Effects of Executive Compensation on Earnings Management. and Cost of Equity Capital

Transcription

1 Effects of Executive Compensation on Earnings Management and Cost of Equity Capital Kanyarat Sanoran Corresponding author: Chulalongkorn Business School, Chulalongkorn University 254 Phayathai Road, Pathumwan, Bangkok 10330, Thailand Leon Wong UNSW Australia UNSW Sydney NSW 2052, Australia leon.wong@unsw.edu.au We thank Ferdinand Gul and Gary Monroe for their helpful comments. 1

2 Effects of Executive Compensation on Earnings Management and Cost of Equity Capital Abstract This study examines how executive compensation in the form of stock option, shareholdings, bonus, and long-term performance plan affect the discretionary accruals and cost of equity capital. We find that executive bonus and long-term performance plan are negatively associated with both discretionary accruals and cost of equity capital. In addition, CEO shareholdings, but not other executives shareholdings, are negatively associated with both discretionary accruals and cost of equity capital. Our additional analyses demonstrate that executive bonus and long-term performance plan affect only income-increasing earnings management, but not income-decreasing earnings management. In sum, our results suggest that executives and shareholders are responsive to different types of executive compensation. Keywords: Executive compensation, Discretionary accruals, Earnings management, Cost of equity capital JEL Classification: G31, M41, M52 1

3 1. Introduction To the extent that executive compensation can influence earnings management, investors may be expected to respond with increased or decreased information risk by demanding higher or lower returns. In this paper, we examine the effects of executive compensation on earnings management and cost of equity capital for four components of executive compensation. Each component of executive compensation differ in its nature, which lead to differences in the executive incentives and investors pricing of information risk. We hypothesize and provide evidence that the effects of executive compensation on earnings management and cost of equity capital are different for each component of executive compensation. Our analyses are based on a sample of 3,436 firm-year observations on the U.S. companies from 2004 to Consistent with prior research, we use discretionary accruals as a measurement for earnings management. We measure the cost of equity capital by using average value of nine alternative methods introduced by prior literature (Easton 2004; Ohlson and Juettner-Nauroth 2005; Gode and Mohanram 2003; Botosan and Plumlee 2002; Gordon and Gordon 1997; Claus and Thomas 2001; Gebhardt, Lee, and Swaminathan 2001; Hail and Leuz 2006; Dhaliwal, Krull, and Li 2007). We find a significantly lower discretionary accruals for bonus and long-term performance plan. Interestingly, we find that CEO shareholdings, but not other executives shareholdings, is negatively associated with the discretionary accruals. By contrast, we do not observe a significant difference in the discretionary accruals for executive stock option. Additionally, we find a significant reduction in cost of equity capital for bonus plans and long-term performance plans, but not for stock options. Remarkably, CEO shareholdings, but not other executives shareholdings, are negatively related to the cost of 2

4 equity capital. Therefore, our analyses indicate that the effect of executive compensation on discretionary accruals and cost of equity capital is not uniform across different types of executive compensation. Our study complement and extend previous studies (i.e., Shaw 2012; Prevost, Devos, and Rao 2013; Boone, Khurana, and Raman 2011; Kabir, Li, and Veld-Merkoulova 2013) by providing evidence on whether each type of executive compensation impact executives accruals reporting behavior and investors benefit in the form of cost of equity capital. This paper may be of interest to regulators, managers, and investors faced with questions about the possible consequences of executive compensation on information risk and shareholder wealth. The rest of this paper is organized as follows. We describe the effects of information risk on cost of equity capital in section 2. We develop the hypotheses in section 3, describe our sample and data in section 4, and present the research design in section 5. We discuss the results of the main and additional tests in sections 6 and 7. We present our conclusions in section Literature review and hypothesis development 2.1. Effects of stock option on earnings management and cost of equity capital Information risk is a non-diversifiable risk factor that is priced by the market (Gray et al., 2009). The higher quality of accruals, the better earnings map into cash flows and hence, the lower the information risk (Francis et al., 2005; Gray et al., 2009). Quality of information, including earnings quality, can mitigate the information risk and thus reduce the cost of equity capital (Aboody et al., 2005; Easley and O'Hara, 2004). Many previous studies (e.g., Aboody et 3

5 al., 2005; Easley and O'Hara, 2004; Francis et al., 2005; Gray et al., 2009) report that higher information risk leads to higher cost of equity capital. Francis et al. (2005) use accrual quality as a proxy for the information risk and find that firms with poorer accruals quality have higher cost of equity capital, compared to firms with better accrual quality. Compensation can influence the executives incentives and earnings management, which affect the firm s cost of equity capital. The effect of each compensation element on the cost of equity capital can be different due to the unique information risk features of each element. Therefore, the hypotheses on the impact of each executive compensation component are separately developed in this study. We expect to detect a positive association between the cost of equity capital and executive compensation components that cause higher information risks. A negative association between the cost of equity capital and executive compensation components that lead to lower information risks is expected. Some incentives for earnings management are the result of executive compensation (Brown, Beekes, and Verhoeven 2011; Pourciau 1993). The opportunistic behavior of executives is primarily associated with increases in executive compensation in the form of stock options and creates the opportunity for executives to manipulate accrual-based earnings (Cohen, Dey, and Lys 2008). Stock option awards may induce compensation incentives that motivate executives to manipulate earnings with the intention of meeting favorable factors of the option grants (Vafeas and Waegelein 2007). Despite this, Laux and Laux (2009) argue that earnings management may not increase even if CEO equity incentives are increased because directors alter their monitoring efforts in response to changes in CEO incentives. In addition, Armstrong, Jagolinzer, and Larcker (2010) find that relatively higher levels of CEO equity incentives lead to a 4

