The effect of Free Cash Flow, Dividend and Leverage to Earnings. Management. Evidence from Malaysia.

The effect of Free Cash Flow, Dividend and Leverage to Earnings Management. Evidence from Malaysia. a Nor Balkish Zakaria, b Zuraidah Mohd Sanusi and c Intan Salwani Mohamed a Accounting Research Institute and Universiti Teknologi MARA Johor, Malaysia. b,c Accounting Research Institute and Universiti Teknologi MARA Shah Alam, Selangor, Malaysia. Acknowledgement: Authors would like to thanks the financial and administrative support from Research Management Institute and Accounting Research Institute of Universiti Teknologi MARA, Malaysia. Abstract Earnings management is said to trifle the application of accounting rules. Despite the hardship to distinguish these allowable practices from earnings fraud, earnings management theoretically still represents the gray area of accounting practices and synonym for earnings fraud. Based on a sample of 273 Malaysian listed firms between 2010-2012, this study investigates the influence of free cash flow, dividend and leverage on earnings management through the discretionary accrual based on Kothari s and Kasznik models. The population firms are the construction, consumer product, industrial product and trading & services. The findings show that free cash flow and leverage are significant to determine the discretionary accrual in Malaysia. These results should be of interest to the stakeholders to oversee the integrity of corporate financial reporting. 1

Keywords: Free cash flow, leverage, dividend, earnings fraud, discretionary accruals 1.0 INTRODUCTION Earnings management (EM) generally implies activities undertaken to smooth earnings over accounting periods or to achieve a designated earnings level, perhaps to meet analysts forecasts. The motivation to manage earnings is due to management s intention to achieve targeted results. The motivation could also come from pressures on management both outside and inside the entity. External pressures also come principally from the capital markets especially in maintaining the shareholders value. Earnings therefore might be used by insiders (management) for their own benefits against the interest of shareholders (Syed Zulfiqar, Hui & Nousheen, 2010). This happened because the information asymmetry between the managers and external information users allow managers to use their discretion in preparing and reporting accounting information for their own advantage (Spohr, 2005). Based on Agency Theory, agency problem occurs because investors and other stakeholders may not be able to make optimal decisions concerning a company when EM distorted economic results and hinder the ability of all stakeholders to make financial decisions. Managers may have personal goals that conflicted shareholders. Moreover, since managers have been empowered by shareholders to make decisions, a conflict of interests has potential agency costs. The information asymmetry between managers and external information users allows managers to use their discretion in preparing and reporting accounting information for their own personal benefits. Although opportunism is limited through the loopholes in 2

accounting standards and flexibility of external auditors, there are some evidences suggesting that managers used their discretion over accounting numbers to achieve personal gains (DeAngelo, DeAngelo & Wruck, 2002; Spohr, 2005). According to Mohammed (2009), the Saudi Stock Market faced an extraordinary crash at the beginning of 2006, which leads the Capital Market Authority (CMA) to suspend the trading of two firms. This event created a serious question about the effectiveness of different monitoring devices that were presumed to protect investors interests in Saudi Arabia. While firms of Real Estate Investment Trusts (REITs) in U.S engaged in real EM to mitigate the effects of the dividend payout regulations since REITs are legally obligated to pay common share dividends equal to at least 90% of taxable income as dividends to common shareholders (Edelstein, 2009). In Malaysia, the 1997 economic crisis had exposed serious weaknesses such as weak financial structure, over-leveraging by companies, lack of transparency, disclosure and accountability (Rashidah & Fairuzana Haneem, 2006). In a response to this economic disaster, Malaysian Code of Corporate Governance (MCCG) was introduced in 2001. The code sets out the principles and best practices on corporate governance to improve the monitoring function of the board of directors, audit committee, and the external audit. MCCG was revised in 2007 with amendments that aimed at strengthening the board of directors and audit committees to ensure that the board of directors and audit committees discharge their roles and responsibilities more effectively. The amendments spell out the eligibility criteria for appointment of directors and the role 3

