Executive compensation and real earnings management in listed companies

(1)

Student name Xiaoshu Shao

Thesis title

Executive Compensation and Real Earnings

Management In Listed Companies

Student number 10828877

Date of final version June 22, 2015

Word count 10，069 words

MSc Accountancy & Control, variant Accountancy

Amsterdam Business School

Faculty of Economics and Business, University of Amsterdam

Supervisor Mr. Wim Janssen

(2)

This document is written by student Xiaoshu Shao 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)

Listed Companies

Abstract

I show that executive compensation is associated with real earnings management. This association varies according to the component of the structured compensation. The compensation that acts as short-term incentive for executives is positively related to real earnings management. As for the compensation that is tied to firm performance in the long run, the association is negative. The results confirms the way of measuring real earnings management, and support principal and agent model.

(4)

1 INTRODUCTION

I

n this paper, I examine the association between executive compensation and real earnings management. Prior literature suggests that short-term executive

compensation，such as cash compensation, motivates executives to engage in

activities lead to real earnings management. While long-term compensation, such as equity-based compensation is tied to firm performance in the long run, and this kind of compensation constrains real earnings management. (Jensen and

Meckling ,1976,Healy 1985) Consistent with this view, I take cash bonus as short-term incentive for executive and take shares owned by executives as the long-term indication. As for the measure of real earnings management, I take both three individual indications of real earnings manipulation and overall index of real earnings management. I suggest that short-term and long-term compensation play different roles in incentives that motivate executives to manipulate real activities. Thus, I expect that bonus compensation of executives is positively associated with real earnings management while shares owned by executives are negatively associated with real earnings management.

Piror literature demonstrates that there are two ways of manipulate earnings: accrual-based and real earnings management. (Roychowdhury 2006; Cohen et al. 2008; Zang 2007). In most early research, accrual-based earnings were the main focus. (Jones 1991; Teoh et al. 1998) And in more recent literature, real earnings

management has also attained much attention.(Roychowdhury 2006; Cohen et al. 2010). Since real earnings management has negative impact on firms’ future cash

(6)

flows and may hurt the firm value in the long run. It is recognized that compared to accrual-based earnings management, real earnings management result in more long-term costs for shareholders to bear. (Roychowdhury 2006; Cohen et al. 2010). This kind of larger long-term cost is originated from price discounts in a limited period, more lenient credit term, less valuable investment in research and

development, less SG&A activities, or more unnecessary inventories of production (Roychowdhury 2006; Gupta et al. 2010). Cohen and Zarowin (2009) added evidence to this view that real earnings manipulation will influence the firm in long

run.However, it costs less if the firms engage in real earnings management activities. It’s because this type of manipulation draws less from auditors or from regulators (Cohen et al. 2008). Auditors and regulators will not change their ideas of the firm in the condition that information of the firm value is disclosed in the right way, even though real earnings management is included in the disclosure. (Kim et al.

2010).Therefore, it’s easy to understand that executives are inclined to manage real earnings rather that accrual-based earnings.(Roychowdhury 2006). And Zang futher developed the relevance of accrual-based and real earnings management. She

suggested in her findings that these two types of earnings management are subsitutes. Consistent with this view of accrual-based and real earnings management, if

accrual-based earnings are constrained, we can have the expectation that firms are more likely to turn to real earnings management. The findings of Ewert and Wagenhofer (2005),also supported this view, as they developed the models stating that in the case that accounting flexibility is reduced, firms will manipulate earnings through real activities.

The importance of compensation was suggested in a large amount of prior literature. Consistent with agency theory, it is stated that executives have incentives to

manipulate earnings to obtain more interests (Jensen and Meckling ,1976). Therefore, executives are motivated to manipulate earnings to benefit themselves regardless of the long-term interest for the firm. And if the compensation for executives is

(7)

consistent with the firm performance in the long run, executives will think more about the interests from the prospect of the firm. Thus, I focus on two components of

executive compensation：cash bonus for the executives and shares owned by

executives. I focus on a sample of 1182 observations from period 2010 to 2012. And I collected that data of the firms that likely have consistent incentives to drive

executives with both cash and shared owned. And I examine the association between real earnings management and these two types of compensation. And real earnings management is measured both on separate component level and aggregated levels. Following Roychowdhury (2006) and Cohen et al. (2008), my proxies for real earnings management consist of a firm’s abnormal cash flows, abnormal production cost (the sum of COGS and change in inventory), abnormal discretionary

expenditures, and a summary measure with these three components combined.

My primary finding is that within the collected sample, the short-term incentive, cash bonus is positively associated with real earnings management. And the long-term incentive of executives, shares owned by executives, is negatively associated with real earrings management. Collectively, my findings are consistent with my prediction that higher short-term of compensation incentive is associated with higher levels of real earnings management for firms that have strong incentives to manage earnings. In contrast, compensation of long-term incentive component is associated with lower level of earnings manipulation through real activities. This paper contributes to the literature by demonstrating that executives of firms have incentives in engaging in more or less real earnings management, and the incentives derive from different part of executive compensation. Past research has exclusively focused on accrual earnings management while more research of real earnings management should be done. My findings regard the positive association between bonus compensation, as well as the negative association between shares owned by executives and real earnings

management. And this finding add piece to literature that the structure of executive is important.

(8)

This paper proceeds as follows. The second section reviews the literature and develops my hypotheses. The third section discusses research design. The fourth section describes our methodology. The fifth section presents empirical results. Finally, the sixth section concludes.

2.Prior literature and background

2.1 Earnings management

2.1.1what is and why is earnings management

Referring to the paper of Healy&Wahlen( 1999), the definition of earnings management is: 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.Why earnings earnings management occur? We usually interpret the two types of market imperfections—information asymmetry and agency costs to formulate the basic conditions for the existence of earnings management. Dye (1988) and

Trueman & Titman (1988) point out that the existence of information asymmetry between managers and shareholders is a necessary condition for earnings

management. Schipper (1989) also supports the view that the condition for earnings management is traced to the persistence of asymmetric information. But she figures out that the blocked communication can be eliminated through the enforcement of contractual arrangement by arguing. From the perspective of a positive association between the conservatism of accounting estimates and corporate disclosure, Imhoff and Thomas (1994) concluded that firms that disclose more information are more

(9)

likely to have conservative accounting estimates (engaging in less earnings management). The second fundamental condition for the existence of earnings management is agency cost. Jensen and Meckling (1976) developed agency theory to explain the nexus between principals (owners/shareholders) and agents (managers). Principals use contracting to motivate agents who would otherwise have conflicts of interest with principals. If the conflicting interests weigh more than the provided motivation, earnings management arises then.

