firm to the so called uncontrollable, external factors for which

(7)

7 they cannot be held accountable. Lastly, this thesis also features societal relevance. Understanding earnings management behavior and how this varies under circumstances (e.g., during a financial crisis) is relevant as this enables capital market participants to anticipate earnings management behavior and correct for this in their decision making. This study is also relevant given the recent debate about the compensation of top-managers. There are people who argue that is not ethical to give high (incentive) compensations to managers, while there are a lot of people who live below the poverty line. According to these people it is not ethical to give such high compensation to these managers, while there is a bigger group of people who need that money more. This research could contribute to this ethical discussion. This study will show if this critique on the size of compensation has changed by these comments over time.

1.4 Structure of the thesis

The remainder of this paper is organized as follow. The next section will explain the available literature about compensation, earnings management and the financial crisis. Also the relationship between these variables will be explained in this section. Section 3 discusses the data used for this research and the empirical approach. Section 4 shows the empirical results of this research. The last section concludes with a summary and discusses directions for future research.

(8)

8

2. Literature review

2.1 Agency problems and compensation

In a public company you have a separation of ownership and control. Public companies are characterized by that the ownership of the company is in the hands of the

shareholders and the control of the company is in the hands of professional managers. A problem which arises from this separation is called the agency problem. The agency relationship is defined as a contract under which one or more persons (the

principal/shareholder) engage another person (the agent/manager) to perform some service on their behalf which involves delegating some decision making authority to the agent (Jensen & Meckling, 1979). If the managers focus on maximizing utility, then you could expect that the managers will not always act in the best of the shareholders. The separation of ownership and control will also cause a difference in the degree of information between the shareholders and managers. This could occur because the manager gets more information from business operations than the shareholders. This is known as information asymmetry. The manager has some information about a subject which could be important for the shareholders. However, the shareholders do not receive this information. This asymmetry of information could create two problems: moral hazard and adverse selection.

The problem of moral hazard may arise when managers participate in sharing the risk, with the condition that the actions taken by the managers will affect the outcome. Optimal risk sharing will not take place, because the manager does not have the right incentives to make the right decisions. That is why often a second-best decision will be made, which trades off some of the risk-sharing benefits for provision of incentives (Hölmstrom, 1979). Because the actions of the managers cannot be observed and contracted upon, this will result in a second-best solution characterized by some residual loss (Jensen & Meckling, 1976). This situation often occurs in insurance, labor contracting and in the delegation of responsibility of decision-making (Hölmstrom, 1979). The adverse selection problem is explained by Akerlof (1970). The adverse selection is about information asymmetry, whereby the manager knows more about a subject (e.g., the current situation and future prospects of the firm) than the

shareholders. This manager retains this information for himself, because this

information could produce value for the manager on the long-term. By not sharing this private information with others, an optimal choice will not be made. This non-optimal choice will have a negative effect on the rest of the company, because a better decision could be made with the private information of the manager.

Corporate governance provides several mechanisms to deal with the problems of information asymmetry and align the interests of the shareholders and managers. One mechanism is incentive compensation. Shareholders cannot control the actions of the managers in an optimal way. As a result, they design compensation contracts that motivate the CEO to take the right actions of his or her own volition. Therefore,

(9)

9 of executive compensation contracts (Conyon & He, 2012). The compensation of

managers should be linked to performance to deal with these problems of the agency relationship. If the incentive compensation of managers is designed in the right way, the managers will make decisions which will benefit the owners of the company (Murphy, 1999).

Managers receive bonuses and equity-based compensation to align the interests of managers with the shareholders. Salary is the base for the compensation of managers and is often not influenced by the performance of the company. Bonuses are part of the incentive compensation that is typically awarded annually and is contingent on financial measures and in some cases supplemented with non-financial measures. Apart from linking CEO’s non-equity flow of compensation to firm performance, an important stream of research in the executive compensation literature predicts that CEO

compensation contracts should also contain equity incentives such as stock options and restricted shares in order to align CEOs’ and owners’ interests and to ameliorate

potential moral hazard problems (Core & Guay, 1999; Holmstrom, 1979; Holmstrom & Milgrom, 1991; Laffont & Martimort, 2002; Conyon & He, 2012). This method of

incentive compensation is growing over time and becomes more important in

compensation contracts. When managers own more shares it is more likely that those managers will act in the interests of the shareholders (Cheng & Warfield, 2005).

Empirical literature on CEO compensation in the US has consistently demonstrated that equity ownership, in the form of stock options, restricted stock, or share ownership are important for resolving agency problems (see reviews by Finkelstein, Hambrick, & Cannella, 2009; Murphy, 1999). If the compensation of managers is influenced by the value of the company, managers will have a strong incentive to make decisions that increase the value of the company. In contrast, managers whose compensation is not influenced by firm performance, in the form of bonuses and equity-compensation, have little incentives to perform the desired tasks. This is because their total compensation and personal wealth will not change when the value of the company reduces.

2.2 Earnings management

The information asymmetry between shareholders and managers means that managers have private information which may be valuable for the shareholders. However, the shareholders don’t receive this information. Besides incentive compensation, another way to deal with the problems associated with asymmetric information is monitoring the managers by financial disclosure.

The stakeholders and shareholders will be able to make a better picture of how the company is performing, after disclosure of the financial statements. These financial statements are made by the managers, and this means that these statements also incorporate information of the managers. The private information of managers is needed to create these financial statements. The financial statements of public companies are created according accounting regulations. The consequences of

(10)

10 the financial statements are free from misstatements and that a minimum level of

disclosure is specified. To provide extra assurance that the information in these financial disclosures is correct, the financial statements need to be audited by an external party. This external party is an auditor, which expresses his opinion on the financial

statements of the firm.

Financial disclosures are for an important part based on accruals. Cash flows are not always informative, because cash flows have timing and matching problems. The result is that cash flows are a ‘noisy’ measure of firm performance. To mitigate these problems, generally accepted accounting principles have evolved to enhance

performance measurement by using accruals to alter the timing of cash flows

recognition in earnings (Dechow, 1994). The recognition of accruals involves discretion of the managers, based on their own interpretation. This discretion can be used by management to signal their private information or to opportunistically manipulate earnings (Dechow, 1994). Managers can influence these accruals to serve their own purposes. This is called opportunistic earnings management. Providing options- and share-based compensation will make the wealth of the managers sensitive to the short-term stock price. These equity-based incentives can lead managers to focus more on short-term stock prices (Cheng & Warfield, 2005). According to Stein (1989) managers use their discretion to manage earnings in order to keep the short-term stock price high (Stein, 1989). This will have a positive influence on the compensation of the managers.