of the nominating committee. On audit committees, the amendments spell out the eligibility criteria for appointment as an audit committee member, the composition of audit committees, the frequency of meetings and the need for continuous training. Motivated with this revised MCCG, this study aims to examine the effect of free cash flow, dividend and leverage to EM. Therefore, data will be collected from 2009 to 2012 where 2008 is believed to be the year that this revised MCCG was steadily been implemented by Malaysian listed firms. However this study will not examine any corporate governance practices among the sample firms as the exploratory factors of interest that will be examined are only the free cash flow, dividend and leverage position with regards to EM. Nonetheless, the interest of this study lies on the EM practices among Malaysian firms with regards to free cash flow, dividend and leverage post this revised MCCG implementation. The remainder of the paper is organised as follows: Section 2 provides the literature review on EM with regard to free cash flow, dividend and leverage and develops the hypotheses; Section 3 describes the research design and methodology; Section 4 presents the results; and finally Section 5 summarises and concludes. 2.0 LITERATURE REVIEW 2.1 Earnings Management According to Healey and Wahlen (1999) EM occurs when managers choose reporting methods and estimation that does not accurately reflects their firms underlying economics. While Leuz, Nanda and Wysocki (2003) defined EM as the alteration of firms reported economic performance by insiders to either mislead some stakeholders 4

or to influence contractual outcomes. In addition, Sun and Rath (2008), as suggested in general, defined EM as when managers exercise judgment for the purpose of hiding true performance in order to either influence the stock performance, to benefit from the contractual terms between the firm and managers, or to influence regulatory decisions. Although EM has been defined in many ways, the basic concept of EM underlies on the change of financial reporting information by the management. There are many reasons why managers tend to manage the earnings. Healey & Wahlen (1999) found that firms manage the earnings to window-dress financial statements prior to public securities offerings, to increase corporate managers compensation and job security, to avoid violating lending contracts, to reduce regulatory costs or to increase regulatory benefits. Duncan (2001) also argued to some factors listed above, for example to meet analysts forecasts, access to debt markets, competition, contractual obligation, a roaring stock market, new financial transactions, market disregard of big charges, merger attractiveness, management compensation, short-term focus, unrealistic plans and budgets, period-end requests from superiors, periods of excessive profit followed by a fear of subsequent decline, concealing unlawful transactions, personal bonuses, promotions, focus on team, and job retention. 2.2 Free Cash Flow and EM Any unused of free cash flow in a company could raises agency problem (Jensen, 1986). This happened because managers may have the choice of any unprofitable investments that benefits their self-interest. As a result company may experience low growth. Adding to this, Chung et al. (2005) posited that a minimal information 5

disclosure by the managers may troublesome the investors to discover any disadvantages of the project especially for the sake of the managers wealth. Based on personal interest, managers overlook the need for projection cash flow and profit forecast. Poor investments will therefore reduce future earnings and result in management turmoil. To avoid this, managers manipulate accounting numbers in order to increase reported earnings (Bukit and Mohd Iskandar, 2009). Mashayekhi,Bagheri and Tahriri (2010) posit that managers of the firms with high free cash flow and low growth probably manage the earnings in order to get some self-interest in the short time. Moreover, Jensen (1986) controls the predicted hypothesis that the financial leverage can be used to adjust that relation for two reasons. First, required debt repayments reduce the cash available to management for non-optimal spending. Second, when a firm employs debt financing, it must undergo the security of lenders and is often subject to lender-induced spending restrictions. Thus, this study hypothesize, H1: Firms with higher level of free cash flow would have higher earnings management. 2.3 Dividend and EM Dividend could be claimed as the distribution of profits to a company's shareholders. It is also known as cash disbursement strategy for public listed that seeks to return cash or assets to their shareholders (Jensen, 1986). According to Farinha and Moreira (2007), dividends would signal not only future profitability but also earnings quality 6