And prior researchers further looked into the incentives for executives to manipulate earnings. For this incentive view, opportunistic behavior and the signaling mechanism are two competing perspectives in earnings management literature. According to Guay et al (1996), earnings management is motivated by opportunism and/or performance measure improvement. J. S.Patell and Wolfson (1984) stated that by both allowing managers being "free of choice" with regard to accounting methods and compensating them accordingly; owners can capitalize upon managers' "local

expertise.” Lambert(1984) demonstrated that an important feature of this analysis is both the principal and the manager are modeled as rational parties. In particular, the principal can make predictions that how the managers will response to any

compensation scheme, and the principal then will take this predicted response into account when he determines to offer the compensation plans. According to this balance, if the principal provides the optimal compensation scheme, the manager will smooth the firm’s income subsequently. In this kind of case, smoothing income make rise of optimal equilibrium behavior. Lambert(1984) and Dye (1988) demonstrated that if a risk-averse manager is preluded from borrowing and lending in the capital markets, the manager will then has the incentive to smooth his firm’s reported income.

Another prospective of the incentives for earnings management is signaling.

Trueman* and Titman offered the opinion that if the managers are allowed o obtain private information abut the firm's future economic earnings before making his

(10)

smoothing decision, the problem will be a more complicated one. Because in this kind of situation, the reported income in current period can act as a signal to investors that how the firm’s future economic earnings will be. And this signal will affect the price at which the firm can sell its securities. This signaling prospect mainly takes the influence on different parties into account. As mentioned above, the compensation structure is complex, and the components of the structure affect the executives differently. The incentive to maximize bonus payments by managing earnings will persist if compensation contracts reward efforts to manipulating earnings (Murphy, 2001).Dye, R(1988) identifies two distinct sources of shareholders' demand for earnings management: an "internal" source, which intended to minimize the expected cost of getting a manager to adopt shareholders' preferred action; and an "external" source, based on current shareholders' desire to influence prospective investors' perceptions of their firm's value.

The incentives of earnings management, basically, can be classified as contractual and market-driven. The theory of contractual incentive is mature. About the contractual incentives for earnings management, Positive Accounting Theory (PAT) exhibits the three classical hypotheses: First is the bonus plan hypothesis, which means that managers are more likely to shift income from future periods to the current period if firms use current earnings in bonus plan. The second one is the debt covenant hypothesis, telling that managers are more likely to shift income from future periods to the current period if firms are closer to the violation of debt covenants. And also the political cost hypothesis, showing that managers are more likely to shift income from the current period to future periods if reporting high profits causes political costs. According to the agency theory of Jensen and Meckling (1976), managers are more likely to make income-increasing manipulation because their compensation is tied to earnings. Healy (1985) pointed out that a boundary exists within which mangers are more likely to choose opportunistic manipulation to meet accounting earning based bonus schemes. Like Healy, Holthausen and Larcker (1995) found evidence

(11)

consistent with the hypothesis that managers manipulate earnings downwards when their bonuses are at their maximum. Unlike Healy, they find no evidence that managers manipulate earnings downwards when earnings are below the minimum necessary to receive any bonus. Second, debt contracts also provide managers incentives to manipulate accounting numbers. The “debt hypothesis” is known as managers will manipulate through discretional exercise to avoid covenant violations. (Watts & Zimmerman, 1978). Many researchers tried to figure out the relationship between earnings management and debt covenants. According to Press and Weintrop (1990), leverage may proxy for factors in addition to the existence of accounting constraints, such as default risk or firms’ investment opportunities, or it may be measuring closeness to constraints. Related researches also include McNichols and Wilson (1988)’s finding that firms manage their earnings by choosing

income-decreasing accruals when income is extreme. Third, accounting figures should take regulations into consideration. Firms are more likely to choose

income-decreasing manipulation to reduce political costs (Watts & Zimmerman, 1978; Holthausen & Leftwich (1983), Cahan(1992) also support the political-cost

hypothesis and state that managers adjust earnings in response to monopoly-related antitrust investigations. And based on the contractual incentives, there exists the argument of efficient contracting. Scott (1997) points out that EM can also be

motivated by efficient contracting purpose. By examining the previous literature, Sun and Rath (2008) raised an interest point of the linkage between efficient contracting and signaling mechanism. Subramanyam (1996) showed that discretionary accruals is priced by the stock market, and the pricing of discretionary accruals arises because mangers use their discretion to improve the ability of earnings to reflect fundamental value. This implicates that information signaling per se is correlated with the level of contracting efficiency.

Another type of incentives for earnings management is capital market-driven motives. Many researches conclude that managers’ manipulation of earnings aims to influence

(12)

short-term stock performance. Consistent with opportunistic accruals management, Teoh, Wong and Rao (1998) found that firms inflate earnings when going public by opportunistically managing current accruals. Erickson & Wang (1999) suggested that in the quarters prior to the merger, acquiring firms manage earnings upward, because they want to exchange less shares in this way. Louis, H. (2004) also confirmed this view. There are some other researches stating that earnings benchmarks provides a strong incentive for earnings management, because failing to beat the benchmark will have significant negative impact on stock valuation (Bartov et al., 2000; Skinner & Sloan, 2002). Bartov and Tsui( 2000) asserted that earnings decreases and losses are frequently managed away. And managers avoid reporting earnings decreases and losses, in order to decrease the costs imposed on the firm in transactions with stakeholders. Burgstahler& Eames(1998) provides new evidence of actions

contributing to the phenomenon that managers avoid reporting earnings lower than analyst forecasts.

2.1.2How real earnings are manipulated

As literature on real manipulations suggested, (Roychowdhury (2004), Gunny (2005)Zang(2005), firms are more likely to engage in real manipulation, in the case that these firms have lower abnormal gains from asset sales and positive total gains from asset sales. And real earnings management often has a cash flow effect. Earnings management are usually defined into two types: accrual-based and real earnings management. Zang contributed fresh view to the relevance of these two kinds of earnings manipulation. According to Zang (2005), when managers make decisions, real earnings manipulation comes before accrual manipulation. And she demonstrated that real and accrual manipulations function as substitutes. After lawsuit fillings, in response to the increase in litigation risk and outside scrutiny, this substitute function works since managers switch from accrual manipulation to real manipulation. And Zang（2005） also provide evidence to her predictions that real manipulation and

(13)

accrual manipulation are negatively related with their own cost determinants and positively correlated.

This paper will focus on real earnings manipulations. Real earnings management occurs when managers undertake actions that deviate from the first best practice to increase reported earnings. Little literature has been done focusing on real earnings manipulation until Roychowdhury(2006), his study provided evidence about manipulations through operating activities. According to Roychowdhury, real earnings manipulations derive from normal operating activities. As he developed the measures to detect real earnings manipulation, he stated that cash flow from

operations (CFO), production costs, and discretionary expenses could capture the effect of real operations. Consistent with this view, this paper alos suggests that firms try to avoid losses by offering price discounts to temporarily increase sales, engaging in overproduction to lower cost of goods sold (COGS), and reducing discretionary expenditures aggressively to improve margins.