A commonly used definition for earnings management is provided by Healy and Wahlen (1999). They define earnings management as follows: 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. Earnings management can be done in different ways.

Real earnings management is managing the earnings of the firm through manipulation of real activities. Real activities manipulation is defined as management actions that deviate from normal business practices, undertaken with the primary objective to mislead certain stakeholders into believing that earnings benchmarks have been met in the normal course of operations (Roychowdhury, 2006). This can be done to meet certain targets. This manipulation of real activities has an effect on the cash flows of the company. An example of real earnings management is reducing discretionary

expenditures, such as a reduction in expenditures on research and development.

However, in this study we will focus on accrual-based earnings management. One means of managing earnings is by manipulation of accruals with no direct cash flow consequences (Roychowdhury, 2006). Accruals can be used to improve the financial position of a company by using accruals as adjustments for revenues that will be earned, but are not yet recorded in the accounts of the company. Accruals can also be used to deteriorate the financial performance of a company by using accruals as adjustments for expenses that have been occurred, but are not yet recorded in the accounts of the

company. Examples of income-increasing accrual-based earnings management are under-provisioning for bad debt expenses and delaying asset write-offs. Managers can

(11)

11 influence the discretionary accruals of a company while non-discretionary accruals cannot be influenced by the managers.

A manager has multiple motives to manage the earnings of the company. The research of Graham et al. (2005) provides several reasons for earnings management. Graham et al. find that CFOs consider earnings, and not cash flows, to be the key metric considered by outsiders. Moreover, meeting benchmarks is one of the most important reasons for managers to manage earnings. The reason to beet benchmarks is that managers think that this will build credibility with the market and helps to increase the stock price of the company (Graham et al., 2005). Furthermore, most CFOs prefer smooth earnings because with the help of smooth earnings, predicting future earnings will be easier. Less predictable earnings command a risk premium in the market. Another reason to manage earnings is that executives think that they are making a right choice when sacrificing value to smooth earnings or to meet benchmarks. Negative earnings surprises can lead to costs due to the turmoil that arises in the market. That is why managers decide to choose to avoid short-term turmoil. The advantages of avoiding short-term turmoil outperform sacrificing long-run value. To sum up, managers have four reasons to meet or beat benchmarks: (i) build credibility with the capital market; (ii) maintain or

increase stock price; (iii) improve the external reputation of the management team; and (iv) convey future growth prospects (Graham et al., 2005). Other literature also shows that managers have incentives to meet or beat benchmarks and managers are rewarded by the market for doing this (Athanasakou et al. 2011; Herrmann et al., 2011; Lopez & Rees, 2002; Skinner & Sloan 2002). Healy and Wahlen (1999) describe also motivations for managing earnings. They state that capital market punish firms if they do not meet the forecasts of analysts. This pushes managers to manage earnings and meet the forecasts. In addition, managers have contracting motivations to manage earnings. The contracts which Healy and Wahlen (1999) discuss are lending contracts and

management compensation contracts. Managers don’t want to violate these lending contracts and use earnings management to follow these contracts.

The aforementioned motivations for earnings management may (indirectly) also serve their personal purposes. For example, meeting benchmarks is not only important for the firm as its builds credibility in the market, Farrel and Whidbee (2003) also document that not performance itself, but a failure to meet the expectations in the market lead to an increased likelihood of CEO turnover. The likelihood of CEO turnover forms an incentive for managers to manage earnings, because income-increasing earnings management is a way to prevent CEO turnover.

Besides serving the interests of the firm, in the survey of Graham et al. (2005) the overall majority acknowledge that their personal incentives shape their reporting

behavior. Prior research confirms this motive for earnings management. For example, Cheng and Warfield (2005) argue that managers with high equity incentives are more likely to report earnings that meat or just beat analysts’ forecast. They also found that managers which have high equity incentives prefer not to report large positive earnings surprises. That is because the wealth of the managers is influenced by future stock

(12)

12 performance. So managers will increase the reserving of current earnings to avoid

future earnings disappointments (Cheng & Warfield, 2005). Furthermore, the papers of Sloan (1996) and Collins and Hribar (2000) provide evidence that managers may be able to manage the capital markets. Both papers show that managers were able to use

accruals to manipulate the market value of their firms. CEO’s sold their stock of the company before very high returns to accruals disappeared. Examples of companies which managed their earnings are Xerox, Enron, Tyco and Waste Management

(Bergstresser & Philippon, 2006). The managers of these companies reported increased earnings and sold their shares and exercised their options subsequently.

In sum, managers with high (equity-based) incentive compensation are

incentivized to increase the performance of the company. Such incentive compensation motivates managers to care about short-term stock prices and to manage earnings. Formally, we hypothesize that:

H1: There is a positive relationship between CEO incentive compensation and the use of earnings management.

2.3 The financial crisis

2.3.1 What is the financial crisis?

The financial crisis started in the US in the subprime mortgage segment and it spread to other market segments in the US and eventually to the whole world (Ackermann, 2008). During the year 2008, the U.S. subprime problem became a global financial crisis, and many financial institutions were affected severely by market deterioration, which resulted in the recognition of large losses (Fiechter & Meyer, 2010). This is related to systematic risk. Systematic risk means that financials are correlated with each other. So the risk of one organization is positively associated with the risk of other organizations and if one organization falls the other organizations will cope with the same problems. Because of this risk, the decrease in trust in the financial market led to a strong decrease in the number of financial transactions. Other causes of the crisis in the confidence was that organizations in unexpected places where making losses, and the uncertainty about potential future losses (Ackermann, 2008).

However, the financial crisis could have such a large impact because not many parties seemed to identify this risk. There are two reasons for that. First, it was difficult for the investors to assess this risk because the financial products were becoming more complex (Barth & Landsman, 2010). Second, the credit rating agencies seemed to have overrated financial products containing subprime mortgages (Benmelech & Dlugosz, 2009).

