Amsterdam Business School
CEO compensation & Earnings
Management:
A comparison between the period before and
during the credit crisis
Name: Soufyan Akoudad Student number: 10660720 Thesis supervisor: dr. P. Kroos Date: 16-6-2016
Word count: 9,942
MSc Accountancy & Control, specialization Control
2
Statement of originality
This document is written by student Soufyan Akoudad 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
Abstract
This paper examines the relation between CEO incentive compensation and the use of earnings management, and whether the financial crisis has moderated this relation. It is hypothesized that there is a positive relation between CEO incentive compensation and the use of earnings management. Furthermore, it is also hypothesized that the credit crisis positively moderates the relationship between CEO incentive compensation and earnings management. Using a sample of 5,599 observations over a period from 2002 until 2010, evidence is provided that total incentive compensation motivates managers to manage the earnings of their firm. This is consistent with prior literature. Managers with high incentive compensation are encouraged to increase the performance of the company. However, no evidence is provided that the credit crisis positively moderates the relationship between CEO incentive compensation and earnings management. This is in contradiction with prior literature. Prior literature has documented that the credit crisis may stimulate managers to use income-increasing accruals to meet earnings benchmarks or to use income-decreasing accruals to defer income to later periods.
4
Table of contents
Statement of originality _____________________________________________________________ 2 Abstract _________________________________________________________________________ 3 1. Introduction __________________________________________________________________ 5 1.1 Background _________________________________________________________________ 5 1.3 Research question ____________________________________________________________ 5 1.2 Motivation and contribution _____________________________________________________ 6 1.4 Structure ___________________________________________________________________ 6 2. Literature review and hypothesis __________________________________________________ 7 2.1 CEO compensation ___________________________________________________________ 7 2.1.1 Separation of ownership and control __________________________________________ 7 2.1.2 Moral hazard and adverse selection __________________________________________ 8 2.1.3 (Incentive) compensation ___________________________________________________ 8 2.2 Financial disclosure and earnings management ____________________________________ 10 2.3 Credit crisis ________________________________________________________________ 12 3. Research methodology ________________________________________________________ 14 3.1 Sample selection ____________________________________________________________ 14 3.2 Empirical model _____________________________________________________________ 14 3.1.1 Main model _____________________________________________________________ 14 3.1.2 Auxiliary model __________________________________________________________ 15 3.3 Measurement of control variables _______________________________________________ 16 4. Results _____________________________________________________________________ 17 4.1 Descriptive statistics _________________________________________________________ 17 4.2 Main analysis _______________________________________________________________ 18 4.3 Robustness analysis _________________________________________________________ 20 5. Conclusion __________________________________________________________________ 23 References _____________________________________________________________________ 245
1. Introduction
1.1 Background
The world is shocked by different scandals, like frauds of Enron and Ahold that follow from the separation of ownership and control. The owners of firms rely on executives, who have a very small share of the firm they manage, to make important decisions (e.g. investment decisions payouts). When the personal financial stake of an executive is relatively unaffected by the value of a firm, there is a possibility that the executive acts in ways that is privately beneficial, while the value to the investors will decrease. These agency problems arise from the separation of control and ownership. There is a conflict of interest as executives aim to increase their own personal wealth, while
shareholders aim to increase the market value of the firm. This means that the interests of the
shareholder and the executive are not aligned, which could decrease the wealth of the shareholder as executive managers may take advantage of their private information at the expense of the shareholder (e.g. misappropriate corporate resources).
Companies often make use of (incentive) compensation plans as a means to align the interests of executive managers with the interests of the shareholders. However, given the inherent discretion in accounting, agency problems may also manifest itself in CEO compensation, since CEOs are
challenged to maximize their income. A good example is presenting financial statements more positive than warranted by the underlying economies. This is called income increasing earnings management. Previous literature (e.g. Healy, 1985) shows that CEO compensation is positively associated with earnings management when the compensation of the CEO depends on the financial performance of the organization. According Healy (1985) CEOs use accruals to maximize their bonus payments, which is called the bonus hypothesis. With respect to bonus, it is often suggested that CEO
compensation contracts are one of the causes of the financial crisis (Fahlenbrach & Stulz, 2011; Tung & Wang, 2011). According to Chang and Chen (2011) CEOs of banks had to engage in high risk taking behavior to be eligible for a bonus. Fahlenbrach and Stulz (2011) argue that the structure of CEO compensation contracts during the financial crisis were not aligned with the interest of the shareholders. The contracts were, according to the authors, not linked to the long-term performance of banks.
More importantly, several studies shows that earnings management in the banking industry is influenced by the economic cycle (Bikker & Metzemaker, 2005; Liu & Ryan, 2006; El Sood, 2012). Banks increase their provisions in times of economic health, while banks might draw on their
provisions in time of economic crisis or use income-decreasing earnings management to shift income to future periods. Liu and Ryan (2006) explain this with income smoothing, which means that banks try to reduce the fluctuations in their reported earnings numbers. Not much research examined this relationship for a broader cross-section of firms.
1.3 Research question
The aim of this paper is to investigate how the credit crisis has affected the relation between CEO compensation and earnings management. So, the relation between CEO compensation and earnings
6
management will be observed before and during the credit crisis. Therefore the research question of this paper will be:‘How does the credit crisis influences the relation between CEO incentive compensation and earnings management?’
In this thesis CEO compensation will be split into different components: cash-based components and an equity-based component as they may provide different incentives for earnings management. The main goal of this research is to investigate whether the financial crisis has moderated the relation between the different CEO compensation components and earnings management.
1.2 Motivation and contribution
CEO compensation has increased significantly in the last decades, making CEO compensation a controversial and much discussed topic in both the media and scientific papers. Furthermore, prior research has documented that CEO compensation is significantly associated with earnings
management (e.g. Jiang et al. 2010). This study contributes to this stream of literature by documenting how the credit crisis moderates the relation between CEO compensation and earnings management.
In addition, there has emerged a literature on economic downturns. The recent credit crisis is according to Brunnermeier (2009) viewed as the worst financial crisis since the great depression in the 1930’s. The financial crisis has a great impact, much more than that anticipated by many. There are trillions of dollars losses, high unemployment rates and several company bankruptcies. Prior literature has documented how banks manage earnings downward during prospective times and manage earnings upward during recessions. This thesis contributes by re-examing this earnings management behavior not solely for financial institutions, but instead for a broader sample of firms.
Lastly, this study also has a societal contribution as the earnings management behavior is important information for capital market participants. This enables the capital market participants to anticipate earnings management behavior and to correct for this behavior in their decision making process. This study could also contribute to the ethical discussion about the compensation of top-managers. It is, according to critics, not ethical to give high compensations to managers in times of economic downturns, while there are lot of people who live below poverty line. This study will contribute to this ethical discussion by showing what are the effects of CEO compensation on their reporting during a crisis.