Although real earnings management draws less scrutiny, this type of earnings

management has negative effect on firm’s future development. Roychowdhury (2006) suggested that real earnings manipulation could reduce firm value. This is because these actions have negative effect on cash flow of the firms. The actions just take the current period into consideration, while this way ignores the influence on firm value in the long run. To be specific, in order to get short-term earnings target, price

discounts are used to increase sales. In this case, customers expect such price in future as well, implying that margins of future sales will be lower. As for the target to

decrease the cost of production, overproduction will be used. Under this condition, excess inventories occur and impose the pressure that the inventories should be sold in following period. At he same time, the holding cost of the firm is also increases to a large amount. Bruns and Merchant (1990) and Graham et al. (2005) conducted

surveys in their studies about earnings manipulations, and they suggested that financial executives are more willing to manipulate earnings through real activities

(14)

instead of accruals. This is because the latter way of manipulation is more likely to attract the scrutiny from auditor or regulator. Their finding has been supported by Cohen et al. (2008). Engaging in accrual-based earnings management is more risky.

2.2 Executive compensation

2.2.1Incentives of executives’ compensation

Executive compensation is always an important issue that is recognized as the reason leading to earnings management. In the case that executive compensation is closely tied to the value of stock and option holdings, according to Bergstressera,, Philipponb， (2004) , manipulation of earnings occur more in firms. And in the period that high accruals occur, executives usually exercise more options and sell large number of shares. This kind of exercise and selling are derived from the executive incentives to manipulate earnings. In addition, their findings also suggest that if the executives are more sensitive to the change of the firms’ share price, they are more inclined to

manipulate earnings. Jensen and Murphy (1990) show that in the period of 1974-1986, executives compensation of stock and option only increase very limited amount

according to the large increase of the shareholder wealth. Hence the executives were not highly motivated. On the other hand, Mehran (1995) finds that the association between firm performance and shares of equity held by executives is positive.

Therefore the way to motivate executives to pursue better performance align with the firm is to provide executives enough incentives. Executives will not choose to benefit the firm more unless they have such incentives. And equity-based compensation is good way to generate such incentives.

But the question is how do the incentives function well to the executives? Since executives are hired as top managers in the firm, they are responsible for all areas of the firm. It’s natural that the compensation of this group has attracted so much

(15)

attention, because of the high compensation and a there’s always the question of “how much is enough”. The findings of Dow and C. Raposo∗（2002）provided some answer to this question. According to their analysis, this kind of pay reflects strategic

discretion more than direct compensation for effort. Jensen and Meckling (1976) suggested that ownership structure, executive compensation structure, and board composition are determined by each other. And at the same time, this kind of pay is also determined by the nature of a firm’s business. The nature includes variables such as business risk, nature of real assets, cash flow pattern, and firm size. And these variables in turn determine firm performance. Executives are often regarded as risk-averse. Consistent with opinion, Harris and Raviv (1979) explained that

managers would like to bear less risk that they prefer their compensation structured. While equity-based compensation is tied to the firm’s stock performance, this

component of the compensation to some degree is beyond managers’ control. Thus, at a certain level of compensation, managers should prefer fixed cash compensation to equity-based compensation. In order to reduce their compensation risk, managers are more likely to engage in activities to reduce the firm’s risk, in turn adversely affecting shareholders’ wealth (Jensen and Meckling, 1976; Amihud and Lev, 1981).

Executives are risk-averse and at the same time are trying to get more benefits . When talking about executive compensation, how the compensation can provide incentives and to what extent the compensation can act as incentives is a key issue. The forms of executive compensation are provided in many different forms,

according to Jensen and Murphy (1990), most of the incentives are originated from holding stock and options. The central conflicting issues in the firm is that firms’ dispersed owner-investors and the executives have different prefers over the

investment projects and decisions. In 1990, Jensen and Murphy stated that executives are not motivated to benefit shareholders unless the compensation for executives is tied to firm performance, the tie is usually in the way of stock option. According to Mehran, executive exposure to stock is just exogenous. And Palia (2001) supported

(16)

this opinion that he demonstrated that incentives of executives are indications of firm nature. In the 1990s, wealth of executives related to stock price increased a lot. Hall and Liebman (1998) demonstrated the average level of executive compensation did increase about three times in 1980s to early1984s. This increase suggested that the compensation of executives was under-incentivized previously. And then comes the time in which executives engage in high level of earnings management.

2.2.2Determinants of compensation

Standard agency theory was developed mainly on the base of incentive issue. An optimal contract connects the principal and agent, and the agent takes over the work with the incentive issue. However, executive compensation has a feature that is different from this standard theory: the contract between principal and agent will change over time, and this change indicates that the firm’s performance is developing over time and its strategy is adjusting as well. There exists an annual pay-setting round in which incentive plans and options can be renegotiated. Executives are members who also participate in the strategy determining process; hence executives are able to influence the compensation contract. Such influence can be a merging business, a restructure of the business, which both are associated with stock option grants.

According to Dow and C. Raposo∗ （2002）,if the strategy of the firm changes, executive compensation would respond to this change to a larger extent than the shareholders expected. This difference between the actual response and the

expectation can be referred to free cash flow theory. Consistent with the managerial empire-building demonstrated by Jensen (1986, 2000), overinvest in public

transportations was argued by Bau- mol (1959), Marris (1967), and Williamson (1964). This kind of overinvestment is viewed as overly dramatic change. And one way to mitigate the overly change is to set ex ante contracting. However, this ex ante

(17)

contracting has its drawback. The drawback is that shareholders may pay the

executives unnecessarily high compensation in advance, even though the executives spend very little effort to in the process. Therefore, this unnecessary pay reminds the shareholders that a wait-and-see approach is a better option if the firm is not going through a dramatic change.

As for ex post contracting, the drawback consists of many uncertain factors. The compensation cannot be determined until the very end. And if executives think more about their own benefits instead of the firm, the conflict of such contracting cannot be ignored. In order to mitigate the conflict of interest in ex post contracting, a good option is to seek better alignment of the interest between shareholders and executives. And for shareholders, the way to achieve this alignment is to provide a well-structured compensation even at early time. And there are two ways to ensure this better alignment. The first one is the ex ante contacting. It’s a way to provide the executives of large amount or percentage of equity share or bonus. In other words, if the executives can get most of the firm value, there’s no doubt that this alignment is confirmed strongly. However, providing the large amount of shares, options or bonus plan is quite expensive. The cost is too high to bear. The second way is to set a ceiling on compensation. It’s in the ex post case when shareholders can make pre-commitments. An upper limit is set to compensation basing on the compensation contract. This way has an advantage that it limits the executives to choose a strategy, which is different from the expectation of shareholders.