6 decreased incidence of accounting irregularities. In contrast, there is considerable evidence pointing to incentive-based pay, particularly the use of stock option compensation, motivating earnings management in order to meet performance targets or thresholds (Shrieves and Gao 2002; Cheng and Warfield 2005; Bergstresser and Philippon 2006; Grant, Markarian, and Parbonetti 2009; Houmes and Skantz 2010) and earnings restatements (Burns and Kedia 2006; Efendi, Srivastava, and Swanson 2007). For example, Bergstresser and Philippon (2006) report a positive association between CEOs equity compensation and accruals management. McAnally, Srivastava, and Weaver (2008) find that, for companies that manage earnings downward, stock option grants are positively associated with the probability of missing earnings targets. Meek, Rao, and Skousen (2007) find a positive relationship between annual CEO stock option compensation and the absolute value of discretionary accruals, implying the likelihood that earnings management increases when CEO compensation in the form of stock options is higher. Cohen, Dey, and Lys (2008) corroborate that an increase in accrual-based earnings management is concurrent with increases in the proportion of equity-based executive compensation. The arguments presented earlier suggest earnings management is positively related with the cost of equity capital. Therefore, there may be either a positive or a negative relation between executive stock option and the cost of equity capital. Hence, the first set of hypotheses is developed as follows: H1a: The proportion of executive stock option compensation is associated with the discretionary accruals. 5

7 H1b: The proportion of executive stock option compensation is associated with the cost of equity capital Effects of executive shareholdings on earnings management and cost of equity capital Another source of incentives for executives is the percentage of managerial shareholdings. Similar to stock option compensation, stock ownership can also impact incentives for earnings management (Cheng and Warfield 2005). Cheng and Warfield (2005) indicate that executives with high stock ownership are more likely to report earnings that meet or just beat analysts forecasts. Another perspective is that managerial shareholdings provide managers with shareholder-like interests so executives tendency to manage earnings may be decreased (Vafeas and Waegelein 2007). Agency theory indicates that management ownership might reduce the conflict of interest between managers and stockholders (Wright et al. 2002). Executives might not be able to diversify away the risk attached with his or her wealth due to stock ownership because his or her human capital is enormously invested in a single place of employment (Smith and Watts 1992). Therefore, executives tolerance for risk is affected by their limited ability to reduce personal risk (Bryan, Hwang, and Lilien 2000). Warfield, Wild, and Wild (1995) report that the level of managerial shareholdings is negatively associated with the level of earnings management. The arguments presented earlier suggest earnings management is positively related with the cost of equity capital. Therefore, there may be either a positive or a negative relation between executive shareholdings and the cost of equity capital. Therefore, the second set of hypotheses is stated as follows: H2a: The proportion of executive shareholdings is associated with the discretionary accruals. 6

8 H2b: The proportion of executive shareholdings is associated with the cost of equity capital Effects of bonus on earnings management and cost of equity capital Previous research provides evidence to strongly support the argument that executives use incentive compensation to their advantage, even though it might disadvantage the shareholders (Devers et al. 2007). CEOs with higher accounting-based compensation are more likely to be involved in opportunistic behavior in order to maximize the value of such compensation (Jiang, Petroni, and Yanyan Wang 2010). Executives can directly manipulate accounting-based measures in many ways such as via changes in debt structure, inventory management, and accounting policies (Murphy 2000). However, it is more difficult for executives to directly manipulate market-based measures (Wiseman and Gomez-Mejia 1998). Cornett, McNutt, and Tehranian (2009) find that there is greater earnings management when an executive s compensation is attached to the firm s performance. Bonus pay creates incentives for executives to choose accounting policies and accruals that make the most of the value of their bonus compensation (Healy 1985; Shrieves and Gao 2002). There is a significant relation between accruals and executives income-reporting incentives under the bonus contracts (Healy 1985; Shrieves and Gao 2002). Gaver, Gaver, and Austin (1995) and Holthausen, Larcker, and Sloan (1995) argue that, with the intention of maximizing their bonus in subsequent periods, executives may manipulate earnings downward when they perceive that their bonuses are at the highest level. Guidry, J. Leone, and Rock (1999) extend the literature by looking at managers within a single firm, which minimizes aggregation effects and removes confounding effects of stock-based pay plans. They provide evidence that managers manipulate 7