(with the sense of earnings with a low degree of manipulation). They also added that an alternative perspective could be that the presence of dividends as an effective managerial monitoring device, the managers are expected to be less tempted to engage in earnings manipulation for self-serving reasons. Agency theory has advanced the different views that dividends may help to reduce agency problems between managers and shareholders (Jensen, 1986). Dividends are able to disperse free cash-flows. However, when firms are close to debt covenants which require dividends to be constrained, managers might be tempted to manipulate earnings upward to maintain dividend (Farinha and Moreira, 2007). Thus, the potential role of dividend policy as a managerial vehicle by which firms can convey to the market a credible signal that their reported income is not being artificially inflated by accounting options (i.e. earnings management). Dividend policy is needed as unpredictable dividend policy would mean surprises to shareholders and might result in a drop in the firm s stock price. Thus, a well-planned dividend policy could prevent these surprises and preserve or even enhance stock price (Fey, Mohd Lukman, Aidil Rizal & Mohd Said, 2008). According to Syed Zulfikar et al. (2010), dividends can be used as a predictor of earnings whereas another view is that earnings can also be used as a predictor of dividends. They added that dividends are considered as a way to mitigate agency problem (i.e earnings management) by giving shareholder their share. Thus, the second hypothesis is, H2: Firms with higher level of dividend payment would have lower earnings management. 7

2.4 Leverage and EM According to Irina (2009), companies that have high leverage may face the risk of bankruptcy if they are unable to make payments on their external debt financing. If high leverage company wishes to take out a new loan, lenders will scrutinize several measures and demand that it keeps its debt within reasonable boundaries. Some previous literature said that leverage increases the potential for managers to EM to avoid debt covenant violations. Sweeney (1994) provided direct evidence to support a debt hypothesis that the larger a firm s debt to equity ratio, the more likely the firm s manager to select income increasing accounting procedures. Studies by Dichev and Skinner (2002), Beatty and Webber (2003) and DeFond and Jiambalvo (1994) found positive relationship that managers manage earnings to avoid debt covenant violations (Jelinek, 2007). Companies that have high leverage may face the risk of bankruptcy if they are unable to make payments on their external debt financing. Some previous literature on leverage and EM said that leverage increases the potential for EM to avoid debt covenant violations. There is evidence that managers make accounting choices to reduce likelihood for the firms to violate debt covenant (Dichev and Skinner, 2002), (Beatty and Webber, 2003). Sweeney (1994) provided direct evidence to support a debt hypothesis that the larger a firm s debt to equity ratio, the more likely the firm s manager to select income increasing accounting procedures. However, based on Malaysian results, Aini, Takiah, Pourjalali, and Teruya (2006) showed that leverage does not affecting EM in Malaysia. They claimed that the corporate sector in 8

Malaysia became more leveraged and heavily dependent on commercial bank financing after the 1997 economic crisis. They added that financial difficulties faced by companies might transpire managers to improve upon their performance through earnings management. Hence, this study hypothesize, H3: Firms with higher level of leverage would have lower earnings management. 3.0 RESEARCH METHODOLOGY 3.1 Sample Selection The data associated with all firms in the four largest industries that are listed on the main board of Bursa Malaysia construction, consumer product, industrial product and trading & services. The study period was from 2010 to 2012; however the data was collected for the period from 2009 to 2012. The departing point was identifying the firms categorised under these 4 sectors from Datastream database as at 31 st July 2013. There were 32 firms under construction, 84 under consumer product, 102 industrial products and finally 73 under trading & services. Nevertheless, 18 firms have to be excluded due to insufficient data. Therefore a final sample became 273 firms that make up 819 firm-years observation. 3.2 Measurement of Variables 3.3.1 Earnings Management This study uses Kothari, Leone and Wasley (2005) discretionary accrual (DA) model as a measurement of earnings management. Discretionary accruals are the residual value of total accruals and non-discretionary accruals. The usage of this measurement 9

is consistent with many studies of earnings management such as Abdul Rahman & Wan Abdullah (2005); Antle (2006); Carramanis & Lenox (2008) and Mohd. Ali, Mohd. Salleh & Hassan (2008). Discretionary accrual or abnormal accrual is widely used because of its ability to capture the quality of accounting information in a common sense and in a universal manner (Choi, Kim, Kim & Zhang, 2010). Kothari et al. estimated the total accruals as below: TA it = [( non-cash current asset it ) - ( current liabilities it excluding the current portion of long term debt) - (Depreciation and amortization it )] / total assets it-1 (3.1) Then discretionary accruals, a proxy for earnings management, are estimated by subtracting nondiscretionary accruals from total accruals, where all accrual variables are scaled by lagged total assets to control for potential scale bias. The use of assets as the deflator is intended to mitigate heteroscedasticity in residuals (Kothari, et al., 2005). The model also included the ROA it or ROA it 1. This approach is designed to provide a comparison of the effectiveness of performance matching including a performance measure in the accruals regression to generate the normal or nondiscretionary proxy (Kothari et al., 2005) with the following equation: TA it = β 0 + β 1 (1 / ASSETS it 1) + β 2 ( SALES it ) + β 3 PPE it + β 4 ROA it (or it 1) + ε it 10