The years 2005 and 2006 were the years which set the basis for the financial crisis. In these years there was an abrupt end of the housing boom and the market for subprime mortgage was slowing down. These events were the first indication of a

downfall of the economy. The events in 2005 and 2006 were the basis of the fall of many subprime lenders in 2007. Many subprime mortgage lenders went bankrupt because of

(13)

13 the losses they were making. Furthermore, in 2007 the sales of homes are falling and banks are exposed of the risk of the subprime mortgage markets. Several banks needed help to continue their business. 2008 was the beginning of the financial crisis (Fiechter & Meyer, 2010). In this year, the crises in the US subprime mortgage market spread to other markets in many countries. A lot of banks over the world were reporting huge losses and eventually went bankrupt. An event what caused worldwide financial panic was the fall of the Lehman Brother bank in 2008. The fall of Lehman Brother was the basis for the fall of other banks and companies. In 2008 also Ireland slides into recession as the first European country. Many other countries will follow Ireland what results in bailouts of several European countries in 2010 and 2011. Furthermore, agreements were made to improve the financial situation. These agreements helped to improve the situation in the markets and the economic environment. The financial crisis eventually ended in 2010. A full overview of the events related to the financial crisis can be found in the Appendix.

The characteristics of the financial crisis are provided by the study of Campello et al. (2009). The results of this study showed that more than half of the respondents of their study will cancel or postpone their planned investment. 86% of the constrained US CFOs said that their investment in attractive projects was restricted during the financial crisis of 2008 (Campello et al., 2009). Firms during a financial crisis are focused on reducing their costs and cash outflows. The reduction in spending behavior of companies and individuals will have negative effects for the government and

environment. Other characteristics of a financial crisis are that firms planned deeper cuts in tech spending, employment, and capital spending (Campello et al., 2009).

Distressed firms also burned more cash, drew more heavily on lines of credit due to the feat that banks would restrict access in the future, and sold more assets to fund their operations. Another argument why financial distress of multiple companies is costly is provided by Opler and Titamn (1994). They state that, financial distress creates a tendency for companies to behave in a way which is harmful to debtholders and non-financial stakeholders, raising costs of stakeholder relationships, and a firm’s weakened condition induces an aggressive response by competitors seizing the opportunity to gain market share.

The financial crisis was a concern for the government and investors amongst others. A decline in the financial performance and spending of companies resulted in such problems that investors lost their money. Because of this fear of losing more, investors tried to sell their assets or withdraw their money form the savings accounts. The government felt the effects of this behavior of investors. This had disastrous consequences for the whole country, because the trust in financial institutions and organizations was fading away. A lot of companies and eventually also countries became financial distressed.

(14)

14

2.3.2 The effect of the financial crisis on the compensation and earnings management relationship

According to Strobl (2013) managers are more likely to manipulate earnings during an economic boom as opposed to a recession. This view is shared by Matejka et al. (2009). They argue that loss-making firms will have a shorter employment horizon, because managers will depart voluntary or forcibly in the near future. That is why companies motivate the long-term effort of their managers by increasing the emphasis on forward-looking non-financial performance measures. So in loss-making firm, you can expect that there is more emphasis on forward-looking measures in CEO bonus plans. Managing earnings and increasing the current performance of the company, will not increase the compensation of the managers. Therefore, managers will be more motivated to create more long-term value and not managing earnings.

However, Burgstahler and Dichev (1997) found evidence for firms which manage earnings to avoid earnings decreases and losses. This behavior would be used by

managers when the bonus-linked incentives of these managers are influenced by meeting targets for earnings. According to Burgstahler and Dichev (1997), a firm does not like to report negative earnings. A reason for firms to present nonparametric evidence, that the distribution of earnings is bunched just above zero. That is why you could expect that managers will manage earnings more and manage earnings upwards during a financial crisis, to avoid negative earnings and to meet benchmarks. Also Farrel and Whidbee (2003) provide arguments why managers would manage their earnings upwards during a financial crisis. The study of Farrell and Whidbee (2003) examines the relationship between CEO turnover and the performance of the company compared to the forecasts of analysts. The authors state that it is the task of the CEO to meet or beat these earnings forecasts. That is why the likelihood of getting fired increases when managers don’t meet these expectations. Their research indicates that boards focus on deviations from expected performance, rather than performance alone, in making CEO turnover decisions (Farrell & Whidbee, 2003). It will be more difficult to meet

benchmarks during the financial crisis because of the bad environmental conditions. Therefore, managers will try to secure their jobs by managing the earnings of the company upwards to meet earnings forecasts.

There are also arguments why managers would engage in income-decreasing earnings management during the financial crisis. Big bath accounting provides such argument. A definition of big bath accounting is given by Walsh et al. (1991), who state that big bath accounting is a negative outlier in an entity’s growth in reported net profit after tax, minority interest and extraordinary items. The authors state that firms take as many losses as possible in a year (apply big bath accounting) to meet expectations in following periods. According to big bath accounting, the market tolerates it that a company is performing poor during a financial crisis. Managers recognize this, so they have the incentives to manage earnings more downwards with the help of accruals. By managing earnings downwards in the financial crisis, they reserve these earnings for the period after the financial crisis. This improvement in performance of the company after the financial crisis will benefit the reputation of the managers. Another reason to

(15)

15 manage earnings downwards is that managers do not have chance to receive a bonus, in the period where the firm is performing poor. That is why the manager defers income to future periods to increase their bonuses in the future. This deferring of income to the future will have no effect on their current compensation. Another argument is provided by the research of Hayn (1995). The author argues that during a financial crisis, financial reporting suffers because losses are less informative than profits about future

performance (Hayn, 1995). The financial disclosure of companies will have less added value to the shareholders and stakeholders, because there are many external factors which influence the financial statements. That is why you could argue that shareholders and stakeholders will pay less attention to the financial statements during a financial crisis. This will incentivize mangers to manage earnings downwards and reserve current earnings to increase future performance.

According to Fiechter and Meyer (2009) managers use earnings management in the financial crisis as overstating losses in order to be in a position to present positive earning in subsequent quarter. For example, DeAngelo et al. (1994) find that managers of troubled firms reduce the company’s income via negative abnormal accruals and discretionary write-offs. So instead of inflating the income of the company, managers will worsen the financial position of the troubled companies. The study of Healy (1985) also supports the argument of big bath accounting. This research shows that managers of poor-performing firms manage the earnings of the firm downwards. The same is found in the study of Han and Wang (1998). Han and Wang (1998) indicate that

petroleum refining firms used accruals to reduce reported earnings during the 1990 Gulf crisis. Petroleum refining firms reported more special write-offs to decrease earnings in this period than in the prior period.