1.4 Structure
The remainder of this thesis is organized as follows. The second chapter discusses the prior literature on CEO compensation, earnings management, and the credit crisis. Chapter two also discusses the hypothesis development. In the third chapter the research design used for this paper will be explained. This will include a description of the sample selection, empirical models and variable measurement. The fourth chapter will present the results. The last chapter, chapter five, will discuss the conclusion and research limitations.
7
2. Literature review and hypothesis
2.1 CEO compensation
2.1.1 Separation of ownership and control
The separation of ownership and control can be defined relative to the owner managed firm. In this situation, the owner makes management decisions and has a claim to the profits of the firm, which means that there is no separation of ownership and control. Separation of ownership and control arises in publicly held corporations. Shareholders own in publicly held corporations residual claims but have little or no direct control over management decisions (Marks, 1999). They, for example, do not engage in the day-to-day management of the firm, neither to direct policy or to set compensation. Consequently, managers of publicly held corporations possess direct control but have relatively small (or none) claim to the profits of the firm.
Several benefits could be derived from the separation of ownership and control. Marks (1999) sees the benefits of separating ownership and control coming from the interaction of three factors. First, under certain conditions and for certain types of decisions, hierarchical decision making may be more efficient than market allocation. In cases of imperfect contracting, market costs can be very high. Therefore, hierarchical decision making can be more efficient. Second, due to economies of scale in both production and decision making, optimal firm size can be large. This is because of economies of scale in production and decision making which is associated with goal congruence of shareholders and management. Finally, shareholders who are not engaged with management tasks, are able to develop an investment strategy where they will be able to diversify and pool. This makes it possible for the shareholders to change their allocations in response to changing market conditions. The synergy from these three factors would give an publicly held company a competitive advantage over other corporate structures (Marks, 1999). Another benefit of separating ownership and control is that
managers are specialized workforce that owners can hire, if they are unable to provide higher marginal product themselves. According to Fama and Jensen (1983) better decisions can be achieved for the organization by delegating decision functions to agents who have relevant specific knowledge, rather than allocating all decision management and control to the residual claimants.
Separation of ownership and control also has several disadvantages. Marks (1999) says that separating ownership and control may be costly. He describes that managers of organizations with a high degree of separation would experience a lack of incentives to run the organization efficiently and make it more profitable. In contrast, they would rather run the organization less efficiently and cash more bonuses. Nor et al. (2010) believe that agency problems may arise from the separation of ownership and control. Agency problems are one of the most distinctive implications of ownership diffusion. An agency problem arises when the goals of management are not in line with those of the shareholders, or when both management and shareholder display different risk taking behavior and because of different risk tolerances, they may take different actions. According to prior studies this might lead to having excessive perk consumption (Jensen and Meckling, 1976), allocating not enough effort towards shareholders’ interest (Fama and Jensen, 1983) and making sub-optimal decisions (Farma and Jensen, 1983). A cause of agency problems is the difference in the degree of information
8
between the managers, shareholders and creditors. Given that executive managers run the firm on daily basis, they have informational advantage over corporate outsiders such as investors and lenders. This is known as information asymmetry. This asymmetry of information could specifically create two problems: moral hazard and adverse selection.2.1.2 Moral hazard and adverse selection
A principal agent problem that occurs in publicly held firms is the moral hazard problem. According to Page (1997) and Dechow (1991) moral hazard problems arise when owners of a company cannot directly control the actions of the manager. Managers have private information about their actions. They may take actions that benefit themselves at the expense of shareholder. An example of a moral hazard problem is described by Dechow and Sloan (1991). They found evidence that CEOs reduce their firm’s research and development (R&D) spending as they approach retirement. Because a retiring CEO cannot be disciplined by the effects of his decisions on future compensation, R&D spending cuts just prior to his retirement increase the proceeds from CEO compensation plans at the expense of shareholders.
Adverse selection is another principal agent problem. This problem of adverse selection arises due the lack of information concerning the value of the managers characteristics (Page, 1991). For example, managers have private information about characteristics such as whether their abilities match the requirements of the firm. If not, it is likely that the managers will not perform well. According to Farrell and Whidbee (2003) bad performance lead to an increased likelihood of CEO turnover. The likelihood of getting fired increases if a CEO is not able to meet or beat the earnings forecasts as this increases the likelihood that a firm may benefit when they replace the incumbent CEO.
2.1.3 (Incentive) compensation
Using executive compensation plans may be in the shareholder’s best interest as a means to align management’s incentives with their incentives (Jensen & Murphy, 1990). These plans incentivize managers to exert more effort and take appropriate risks, which should lead to an increase in firm value (Indjejikian, 1999). According to Scot (2009) an executive compensation plan is an agency contract between the company and its manager that attempts to align the interests of shareholders and managers by basing the manager’s compensation on one or more measures of the manager’s performance in operating the company. A compensation plan ensures that the effort of the manager is aligned with the goals of an organization. A manager’s pay is generally based on a fixed part and an variable part. Jensen and Murphy (1990) say that the variable part aligns specific incentives between the manager and the firm. The variable part incentivizes the manager to exert more effort and take more risks, because the variable part leads the manager to a higher bonus (Gibbons, 1998).
The first step in this process is measuring the effort of the manager. The principal needs to determines certain measures which could measure the effort of the manager. Prendergast (2002) says that effort can be measured in two ways: measuring the input and output. An example of measuring the input is the amount of hours worked of specific aspects by the manager. An example of measuring the output is the amount of targets achieved by the manager. The amount of specific knowledge the
9
agent possesses determines the use of input or output measures (Hwang et al., 2009). The measured effort leads to the manager’s compensation.The resemblance of the manager’s current performance using proper performance measures is an important factor in the compensation plan (Indjejikian & Nanda, 2002). Performance measures such as earnings or customer satisfaction incorporate information about whether the managers have taken the desired actions and whether they are sufficiently able. The agency theory is according to Indjejikian (1999) based on the trade-off between incentives and risk. Managers are risk averse, which means that compensation contracts needs to make sure that providing more effort results in a higher payment to compensate the risk induced upon the manager (Indjejikian, 1999). These two
components are the fundamentals for designing compensation contracts. Performance measures are not perfect in the sense that they also incorporate the impact of exogenous factors (e.g. economic climate) and that sometimes an increase in the performance measure does not always correspond with an increase in firm value. There are four criteria for the evaluation of performance measures: (i) to what extent can executives influence performance measures (sensitivity); (ii) to what extent are measures influenced by other factors (noise); (iii) to what extent does an increase in measure leads to increase in firm value (congruence); and (iv) to what extent is a measure susceptible to manipulation. Typically, performance measures are more prominently used in incentive contracts when measures are more sensitive, less noisy, more congruent, and less susceptible to manipulation.