The typical compensation package in firms nowadays has components of stock option grant and annual salary, as well as annual bonus. And the salary is renegotiated each year and executives always want to have more options to an existing portfolio. Because of this kind of renegotiation, it’s really hard to realize the pre-commitment. Dow and C. Raposo∗（2002）pointed out this pre-commitment may be possible at the social level. A widespread use of stock options was noticed in most industry groups instead of in utilities, according to the findings of Murphy (1999). One explanation to

(18)

this difference is that executives have more opportunities to make choices of changes to influence firm strategies, due to their own interests, especially in the industries. Following the opinions mentioned above, Demsetz and Lehn (1985) and Core and Guay (2002) demonstrated that firm performance is positive associated with the sensitivity of executive pay to performance. Dow and C. Raposo（2002） also pointed out that the reason for the higher pay granted to executives can be derived to the more opportunities they have to influence the strategic development of the firms. While other managers were constrained to the direct relation between what they have done and what extent they should have been compensated.

2.2.3 How executives’ compensation firm performance

The typical compensation package in firms nowadays has components of stock option As mentioned in previous part, stock option and annual salary and bonus combine to make executive compensation in firms. And each year, executives renegotiate the compensation. It’s already widely accepted that executive compensation is associated with firm performance. It’s about the scrutiny from shareholders. If the firm is doing well, shareholders will concentrate less on the high compensation pay of executives. There are some cases that pay for luck arises. For example, if oil price has a sudden rise, the firm performance is a good picture. This better performance cannot be attributed to executives, but executives are compensated more because of luck.

Skimming happens in the same way. In firms that are well governed, skimming is less prevalent. This is because in such firms, the shareholders on board will not provide executives that many opportunities or power to weigh in compensation determination. Consistent with this view, Bertrand and Mullainathan (2005) suggest that paying for luck occur less in well-governed firms. They gave the reason that shareholders will pay more attention to executives if the firm performance is not good. A conflicting situation related to executive compensation exists in the oil industry. If the crude oil price goes up, executive compensation increases too. But if the price goes down,

(19)

executive compensation doesn’t decrease. Then comes the question that why this asymmetry occurs? Why is human capital of executive worth more when there’s no luck? While actually executive human capital should be worth more when there’s no luck? A possible explanation is that executives are paid for luck because this pay can be seen as incentives for them to forecast luck. Bertrand and Mullainathan hence demonstrated that skimming view predictions are consistent with the performance in poorly governed firms; in contrast, contracting view is consistent with the

performance in well-governed firms. t Bertrand and Mullainathan (2000b). Besides the factor of whether the firm is well governed or not, difference job

classifications lead to different incentives of executives as well. Firstly, it’s important for the executives to recognize that the incentives for them should align with their responsibilities in the firm. A lot of researches like Garen (1994), Aggarwal and Sam- wick (1999, 2002) and Jin (2002) show that the agency theory can explain why incentives in firms are different. However, just comparing the situation in one firm cannot solve the problem whether managerial responsibilities have effect on incentives. Since the association of pay to performance drives incentives, and this association acts as structured motivation for executives, Aggarwal and Samwick (2003) stated this is consistent with the extent of how shareholders recognize executives’ effort. Shareholders will measure the effort basing on performance through diverse signals. According to their findings, if the signal of the measure of quite clear and accurate, executives will be less sensitive to the performance. If the signals are measure in a more combined way, executives may be more sensitive to the performance. Therefore, it’s very important that the incentives are not motivating as separate individuals. According to findings of Aggarwal and Samwick(2003), the more precise the performance indication is, the less sensitive the executives are to the associate between pay and firm performance, even the executives have their own responsibilities. However, in the situation related to short-term compensation, the sensitivity is different. As for the short-term compensation, regarding their pay,

(20)

executives are more sensitive to firm performance if the measure is more accurate and clear. They also supported the agency theory that they confirm the prediction from the theory that if compensation is structured, the risks of the executives are shared and more acceptable one in this way. In addition, they demonstrated that the extent of how risks are shared is related to executives’ responsibilities. At the same time, the risk is a component determining how sensitive the executives are to the firm performance. Aggarwal and Samwick(2003) address the importance of how executives are sensitive to firm performance in a firm. And the problem between principal and agent can be taken to analyze among the executives in the firms, not just constrained to CEOs.

3 Hypothesis developments

Earnings management is the issue that executives have to deal with in firms. Why earnings management occurs? Because cash flows are not always timely reflecting the exact value of the firm, the possibility of managing earnings can be realized through manipulation in reported income. Cash flows are easy to measure and observe, while some changes in firm value are not reflected in cash flows, calculation the firm value is something that involves discretion. The gap between cash flows and reported firm value generates manipulation. And there are some manipulations often used following. For instance, the firm can reduce its reported earnings through assuming a higher rate of depreciation. Or the firm can increase its reported earnings through a lower rate of depreciation. Another way of manipulation is to take the expected expenses that cannot generate future cash flows, and treat such expenses as investment

expenditures.

Because of the components of executive compensation are various and the incentives of earnings manipulation function differently, it hasn’t reached a consensus that how

(21)

executive compensation is associated to earnings management. Some researchers point out that compensation motivates executives to manipulate earnings. Because executives will be compensated more if manage earnings to meet targets. Healy(1985) and Watts(1996) supported this view in their studies. Murphy(1998) stated that

compensation plans do play the role as incentives, and such incentives will lead to at least one type of earnings management. Consistent with this opinion, after analyzing the firms in Japna, Shuto (2007) said that it’s significant that if more options can be manipulated, more incentives of earnings management will occur. Shuto (2007) also demonstrated that the association between discretionary accruals and executive bonus varies depending upon the circumstances of the firm. Carter et al. (2005) suggested in his findings that the structure of bonus compensation varies according to the changes in earnings manipulation activities. Some other researches held different opinion on this association between compensation and earnings management among executives. As Chen et.al （2002）showed that the relation between abnormal returns and changes in ownership concentration is significantly negative, using the sample from Singapore. Actually this difference is originated from the different nature of the components of the compensation. Therefore, the components in executive compensation should be analyzed in specific way.