(3.2) Where: ROA it (or it 1) = Net income in current year (or previous year) divided by total assets in the current year; and all other variables are as previously defined. Finally, the residual (ε it ) from the regression is the discretionary accruals (DAC). The independent variables in this current study are free cash flow, dividend and leverage. 3.3.2 Independent Variables a) Free Cash Flow Prior studies used free cash flow (Lehn & Poulsen, 1989) to reflects the surplus free cash flow and this also comes together with growth prospect indicated by the book ratio (Skinner, 1993). Lehn and Poulsen (1989) measured free cash flow as operating income before depreciation minus expenses such as tax expenses, interest expense and dividend. Chung et al. (2005) categorise companies as having potential free cash flow when the free cash flow of the companies is above median and price to book ratio is below median. This study computed free cash flow (FCF) by divided the ratio of free cash flow by total asset. b) Dividend According to Zunaidah and Fauzias (2008), dividend was primarily measured by dividend yield (dividend-to-price ratio). Hence, the dividend yield is the dividend per 11

share (DPS) divided by closing market price per share (MPS), that is, DYLD = DPS / MPS. Schooley and Barney (1994) posit that the dividend yield was used rather than the payout ratio (dividends to earnings) for two reasons. Firstly, the denominator in dividend yield is a market measure (share price) compared to an accounting measure (net income). Secondly, to avoid problems of negative payout ratios are resulting from negative earnings or excessively high payout ratios resulting from income being close to zero. This study measures dividend (DIVRATIO) as dividend per share (DPS) divided by closing market price per share (MPS) in year t (DPS / MPS c) Leverage There are different opinions on whether leverage increases or decreases the potential for EM to happen. Some literature such as Jensen (1986), Stulz (1990), and Hart and Moore (1995) suggested that debt discourages free cash flows over-investment by self-serving managers and it can be monitored by lenders. Hence it will reduce the tendency for EM to happen. Others somehow, argued that leverage increases the potential for EM (Dichev and Skinner, 2002), (Beatty and Weber, 2003), (Sweeney, 1994), (Watts and Zimmerman, 1990), (DeFond and Jiambalvo, 1994). The current study is keener to the second opinion in the sense that managers tend to implement EM to avoid debt covenant violation. The measurement of leverage (LEVERAGE) in this study followed the measurement of Kim and Yoon (2008) and Rashidah and Fairuzana Haneem (2006), thus leverage (LEVERAGE) is measured as total debt divided by total assets. 3.3.3 Control Variables 12

Accruals may vary across industries (Gu, Lee and Rosset, 2005; Barth, Cram and Nelson, 2001). Jelinek (2007) also coded industry indicator variables according to 43 Fama and French (1997) industry groupings. The sample firms were categorised according to four big industries provided by Bursa Malaysia (Bursa Malaysia Berhad, 2012). Four industries controlled in this current study are construction (CONSTRUCTION), consumer products (CONSPROD), industrial products (INSPROD) and trading and services (TRADINDSER). 3.4 Research Method The current study uses the descriptive statistics and standard regression analysis. SPSS statistical package is used for all analysis. To test the research hypotheses, a multivariate model was developed. Absolute discretionary accruals are measured based on Kothari s et al. (2005) model (KOTHDACC) which acted as the dependent variable. Free cash flow (FCF), Dividend (DIVRATIO) and leverage (LEVERAGE) appeared as independent variables; while CONSTRUCTION, CONSUMER PRODUCTS, INDUSTRIAL PRODUCT AND TRADING & SERVICES appeared as control variables. KOTHDACC it = β o + β 1 FCF it + β 2 DIVRATIO it + β 3 LEVERAGE it + Β 4 CONSTRUCTION + β 5 CONSPROD + β 6 INSPROD + β 7 TRADINGSER + ε it (3.3) Where : KOTHDACC it = Absolute Discretionary Accruals of firm i in year t based on 13