In the academic literature there is a distinction made between different compensations compenents (Merchant and Van der Stede 2007): fixed salaries, short-term incentives and long-term incentives. The fixed salary is the only element of compensation that does not depend on
performance. Managers typically desire a certain fraction of their compensation to be fixed. Managers that are risk averse have a relatively larger part of fixed salary as part of their total compensation (Murphy, 1999).
The annual (cash) bonus is an example of a short-term incentive. Nowadays most managers have a bonus plan that contains an individual performance portion (Murphy, 1999). Managers can achieve these bonuses through attaining predetermined targets. This can be both financial and nonfinancial measurements. A survey of Towers Perrin (2007) shows that 65 of the 68 sample companies using only one performance measure used accounting indicators as performance
measurement. 62% of the selected companies using multiple performance measures used accounting indicators. An advantage of accounting performance measures is the low susceptibility to noise, which means that exogenous factors have relatively small impact. However, accounting measures are often criticized because of weak congruence. Accounting measures are referred as backward looking measures, because it reflect economic impact of decisions made in the past. Therefore, nonfinancial measures are often supplemented to accounting measures in executive bonus plans. Nonfinancial measures are also limited susceptible to noise and are intended to improve the congruence of the annual bonus plan because of their long term focus. However, nonfinancial measures are perceived to be susceptible to manipulation.
Equity based compensation is offered to management as a long term incentive for the firm. Mostly, this compensation can be exercised only over a period of time and is non-tradable (Cheng
10
and Warfield 2005). The degree of market performance measures in managerial compensationcontracts has increased over the years. Murphy (1999) states that stock-based compensation was the most important development in executive compensation in the 1980s and 1990s. Market performance measures are relatively sensitive measures for executives. However, market performance measures are relatively noisy because of strong impact of external factors, which means that much risk imposed on managers. Finally, market performance measures are relatively congruent if the current price is adequate reflection of true economic value of the firm.
2.2 Financial disclosure and earnings management
Firms periodically disclose information to capital market. They are mandated to do so to address the moral hazard and adverse selection problems that follow from the separation of ownership and control. Through the disclosed information, investors can verify whether the actions of executive managers are in their best interest and they can assess the future prospects of the firm to make informed decisions whether they should increase, maintain or decrease their current stock ownership. According to Bushee and Noe (2000) increased disclosure may attract institutional investors who actively manage their portfolios, reducing share price volatility and hence systematic risk. Therefore, companies might be more inclined to voluntarily disclose as much information as possible to investors in order to reduce their cost of capital.
Financial statements are created according to accounting regulations. Accounting regulation specify a minimum level of disclosure and also aims at increasing the credibility of the information by specify rules witch information should be disclosed in which periods in which form. The financial statements need to be audited by an external party to provide assurance that the information in these financial disclosures is correct. This external party is an auditor, which express his opinion on the financial records and statements. The role of the auditor is to enforce the application of proper accounting policies. Auditors may go along with the earnings management behavior of managers in order to avoid dismissal. However, Francis and Wang (2008) describe that auditor incentives change as investor protection regimes become stricter. There is a greater likelihood that client misreporting is detected and auditors are punished. According to Francis and Wang (2008) Big 4 auditors are more sensitive to the cost of client misreporting and its effect on auditor reputation and are more likely to enforce higher earnings quality as investor protection regimes become stronger.
Financial disclosures are for an important part based on accrual accounting. Accruals are to demonstrate the true performance of the firm by recording revenues and costs in the relevant period, rather than presenting the cash in- and outflows. An example of an accrual is deferred revenue which is reported when the cash flow from the sale is received before the recording of the sale. This shows that cash flows are a relative noisy measure of firm performance. To moderate these problems, generally accepted accounting principles have evolved to enhance performance measurement by using accruals to deviate from cash flows in recognizing revenues and expenses (Dechow, 1994). However, the recognition of accruals involves discretion of executive managers and is actually based on interpretation, estimates and judgment. Managers can use this discretion to signal their private information or to opportunistically manipulate earnings (Dechow, 1994). Managers can influence these
11
accruals to serve their own purposes at expense of shareholders (Graham et al, 2005). This is called opportunistic earnings management. Examples of accruals that require discretion are losses from bad debts, asset impairments and the salvage value of long-term assets. If these estimates are biased in order to affect the underlying true economic performance, opportunistic earnings management has been applied (Healy and Wahlen 1999).Healy and Wahlen (1999) reviewed the earnings management literature. 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”. According to Ronen and Yaari (2008) the definition of earnings management of Healy and Wahlen (1999) is a practice of using manipulation to misrepresent or reduce transparency of the financial reports. It has also been argued that inside managers may use their discretion (i.e., manage earnings) to signal private information to the capital market. In any case, Schrand and Zechman (2012) show how firms that are actively managing earnings (i.e. within GAAP earnings management )are more likely to engage in fraudulent activities later in time.
Earnings management can be divided into two parts: real earnings management and accrual-based earnings management. Real earnings management is managing the earnings of the firm through manipulation of real activities. Roychowdhury (2006) defines real activities manipulation as management actions that mislead stakeholders into believing that earnings benchmarks have been met in the normal course of operations. These activities has an effect on the cash flows of the firm. An example is to postpone a certain expense regarding for example R&D. According to Roychowdhury (2006) accrual-based earnings management has no direct cash flow consequences. Accruals can be divided in discretionary and non-discretionary accruals. Discretionary accruals are concerned with decisions of executives to manage earnings but cannot be explained from normal activities. Non discretionary accruals can be explained from normal activities. Firms typically depreciate more if they have more assets in place. Cohen et al. (2008) exposed that accrual-based earnings management declined significantly since the passage of SOX. However, firms continued to manage their earnings by adopting real earnings management instead. A reason that managers adopt real earnings
management over accrual-based earnings management is the lower probability of drawing auditor or regulatory inspection. When it is properly disclosed in the financial statements, real earnings
management cannot influence the auditors’ opinion (Kim et al. 2010).