9 earnings to maximize their short-term bonus compensation. The arguments presented earlier suggest earnings management is positively related with the cost of equity capital. Therefore, there may be a positive relation between executive bonus and the cost of equity capital. Thus, the third set of hypotheses is: H3a: The proportion of executive bonus compensation is positively associated with the discretionary accruals. H3b: The proportion of executive bonus compensation is positively associated with the cost of equity capital Effects of long-term performance plan on earnings management and cost of equity capital A long-term performance plan is implemented to encourage executives to concentrate on enhancing firm performance over several years instead of focusing on short-term performance only (Richardson and Waegelein 2002). Long-term performance plans encourage executives to increase their decision-making perspectives and make decisions that align with shareholders interests (Larcker 1983; Tehranian, Travlos, and Waegelein 1987b, 1987a). Also, performance plans can decrease the manager s exposure to risk by reducing the extraneous variability in the manager s performance measures (Richardson and Waegelein 2002). Prior literature (e.g.,(richardson and Waegelein 2002; Vafeas and Waegelein 2007) indicates that companies with long-term performance plans undertake less earnings management than do companies that have only short-term bonus plans. Long-term performance plans encourage executives to increase their decision-making perspective because there is less incentive to manage annual 8

10 accounting earnings (Vafeas and Waegelein 2007), because executives pay more attention on long-term earnings growth rather than managing the accounting earnings of each period to take advantage of their annual benefit (Richardson and Waegelein 2002). The arguments presented earlier suggest earnings management is positively related with the cost of equity capital. Therefore, there may be a negative relation between long-term performance plan and the cost of equity capital. This leads to the last set of hypotheses as follows: H4a: The proportion of executive long-term performance compensation is negatively associated with the discretionary accruals. H4b: The proportion of executive long-term performance compensation is negatively associated with the cost of equity capital. 3. Sample selection and data The sample is comprised of U.S. companies during the period from 2004 to The data about executive compensation is collected from the Compustat ExecuComp database. We collect the compensation data and run the tests separately for CEOs, CFOs, and top five highlypaid executives because the CFOs manage financial system of the company, while CEOs have the power to replace CFOs who do not follow the CEOs preferences (Mian 2001; Fee and Hadlock 2004). CEOs were identified based on executives titles that include any of the following phrases: CEO, chief executive, and managing director. CFOs were identified based on executives titles that include any of the following phrases: CFO, chief financial, finance, controller, and treasurer. Auditor data was obtained from AuditAnalytics. The market-based data was obtained from CRSP. The analysts earnings forecasts were obtained from I/B/E/S. 9

11 Stock price and financial data was collected from CRSP Daily Prices and CRSP Compustat Merged Annual data. A firm-year observation is only included if all data items are available for the firm-year observation. Consistent with prior research and to eliminate confounding industry effects, observations in the banking and financial sector (SIC code ) are excluded from this study. Furthermore, all continuous variables are winsorized at the 1 st and 99 th percentile of their values. The final sample for the tests consists of 3,436 observations. [insert Table 1 here] Panel A of Table 1 details the distribution of samples in each year. The number of sample observations is highest in 2010, with 576 firm-year observations, and lowest in 2008, with 452 firm-year observations. Panel B of Table 1 details the distribution of samples across industries. Manufacturing is the largest represented industry, making up percent of the sample. This is followed by services, which represented percent of the sample. 4. Variable measurement and research design 4.1. Variables of interest The variables of interest involve four measures of executive compensations. Stock option STOPCEO = Percentage of compensation in the form of grants of stock options to the CEO during the current financial year; 10

12 STOPCFO = Percentage of compensation in the form of grants of stock options to the CFO during the current financial year; and STOPALL = Percentage of compensation in the form of grants of stock options to the top highly paid executives during the current financial year. Managerial shareholdings SHARECEO = Percentage of total shares outstanding held by the CEO at the balance sheet date, excluding options; SHARECFO = Percentage of total shares outstanding held by the CFO at the balance sheet date, excluding options; and SHAREALL = Percentage of total shares outstanding held by the top highly paid executives at the balance sheet date, excluding options. Bonus BONCEO = Percentage of compensation in the form of a bonus earned by the CEO during the current financial year; BONCFO = Percentage of compensation in the form of a bonus earned by the CFO during the current financial year; and BONALL = Percentage of compensation in the form of a bonus earned by the top highly paid executives during the current financial year. 11

13 Long-term performance plans LTPPCEO = Percentage of compensation in the form of the amount paid to the CEO during the current financial year under the company's long-term incentive plan; LTPPCFO = Percentage of compensation in the form of the amount paid to the CFO during the current financial year under the company's long-term incentive plan; and LTPPALL = Percentage of compensation in the form of the amount paid to the top highly paid executives during the current financial year under the company's long-term incentive plan Model specification Earnings management Following previous studies (e.g., Ball and Shivakumar 2008; Ball and Shivakumar 2006; Dechow and Dichev 2002; Dechow, Sloan, and Sweeney 1995; Gul, Fung, and Jaggi 2009; Kothari, Leone, and Wasley 2005), we use discretionary accruals as a measurement for earnings management. We measure discretionary accruals as the residual from the model by following Ball and Shivakumar (2006) and Ball and Shivakumar (2008). The model is as follows: TACC = β 0 + β 1 SALES + β 2 PPE + β 3 OCF + β 4 NOCF + β 5 OCF*NOCF + ε (1) The variables are defined as follows: TACC = total accruals, calculated as net income before extraordinary items less cash flow from operations; 12