Kothari s et al. model as (3.2); FCF it = Free cash flow by divided the ratio of free cash flow by total asset. DIVRATIO it = Ratio of dividend per share (DPS) divided by closing market price per share (MPS) in year t (DPS / MPS); LEVERAGE it = Ratio of total debt over total asset for firm i in year t; CONSTRUCTION it = takes the value of 1 if the firm is construction firm; 0 otherwise; CONSPROD it = takes the value of 1 if the firm is consumer product; 0 otherwise INSPROD it = takes the value of 1 if the firm is industrial product; 0 otherwise. TRADINGSER it = takes the value of 1 if the firm is trading and services; 0 otherwise. ε it = Residual error. 4.0 RESULTS 4.1 Descriptive Analysis Several test of the ordinary least square (OLS) assumptions were conducted to ensure that significant multicollinearity between independent variables did not exist; there was no problem of heteroscedasticity; a linear relationship existed between dependent and independent variables; the values of dependent variables were reasonably normal for each of the independent variables; and that specification and measurement errors 14

did not exist. The regression diagnostics include an analysis of correlation matrix, variance inflation factor (VIF), residuals, Q-Q plot, skewness and kurtosis. Nonetheless, the analysis of skewness and kurtosis shows that the absolute discretionary accruals measures based on Kothari et al. (2005) model is significantly non-normally distributed with skewness outside the range of ±1.96 and kurtosis outside the range of ±2.0. To solve this problem, a method proposed by Cooke (1998) to normalise the variables, which is referred as Van der Waerden transformation is approached. Using this transformation procudure, variables from the actual observations were transformed to normal distributions by dividing the distribution into the number of observations plus one region on the basis that each region has equal probability (Cooke, 1998). The same transformation procedures were done for the independent variables with a non-normality problem, which were free cash flow (FCF) and dividend ratio (DIVRATIO). The empirical analysis used these normal scores to replace the actual values of the variables. Table 1: Descriptive Statistics of Variables. 15

CONTINUOUS VARIABLES Min Mean Maximum Median SD DISCRETIONARY ACCRUAL KOTHDACC 0.000 0.001 3.031 0.002 0.993 (KHOTARI) FREE CASH FLOW FCF 0.000 0.031 3.031 0.002 0.994 DIVIDEND RATIO DIVRATIO 0.000 0.200 3.031 0.172 0.927 LEVERAGE LEVERAGE 0.000 0.199 0.741 0.172 0.166 Frequency Percentage DICHOTOMOUS VARIABLES No=0 Yes=1 No=0 Yes=1 CONSTRUCTION CONSTRUCTION 753 66 91.9 8.1 CONSUMER PRODUCT CONSPROD 582 237 71.1 28.9 INDUSTRIAL PRODUCT INSPROD 521 298 63.6 36.4 TRADING AND SERVICES TRADINGSER 601 218 73.4 26.6 The sample consists of 819 firm-year observation for the period 2010-2012 corresponding to 273 firms. KOTHARIDACC is the absolute discretionary accruals estimated using Kothari et al. (2005) model. FCF is the free cash flow obtained from operation income before depreciation minus taxation, interest and dividend. DIVRATIO is the dividend per share divide by market price per share. LEVERAGE is total debt divide by total asset. CONSTRUCTION takes the value of 1 if the firm is construction firm; 0 otherwise. CONSPROD takes the value of 1 if the firm is consumer product; 0 otherwise. INSPROD takes the value of 1 if the firm is industrial product; 0 otherwise. TRADINGSER takes the value of 1 if the firm is trading and services; 0 otherwise. From Table 4.1 the descriptive statistics show that the average of free cash flow among the sample firms is 0.031. The mean and median of dividend ratio, as measured by DPS/market share price, is 0.200 and 0.172 respectively. The average leverage 0.199. The table also shows that majority of the sample firms (36.39%) comes from Industrial Product industry followed by consumer product (28.94%). 4.2 Point-Biserial Correlation Coefficient Since there are several dichotomous variables tested, point-biserial correlation will be more suitable to be used (Linacre, 2008). Table 4.2 reports the correlation coefficients of the EM determinants variables. The results show that the free cash flow (FCF) has 16