Graham et al. (2005) describe several reasons and motives to manage earnings. One of the most important reason is meeting benchmarks. According to Graham et al. (2005) managers have four main 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. Furthermore, most CFOs have a preference for smooth earnings. Graham et al. (2005) describe that one of the benefits of smooth earnings is that the prediction of future earnings will be easier. Less predictable earnings command a risk premium in the market. Healy and Wahlen (1999) describe that companies are punished by capital markets if they do not meet forecasts of analysts. To meet these forecasts, managers are forced to manage earnings. Besides serving the
12
interests of the firm, managers use earnings management to serve their personal incentives (Graham et al., 2005). For example, Healy and Wahlen (1999) describe that managers use earningsmanagement to increase the proceeds from their compensation contracts.
The relationship between executive compensation and earnings management has been researched extensively. Gao and Schrieves’s (2002) show strong evidence that earnings management is influenced by compensation contracts. They found evidence that annual bonuses and stock options are positively related to earnings management. A similar kind of research is done by Cheng and Warfield (2005). This research examined the relation between CEO equity compensation and earnings management. They demonstrate that managers with high equity incentives prefer not to report large positive earnings surprises, because the wealth of the manager is influenced by future stock
performance. Finally, Bergstresser and Philoppon (2006) provide evidence that in companies where CEOs compensation is more closely linked to the value of stocks and options, the use of discretionary accruals to manipulate earnings increases. In sum, managers with high incentive compensation are encouraged to increase the performance of the company. This motivates managers to care about short-term stock prices and to manage the reported earnings of the company. Therefore, the first hypothesis is as follows:
H1: There is a positive relation between CEO incentive compensation and the use of earnings management.
2.3 Credit crisis
The credit 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). There has been a massive growth in subprime mortgage contract around 2002 (Chang & Chen, 2011). These mortgages allowed banks to provide a mortgage to purchase a house, even though the borrower has a low credit score, little documentation of his income and/or a small down payment. Home owners had because of these loans the chance to refinance loans and withdraw cash from houses that had appreciated in price. The risk for the banks of not being fully repaid increased through these loans (Kiff & Mills, 2007). This went wrong because house prices decreased. As a result, financial institutions suffered huge losses, downsized or went bankrupt (Chang & Chen, 2011).
The basis for the credit crisis was set in the years 2005 and 2006. In these years the first indications of the downfall of the economy were visible. There was an abrupt end of the housing boom and the market for subprime mortgage was slowing down. In 2007 many subprime mortgage lenders went bankrupt because of the losses they made, or needed help to continue their business. According to Fiechter and Meyer (2010) the credit crisis started in 2008, because the crisis in the US subprime mortgage market spread to other countries. The fall of the Lehman Brother bank in 2008 was an event that caused worldwide financial panic. This bank was the fourth-largest investment bank in the United States. Confidence in the banking industry dropped heavily. According to the well-known economist Alan Blinder no financial institution seemed safe after the collapse of Lehman Brothers.
13
Campello et al. (2009) provided characteristics of the credit crisis. Their study showed that more than half of the CFOs that were surveyed canceled or postponed their planned investments. 86% of the constrained US CFOs said that their investment in attractive projects was restricted during the credit crisis of 2008 (Campello et al., 2009). During a credit crisis firms reduce their costs and cash outflows, which have negative effects for the government and environment. Firms postponedexpenses regarding tech, employment and capital at the expense of long term value (Campello et al., 2009). Because of the bad financial performance of companies investors lost their money. Investors became fearful of losing more money, which resulted in asset sales or money withdraws from the savings accounts. This behavior of investors had catastrophic consequences for countries, which resulted in mistrust in financial institutions and organizations. As a result, companies and countries became financial distressed.
Previous literature has developed arguments why managers are more likely to manipulate earnings in periods of economic crisis. Burgstahler and Dichev (1997) describe that managers will manage earnings to avoid earnings decreases and losses when the incentives are contingent on earnings targets. Further, managers will manage earnings upwards during a credit crisis to meet benchmarks. This view is shared by Farrell and Whidbee (2003). They argue that it will be more difficult to meet benchmark during a credit crisis because of the bad environmental conditions. As a result that managers will manage the earnings to meet earnings targets and to secure their jobs.
However, previous literature also found arguments why managers would engage in income-decreasing earnings management during periods of economic crisis. Walsh et al. (1991) describe big bath accounting as such an argument. Big bath accounting is a negative outlier in an entity’s growth in reported net profit after tax, extraordinary items and minority interest (Walsh et al., 1991). The market tolerates, according to big bath accounting, bad performance of companies in periods of economic crisis. Therefore managers will be incentivized to manage earnings downwards through the accruals. By managing earnings downwards in periods of economic crisis, they reserve earnings for periods after the crisis. This improvement of performance after the crisis will benefit the reputation of the manager. Hayn (1995) provide another argument why managers would engage in income-decreasing earnings management in periods of economic crisis. She describe that in periods of economic crisis, financial reporting suffers because losses are less informative than profits about future performance. The financial statements of companies will add less value to external parties because the influence of external factors. Therefore shareholders and stakeholders will pay less attention to financial
statements in periods of economic crisis. This incentivizes managers to manage earnings downwards and reserve current earnings to increase future performance.
Based on the arguments that the credit crisis may stimulate managers to use income-increasing accruals to meet earnings benchmarks or to use income-decreasing accruals to defer income to later periods, the second hypothesis is formulated as follows:
H2: The credit crisis positively moderates the relationship between CEO incentive compensation and earnings management.
14
3. Research methodology
3.1 Sample selection
The initial sample consists of all firms included in the Execucomp database from 2002 to 2010. The Execucomp database roughly corresponds with the S&P1500. The S&P1500 combines three leading indices, the S&P500, the S&PMidCap400 and the S&PSmallCap600. These indices cover
approximately 90% of the United States market capitalization. The total sample consists of ‘pre-crisis years’ and ‘crisis years’. The ‘pre-crisis years’ are observations of the years 2002 until 2007, while the ‘crisis years’ consists the years 2008 until 2010. Previous literature shows that the year 2008 is the year in which the credit crisis started (French, Leyshon & Thrift, 2009; Fiechter & Meyer, 2010; Erkens, Hung & Matos, 2012).
For this study the data concerning CEO incentive compensation will be collected from Execucomp. The initial sample of this study consists of 17,619 firm-year observations. For the estimation of discretionary accruals and control variables the data will be collected from Compustat. Incomplete and missing data from Compustat leads to the exclusion of 10,912 observations. Financial organizations with SIC codes between 6000 and 6999 and utilities with SIC codes between 4500 and 4999 will be excluded from the sample. This leads to the exclusion of 1,108 observations. Finally, all variables at the 1st and 99th percent of distribution are winsorized to address any potential impact of outliers. The final sample consists of 5,599 observations.