14 SALES = change in sales; PPE = gross property, plant, and equipment; OCF = cash flow from operations; NOCF = 1 when OCF < 0, 0 otherwise; and OCF*NOCF = an interaction term of OCF and NOCF. The above variables are standardized by average total assets. We estimate this model for each firm in cross-sectional regressions by 2-digit SIC industry code. We require each industry to have at least 20 observations in any given year. Then, we use the following model to test our hypotheses regarding the discretionary accruals: DACC = β 0 + β 1 SIZE + β 2 LEV + β 3 ROA + β 4 LOSS + β 5 OCF + β 6 INVREC + β 7 GROWTH + β 8 BP + β 9 BETA + Variables of interest + ε (2) The variables are defined as follows: DACC = residual from TACC model; SIZE = size measured by the natural logarithm of the market value of common equity at the end of the fiscal year; LEV = financial leverage measured by the ratio of total debt to total assets at the end of the fiscal year; 13

15 ROA = return on assets calculated as the ratio of earnings before interest and tax divided by total assets; LOSS = 1 for firms with negative net income, 0 otherwise; OCF = cash flow from operations; INVREC = total inventories plus total receivables divided by total assets; GROWTH = earnings growth measured as the difference between the mean analysts earnings forecasts for four and three years ahead divided by the mean of three year ahead earnings forecasts; BP = ratio of book value of equity to market value of equity at the end of the fiscal year; and BETA = share beta (systematic risk) calculated over 36 months to the fiscal year-end. We include several control variables in order to control for factors affecting discretionary accruals. SIZE is included in the model because the accrual behavior of executives of large and small firms is different. Specifically, executives of larger firms report more stable discretionary accruals (Dechow and Dichev 2002). Dechow, Sloan, and Sweeney (1995) document that overestimated accruals may resulted from poorly performing firms. As a result, LEV, ROA, and LOSS are included in the model to capture the financial condition of the firm. We include OCF in the model because Dechow (1994) report the negative association between cash flows and accruals. We include INVREC in the model because firms may report higher proportions of inventory and receivables due to greater opportunities for earnings management (Ittonen, 14

16 Vähämaa, and Vähämaa 2013). Growth firms may report systematically high levels of discretionary accruals (Gul, Fung, and Jaggi 2009; Ittonen, Vähämaa, and Vähämaa 2013). Therefore, GROWTH and BP are included to capture the difference in the accruals behavior between high growth firms and low growth firms. We include BETA in the model because previous studies find that the level of earnings management decreases with BETA Cost of equity capital Following previous research (i.e., Boone, Khurana, and Raman 2011; Hail and Leuz 2006; Dhaliwal, Heitzman, and Zhen Li 2006), to mitigate the effects of specific assumptions that underlie each method on the results, the dependent variable is a measure of the average expected cost of equity capital from different alternative approaches. Boone, Khurana, and Raman (2011) use the average of the four alternative approaches as r GLS, r CT, r OJN, and r MPEG. Nevertheless, in this paper, we measure the average value of the expected cost of equity capital by using nine estimates used in prior literature (Easton 2004; Ohlson and Juettner- Nauroth 2005; Gode and Mohanram 2003; Botosan and Plumlee 2002; Gordon and Gordon 1997; Claus and Thomas 2001; Gebhardt, Lee, and Swaminathan 2001; Hail and Leuz 2006; Dhaliwal, Krull, and Li 2007). These are r PEG, r PEGST, r MPEG, r OJN, r GM, r BP, r GG, r CT, and r GLS, as defined in Table 2. [insert Table 2 here] We use the following model to test our hypotheses regarding the cost of equity capital: 15

17 COE = β 0 + β 1 LEV + β 2 BP + β 3 BETA + β 4 GROWTH + β 5 SIZE + β 6 VAR + β 7 ROA + β 8 EQ + Variables of interest + ε (3) The variables are defined as follows: COE = cost of equity capital estimated by nine models as listed in Table 1; LEV = financial leverage measured by the ratio of total debt to total assets at the end of the fiscal year; BP = ratio of book value of equity to market value of equity at the end of the fiscal year; BETA = share beta (systematic risk) calculated over 36 months to the fiscal year-end; GROWTH = earnings growth measured as the difference between the mean analysts earnings forecasts for four and three years ahead divided by the mean of three year ahead earnings forecasts; SIZE = size measured by the natural logarithm of the market value of common equity at the end of the fiscal year; VAR = earnings variability measured by the standard deviation of analysts earnings forecasts available on I/B/E/S International during the fiscal year-end month; ROA = return on assets calculated as the ratio of earnings before interest and tax divided by total assets; and EQ = earnings quality measured as the absolute value of residuals using the Dechow and Dichev (2002) approach. 16

18 Consistent with prior research, we include commonly used control variables to capture the effects of other factors that impact cost of equity capital. Prior research (e.g., Fama and French 1992; Gebhardt, Lee, and Swaminathan 2001) find a positive relation between the perceived risk associated with leverage (LEV) and the cost of equity capital. Previous studies (e.g., Fama and French 1997, 1992; Khurana and Raman 2004; Boone, Khurana, and Raman 2008) indicate that book-to-price ratios (BP) is positively associated with the cost of equity capital. A firm s systematic risk (BETA) is positively related to its cost of equity capital (Botosan and Plumlee 2005). Beaver, Kettler, and Scholes (1970) and La Porta (1996) document a positive association between growth (GROWTH) and the cost of equity capital because earnings from growth opportunities are riskier than normal earnings. As in prior research (e.g., Khurana and Raman 2004; Botosan and Plumlee 2005; Boone, Khurana, and Raman 2008; Brennan and Subrahmanyam 1996; Fama and French 1997; Gebhardt, Lee, and Swaminathan 2001), firm size (SIZE), as measured by market capitalization, is included in the model. Earnings variability (VAR) is controlled for in the model because Gebhardt, Lee, and Swaminathan (2001) present that stable and increasing earnings lead to lower risk premiums. Return on assets (ROA) has implications for the financial health of the firm and reported earnings directly affect the cost of capital through investors expectations of returns (Gebhardt, Lee, and Swaminathan 2001; Gode and Mohanram 2003). Francis et al. (2005) report that firms with good earnings quality (EQ) have a lower cost of capital than firms with poor earnings quality. 5. Results 5.1. Descriptive statistics 17