Determining executive compensation is highly related the firm performance. When judging firm performance, the persistence of earnings is an important indicator. And it’s reasonable that if executives have more incentives, and these executives are sensitive to the share price of the firm, such incentives and sensitivity will lead to more manipulations of earnings. And horizon problem is typical in this kind of circumstance. If executives are able to determine the firms’ development decisions, horizon problem appears in the case that they will leave before the firm’ optimal investment horizon comes. (Smith and Watts, 1982; Johnson, 1987; Dechow and Sloan, 1991; Ittner et al., 1997). Smith and Watts(1982) demonstrated that executives will sacrifice firm’s profitability in long-term to gain short-term profit. And this kind

(22)

of “sacrifice” is always criticized because evaluating firm performance is based on accounting earnings (Smith and Watts, 1982; Dechow and Sloan, 1991). Bergstresser and Philippon (2004) also stated the association between that the persistence of earnings and executive compensation. Particularly, they pointed out that earnings persistence is relevant to cash salary and bonuses instead of stock-based, or other compensation components.

Based on previous studies we know that efficient contracts act as incentives for managers while at the same time , efficient contracts also play the role of constraining managers to maximize their own benefits. When the role of incentive weighs more, executives will manipulate earnings to get higher compensation. And as mentioned above, according to Jensen and Meckling（1976），executives are risk-averse, and they prefer to engage in activities with less risk. And compared with equity-based

compensation, cash compensation is something that is within their control. And consistent with horizon problem（Smith and Watts，1982），executives are inclined to get the short-term compensation through earnings manipulations.

Thus I expect that cash bonus component of executive compensation is positively associated with real earnings management. And I develop my first hypothesis as below:

H1 Earnings management is positively related to executives’ compensation And ownership concentration, or equity held by executives is one type of

constraining method. And when the role of constraining of efficient contracts weighs more, what executives can benefit from the firms is more closely related to firms’ performance in long-term. In addition, Under this situation, earnings management is less, because executives will focus on the

sustainability of firms’ development. Executives will pay more attention to the profitability in long run, and are less motivated to sacrifice long-term earnings to gain short-term items. They will balance between their own compensation

(23)

in one period and firm’s value in long term. It cannot be guaranteed that this choice will end with the longer period consideration, but this consideration does play the role of constraining earnings management. Since executives must take into account the influence on their shares when they leave.

According to Harris and Raviv (1979), equity-based compensation is tied to the firm’s stock performance and to some degree beyond managers’ control. In this case executives will choose to align with the firm performance in long run. While real earnings management has negative influence on future firm value, executives are not likely to engage in real earnings manipulations. Therefore, I expect that the executive ownership is negatively associated with earnings management as the second hypothesis.

H2 Earnings management is negatively related to executive ownership

4. Research design

4.1Compensation metrics

Compensation will be analyzed from following dimensions.

Variable Definition

COMPt Annual bonus pay of executives in period t

ΔCOMPt/ ΔROE t Sensitivity of executives bonus

compensation to firms’ performance in period t

MGTt Shares held by managers in period t

(24)

I selected the annual bonus of executives as the first measure of executive

compensation. As developed in the hypothesis, cash compensation plays an important role in incentivizing executives to manipulate earnings.

ΔCOMPt/ ΔROE t :

This measure is the ratio of the changes of bonus to changes of return on equity. According to the contract theory, if compensations structure is more related to the firms’ short-term earnings, the executives would have more incentives to manipulate earning. According to this theory and my hypothesis, I use this ratio measure as the second indication of compensation. And whether stock options, restarted stock and appreciation rights are granted to executives is a way to indicate the effect of

restriction of stock on executives’ earnings management. But due to the limited time arrangement of developing this paper, this variable of restricted stock and options is not taken as a variable.

MGTt:

It’s a measure of the incentive of executive compensation in mid-term or long-term. It represents the shares owned by the executives in firms in period t, and the shares are the result of accumulation of previous periods owned shared. Therefore, it’s a measure about how the shares owned by executives can act as incentives for executives to manipulate earnings.

4.2 Real earnings management

According to prior studies, I develop the proxies for real earnings manipulation in a aggregated way. As in Roychowdhury (2006), the three metrics are used to study the level of real activities manipulations: the abnormal levels of cash flow from

operations (CFO), abnormal discretionary expenses, and abnormal production costs. Some subsequent studies followed the construct of these three proxies, and evidence was provided. Just like Gunny(2005) and Zang (2006) did.

(25)

4.2.1. Components in real earnings management

This paper will focus on real earnings management; therefore the variables will develop in the way consistent with the features of real earnings management. As we know that real earnings management normally is manipulated through firms’

activities of operating activities, financing activities and investment. Operating

activities are activities that manipulate the production, inventories and sales as well as the trade of long-term assets，and discretional expenses. To be specific, the first way is increasing price discounts or providing more lenient credit terms. This way will temporarily accelerate sales cycles, resulting in temporary increases of sales. However, this increase disappears once the price goes back. What’s more, there’re limited cashes supporting this kind of increase, and this lack of cash support will result in lower cash flows compared to the cash support in the same period in other years. The second way is to report lower cost of goods through increase of production. When the production increases, the fixed cost that each unit bears will be lower, thus average cost will be lower, then higher margins will be reported. However, production costs from other products will result in higher annual production costs to sales, also the lower operating cash flows. And the last way relates to decreases in discretionary expenses, including advertising, R&D, and SG&A expenses. Reducing such expenses will boost current period earnings. To be specific, if the firm pays for this kind of expenses in cash, higher current period cash flows occur, while lower future cash flows are the other side of this sword.

4.2.2 real earnings management metrics

According to the analysis above, the effect of real earnings management can be seen through cash flows from operations, production costs and discretional expenses. This paper develops basing on the model used in Roychowdhury (2006), and

(26)

of the regression model to represent the extent of real earnings management. The calculations of these three variables are as follows.

Cash flow from abnormal operation

The normal levels of CFO, discretionary expenses, and production costs are generated by using the model developed by Dechow et al. (1998) , which were also implemented in Roychowdhury (2006). According to the model, normal CFO is a linear function of sales and change in sales. The following cross-sectional regression is run to estimate the model for each year:

!"#$ !"!!=k1 ! !"!! + k2 !"#$!% !"!! + k3 △!"#$!% !"!! +εit,（1.1.）

Abnormal CFO (RCFO) is actual CFO minus the normal level of CFO calculated using the estimated coefficients from 1.1.

(2) Abnormal production cost

According to the method developed or implemented in Dechow et al.