3.2 Empirical model
3.1.1 Main model
The following empirical model will be used to test the hypotheses:
|EM| = β0 + β1*INC_COMP+ β2*CRISIS + β3*INC_COMP*CRISIS + CONTROLS + ε
This study will focus on accrual based earnings management which is indicated by |EM|. |EM| indicates the absolute value of accrual based earnings management as earnings management may involve income-increasing or income-decreasing earnings management. The estimation of accrual based earnings management will be further described in section 3.1.2. The degree of incentive compensation is indicated by INC_COMP. Further, CRISIS is a dummy variable equal to 0 for the ‘pre-crisis years’ and equal to 1 for the ‘‘pre-crisis years’. The relation between incentive compensation and earnings management in the ‘pre-crisis years’ is represented by β1, while the sum of coefficients β1+ β3 represents the relation between incentive compensation and earnings management in the ‘crisis years’. On the basis of the first hypothesis, I expect β1 is greater than 0. The second hypothesis can be tested by looking at β3. That is, if the credit crisis positively moderates the relationship between CEO compensation and earnings management, the value of β1 + β3 must be larger than the value of β1. So, hypothesis 2 is supported if β3 is greater than 0.
As stated in section 2.1.3 incentive compensation can be divided into two components: the annual (cash) bonus incentive (short-term) and the equity based incentive (long-term). Gao and
15
Shrieves (2002) describe that the size of a bonus depends on the outcomes of the contractedperformance measures, the performance standards, and the structure of the pay-performance relation. Here, the degree of incentives is measured by looking at the realized (ex-post) compensation. The main measure of this research is the total incentive compensation. This is formulated as follows:
INC_COMP = Amount of bonus + Equity flow / Total compensation
I will also decompose this further by looking separately at bonus and equity incentives. The definition for the bonus incentives will be as follow:
INC_COMP (BONUS) = Amount of bonus / Salary + Bonus
Equity based incentives of CEOs will be measured by looking at the grants of shares and options awarded in a specific year. The definition of the equity-based compensation is as follow:
INC_COMP (EQUITY) = Equity flow / Total compensation
3.1.2 Auxiliary model
Several models can be used to measure earnings management. According to Dechow et al. (1995) the Modified Jones Model provided the strongest empirical test of detecting earnings management. Therefore this study will use this model. The Modified Jones Model measures earnings management as follow:
TAi,t = Total accruals calculated as IBCi,t – (OANCFi,t – XIDOCi,t) of firm i in year t
IBCi,t = Income before extra items of firm i in year t
OANCFi,t = Net cash flow of operating activities of firm i in year t
XIDOCi,t = Cash flow of extraordinary items and discontinued operations of firm i in year t
Assetsi,t-1 = Total assets of the preceding year of firm i in year t
ΔREVi,t = Change in revenues from the preceding year of firm i in year t
ΔARi,t = Change in account receivables from the preceding year of firm i in year t
PPEi,t = Gross value of property, plant, and equipment of firm i in year t
The total level of accruals will be estimated by subtracting income before extra items (IBC) from the net cash flow of operating activities (OANCF), cash flow of extraordinary items and discontinued operations (XIDOC). The Modified Jones Model will be used to distinguish between discretionary and non-discretionary accruals, and measures earnings management as the absolute value of the residual
16
(the discretionary accruals). The residual will be retrieved by performing a regression analysis of the auxiliary model per 2-digit SIC code. Regressions are performed when there are at least 10observations for each industry consistent with prior literature (e.g.,Cheng and Warfield, 2005).
3.3 Measurement of control variables
Control variables will be used in this research to control for possible alternative explanation for the findings. The control variables used for this research are: leverage, market-to-book ratio, firm size, accounting performance (ROA) and market performance. Prior research have found relations between earnings management and these variables (e.g., Bergstresser and Philippon, 2004; Gao and Shrieves ,2002).
According to Kostiuk (1990) compensation policy is an important variable for dealing with financial distress. Therefore leverage will be used as control measure for this research, because leverage is an often used proxy for financial distress. Leverage is measured by the ratio between equity and debt within a firm. Bergstresser and Philippon (2004) describe that the market-to-book ratio is a proxy for the growth options of a firm that may also affect compensation design choices. Market-to-book ratio is defined as the market value of equity divided by the book value of equity. Furthermore, the natural logarithm of total assets will be included as an indicator for the variable firm size. According to Watts and Zimmerman (1990) there is a negative association between the levels of earnings
management and the size of firms, since larger firms are under more scrutiny of the outside world. Also firm performance will be used as a control variable. To control for differences in performance, the yearly return on assets measured by earning divided by lagged total assets (ROA) will be included (Dechow et al., 1995). Finally, also market performance is included defined as the end of year market value minus the beginning of year market value, divided by beginning of year market value. I control for industry effects by including two-digit SIC indicators.
17
4. Results
4.1 Descriptive statistics
Table 1: Descriptive statistics Variables N mean median sd p25 p75 EM 5599 0,079 0,043 0,131 0,019 0,085 INC_COMP_BONUS 5599 0,438 0,499 0,252 0,278 0,636 INC_COMP_EQUITY 5599 0,428 0,480 0,266 0,233 0,639 INC_COMP 5599 0,657 0,739 0,241 0,556 0,828 CRISIS 5599 0,627 1,000 0,484 0,000 1,000 LEV 5599 0,171 0,153 0,156 0,008 0,274 MTB 5599 2,397 2,042 1,499 1,358 3,096 LNSIZE 5599 7,274 7,160 1,513 6,176 8,270 ROA MARKETPERF 5599 5599 0,050 0,141 0,059 0,059 0,104 0,619 0,016 -0,228 0,099 0,365 In this section the descriptive statistics will be discussed. Table 1 shows that the mean of the
discretionary accruals (EM) is more or less the same as in previous literature (Gao & Shrieves, 2002; Meek et al. 2007). Furthermore, the total compensation of CEOs consists for about 66% of incentive compensation. On the other hand, equity incentives has a percentage of 43% which indicates that the majority of the incentives originate from equity incentives opposed to bonus incentives. 63% of the firm-year observations are from 2008 onwards.1 The mean of leverage is 0,171, which indicates that firms finance their assets with a small part of debt. The market-to-book ratio has a value of 2,397. This indicates the average growth of the firms and suggests a high availability of growth opportunities. The variable size is a natural log number of total assets. The mean of 7,274 corresponds with an average firm size of 1442 million dollars. The ROA measures the performance of a firm and is around 5% in this sample. Finally, the mean market performance is 0,141 which indicates that the market value on average increased by 14% over sample period.