19 We report descriptive statistics for the full sample in Table 3. The mean (median) discretionary accruals is 0.02 (0.02) with a standard deviation of The mean (median) cost of equity capital for the average value of nine estimates is 0.10 (0.09) with a standard deviation of Boone, Khurana, and Raman (2011) report lower mean and median cost of equity capital of and respectively, which may be attributable to the different number of estimates used to calculate average value and their larger sample. [insert Table 3 here] 5.2. Univariate correlations We report Pearson (below the diagonal) and Spearman (above the diagonal) pairwise correlations among the variables for the full sample in Table 4. As expected, discretionary accruals and cost of equity capital are strongly correlated with many executive compensation variables. For Pearson correlation, we find a significantly positive correlation between discretionary accruals and STOPCEO, STOPCFO, and STOPALL. We find a significantly negative correlation between discretionary accruals and BONCEO, BONCFO, and BONALL. However, the Pearson correlation between discretionary accruals and SHARECEO, SHARECFO, SHAREALL, LTPPCEO, LTPPCFO, LTPPALL is not reliably different from zero. In addition, we find a significantly negative Pearson correlation between cost of equity capital and SHARECEO, BONCEO, LTPPCEO, BONCFO, LTPPCFO, BONALL, and LTPPALL. However, the Pearson correlation between cost of equity capital and STOPCEO, STOPCFO, SHARECFO, STOPALL, and SHAREALL is not reliably different from zero. 18

20 For Spearman correlation, we find a significantly positive correlation between discretionary accruals and STOPCEO, STOPCFO, and STOPALL. We find a significantly negative correlation between discretionary accruals and SHARECEO, BONCEO, SHARECFO, BONCFO, SHAREALL, and BONALL. However, the Spearman correlation between discretionary accruals and LTPPCEO, LTPPCFO, LTPPALL is not reliably different from zero. In addition, we find a significantly negative Spearman correlation between cost of equity capital and BONCEO, LTPPCEO, BONCFO, LTPPCFO, BONALL, and LTPPALL. However, the Spearman correlation between cost of equity capital and STOPCEO, SHARECEO, STOPCFO, SHARECFO, STOPALL, and SHAREALL is not reliably different from zero. We diagnose multicollinearity in the regressions using variable inflation factors (VIFs) as discussed in the section of regression results. [insert Table 4 here] 5.3. Regression results Discretionary accruals Table 5 presents regression results of the effect of executive compensation on the discretionary accruals, testing H1a, H2a, H3a and H4a with respect to earnings management. We perform separate analyses for CEO, CFO, and top five highest-paid executives. Each model shows the R-squared value of , and , respectively. The low VIFs suggest that collinearity is not likely to be a problem in interpreting the regression results. The coefficient estimates of BONCEO, LTPPCEO, BONCFO, LTPPCFO, BONALL, and LTPPALL are significant with a negative sign, showing that bonus and long-term performance plan of CEO, 19

21 CFO, and top-five highly paid executives are associated with a lower discretionary accruals. Remarkably, the coefficient of SHARECEO is also significant with a negative sign, indicating that CEO shareholdings is also related to a lower discretionary accruals. The coefficient estimates of STOPCEO, STOPCFO, SHARECFO, STOPALL, SHAREALL are not statistically significant, suggesting that CEO stock option, CFO stock option, top-highly paid executives stock option, CFO shareholdings, and top-highly paid executives shareholdings are not related to the change of discretionary accruals. Overall, the results support H3a and H4a. [insert Table 5 here] Cost of equity capital Table 6 reports regression results of the effect of executive compensation on the cost of equity capital, testing H1b, H2b, H3b and H4b. We perform separate analyses for CEO, CFO, and top five highest-paid executives. Each model shows the R-squared value of , and , respectively. The low VIFs suggest that collinearity is not likely to be a problem in interpreting the regression results. The coefficient estimates of BONCEO, LTPPCEO, BONCFO, LTPPCFO, BONALL, and LTPPALL are negative and significant, showing that bonus and long-term performance plan of CEO, CFO, and top-five highly paid executives are associated with a lower cost of equity capital. Remarkably, the coefficient of SHARECEO is also significant with a negative sign, indicating that CEO shareholding is also related to a lower cost of equity capital. The coefficient estimates of STOPCEO, STOPCFO, SHARECFO, STOPALL, SHAREALL are not statistically significant, suggesting that CEO stock option, CFO stock option, top-highly paid executives stock option, CFO shareholdings, and top-highly paid executives shareholdings are 20