Based on the prior line of arguments that the financial crisis may induce

managers to use income-increasing accruals as a mean to meet earnings benchmarks or to use income-decreasing accruals to defer income to subsequent periods, I formulated the hypothesis as follows:

H2: The financial crisis positively moderates the relationship between incentive compensation and earnings management.

(16)

16

3. Research methodology

3.1 Sample selection

This paper’s sample will be based on the U.S market from 2004 up to 2010. The U.S market is chosen because of the access to accounting information on U.S firms. My initial sample is of all firms included in the Execucomp database from 2004 up to 2010. The coverage of the Execucomp database roughly corresponds with the S&P1500. I have an initial sample of 10501 firm-year observations. I collect data on CEO compensation from Execucomp and drop all missing data. This results in a sample of 9325 observations. I collect data for the estimation of discretionary accruals and control variables from Compustat. After dropping incomplete observations and merging Compustat data with Execucomp data, the sample size is reduced to 7406 observations. I exclude financial organizations (SIC codes between 6000 and 6999) and utilities (SIC codes between 4500 and 4999) from the sample, because managers in these organizations could have

different motivations to manage their earnings (Burgstahler & Eames, 2003, Cheng & Warfield, 2005). 1170 observations are dropped from the sample. I winsorized all variables at the 1st_{and 99}th_{percent of distribution to address any potential impact of} outliers. My final sample consists of 6236 firm-year observations from 1490 unique firms.

3.2 Empirical model

3.2.1 Main model

I test my hypotheses by means of the following empirical model:

EM= β0 + β1*INC.COMP+ β2*CRISIS + β3*INC.COMP*CRISIS + CONTROLS + ε

Where EM denotes accrual-based earnings management, INC.COMP represents the degree of incentive compensation and crisis is a dummy variable equal to one for the years 2008-2010 and equal to zero for the years 2004-2007. β1 represents the

relationship between incentive compensation and earnings management in the pre-crisis years, and (β1 + β3) represents the relationship between incentive compensation

and earnings management in the post crisis years.

To test the first hypothesis, the value of β1 has to be known. A positive value of β1

indicates that there is a positive relationship between CEO incentive compensation and earnings management and that the first hypothesis is supported. So on the basis of H1, I expect that β1 > 0.

To test the second hypothesis, the values of β1 and β3 are important. Because I

expect that managers manage earnings more during the financial crisis, I expect that β3

has a positive value. If the financial crisis positively moderates the relationship between incentive compensation and earnings management, than the sum of β1 and β3 must be

larger than the value of β1. So on the basis of H2, I expect that β1 + β3 > β1. Simplifying, H2

(17)

17 Incentive compensation can be measured in several ways. According to Gao and

Shrieves (2002) the size of a bonus depends on the performance measures, performance standards, and the structure of the pay-performance relation. In this research, bonus is measured as part of the total compensation of the CEO. In this way, a rate is provided which indicates which part of the actual compensation is in the form of a bonus. The formula to measure bonus incentives is as follows:

𝐼𝑁𝐶. 𝐶𝑂𝑀𝑃 (𝐵𝑂𝑁𝑈𝑆) = 𝑎𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝑏𝑜𝑛𝑢𝑠 𝑡𝑜𝑡𝑎𝑙 𝑐𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑖𝑜𝑛

Stock and options are equity-based incentives for managers. Equity-based incentive for the CEO is measured by looking at the flow of options and shares awarded to the CEO in one year. The equity flow is divided by the total compensation of the CEO. This results in a rate of equity flow in relation with total compensation. The formula to measure equity-based incentive compensation is as follows:

𝐼𝑁𝐶. 𝐶𝑂𝑀𝑃 (𝐸𝑄𝑈𝐼𝑇𝑌) = 𝑒𝑞𝑢𝑖𝑡𝑦 𝑓𝑙𝑜𝑤 𝑡𝑜𝑡𝑎𝑙 𝑐𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑖𝑜𝑛

The total incentive compensation is my main measure and is measured by the sum of bonus and equity flow of CEOs divided by the total compensation of the CEO. In this way the part of incentive compensation is calculated which could be an incentive for managers to manage earnings. The formula to measure the total incentive compensation of CEOs is as follows:

𝐼𝑁𝐶. 𝐶𝑂𝑀𝑃 =𝑎𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝑏𝑜𝑛𝑢𝑠 + 𝑒𝑞𝑢𝑖𝑡𝑦 𝑓𝑙𝑜𝑤 𝑡𝑜𝑡𝑎𝑙 𝑐𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑖𝑜𝑛

3.2.2 Auxiliary model

To measure the level of accruals, data from the income statements of companies are used. Different models to empirically estimate the (discretionary) accruals are

developed over time to measure earnings management. Dechow et al. (1995) discussed these different models. They concluded that the Modified Jones model provided the strongest proof of detecting earnings management. That is why in this research the Modified Jones model is used to measure earnings management. The Modified Jones model measures earnings management as follows:

TAi,t = Total accruals calculated as IBCi,t – (OANCFi,t – XIDOCi,t) of firm i in yeart IBCi,t = Income before extra items of firm i in yeart

OANCFi,t = Net cash flow of operating activities of firm i in yeart

XIDOCi,t = Cash flow of extraordinary items and discontinued operations of firm i in yeart

(18)

18 ∆REVi,t = Change in revenues from the preceding year of firm i in yeart

∆ARi,t = Change in account receivables from the preceding year of firm i in yeart PPEi,t = Gross value of property, plant, and equipment of firm i in yeart

I first estimate the total level of accruals by subtracting the net cash flow of operating activities (OANCF) and cash flow of extraordinary items and discounted operations (XIDOC) from income before extra items (IBC). To distinguish between discretionary and non-discretionary accruals, I use the Modified Jones Model. The Modified Jones Model measures earnings management as the absolute value of the residual. The residual is retrieved by performing a regression analysis of the auxiliary model per 2-digit SIC code. This residual represents the discretionary accruals, and that is what is used in this paper as a measure of earnings manipulation

3.4 Measurement of control variables

To research the relationship between earnings management and compensation, control variables will be used to control for alternative explanations for my findings (so called confounding variables). By using these variables in the regression model, the chance of potential external influences will be reduced and a potential alternative explanation for the results will be excluded. The control variables used in this paper are: leverage, market to book ratio, firm size, firm performance and industry. Leverage, market to book ratio and firm size are being used to control for growth. Prior literature suggests that compensation policy is an important part for dealing with financial distress

(Kostiuk, 1990). To account for this, the control variable leverage will be used which is an economic determinant of financial distress. Furthermore, by using firm size and market-to-book ratio as control measures the model will suggest that the result are not driven by the more volatile operating environments of firms that use a lot of stock-based compensation (Bergstresser & Philippon, 2006).The natural logarithm of total assets will be included as a proxy of the size of a company. The research of Watts and Zimmerman (1990) suggest that there is a negative association between the levels of earnings management and the size of firms, because larger firms are under more scrutiny of the outside world (i.e. investors, financial press, etc.). One could expect that there would be less opportunity for managers to manage earnings of these firms. Also firm performance is being used as a control variable. The yearly return on assets as measured by earnings divided by lagged total assets (ROA) will be included to control for differences in performance (Dechow et al., 1995). Lastly, industry-effects are addressed to control for differences between different industries. CEO compensation and earnings management could be different among industries and controlling for industry-effects would improve the robustness of the findings.