(1998) ，’Roychowdhury (2006).Cohen(2010) as well. production costs are defined as the sum of COGS and change in inventory during the year. COGS is taken as a linear function of contemporaneous sales, and model inventory growth as a linear function of the contemporaneous and lagged change in sales. Hence to estimate the normal level of production costs the following model is used:

!"#$% !"!! =k1 ! !"!! + k2 !"#$!% !!!! + k3 △!"#$S! !"!! + k4 △!"#$!%!! !"!! +εit,(1.2)

Abnormal production costs (RPODt) is computed as the difference between the actual vales and the normal levels predicted from 1.2

(3) Discretional expenses

And discretionary expenses are taken as a function of lagged sales and estimate the following model to derive ‘normal’ levels of discretionary expenses. Taking the issue that manipulation of sales may go upwards leading the increase of reported earnings

(27)

in a specific year, the model takes discretional expenses as a function of lagged sales. And the following model is used the estimate the “normal” levels of discretional expenses. !"#$% !"!!=k1 ! !"!! + k2 !"#$!%!! !"!! +εit,（1.3）

Abnormal discretionary expenses(RDISXt) is computed as the difference between the actual values and the normal levels predicted from 1.3

And the three individual measures combined can capture the metric of real earnings management activities. Following Zang(206) as well as Cohen(2010)，the higher amount of discretional expenses, the more likely that the firm is manipulating the discretional expenses downwards. And add it to abnormal production costs. Also, consistent with Zang(2006) and Cohen(2010), abnormal cash flows from operations and abnormal discretionary expenses are multiplied by negative one. And the higher these amounts, the more likely that the firm is manipulating earnings upwards through sales and discretional expenditures.

Chi et.al.(2011) developed a comprehensive measure of real earnings management by combining three individual measures. It is an index of real earnings management of the sum of three components as following:

RPROXYt = PRODt — RCFOt — RDISXt

Just as this index shows, Chi et al(2011) demonstrates that higher levels of PROXY indicate higher levels of overall real earnings management.

And these are the three situations more likely to incur in firms engaging in earnings management: abnormal low cash flow from operation activities, abnormal high production costs, and abnormal low discretional expenses. When one of these three or all three situations comes into appearance, this overall index will increase. Therefore, it’s an indicator of the extent of real earnings management.

(28)

4.2.3 Control variables

Variable Definition

DEBTRt Measure of debt, ratio of Long-term debt

to total asset in period t

ROEt Return on equity in period t

YEAR

Dummy variable

The variables listed above are often used in literature as control variables in contract theory problems. DEBTRt is the measure of debt level, controlling the incentives for executives to manipulate earnings. According to contract theory, the higher debt level of the firm is, the more likely the contract may be violated. In this case, in order to reduce the debt level, the firms will be more likely to manipulate earnings upwards. The dummy variable of the year is derived from the change of the market situation and change of the scrutiny extent from related departments. Due to such kind of changes, the effect of year should also be taken into consideration. The competition situations of different industries are different, and competition will add the difficulty in making earnings, resulting in different extent and way of manipulating earnings. But this paper doesn’t take the year as dummy variable because of limited time. In the sample selection part, the financial institutions are all deleted because the operations in such firms are not suitable for the models to describe.

4.3 Models

Basing on these metrics, to test my hypothesis, I run the following regressing model for the sample of firms.

(29)

In model 2.0, the index of real earnings management and the three separate measures of real management are all analyzed to test how the incentives of compensation drive manipulations through real earnings activities.

RPROXYit=αit +β1 COMPit+β2 ΔCOMPit/ΔROEit+β3MGTit+CONTROL (2.0) Model 2.1 is used to capture the association between abnormal cash flow and bonus compensation and shared owned by executives.

RCFOit=αit +β1 COMPit+β2 ΔCOMPit/ΔROEit+β3MGTit+CONTROL (2.1) Model 2.2 is used to capture how abnormal production cost is associated with cash bonus and shared owned by executives.

RPRODit=αit +β1 COMPit+β2 ΔCOMPit/ΔROEit+β3MGTit+CONTROL (2.2) Model 2.3 is used to capture the association between abnormal discretional expenses and bonus compensation and shared owned by executives.

RDISXit=αit +β1 COMPit+β2 ΔCOMPit/ΔROEit+β3MGTit+CONTROL (2.3) Because I think that these three components of real earnings management should be tested both separately and in an aggregated way. The aggregated measure is a mixed indicator, and it’s clearer to explain the association through comparing the individual effect and mixed effect.

5.Analysis

5.1Sample description

The data was selected from the COMPUSTAT in period 2010 to 2012, and the firms were the North America firms. I eliminate the data of firms in which the executives left during 2010-2012period, and the data of financial industry firms was eliminated as well. The sample was tested with 1182.

(30)

5.2 Descriptive statistics

(1) Description of executive annual bonus compensation Table 1

Mean Media Max Min Std. dev N

2010.00 148.48 0.00 4355.00 0.00 508.84 340.00 2011.00 159.21 0.00 7356.00 0.00 659.25 340.00 2012.00 188.56 0.00 13060.85 0.00 909.65 340.00 The table summarizes descriptive statistics of bonus compensation of executives for our sample. The sample description is listed from 2010-2012.

From table 1,we can see that the bonus of executives is increasing in this 2010 to 2012 period, the maximum value of this variable is increasing as well. For the difference of the standard deviation in 3 years is getting larger, stating that the issue of executive compensation, to be specific in bonus, is always focused. What’s more, the data demonstrates that compensation provided among difference executives from different firms is quite diverse. In this case, the executive bonus compensation is going upwards, suggesting the expectation that executives could create more interests.

(31)

(2) Description of all variables Table2

Variables Mean Media Max Min Std. dev

RPROXYt 1511.815 53.40077 233348.2 -160160.4 16142.27 RCFOt 591.5257 15.94367 84998.59 -42547.89 4333.647 RPRODt 3368.589 410.1599 252000 -70265.9 14980.73 RDISXt 1265.247 205.3809 36673.49 -5009.224 3874.973 RCFOt 591.5257 15.94367 84998.59 -42547.89 4333.647 ROEt 0.460854 0.1 70.38 -12.34 4.216657 DEBTt 0.270245 0 16.81 0 1.068263 COMPt 158.0275 0 13060.85 0 676.5252 MGTt 617.0084 183.075 52379.04 0 2674.16 COMPt/ROEt -2107.903 0 224274.8 -3449297 104231.4

According to the description of the all the variables, we can see that standard deviation of the index of aggregated level of real earnings management is distinguishable. The standard deviation value of shares owned by executives and executives’ sensitivity of compensation to firms’ performance is distinguishable as well. This kind of non-negligible value demonstrates that real manipulation activities are diverse across different firms. And this value also states that how sensitive the executives respond to compensation vary a lot in different firms.