Table two gives an overview of the Pearson correlations. The Pearson correlation shows the relation between the variables. Correlations with a coefficient of 1 means that two variables perfectly correlate with each other, while a coefficient of -1 means that the variables perfectly negative correlate with each other. Bold correlation coefficients are significant at the 10% level. The table shows that equity-based incentives and non-equity-based incentives are positively correlated with each other. This indicates that they are not used as substitutes but as complements. Non-equity-based incentives and earnings management are negatively correlated with each other, which means that increases in bonus incentives are associated with less earnings management. Total incentive compensation and equity-based incentives are not significantly correlated with earnings management. This preliminary evidence suggests that there is not enough evidence to state that total incentive compensation
influences earnings management. Further, crisis is negatively related with non-equity-based incentives
1
18
(-0,03), while it has a positive correlation with equity-based incentives (0,04). This means that during the credit crisis there is more focus on equity-based incentives instead of non-equity-based incentives. Finally, managers of bigger and better performing firms will manage their earnings less than managers of smaller and less performing firms.Table 2: Pearson correlation matrix
Variables 1 2 3 4 5 6 7 8 9 10 1.EM 1,00 2.INC_COMP_BONIUS -0,03 1,00 3.INC_COMP_EQUITY 0,01 0,07 1,00 4.INC_COMP -0,01 0,63 0,76 1,00 5.CRISIS 0,01 -0,03 0,04 0,03 1,00 6.LEV 0,01 0,10 0,09 0,12 -0,01 1,00 7.MTB -0,01 0,23 0,11 0,20 -0,17 0,08 1,00 8.LNSIZE 9.ROA 10.MARKETPERF -0,06 -0,12 0,00 0,40 0,33 0,21 0,32 -0,01 -0,05 0,43 0,18 0,10 0,00 -0,10 0,04 0.35 -0,21 -0,05 0,67 0,41 0,23 1,00 0,12 0,04 1,00 0,14 1,00
4.2 Main analysis
Table 3: OLS regression results H1
Variables
Model 1A Model 1B Model 1C
Pred. Intercept ,102*** (,009) ,105*** (,009) ,104*** (,009) INC_COMP + ,016* (,008) -- -- INC_COMP_BONUS + -- ,018** (,008) -- INC_COMP_EQUITY CRISIS + -- ,003 (,004) -- ,003 (,004) ,008 (,007) ,003 (,004) LEV ,000 (,012) ,000 (,012) ,001 (,012) MTB ,006*** (,001) ,006*** (,001) ,006*** (,001) LNSIZE -,006*** (,001) -,006*** (,001) -,005*** (,001) ROA MARKETPERF Industry controls Adjusted R-squared F-value -,179*** (,020) ,001 (,003) Yes 0,019 16,399*** -,188*** (,020) -,002 (,003) Yes 0,019 16,595*** -,174*** (,020) ,000 (0,003) Yes 0.018 16,056***
Coefficients and standard errors (between brackets) are reported. ***, **, * denotes 1%, 5%, and 10% significance levels respectively
19
Table 3 shows the Ordinary Least Squares (OLS) regression results for hypothesis 1. It examines whether CEO incentive compensation positively influences earnings management. Table 3 contains three models (1A, 1B, and 1C), which subsequently describes the result for the total incentives (INC_COMP), the bonus incentives (INC_COMP_BONUS) and the equity incentives(INC_COMP_EQUITY).
The main measure of this research is the total incentive compensation. Model 1A in table 3 shows that there is a positive and significant relation between CEOs’ total incentive compensation (INC_COMP) and earnings management. This is consistent with prior literature (Cheng & Warfield, 2005; Collins & Hribar, 2000; Healy, 1985). When I further distinguish between bonus incentives and equity incentives, I find that this result is driven by the bonus incentives (INC_COMP_BONUS). Model 1B in table 3 shows that there is a positive and significant relation between bonus incentives
(INC_COMP_BONUS) and earnings management. This is consistent with both the prediction and prior literature. Prior literature has documented that the main reason for earnings management is personal interest, as a result of the compensation of CEOs (Bauman & Shaw, 2006; Bergstresser & Philippon, 2006). Furthermore, model 1C in table 3 shows that the relation between equity-based incentives (INC_COMP_EQUITY) and earnings management is not significant (e.g., in contrast with the findings of Cheng and Warfield (2005) and Bergstresser and Philippon (2006)). This means that for both the total incentives and bonus incentives hypothesis 1 is supported. Finally, with respect to the control variables, the market-to-book ratio, firm size and ROA are significantly related with earnings
management. The results generally show that the market-to-book ratio is positively related, while firm size and ROA are negatively related with earnings management. Note that the crisis does not have a direct effect on accrual based earnings management. In the subsequent analysis, I will investigate whether the crisis affects (moderates) the documented relationship between CEO incentives and earnings management.
With respect to the model performance of table 3, model 1A, 1B and 1C are significant (F-value of resp. 16,399***, 16,595*** and 16,056***). However, according to the adjusted R-squared of all three models the independent variables explain only about 2% of earnings management. This means that the majority of earnings management is explained by factors not present in this model.
Table 4 shows the Ordinary Least Squares (OLS) regression results for hypothesis 2. It examines whether CEO incentive compensation positively influences earnings management and whether the credit crisis moderates this relation. Model 2A of table 4 shows that INC_COMPxCRISIS has a negative value. However, this coefficient is not significant. So, with respect to total incentive compensation, there is no evidence that the credit crisis positively moderates the relationship between CEO incentive compensation and earnings management. Models 2B and 2C of table 4 shows also an insignificant negative value for INC_COMP_BONUSxCRISIS and INC_COMP_EQUITYxCRISIS. Therefore hypothesis 2 is not supported. Consistent with my earlier results, I again find a positive and significant relation between total CEO incentive compensation and earnings management (model 2A, table 4). So, this indicates that there is a relationship between CEO incentives and earnings
20
INC_COMP*CRISIS is not significant). Furthermore, with respect to the control variables, firm size and ROA are negatively and significantly related with earnings management. This means that managers of bigger and better performing firms will manage their earnings less. This is consistent with priorliterature. According to Watts and Zimmerman (1990) there is a negative association between the levels of earnings management and the size of firms, since larger firms are under more scrutiny of the outside world. Finally, leverage and the market-to-book ratio are positively and significantly related with earnings management.