(19)

19

4. Results

4.1 Descriptive statistics

Table 1: Descriptive statistics

N mean median sd p25 p75 Variables EM 6236 0,001 0,005 0,070 -0,030 0,038 INC.COMP(BONUS) 6236 0,223 0,203 0,168 0,103 0,314 INC.COMP(EQUITY) 6236 0,443 0,0,464 0,251 0,262 0,631 INC.COMP 6236 0,664 0,713 0,212 0,540 0,826 CRISIS 6236 0,474 0 0,499 0 1 LEV 6236 0,512 0,493 0,508 0,334 0,627 MTB 6236 2,917 2,259 25,562 1,469 3,538 SIZE 6236 7,266 7,157 1,574 6,168 8,300 ROA 6236 0,038 0,056 0,432 0,019 0,095

Table 1 reports descriptive statistics on variables used in the main empirical model. It shows that the average discretionary accruals (EM) is more or less comparable with other studies (Gao & Shrieves, 2002; Meek et al. 2007). Furthermore, 66% of the total compensation of CEOs consists of incentive compensation. From this 66% of incentive compensation, about 44% is determined by equity compensation. The sample consists for 47% out of firm years during the financial crisis, what means that both periods are equally represented in this sample. The mean leverage of the firms is 51%, which indicates that companies finance half of their assets through debt. The Market-to-Book value is about 2,9 which indicates the average growth of the companies. This suggests a high availability of growth opportunities for the firms in this sample. Moreover, the size of companies has an average value of 7,3, which corresponds with an average company size of (average assets) of about 5,5 billion dollars. The performance of a firm is

measured by the return on assets (ROA), which is around 4% for the firms in this sample.

Table two provides the descriptive statistics regarding incentive compensation and earnings management over industries, based on a classification by two-digit SIC codes. As can been seen from the table, the sample appears to be concentrated on

manufacturing firms. 57,89% of the companies in the sample are manufacturing firms (SIC 20-39). There are no agriculture, forestry, fishing and hunting firms (SIC 00-09) and international affairs & non-operating establishments firms (SIC 90-99) in this sample. With respect to the averages of my proxies for accrual-based earnings management and incentive compensation no large discrepancies are found related to discretionary accruals and incentive compensation.

(20)

20

Table 2: Descriptive statistics per industry

Table three gives an overview of the Pearson correlations. Bold correlation coefficients are significant at the 10% level. The table shows that non-equity based incentive compensation and equity-based incentive compensation are negatively related with each other (-0,55), what indicates that they are used more likely as substitutes. Moreover, non-equity based incentive compensation is positively correlated with earnings management (0,08), what means that managers manage their earnings more if there are more bonus incentives in their compensation contracts. While bonus

incentives are positively correlated with earnings management, equity-based incentives show a negative correlation with earnings management (-0,06). This indicates that managers manage their earnings more if there are less equity incentives in their compensation contracts. In contradiction with bonus- and equity-incentives, total incentive compensation is not significantly correlated with earnings management. This means that there is not enough evidence to state that total incentive compensation influences earnings management in a certain way. Furthermore, incentive compensation is positively related with the financial crisis (0,11), what means that during the financial crisis there is more focus on incentive compensation in the total compensation of CEOs. Incentive compensation (both equity-based and non-equity based) and a companies’ size are positively correlated what means that in larger companies managers receive more incentive compensation. The table also shows that managers will manage their earnings more in bigger firms and better performing firms (0,04 & 0,17). But in firms where in companies which finance their assets more through debt, engage less in

earnings management. This because leverage and earnings management are negtatively correlated (-0,08). Finally, the magnitude of correlations does not warrant concerns about multicollinearity.

Table 3: Pearson correlation matrix

Variables 1 2 3 4 5 6 7 8 9 1.EM 1,00 2.INC.COMP(BONUS) 0,08 1,00 3.INC.COMP(EQUITY) -0,06 -0,55 1,00 4.INC.COMP -0,01 0,14 0,75 1,00 5.CRISIS -0,02 0,00 0,09 0,11 1,00 6.LEV -0,08 0,03 0,00 0,03 0,00 1,00 7.MTB 0,00 0,00 0,01 0,01 -0,02 0,00 1,00 8.SIZE 0,04 0,09 0,23 0,35 0,04 0,07 0,01 1,00 9.ROA 0,17 0,05 -0,01 0,03 -0,03 -0,39 0,02 0,09 1,00

N (% of total) Mean EM Mean INC.COMP Industry

Mining & construction (SIC: 10-19) 478 (7,67%) 0,002 0,711 Manufacturing (SIC: 20-39) 3610 (57,89%) 0,001 0,662 Transportation (SIC: 40-49) 101 (1,62%) -0,004 0,617 Wholesale & retail trade (SIC: 50-59) 789 (12,65%) 0,000 0,654 Services (SIC: 70-89) 1258 (20,17%) 0,001 0,659

(21)

21

4.2 Main analyses

Table 4: OLS regression results H1

Model 1A Model 1B Model 2 Variables Pred. Intercept 0,003 (0,006) 0,009 (0,006) 0,008 (0,006) INC.COMP(BONUS) + 0,023*** (0,007) -- -- INC.COMP(EQUITY) + -- -0,014* (0,004) -- INC.COMP + -- -- -0,007 (0,005) LEV -0,005*** (0,002) -0,005** (0,002) -0,005** (0,002) MTB 0,000 (0,000) 0,000 (0,000) 0,000 (0,002) SIZE -0,001 (0,001) 0,000 (0,001) 0,000 (0,001) ROA 0,018*** (0,002) 0,018*** (0,002) 0,018*** (0,002)

Industry controls Yes Yes Yes

Adjusted R-squared 0,029 0,029 0,026

F-value 20,68*** 20,54*** 18,76***

***, **, * denotes 1%, 5%, and 10% significance levels respectively

Table 4 presents the regression summary statistics for the empirical model that is estimated by means of Ordinary Least Squares (OLS). It examines the relationship that incentive compensation positively influences earnings management and that the

financial crisis positively moderates this relationship.Each column describes the result for a different part of the incentive compensation, that is, the bonus incentives (1A), the equity incentives (1B), and the total incentives (2). For the first hypothesis we first analyze the relationships of the incentives with earnings management in the pre-crisis period. So, here I focus on the coefficient on INC.COMP(BONUS), INC.COMP(EQUITY) and INC_COMP respectively.