(32)

(3)

Table3 Correlation Matrix

RPROXY RPROD ROE RDISX RCFO DEBT COMP MGT COMP/ROE RPROXY RPROD 0.9115 *** ROE 0.0178 0.0085 RDISX -0.2835*** 0.0423 -0.0248 RCFO -0.3204*** 0.0238 -0.0147 0.3081 *** DEBT -0.1343*** -0.0908*** 0.0236 0.1688 *** 0.036 COMP 0.01 0.0988 *** 0.0143 0.1439 *** 0.2504*** 0.1249 *** MGT -0.007 0.0078 0.0816 0.0124 0.041 0.028 0.1649*** COMP/ROE 0.044 0.0562* 0.0026 0.0137 0.019 -0.0862 *** 0.035 0.0198

(33)

Table 4

Model2.0 Model 2.1 Model 2.2 Model 2.3

RPROXY RCFO RPROD RDISX

Intercept 2027.153 341.2547 3390.743 1022.335 (0.0001)*** （0.0094）*** （0.00）*** （0.00）*** COMP 0.148102 1.597098 2.465714 0.720514 -0.8335 （0.00）*** （0.0002）*** （0.00）*** ΔCOMP/ΔROE 0.00497 0.000471 0.006298 0.000858 -0.2689 -0.6899 -0.1306 -0.4206 MGT -0.039222 0.002039 -0.052819 -0.015636 -0.825 -0.9651 -0.7479 -0.7097 DEBT -2004.455 23.7366 -1414.577 566.1411 (0.00)*** -0.8178 （0.0006）*** （0.00）*** ROE 81.65338 -19.08464 35.46592 -27.10281 -0.4618 -0.5122 -0.7301 -0.3025

Year Dummies Yes Yes Yes Yes

n 1182 1182 1182 1182 Adjust R^2 0.015414 0.059198 0.018482 0.041301 F(sig.) 4.69768 15.86235 5.447693 11.17558 （0.000298） *** (0.00)*** (0.000058)*** (0.00)***

*, **, *** Denote 0.1, 0.05, and 0.01 significance levels, respectively. Another point to mention in the result is the value of R2. The value is much smaller than the R2 in other

(34)

related researches. I take the reason of this smaller value as the limited observations I had in my sample.

5.3 Empirical findings

Table 4presents my main result. I find a positive coefficient of 0.148 on Comp in model 2.0,suggesting that the bonus compensation of executives is positively related to the overall level of real earnings manipulation. And I find a negative 0.039 coefficient MGT in model 2.0,suggesting that shares owned by executives are negatively related to real earnings management activities. And the coefficient on DEBT is also significant, suggesting that debt level is also closely related to real manipulation activities. And the main relation presented in this result is consistent with my hypothesis.

Specifically, I also find the significantly positively coefficient on COMP in regression of RCFO, RPROD, RDISX. This coefficient suggests the relationship between executive bonus compensation and three different aspects of real earnings management. But I fail to distinguish the difference among these three aspects from getting the sign of positive and negative. Since according to the expectation from hypothesis development, abnormal discretional expenses and abnormal cash flows are more likely to be manipulated downwards in earnings manipulation activities. And abnormal high production costs are more likely to occur in real earnings manipulations. This lack of variation of positively and negatively relating indicators is the weakness of my results. However, I also expect that as the overall index of real earnings management, PROXY is a mixed indicator. Therefore, the three separate elements of real earnings management can already suggest the significant association between real earnings management and bonus compensation.

Collectively, to some extent, these results provide the evidence that the bonus part of executive compensation is positively related to real earnings management. And this is consistent with my hypothesis, which is derived from the view that compensation of

(35)

executives, especially the cash part, is often an important driver of incentives of executives of firms to manage reported earnings. They do this to earn more compensation for their own interests. According to prior research，age of executives weighs a lot this incentive. It’s an issue that we should take into serious consideration, although this paper didn’t focus on the age issue.

As for the other main variable measuring executive compensation, shares owned by executives (MGT). To be specific, the negative coefficient 0.0528, suggesting that it is negatively related to abnormal production cost, consistent with the hypothesis that for executives’ long-term interest, executives are less likely to engage in activities that reduce abnormal production costs. And the coefficient of MGT to RCFO also suggests that in long-term view, executives are less likely to manipulate real earnings activities through abnormal cash flows. These results suggest that shared owned by executives are associated with less real earnings management. As for the regression result of the shares owned by executives, they are not that significant, but the association is consistent with hypothesis as well. Regarding the reason for these less significant results, compared to cash bonus, which can be easily observed and measured accurately, shares owned by executives are more complex. The influence from the market also contributes to the situation of the amount of value of the shares. Hence, it’s difficult to measure this association with taking all other related variables. And I suggest that further studies to develop the controlling variables to test the role of shares owned by executives more precise.

This finding adds the important insights into the discussion whether shares owned by executives act differently from other forms of compensation. Prior research has focused a lot how to motivate executives to make more contribution to firms, and developed a lot how to provide the executives the right incentive in long-term view. This paper adds another piece in the literature that different forms of compensation do have divers incentives for executives, from the view of earnings manipulations

(36)

6.Conclusions

The literature has focused and discussed the incentives of executives to make contribution to the firms as well as the incentives of executives to obtain their own interests, besides the interests of firms. As both accrual-based and real earnings management influence how the firms operate, and these two kinds of earnings manipulations are differently derived and lead to different situations of the firms. I examine whether the consequence originated from the incentives for executives of real earnings management are just the different components of compensation. I try to find that whether the different components in compensation play different roles in firms operating activities. And using the proxy as the measure of overall level real earnings management, my findings are consistent with the hypothesis. My findings also suggest that shareholders of the firm should monitor executive compensation, as it’s a kind of incentives for executives that finally have influence on the firms. And the three individual aspects of real earnings manipulation are all significantly associated with the overall index. Particularly, the abnormal cash flow as well as the discretional expense is significantly negatively associated with the overall index. And the production cost is significantly positively associated with the overall index. This significant association confirms that the overall index of real earnings management can be used as a measure of these manipulations. The sensitivity of executives’ bonus to firms’ performance is negatively associated to overall level of executive

compensation. Using both three individual measure of real earnings management and the aggregated index of real earnings management, I confirm that ,besides the

elements of compensation, the indication of firm performance, such as debt level of , is associated with the real earnings manipulations. The implication of compensation, both short-term and long-term interests included, should be taken into account when determining the compensation structure.

(37)

Upward real earnings management has been focused on a lot in past research (Roychowdhury 2006; Cohen 2010). Past research has focused on executive

incentives and compensation structure as well. But the relation between real earnings management and executive compensation hasn’t been discussed much. Therefore, more related researches are needed to provide to the determination of executive compensation, and this kind of determination will be useful to motivate the executives and will lead the firms to address better firm performance. We confirm that the

driving incentives of compensation lead to real earnings management. I’ve acknowledged that the way to provide executive incentive is important. I’ve confirmed the importance of a well-structured compensation, and the key issue to motivate the executives to act for the benefits of the firm is to make the interests align with executives’ benefits.

(38)

Reference list

Amy Y. Zang*,2005, “Evidence on the Tradeoff between Real Manipulation and Accrual Manipulation”, Unpublished dissertation, Duke University.