22 not related to the change of cost of equity capital. Taken together, the results support H3b and H4b. [insert Table 6 here] 6. Additional analyses We partition the sample into two groups based on the sign of DACC to examine whether there is a differential relation between executive compensation components and earnings management behavior, which result in cost of equity capital, conditional on whether firms engage in income-increasing or income-decreasing accruals. Table 7 reports regression results for discretionary accruals that estimated separately for the subsamples of income-increasing (Panel A) and income-decreasing accruals (Panel B). We find that executive compensation components affect only income-increasing earnings management, but not income-decreasing earnings management. Focusing on the variables of interest tested in discretionary accruals model, the results for income-increasing accruals subsample are in line with the results for full sample, except for CEO shareholdings that becomes statistically insignificant. [insert Table 7 here] The results of cost of equity capital are consistent with the results of discretionary accruals. Table 8 reports regression results for cost of equity that estimated separately for the subsamples of income-increasing (Panel A) and income-decreasing accruals (Panel B). We find that executive compensation components of firms with income-increasing earnings 21

23 management and income-decreasing earnings management affect cost of equity capital. Focusing on the variables of interest tested in cost of equity model, the results for both incomeincreasing accruals and income-decreasing accruals subsamples are in line with the results for full sample, except for CEO shareholdings of firms with income-increasing accruals that becomes statistically insignificant and CFO shareholdings of firms with income-decreasing accruals that becomes statistically significant. [insert Table 8 here] 7. Conclusion This study examines how executive compensation components affect earnings management and cost of equity capital. We hypothesize that the different types of executive compensation lead to different effects on earnings management and cost of equity capital. We test our hypotheses by using 3,436 firm-year observations of U.S. firms over the years 2004 to The results of discretionary accruals are consistent with the results of cost of equity capital. We find that executive bonus and long-term performance plan lead to the reduction in both earnings management and cost of equity capital. We also find that only executive bonus and long-term performance plan of firms that engage in income-increasing accruals exhibit significantly reduction in earnings management. Our findings reveal that shareholders realize the information risks resulted from the type of executive compensation and value these risks into the cost of equity capital. These findings would be of interest to regulators, standard setters, managers and investors who are interested in executive compensation design, earnings management, and cost of equity capital. 22

24 References Armstrong, C. S., A. D. Jagolinzer, and D. F. Larcker, 2010, Chief executive officer equity incentives and accounting irregularities, Journal of Accounting Research 48, Ball, R., and L. Shivakumar, 2008, Earnings quality at initial public offerings, Journal of Accounting and Economics 45, Ball, R. A. Y., and L. Shivakumar, 2006, The role of accruals in asymmetrically timely gain and loss recognition, Journal of Accounting Research 44, Beaver, W., P. Kettler, and M. Scholes, 1970, The association between market determined and accounting determined risk measures, Accounting Review, Bergstresser, D., and T. Philippon, 2006, CEO incentives and earnings management, Journal of financial economics 80, Boone, J. P., I. K. Khurana, and K. K. Raman, 2008, Audit firm tenure and the equity risk premium, Journal of Accounting, Auditing & Finance 23, Boone, J. P., I. K. Khurana, and K. K. Raman, 2011, Investor pricing of CEO equity incentives, Review of Quantitative Finance and Accounting 36, Botosan, C. A., and M. A. Plumlee, 2002, A re-examination of disclosure level and the expected cost of equity capital, Journal of Accounting Research 40, , 2005, Assessing alternative proxies for the expected risk premium, Accounting Review 80, Brennan, M. J., and A. Subrahmanyam, 1996, Market microstructure and asset pricing: On the compensation for illiquidity in stock returns, Journal of financial economics 41, Brown, P., W. Beekes, and P. Verhoeven, 2011, Corporate governance, accounting and finance: A review, Accounting & Finance 51, Bryan, S., L. Hwang, and S. Lilien, 2000, CEO stock-based compensation: An empirical analysis of incentive-intensity, relative mix, and economic determinants, The Journal of Business 73,

25 Burns, N., and S. Kedia, 2006, The impact of performance-based compensation on misreporting, Journal of financial economics 79, Cheng, Q., and T. D. Warfield, 2005, Equity incentives and earnings management, The Accounting Review 80, Claus, J., and J. Thomas, 2001, Equity Premia as Low as Three Percent? Evidence from Analysts' Earnings Forecasts for Domestic and International Stock Markets, The Journal of Finance 56, Cohen, D. A., A. Dey, and T. Z. Lys, 2008, Real and accrual-based earnings management in the pre- and post-sarbanes-oxley periods, The Accounting Review 83, Cornett, M. M., J. J. McNutt, and H. Tehranian, 2009, Corporate governance and earnings management at large U.S. bank holding companies, Journal of Corporate Finance 15, 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, Dechow, P. M., and I. D. Dichev, 2002, The quality of accruals and earnings: The role of accrual estimation errors, The Accounting Review 77, Dechow, P. M., R. G. Sloan, and A. P. Sweeney, 1995, Detecting earnings management, The Accounting Review 70, Devers, C. E., A. A. Cannella, G. P. Reilly, and M. E. Yoder, 2007, Executive compensation: A multidisciplinary review of recent developments, Journal of Management 33, Dhaliwal, D., L. Krull, and O. Z. Li, 2007, Did the 2003 Tax Act reduce the cost of equity capital?, Journal of Accounting and Economics 43, Dhaliwal, D. A. N., S. Heitzman, and O. Zhen Li, 2006, Taxes, leverage, and the cost of equity capital, Journal of Accounting Research 44, Easton, P. D., 2004, PE Ratios, PEG Ratios, and Estimating the Implied Expected Rate of Return on Equity Capital, The Accounting Review 79,