Bonus incentives are part of the total incentive compensation and the coefficient that describes the relationship with earnings management is INC.COMP(BONUS) in model 1A. This relationship is positive and significant (coefficient 0,023 and p<0,05). This result is consistent with prior literature who state that the main reason for earnings management is personal interest, as a result of the compensation of CEOs (Bauman & Shaw, 2006; Bergstresser & Philippon, 2006). Consistent with my predictions, I find that bonus incentives are positively related with earnings management. Equity incentives are also part of the total incentive compensation and consist of shares and options awarded to the CEO. The coefficient that describes the relationship between equity incentives and earnings management is INC.COMP(EQUITY) in model 1B. This relationship is negative and significant (coefficient -0,014 and p<0,05). In contradiction with the studies of Cheng and Warfield (2005) and Bergstresser and Philippon (2006), I find that equity incentives are negatively related with earnings management.

The main measure for incentive compensation is total incentive compensation, combining bonus and equity incentives. The coefficient that describes the relationship between total incentive compensation and earnings management is INC.COMP in model 2. Based on the sample of 3283 firm-years over a period ranging from 2004 to 2007, the results do not indicate a significant relationship between CEO incentive compensation

(22)

22 and the use of earnings management. This because β1 = -0,007 (p>0,05). The results are in contradiction with prior results (Cheng & Warfield, 2005; Sloan, 1996; Collins & Hribar, 2000; Healy, 1985). This means that H1 is not supported.

Furthermore, the leverage and ROA of a company in the pre-crisis period are significantly related with earnings management. The model provides evidence that leverage is negatively related with earnings management (-0,005) and that ROA is positively related with earnings management (0,018).

Table 4: OLS regression results H2

Model 1A Model 1B Model 2 Variables Pred. Intercept -0,012*** (0,004) 0,000 (0,005) -0,002 (0,005) INC.COMP(BONUS) + 0,030*** (0,005) -- -- INC.COMP(EQUITY) + -- -0,022*** (0,004) -- INC.COMP + -- -- -0,010* (0,006) CRISIS 0,005 (0,005) -0,009* (0,005) -0,004 (0,006) INC.COMP*CRISIS + -0,009 (0,007) 0,012 (0,007) 0,004 (0,008) LEV -0,002 (0,002) -0,002 (0,002) -0,002 (0,002) MTB 0,000 (0,000) 0,000 (0,000) 0,000 (0,000) SIZE 0,001* (0,001) 0,002*** (0,001) 0,001** (0,001) ROA 0,026*** (0,002) 0,027*** (0,002) 0,027*** (0,002)

Adjusted R-squared 0,035 0,035 0,030

F-value 33,14*** 33,09*** 28,92***

***, **, * denotes 1%, 5%, and 10% significance levels respectively

The impact of the financial crisis on the relationship between incentive compensation and earnings management is provided by the value of INC.COMP*CRISIS.

INC.COMP*CRISIS has a positive value for equity incentives and total incentive compensation (0,012 & 0,004) and a negative value for bonus incentives (-0,009). However, all these coefficients are not significant because p>0,05. This means that I do not find support that the financial crisis moderates the relationship between incentive compensation and earnings management in a positive way. This in contradiction with my predictions and no evidence is found for big bath accounting. The results do also not support the study of Farrell and Whidbee (2003), where managers manage their

earnings more during financial distress in order to secure their jobs. This means that H2 is not supported.

Furthermore, the performance of a company (ROA) is positively related with earnings management in the period 2004 to 2010 (0,027). This indicates that managers in better performing firms manage their earnings more.

With respect to the model performance, both models are significant (F = 18,76 & F = 28,92, p<0,01). However, according to the adjusted R-squared of both models only 3% of earnings management is explained by the independent variables. This means that the majority of earnings management is explained by factors not present in this model.

(23)

23

4.3 Robustness analysis

The sensitivity of the results is assessed in the following way. A robust regression is applied, which controls for heteroscedastic errors and outliers. First, by performing a robust regression, estimation biases due to influential statistics are limited. These influential statistics usually are compromised out of outliers and high leverage

observations. There is no reason to exclude observations from the sample, because it is unlikely that the influential statistics in the data set are data that is from a different population or are a result of data entry errors. Because, the robust regression weights observations on a different manner, the robust regression is differently based on how behaved the observations are. This will give other results than an OLS regression. The results of the robust regression are summarized in table 5 and 6.

Table 5: Robust regression results H1

Model 1A Model 1B Model 2 Variables Pred. Intercept 0,007 (0,004) 0,008* (0,004) 0,010** (0,005) INC.COMP(BONUS) + -0,002 (0,006) -- -- INC.COMP(EQUITY) + -- -0,007** (0,004) -- INC.COMP + -- -- -0,012*** (0,004) LEV 0,005 (0,003) 0,004 (0,003) 0,005 (0,004) MTB 0,000 (0,000) 0,000 (0,000) 0,000 (0,000) SIZE -0,002** (0,001) -0,001* (0.,001) -0,001 (0,001) ROA 0,114*** (0,005) 0,116*** (0,005) 0,124*** (0,005)