Table 4: OLS regression results H2
Variables
Model 2A Model 2B Model 2C
Pred. Intercept ,114*** (,010) ,116*** (,010) ,110*** (,010) INC_COMP + ,015* (,008) -- -- INC_COMP_BONUS + -- ,019** (,008) -- INC_COMP_EQUITY CRISIS INC_COMPxCRISIS INC_COMP_BONUSxCRISIS INC_COMP_EQUITYxCRISIS + + + + -- ,003 (,004) -,021 (,015) -- -- -- ,003 (,004) -- -,032 (,014) -- ,008 (,007) ,003 (,004) -- -- -,011 (,013) LEV ,001 (,010) ,000 (,012) ,001 (,012) MTB ,006*** (,001) ,006*** (,001) ,006*** (,001) LNSIZE -,006*** (,001) -,006*** (,001) -,005*** (,001) ROA MARKETPERF Industry controls Adjusted R-squared F-value -,170*** (,020) ,001 (,003) Yes 0,019 14,618*** -,184*** (,020) -,002 (,003) Yes 0,020 15,040*** -,174*** (,020) ,000 (0,003) Yes 0.018 14,051***
Coefficients and standard errors (between brackets) are reported. ***, **, * denotes 1%, 5%, and 10% significance levels respectively
With respect to the model performance of table 4, model 2A, 2B and 2C are also significant (F-value of resp. 14,618***, 15,040*** and 14,051***). However, according to the adjusted R-squared of all three models the independent variables explain, just like the models in table 3, only about 2%of earnings management. Also, this means that the majority of earnings management is explained by factors not present in this model.
4.3 Robustness analysis
A subsequent regression analysis is performed to assess the sensitivity of the results. It controls for outliers as these may have a strong influence on the final regression results. For this robust regression Cook’s D is used. Cook’s D identifies outliers and measures the effect of deleting a given observation. Observations with large Cook’s distance in the data might be an outlier. According to Bollen and
21
Jackman (1990) the cut-off values that should be used to spot the high influential values is 4/N, where N is the number of observations. Therefore the cut-off value that is used in table 5 is 0,0007. The results of the robust regression are summarized in table 5.Table 5: Robust regression results H2 Variables
Model 2A Model 2B Model 2C
Pred. Intercept ,088*** (,005) ,089*** (,005) ,084*** (,005) INC_COMP + ,009** (,004) -- -- INC_COMP_BONUS + -- ,012*** (,004) -- INC_COMP_EQUITY CRISIS INC_COMPxCRISIS INC_COMP_BONUSxCRISIS INC_COMP_EQUITYxCRISIS + + + + -- ,004** (,002) -,001 (,007) -- -- -- ,005** (,002) -- -,006 (,007) -- ,003 (,003) ,003 (,004) -- -- ,001 (,006) LEV -,006 (,006) -,007 (,006) -,006 (,006) MTB ,002*** (,001) ,002*** (,001) ,002*** (,001) LNSIZE -,003*** (,001) -,003*** (,001) -,003*** (,001) ROA MARKETPERF Industry controls Adjusted R-squared F-value -,125*** (,010) -,001 (,001) Yes 0,036 26,319*** -,184*** (,011) -,001 (,001) Yes 0,040 29,511*** -,128*** (,010) ,000 (0,001) Yes 0.038 27,856***
Coefficients and standard errors (between brackets) are reported. ***, **, * denotes 1%, 5%, and 10% significance levels respectively
Note that the main inferences are not affected. Model 2A in table 5 again shows that a positive relation between CEOs’ total incentives compensation (INC_COMP) and earnings management. However, this relation denotes a 5% significance level instead of the 10% significance level in the OLS regression. So, the results are slightly more pronounced in this analysis. The same goes for the positive relation between bonus incentives (INC_COMP_BONUS) and earnings management. Furthermore, the coefficients of INC_COMPxCRISIS, INC_COMP_BONUSxCRISIS and INC_COMP_EQUITYxCRISIS remain insignificant. This means that when using a robust regression there is still no evidence that the credit crisis positively moderates the relationship between CEO incentive compensation and earnings management. This is again in contradiction with both the prediction and prior literature. However, CRISIS denotes a positive and significant value in model 2A and 2B of table 5. This means that there is more accrual-based earnings management in ‘crisis years’ than in ‘pre-crisis years’ independent of the degree of incentive compensation. Finally, with respect to the control variables, the market-to-book ratio, firm size and ROA are still significantly related with earnings management.
With respect to the model performance of table 5, model 2A, 2B and 2C are significant (F-value of resp. 26,319***, 29,511*** and 27,856***). However, according to the adjusted R-squared of
22
all three models the independent variables explain only about 4% of earnings management. Again, this means that the majority of earnings management is explained by factors not present in this model.23
5. Conclusion
The aim of this paper is to investigate how the credit crisis has affected the relation between CEO incentive compensation and earnings management. Based on prior research it is predicted that (equity-based) incentive compensation will motivate managers to manage earnings. It is also expected that managers will manage earnings more during the credit crisis than before the credit crisis. This expectation is based on the arguments that the credit crisis may stimulate managers to use income-increasing accruals to meet earnings benchmarks or to use income-decreasing accruals to defer income to later periods.
Not all the findings of this research are consistent with the predictions. Using a sample of 5,599 observations over a period from 2002 until 2010, evidence is provided that total incentive compensation motivates managers to manage the earnings of their firm. However, no evidence is provided that the credit crisis positively moderates the relationship between CEO incentive compensation and earnings management. Moreover, the robustness test shows that the positive relation between total incentive compensation and earnings management is lower and stronger. The robustness test also shows that there is no evidence that the credit crisis positively moderates the relationship between CEO incentive compensation and earnings management. This is again in contradiction with both the prediction and prior literature.
This study contributes to prior literature by re-examing the earnings management behavior not solely for financial institutions, but instead for a broader sample of firms. This is 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 credit crisis on the relationship between incentive compensation and earnings management. Additionally, this study also has a societal contribution as the earnings management behavior is important information for capital market participants. This enables the capital market participants to anticipate earnings management behavior and to correct for this behavior in their decision making process.
The results of this study contain a number of limitations. First of all, the data for CEO incentive compensation is obtained from large publicly listed US companies. Which means that generalization of the findings for small and mid-sized firms is limited. Further research may investigate whether these findings are also valid for companies in different institutional settings. Secondly, prior literature states that the behavior of managers may be affected by other influences. This study used proxies that are limited to bonus, grants of shares and grants of options, while elements such as responsibilities of managers, future promotions and personal relations are not taken into consideration. These other influences could be used in future research. Lastly, the adjusted R-squared of the models is very low, which means that the independent variables used for this research may not be proper. Future
research could use different independent variables to examine how the credit crisis has affected the relation between CEO incentive compensation and earnings management.
24
References
Ackermann, N. (2008). The subprime crises and its consequences. Journal of Financial Stability, Vol. 4, pp. 329-337.
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.
Bergstresser, D., Philippon, T. 2006. CEO Incentives and Earnings Management. Journal of Financial Economics, 80 (3): 511-529.