26 Efendi, J., A. Srivastava, and E. P. Swanson, 2007, Why do corporate managers misstate financial statements? The role of option compensation and other factors, Journal of financial economics 85, Fama, E. F., and K. R. French, 1992, The cross-section of expected stock returns, the Journal of Finance 47, , 1997, Industry costs of equity, Journal of financial economics 43, Fee, C. E., and C. J. Hadlock, 2004, Management turnover across the corporate hierarchy, Journal of Accounting and Economics 37, Francis, J., R. LaFond, P. Olsson, and K. Schipper, 2005, The market pricing of accruals quality, Journal of Accounting and Economics 39, Gaver, J. J., K. M. Gaver, and J. R. Austin, 1995, Additional evidence on bonus plans and income management, Journal of Accounting and Economics 19, Gebhardt, W. R., C. M. C. Lee, and B. Swaminathan, 2001, Toward an Implied Cost of Capital, Journal of Accounting Research 39, Gode, D., and P. Mohanram, 2003, Inferring the Cost of Capital Using the Ohlson-Juettner Model, Review of Accounting Studies 8, 399. Gordon, J. R., and M. J. Gordon, 1997, The Finite Horizon Expected Return Model, Financial Analysts Journal 53, 52. Grant, J., G. Markarian, and A. Parbonetti, 2009, CEO risk-related incentives and income smoothing, Contemporary Accounting Research 26, Guidry, F., A. J. Leone, and S. Rock, 1999, Earnings-based bonus plans and earnings management by business-unit managers, Journal of Accounting and Economics 26, Gul, F. A., S. Y. K. Fung, and B. Jaggi, 2009, Earnings quality: Some evidence on the role of auditor tenure and auditors industry expertise, Journal of Accounting and Economics 47, Hail, L., and C. Leuz, 2006, International Differences in the Cost of Equity Capital: Do Legal Institutions and Securities Regulation Matter?, Journal of Accounting Research 44,

27 Healy, P. M., 1985, The effect of bonus schemes on accounting decisions, Journal of Accounting and Economics 7, Holthausen, R. W., D. F. Larcker, and R. G. Sloan, 1995, Annual bonus schemes and the manipulation of earnings, Journal of Accounting and Economics 19, Houmes, R. E., and T. R. Skantz, 2010, Highly valued equity and discretionary accruals, Journal of Business Finance & Accounting 37, Ittonen, K., E. Vähämaa, and S. Vähämaa, 2013, Female auditors and accruals quality, Accounting Horizons 27, Jiang, J., K. R. Petroni, and I. Yanyan Wang, 2010, CFOs and CEOs: Who have the most influence on earnings management?, Journal of financial economics 96, Kabir, R., H. Li, and Y. V. Veld-Merkoulova, 2013, Executive compensation and the cost of debt, Journal of Banking & Finance 37, Khurana, I. K., and K. Raman, 2004, Litigation risk and the financial reporting credibility of Big 4 versus non-big 4 audits: Evidence from Anglo-American countries, The Accounting Review 79, Kothari, S. P., A. J. Leone, and C. E. Wasley, 2005, Performance matched discretionary accrual measures, Journal of Accounting and Economics 39, La Porta, R., 1996, Expectations and the cross-section of stock returns, the Journal of Finance 51, Larcker, D. F., 1983, The association between performance plan adoption and corporate capital investment, Journal of Accounting and Economics 5, Laux, C., and V. Laux, 2009, Board committees, CEO compensation, and earnings management, The Accounting Review 84, McAnally, M. L., A. Srivastava, and C. D. Weaver, 2008, Executive stock options, missed earnings targets, and earnings management, The Accounting Review 83, Meek, G. K., R. P. Rao, and C. J. Skousen, 2007, Evidence on factors affecting the relationship between CEO stock option compensation and earnings management, Review of Accounting and Finance 6,

28 Mian, S., 2001, On the choice and replacement of chief financial officers, Journal of financial economics 60, Murphy, K. J., 2000, Performance standards in incentive contracts, Journal of Accounting and Economics 30, Ohlson, J. A., and B. E. Juettner-Nauroth, 2005, Expected EPS and EPS Growth as Determinantsof Value, Review of Accounting Studies 10, Pourciau, S., 1993, Earnings management and nonroutine executive changes, Journal of Accounting and Economics 16, Prevost, A. K., E. Devos, and R. P. Rao, 2013, The effects of relative changes in CEO equity incentives on the cost of corporate debt, Journal of Business Finance & Accounting 40, Richardson, V. J., and J. F. Waegelein, 2002, The influence of long-term performance plans on earnings management and firm performance, Review of Quantitative Finance and Accounting 18, Shaw, K. W., 2012, CEO incentives and the cost of debt, Review of Quantitative Finance and Accounting 38, Shrieves, R. E., and P. Gao, 2002, Earnings management and executive compensation: A case of overdose of option and underdose of salary?, SSRN elibrary. Smith, C. W., and R. L. Watts, 1992, The investment opportunity set and corporate financing, dividend, and compensation policies, Journal of financial economics 32, Tehranian, H., N. G. Travlos, and J. F. Waegelein, 1987a, The effect of long-term performance plans on corporate sell-off-induced abnormal returns, the Journal of Finance 42, , 1987b, Management compensation contracts and merger-induced abnormal returns, Journal of Accounting Research 25, Vafeas, N., and J. Waegelein, 2007, The association between audit committees, compensation incentives, and corporate audit fees, Review of Quantitative Finance and Accounting 28,