F-value 186,24*** 197,37*** 226,59***

***, **, * denotes 1%, 5%, and 10% significance levels respectively Table 6: Robust regression results H2

Model 1A Model 1B Model 2 Variables Pred. Intercept -0,004 (0,004) -0,002 (0,004) 0,004 (0,004) INC.COMP(BONUS) + -0,002 (0,004) -- -- INC.COMP(EQUITY) + -- -0,009*** (0,003) -- INC.COMP + -- -- -0,019*** (0,004) CRISIS 0,020*** (0,004) 0,016*** (0,004) 0,009* (0,005) INC.COMP*CRISIS + -0,024*** (0,005) -0,017*** (0,006) -0,007 (0,007) LEV 0,004** (0,002) 0,003 (0,002) 0,003* (0,002) MTB 0,000* (0,000) 0,000* (0,000) 0,000* (0,000) SIZE -0,001 (0,000) 0,000 (0,000) 0,000 (0,000) ROA 0,194*** (0,004) 0,191*** (0,004) 0,195*** (0,004)

F-value 605,86*** 604,17*** 627,40***

(24)

24 The results of the robust regression differ with those of the OLS regression results. The relationship between total incentive compensation and earnings management in the pre-crisis period is significantly negative at -0,012 (p<0,01). So managers will manage the earnings of the company more when there are fewer incentives in the compensation contracts. Based on the results of the robust regression H1 is rejected. Furthermore, leverage is insignificant in explaining earnings management in the pre-crisis period when using robust regression.

The INC.COMP*CRISIS coefficient on total incentive compensation during the financial crisis stays insignificant when using robust regression. This means that there is not enough evidence to reject or support H2 and this is again in contradiction with my predictions. However, there is a significant negatively relationship with earnings management if bonus- and equity incentives are studies independently when using robust regression. This is in contrast with the expectations.

(25)

25

5. Conclusion

This paper examines the relationship between incentive compensation and earnings management and what the effect of the financial crisis is on this relationship. Based on the literature of earnings management before and during the financial crisis, it is predicted that (equity-based) incentive compensation will motivate managers to manage earnings of their company. Also, it is expected that managers manage the earnings of the company more during the financial crisis than before the financial crisis. This expectation is based on the line of arguments that the financial crisis may induce managers to use more income-increasing accruals as a mean to meet earnings

benchmarks and to use more income-decreasing accruals to defer income to subsequent periods.

The findings are not consistent with the projections. Using a sample of 6236 firm-year observations over a period from 2004 up to 2010, no evidence is provided that incentive compensation motivates managers to manage the earnings of their company. Moreover, the results provide no evidence that the financial crisis moderates the relationship between incentive compensation and earnings management in any way, because the results of this paper were not significant. However, after the robustness test I found a negative relation between incentive compensation and earnings management, what was in contradiction with my predictions. A possible explanation for these findings is that managers have the incentives to time the release of good and bad news.

This study contributes to prior literature by focusing on the less researched area of CEO compensation and earnings management. This paper also extends prior

literature by answering the question what the impact is of the financial crisis on the relationship between incentive compensation and earnings management. This is not a priori clear and this study gives insights to this. Additionally, this paper could be used for societal ends. Understanding earnings management behavior and how this varies under circumstances (e.g., during a financial crisis) is relevant as this enables capital market participants to anticipate earnings management behavior and correct for this in their decision making.

Results of this study are subjected to the following limitations. First, the data for incentive compensation of CEOs are obtained from large publicly listed American companies. That is why generalization of the findings for small and mid-sized firms is limited for this study. Further research could investigate whether these findings are also valid for companies in different institutional settings. Second, the proxies used in this study for incentive compensation may not be perfect. The proxies are limited to bonus plans, stock grants and option grants. However, prior literature states that the behavior of managers might be subjected to other elements of compensation such as: contribution to pension plans and perquisites. Also the decision of managers to manage the earnings of their firm may be affected by other influences. For example responsibilities, future promotions and personal relations could influence the behavior of managers. These influences are not taken into consideration in this study. The other elements of compensation and other influences could be used in future research to get a better picture of the actions of the managers. Third, I repeat that there is no evidence for the

(26)

26 relation of earnings management before and during the financial crisis. A possible

reason for this can be the variables used to research this relationship. The adjusted R-squared of the models are low, what implies that the independent variables used to research this relationship may not be proper. Future research could use different independent variables to examine this relationship. Lastly, this study uses limited control variables. Further research could take bonus and equity control variables into account to control for alternative explanations for the findings.

(27)

27

6. References

- Ackermann, N. (2008). The subprime crises and its consequences. Journal of Financial Stability, Vol. 4, pp. 329-337.

- Akerlof, G. (1970). The market for "lemons": quality uncertainty and the market mechanism. The Quarterly Journal of Economics, Vol. 84, pp. 488-500.

- Athanasakou, V., Strong, N. & Walker, M. (2011). The market reward for achieving analyst earnings expectations: Does managing expectations or earnings matter? Journal of Business Finance & Accounting, Vol. 38, pp. 58-94.

- Baker, T., Collins, D. & Reitenga, A. (2003). Stock option compensation and earnings management incentives. Journal of Accounting, Auditing & Finance, Vol. 18, pp. 557-582.

- Barth, M. & Landsman, W. (2010). How did Financial Reporting Contribute to the Financial Crisis? European Accounting Review, Vol. 19, pp. 399-423.

- Barua, A., Legoria, J. & Moffitt, J. (2006). Accruals management to achieve earnings benchmarks: A comparison of pre-managed profit and loss firms. Journal of Business Finance & Accounting, Vol. 33, pp. 653-670.

- Bauman, M., & Shaw, K. (2006). Stock option compensation and the likelihood of meeting analysts’ quarterly earnings targets. Review of quantitative finance and accounting, Vol. 26, pp. 301- 319.

- Benmelech, E. & Dlugosz, J. (2009). The Credit Rating Crisis. NBER Macroeconomics Annual 2009, Vol. 24, pp. 161 – 207.

- Bergstresser, D. & Philippon, T. (2006). CEO incentives and earnings management. Journal of Financial Economics, Vol. 80, pp. 511-529.

- Burgstahler, D. & Dichev, I. (1997). Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics, Vol. 24, pp. 99–126.

- Burgstahler, D. & Eames, M. (2003). Earnings management to avoid losses and earnings decreases: Are analysts fooled? Contemporary Accounting Research, Vol. 20, pp. 253–294.

- Campelo, M., Graham, J.R. and Harvey, C.R. (2011). The real effects of financial constraints: evidence from a financial crisis. Journal of Financial Economics, Vol. 97, pp. 470 – 487.

- Cheng, Q. and Warfield, T. (2005). Equity Incentives and Earnings Management. The Accounting Review, Vol. 80, pp. 441-476.