Andrei Sheleifer ,and Robert W. Vishny *,1997，“The Limits of Arbitrage”，Journal of Finance.Vol. LII, No.1,(Mar)

Akinobu Shuto ,2007,“Executive compensation and earnings management: Empirical evidence from Japan”, 16 (2007) 1–26.

Bartov, E., Givoly, D. & Hayn, C. 2002, “The rewards to meeting or beating earnings expectations”. Journal of Accounting and Economics, 33, 173-204.

Bartov, E., Gul,F., and Tsui,J. 2000. "Discretionary-accruals models and audit qualifications." Journal of Accounting and Economics, 30, 421-452.

Beatty, A., Ke, B., and Petroni, K. 2002. "Earnings management to avoid earnings declines across publicly and privately-held banks." The Accounting Review, 77, 547- 570.

Burgstahler, D., and Dichev, I., 1997, “Earnings management to avoid earnings decreases and Losses”, Journal of Accounting and Economics, 24, 99-126.

Burgstahler, D., Eames, M. 1998, “Management of earnings and analysts’ forecasts to achieve zero and small positive earnings surprises”, University of Washington

Working Paper.

Bradshaw, M.T., Richardson, S.A. and Sloan, R.G. 2001, “Do analysts and auditors use information in accruals?” Journal of Accounting Research, 39, 45-74.

Cahan, S. F.1992. "The Effect of Antitrust Investigations on Discretionary Accruals: A Refined Test of the Political-Cost Hypothesis." Accounting Review, 67, 77-95. Clinch, G. and Margliolo, J. 1993, “CEO compensation and components of earnings in bank holding companies”, Journal of Accounting and Economics, 16, 241-272.

(39)

DeAngelo, L., and Skinner, J. 1994. “Accounting Choice in Troubled Companies”. Journal of Accounting and Economics, 17, 113-143.

Dechow, P. M.1994. “Accounting earnings and cash flows as measures of firm performance: the role of accounting accruals”. Journal of Accounting and Economics, 18, 3-42.

Dechow, P.M, and Sloan, R. 1991, “Executive incentives and the horizon problem”, Journal of Accounting and Economics, 14, 51-89.

Demski, J. S., Patell, J. M. & Wolfson, M. A. 1984, “Decentralized choice of monitoring systems”. The Accounting Review, 59, 16-34.

Dye, R. 1988, “Earnings management in an overlapping generations model”, Journal of Accounting Research, 26, 195-235.

Easton, P. D. & Zmijewski, M. E. 1989, “Cross-sectional variation in the stock market response to accounting earnings announcements”. Journal of Accounting and Economics, 11, 117-141.

Erickson, M. and Wang, S. 1999, “Earnings management by acquiring firms in stock for stock mergers”, Journal of Accounting and Economics, 27, 149-176.

Guay, W., Kothari, S. and Watts, R. 1996, “A market-based valuation of discretionary- accrual models”, Journal of Accounting Research, 34, 83-105. Hamid Mehran,1995,” Executive compensation structure, ownership, and firm performance”, Journal of Financial Economics 38 ,163-184.

Healy,P.M. 1985, “The effect of bonus schemes on accounting decisions”, Journal of Accounting and Economics, 7, 85-107.

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

Henock Louis,2003,“Earnings management and the market performance of acquiring firms”,Journal of Financial Economics ,74 (2004) 121–148.

(40)

Holthausen, R, W. and Leftwich, R. W. 1983, “The economic consequence of accounting choice: Implications of costly contracting and monitoring,” Journal of Accounting and Economics, 5, 77-117.

Holthausen, R. W. and Larcker, D.F.1995. "Annual bonus schemes and the manipulation of earnings." Journal of Accounting and Economics, 19, 129-74. James Dow and Clara C.Raposo*_{,2005,”CEO compensation, change ,and corporate} strategy”, The Journal of Finance.Vol.LX.No.6(Dec).

Jennifer J.Jones,1991,“Earnings management during important relief investigations”, Journal of Accounting Research,Vol.29 No.2,(Autumn)

Jensen,M.C. and Meckling, W.H.1976, “Theory of the firm: managerial behavior, agency costs, and ownership structure,” Journal of Financial Economics, 3, 305-360. John R.Deckop* _{, 1988,”Determinants of chief executive officer compensation”,} Industrial and Labor Relations Review,Vol.41,No.2, (Jan).

Kevin J. Murphy* ,1985,”Corporate performance and managerial remuneration”, Journal of Accounting and Economics,7,11-42

Lambert, R. A.1984. "Income smoothing as rational equilibrium behavior." The Accounting Review, 59, 604-618.

Louis, H. 2004, “Earnings management and the market performance of acquiring firms”, Journal of Financial Economics, 74, 121-148.

Marianne Bertrand and Sendhil Mullainathan,”Are CEOs rewarded for luck? The ones without principals are”, The Quarterly Journal of Economics, Vol. 116, No. 3 (Aug., 2001), pp. 901-932

McNichols,M. and Wilson, G.. 1988, “Evidence of earnings management from the provision for bad debts”, Journal of Accounting Research, 26, 1-31.

Rangan, S. 1998, “Earnings management and the performance of seasoned equity offerings”, Journal of Financial Economics, 50, 101-122.

(41)

Rajesh K. Aggarwal and Andrew A. Samwick*, 2003,“Performance Incentives within Firms: The Effect of Managerial Responsibility”, The Journal of Finance.Vol. LVIII, NO. 4 ( AUG).

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, 7, 289-312.

Subramanyam, K. R. 1996, “The pricing of discretionary accruals”, Journal of Accounting and Economics, 22, 249-281.

Sugata Roychowdhury*,2006,“Earnings management through real activities manipulation”，Journal of Accounting and Economics 42,335–370

Teoh, S., Welch, I., and Wong, T. 1998b, “Earnings management and the post-issue performance of seasoned equity offerings”, Journal of Financial Economics, 50, 63- 99.

Teoh, S., Wong, T. and Rao, G. 1998. “Are accruals during initial public offerings opportunistic?” Review of Accounting Studies, 3, 175–208.

Watts, R.L., and Jerold L. Zimmerman, 1986, “Positive Accounting Theory: A Ten Year Perspective”, The Accounting Review, 1 January 1990, Vol.65(1), pp.131-156 Watts, R.L., and Zimmerman, J.L 1978 “Towards a positive theory of the

determination of accounting standards”, The Accounting Review,53, 112-134. Wang, Z. and Williams, T.H. 1994, “Accounting income smoothing and stockholder wealth”, Journal of Applied Business Research, 10, 96-110.

Xavier Gabaix and Augustin Landier,2008,”Why has CEO pay increased so much” ,The Quarterly Journal of Economics, February.

Executive compensation and real earnings management in listed companies

Student name Xiaoshu Shao

Thesis title