29 Warfield, T. D., J. J. Wild, and K. L. Wild, 1995, Managerial ownership, accounting choices, and informativeness of earnings, Journal of Accounting and Economics 20, Wiseman, R. M., and L. R. Gomez-Mejia, 1998, A behavioral agency model of managerial risk taking, The Academy of Management Review 23, Wright, P., M. Kroll, A. Lado, and B. Van Ness, 2002, The structure of ownership and corporate acquisition strategies, Strategic Management Journal 23, Table 1 Descriptive information on year and industry distribution Panel A: Distribution of sample firms by year Year N Percent Total 3, Panel B: Distribution of sample firms by industry SIC Industry N Percent Manufacturing 1, Services Transportation and Public Utilities Retail Trade Wholesale Trade Mining Construction Agriculture, Forestry, and Fishing Public Administration Total 3,

30 Table 2 Cost of equity capital equations COE proxies (Easton 2004) (Easton 2004) (Easton 2004) Equation (Ohlson and Juettner- Nauroth 2005) (Gode and Mohanram 2003) (Botosan and Plumlee 2002) (Gordon and Gordon 1997) (Claus and Thomas 2001) (Gebhardt, Lee, and Swaminathan 2001) ,, = cost of equity capital estimated approaches introduced by Easton (2004); = cost of equity capital estimated approach introduced by Ohlson and Juettner-Nauroth (2005); = cost of equity capital estimated approach introduced by Gode and Mohanram (2003); = cost of equity capital estimated approach introduced by Botosan and Plumlee (2002); = cost of equity capital estimated approach introduced by Gordon and Gordon (1997); = cost of equity capital estimated approach introduced by Claus and Thomas (2001); = cost of equity capital estimated approach introduced by Gebhardt, Lee, and Swaminathan (2001); eps = earnings per share; P = stock price at fiscal year end; dps = dividend per share; bps = book value per share; and roe = return on equity. 29

31 Table 3 Descriptive statistics for sample companies Total Sample (n = 3,436) Mean Std. Deviation Min 25% Median 75% Max Dependent Variables DACC COE Control Variables LOSS OCF INVREC LEV BP BETA GROWTH SIZE VAR ROA EQ Variables of interest STOPCEO SHARECEO BONCEO LTPPCEO STOPCFO SHARECFO BONCFO LTPPCFO STOPALL SHAREALL BONALL LTPPALL DACC = residual from TACC model; COE = cost of equity capital estimated by nine models as listed in Table 1; LOSS = 1 for firms with negative net income, 0 otherwise; OCF = cash flow from operations; INVREC = total inventories plus total receivables divided by total assets; LEV = financial leverage measured by the ratio of total debt to total assets at the end of the fiscal year; BP = ratio of book value of equity to market value of equity at the end of the 30

32 fiscal year; BETA = share beta (systematic risk) calculated over 36 months to the fiscal year-end; GROWTH = earnings growth measured as the difference between the mean analysts earnings forecasts for four and three years ahead divided by the mean of three year ahead earnings forecasts; SIZE = size measured by the natural logarithm of the market value of common equity at the end of the fiscal year; VAR = earnings variability measured by the standard deviation of analysts earnings forecasts available on I/B/E/S International during the fiscal yearend month; ROA = return on assets calculated as the ratio of earnings before interest and tax divided by total assets; EQ = earnings quality measured as the absolute value of residuals using the Dechow and Dichev (2002) approach; STOPCEO = Percentage of compensation in the form of grants of stock options to the CEO during the current financial year; SHARECEO = Percentage of total shares outstanding held by the CEO at the balance sheet date, excluding options; BONCEO = Percentage of compensation in the form of a bonus earned by the CEO during the current financial year; LTPPCEO = Percentage of compensation in the form of the amount paid to the CEO during the current financial year under the company's long-term incentive plan; STOPCFO = Percentage of compensation in the form of grants of stock options to the CFO during the current financial year; SHARECFO = Percentage of total shares outstanding held by the CFO at the balance sheet date, excluding options; BONCFO = Percentage of compensation in the form of a bonus earned by the CFO during the current financial year; LTPPCFO = Percentage of compensation in the form of the amount paid to the CFO during the current financial year under the company's long-term incentive plan; STOPALL = Percentage of compensation in the form of grants of stock options to the top highly paid executives during the current financial year; SHAREALL = Percentage of total shares outstanding held by the top highly paid executives at the balance sheet date, excluding options; BONALL = Percentage of compensation in the form of a bonus earned by the top highly paid executives during the current financial year; and LTPPALL = Percentage of compensation in the form of the amount paid to the top highly paid executives during the current financial year under the company's long-term incentive plan. 31