- Collins, D. and Hribar, P. (2000). Earnings-based and accrual-based market anomalies: one effect or two? Journal of Accounting and Economics, Vol. 29, pp. 101-123.

- Conyon, M. and He, L. (2012). CEO compensation and corporate governance in China. Corporate governance, an international review, Vol. 20, pp. 575 - 592

- Core, J. and Guay, W. (1999). The use of equity grants to manage optimal equity incentive levels. Journal of Accounting and Economics, Vol. 28, pp. 151–184.

- DeAngelo, H., DeAngelo, L., and Skinner, D.J. (1994). Accounting choice in troubled companies. Journal of Accounting and Economics, Vol. 17, pp. 113-143.

- Dechow, P. (1994). Accounting earnings and cash flows as measures of firm performance: The role of accounting accruals. Journal of Accounting and Economics, Vol. 18, pp. 3 – 42.

(28)

28

- Dechow P, Sloan, R, & Sweeny, A. (1995). Detecting Earnings Management. The accounting Review, Vol.70, pp. 193-225.

- Farrell, K. & Whidbee, D. (2003). Impact of firm performance expectations on CEO turnover and replacement decisions. Journal of Accounting and Economics, Vol. 36, pp. 165-96

- Fiechter, P. & Meyer, C. (2010). Big Bath Accounting Using Fair Value Measurement Discretion during the Financial Crisis. Working paper, University of Zurich

- Finkelstein, S., Hambrick, D. & Cannella, A. (2009). Strategic leadership: Theory and research on executives, top management teams, and boards. Oxford: Oxford University Press.

- Gilson, S. (1989). Management turnover and financial distress. Journal of Financial Economics, Vol. 25, pp. 241-262

- 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.

- Hall, B. & Liebman, J. (1998). Are CEOs Really Paid Like Bureaucrats? Quarterly Journal of Economics, Vol. 113, pp. 653–691.

- Han, J. and Wang, S. (1998). Political Costs and Earnings Management of Oil Companies during the 1990 Persian Gulf Crisis. The Accounting Review, Vol. 73, pp. 103-117.

- Hayn, C. (1995). The information content of losses. Journal of Accounting and Economics, Volume 20, pp. 125-153.

- Healy, P. M. (1985). The Effect of Bonus on Accounting Decisions. Journal of Accounting and Economics, Vol. 7, pp. 85-107.

- Healy, P. M. & Wahlen, J. M. (1999). A Review of the Earnings Management Literature and Its Implications for Standard Setting. Accounting Horizons, Vol. 13, pp. 365-383.

- Herrmann, D., Hope, O., Payne, J. & Thomas, W. (2011). The market's reaction to unexpected earnings thresholds. Journal of Business Finance & Accounting, Vol. 38, pp. 34-57.

- Holmstrom, B., and Milgrom, P. (1991). Multitask principal-agent analyses: Incentive contracts, asset ownership and job design. Journal of Law, Economics and Organization, Vol. 7, pp. 24–52.

- Hölmstrom, B. (1979). Moral hazard and observability. The Bell Journal of Economics, Vol. 10, pp. 74-91. - Iatridis, G., & Dimitras, A.I. (2013). Financial crisis and accounting quality: Evidence from five European countries. Advances in Accounting, incorporating Advances in International Accounting, Vol. 29, pp. 154-160.

- Ivashina, V., and Scharfstein, D. (2008). Bank Lending During the Financial Crisis of 2008. Journal of Financial Economics, Vol. 97, pp. 319 – 338.

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

- Kostiuk, P. (1990). Compensating differentials for shift work. Journal of Political Economy, Vol. 98, pp. 1055–1075.

- Laffont, J. and Martimort, D. (2002). The theory of incentives: The principal-agent model. Princeton, NJ: Princeton University Press.

(29)

29

- Liberty, S. E. and Zimmerman J. L. (1986). Labor Union Contract Negotiations and Accounting Choices. The Accounting Review, Vol. 61, pp. 692-712.

- Lopez, T. and Rees, L. (2002). The effect of beating and missing analysts' forecasts on the information content of unexpected earnings. Journal of Accounting, Auditing & Finance, Vol. 17, pp. 155-184. - Matejka, M., Merchant, K. & Van der Stede, W. (2009). Employment horizon and the choice of

performance measures: Empirical evidence from annual bonus plans of loss-making entities. Management Science, Vol. 55, pp. 890 – 905.

- Meek, G., Rao, R. & Skousen, C. (2007). Evidence on factors affecting the relationship between CEO stock option compensation and earnings management. Review of Accounting and Finance, Vol. 6, pp. 304-23. - Mehran, H. (1995). Executive compensation structure, ownership, and firm performance. Journal of Financial Economics, Vol. 38, pp. 163–184.

- Mishkin, F. S. (1992). Anatomy of a financial crisis. Journal of Evolutionary Economics, Vol. 2, pp. 115 – 130.

- Murphy, K. J. (1999). Executive Compensation, Handbook of Labor Economics, pp. 2485-2563.

- Opler, T.C., Titman, S. (1994). Financial Distress and Corporate Performance. Journal of Finance, Vol. 49, pp. 1015–1040.

- Roychowdhury, S. (2006). Earnings management through real activities manipulation. Journal of Accounting and Economics, Vol. 42, pp. 335-370.

- Skinner, D. and Sloan, R. (2002). Earnings surprises, growth expectations, and stock returns or don’t let an earnings torpedo sink your portfolio. Review of Accounting Studies, Vol. 7, pp. 289-312.

- Sloan, R. (1996). Do stock prices fully reflect information in accruals and cash flows about future earnings? Accounting Review, Vol. 71, pp. 289–316.

- Stein, J. C. (1989). Efficient capital markets, inefficient firms: A model of myopic corporation behavior. Quarterly Journal of Economics, Vol. 104, pp. 655–669.

- Strobl, G. (2013). Earnings manipulation and the cost of capital. Journal of Accounting Research, Vol. 51, pp. 449 – 473.

- Walsh, P., Russel, C & Clarke, F. ( 1991). Big bath accounting using extraordinary items adjustments: Australian Empirical evidence. Journal of Business Finance & Accounting, Vol. 18, pp. 173-189.

- Watts, R. & Zimmerman, J. (1990). Positive accounting theory: a ten year perspective. The Accounting Review, Vol. 65, pp. 131–157

The financial crisis, earnings management and CEO compensation

Accountancy & Control