Bertrand, M. & Mullainathan, S. 2001. Are CEOs rewarded for luck? The ones without principles are. The Quarterly journal of economics, august 2001
Bikker, J.A., Metzemakers, P.A.J. 2005. Bank provisioning behaviour and procyclicality. Journal of International Financial Markets, Institutions and Money, 15 (2): 141-157.
Bollen, K.A., Jackman, R.W. 1990. Regression Diagnostics: An Expository Treatmant of Outliers and Influential Cases. Modern Methods of Data Analysis: 257-91.
Brunnermeier, M. K. 2008. Deciphering the Liquidity and Credit Crunch 2007-08. Princeton University - Department of Economics.
Burgstahler, D. & Dichev, I. (1997). Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics, Vol. 24, pp. 99–126.
Campelo, M., Graham, J.R. and Harvey, C.R. (2009). The real effects of financial constraints: evidence from a financial crisis. Journal of Financial Economics, Vol. 97, pp. 470 – 487
Chang, G. D., Chen, F. W. 2011. CEO behavior and Subprime Mortgage Crisis. The International Journal of Business and Finance Research, 7 (3): 13-25.
Cheng, Q., and T. Warfield. 2005. Equity incentives and earnings management. Accounting Review, 80: 441–476.
Cohen, D.A., Dey. A., and Lys. T.Z. 2008. Real and Accrual-based earnings management in the Pre- and Post-Sarbanes-Oxley Perio. The Accounting Review , 83 (3), 757-787.
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.
25
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.
Dechow, P.M., and R.G. Sloan. (1991). Executive incentives and the horizon problem: An empirical investigation. Journal of Accounting and Economics, vol. 14, pp. 51-89.
El Sood, H.A. 2012. Loan Loss Provisioning and Income Smoothing in US Banks Pre and Post the Financial Crisis. International Review of Financial Analysis, 25 (4): 64-72.
Erkens, D.H., Hung, M., Matos, P. 2012. Corporate Governance in the 2007-2008 Financial Crisis: Evidence from Financial Institutions Worldwide. Journal of Corporate Governance, 18,(2) 207-412.
Fahlenbrach, R., Stulz, R.M. 2011. Bank CEO Incentives and the Credit Crisis. Journal of Financial Economics, 99: 11-26.
Fama, E.F., Jensen, M.C. 1983. Separation of Ownership and Control. Journal of Law and Economics, Vol. 26, June.
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
Francis, J. R., Wang, W. 2008. The Joint Effect of Investor Protection and Big 4 Audits on Earnings Quality around the World. Contemporary Accounting Research, Vol. 25 No. 1.
French, S., Leyshon, A., Thrift, N. 2009. A very geographical crisis: the making and breaking of the 2007–2008 financial crisis. Cambridge Journal of Regions, Economy and Society, 2 (2): 287-302 Gao, P., Shrieves, R. 2002. Earnings Management and Executive Compensation: A Case of Overdose of Option and Underdose of Salary? Working paper, Northwestern University, University of Tennessee at Knoxville.
Gibbons, R. 1998. Incentives in organizations. The journal of economic perspectives, Vol. 12, No. 4, 115-132.
26
Graham, J., C. Harvey & Rajgopal, S. (2005). The economic implications of corporate financialreporting. Journal of Accounting and Economics, 40 (1-3), 3–73.
Hayn, C. (1995). The information content of losses. Journal of Accounting and Economics, Volume 20, pp. 125-153.
Healy, P.M. 1985. The impact of bonus schemes on the selection of accounting principles. Journal of Accounting and Economics, 7 (1-3): 85-107.
Healy, P. M. & Wahlen, J. (1999). A review of the earnings management literature and its implications for standard setting. Accounting Horizons, 13 (4), 365–383
Hwang, Y. & Erkens, D. H. & Evans III, J. H. 2009. Knowledge sharing and incentive design in production environments: Theory and evidence. The accounting review, Vol. 84, No. 4, 1145-1170.
Indjejikian, R.J. 1990. Performance evaluation and compensation research: An agency perspective. Accounting Horizons, 13 (2): 147-157.
Indjejikian, R. J. & Nanda, DJ. 2002. Executive target bonuses and what they imply about performance standards. The accounting review, Vol. 77, No. 4, 793-819.
Jensen, M. Meckling, W. 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3: 305-360.
Jensen, M.C., Murphy, K.J. 1990. Performance Pay and Top-Management Incentives. Journal of International Financial Economics, 98 (2): 225-264.
Jiang, J., Petroni, K.R., Wang, I.Y. 2010. CFOs and CEOs: Who have the most influence on earnings management? Journal of International Financial Economics, 96: 513-526.
Kiff, J., Mills, P. (2007) ‘Money for nothing and checks for free: recent developments us U.S. subprime mortgage markets’, IMF Working paper07/188.
Kim, B., L.L. Lisic, and M. Pevzner (2010). Debt covenant slacks and real earnings management. Working Paper George Mason University .
Kostiuk, P. (1990). Compensating differentials for shift work. Journal of Political Economy, Vol. 98, pp. 1055–1075.
27
Liu, C.C., Ryan, S.G. 2006. Income Smoothing over the Business Cycle: Changes in Banks'Coordinated Management of Provisions for Loan Losses and Loan Charge-Offs from the Pre-1990 Bust to the 1990s Boom. The Accounting Review, 81 (2): 421-441.
Marks, S.G. 1999 . Separation of Ownership and Control. Encyclopedia of Law and Economics, chapter no. 5630.
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.
Prendergast, C. 2002. The Tenuous trade-off between risk and incentives. Journal of political economics, Vol. 110, No. 5, 1071-1102.
Ronen, J. and V. Yaari (2008). Earnings management, emerging insights in theory, practice and research. Springer Science Business Media, .
Roychowdhury, S. (2006). Earnings management through real activities manipulation. Journal of Accounting and Economics, Vol. 42, pp. 335-370.
Schrand, C. M., Zechman, S. I. C., 2012. Executive overconfidence and the slippery slope to financial misreporting. Journal of Accounting and Economics, 53, 311-329.
Scott, W.R. 2009. Financial accounting theory. Fifth edition, Pearson. ISBN-10: 0-13-207286-6.
Stolowy, H., Breton, G.2004. Accounts Manipulation: A Literature Review and Proposed Conceptual Framework. Review of Accounting and Finance, 3 (1): 5-66.
Strobl, G. (2013). Earnings manipulation and the cost of capital. Journal of Accounting Research, Vol. 51, pp. 449 – 473.
Tung, F., Wang, X. 2011. Bank CEO’s, inside debt compensation, and the gobal financial crisis. Boston Univ. School of Law Working Paper, No. 11-49.
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.