a positive and significant relationship with the dividend ratio (DIVRATIO) while dividend ratio (DIVRATIO) is negative and significant with leverage (LEVERAGE). CONTRSUTION industry shows negative and significant correlation results to other control variables consumer product (CONSPROD), industrial product (INSPROD) and trading & services (TRADINGSER). Overall the correlation values are all far below 0.70, suggesting that there is no problem of multicollinearity (Pallant, 2007). 17

Table 2: Correlation matrix of EM determinants variables. FCF DIVRATIO LEVERAGE CONSTRUCTION CONSPROD INSPROD DIVRATIO.335** LEVERAGE -.023 -.180** CONSTRUCTION.035.045.021 CONSPROD -.050.012.009 -.189** INSPROD -.022 -.047.025 -.224** -.483** TRADINGSER.053.012 -.049 -.178** -.384** -.455** The sample consists of 819 firm-year observation for the period 2010-2012 corresponding to 273 firms. The correlation coefficients are based on Pearson product correlations. * and ** represents statistical significance at 0.01 and 0.05 level respectively. FCF is the free cash flow obtained from operation income before depreciation minus taxation, interest and dividend. DIVRATIO is the dividend per share divide by market price per share. LEVERAGE is total debt divide by total asset. CONSTRUCTION takes the value of 1 if the firm is construction firm; 0 otherwise. CONSPROD takes the value of 1 if the firm is consumer product; 0 otherwise. INSPROD takes the value of 1 if the firm is industrial product; 0 otherwise. TRADINGSER takes the value of 1 if the firm is trading and services; 0 otherwise. 18

4.3 Regression Analysis This section reports the testing of the three (3) hypotheses developed in Chapter 2. The objective of this analysis is to examine the relationship between EM (discretionary accruals) and each independent variable for study period (2010-2012). The regression results are presented in Table 4.3. Based on a pool data, this study mainly regress the variables using regression model 3.4. Besides that, the Mahalanobis distance is used in order to detect the outliers. Table 3: Regression results of discretionary accruals estimated using Kothari et al. (2005) model. NKOTHARIDACC = β o + β 1 (1/T.A it-1) + β 2 ( SALES it ) + β 3 PPE it + β 4 ROA i + ε it Coefficients t-stat. VIF Intercept β 0 0.170** 2.346 FCF β 1 0.138* 2.938 1.895 DIVRATIO β 2-0.048-1.129 1.363 LEVERAGE β 3-0.553** -2.427 3.079 CONSTRUCTION β 4 0.059 0.460 1.166 CONSPROD β 5 0.075 0.953 1.569 INSPROD β 6 0.095 1.206 1.453 TRADINGSER β 7 0.170** 2.346 1.675 Adjusted R-Squared 0.052 N=819 Notes: * and ** represents statistical significance at 0.01 and 0.05 levels, respectively (two-tailed test). NKOTHARIDACC is the normal score of KOTHARIDACC using Van der Waerden s formula. KOTHARIDACC is the absolute value of firm discretionary accruals estimated using Kothari et al. (2005) model. FCF is the normal score of free cash flow using Van der Waerden s formula. DIVRATIO the normal score of dividend ratio using Van der Waerden s formula. LEVERAGE is the total debt divide by total assets. CONSTRUCTION takes the value of 1 if the firm is construction firm; 0 otherwise. CONSPROD takes the value of 1 if the firm is consumer product; 0 otherwise. INSPROD takes the value of 1 if the firm is industrial product; 0 otherwise. TRADINGSER takes the value of 1 if the firm is trading and services; 0 otherwise. 19

Table 3 presents the results of the ordinary least square regression used to test the relationship between earnings management (NKOTHARIDACC) and free cash flow (FCF), dividend (DIVRATIO) and leverage (LEVERAGE). The absolute value of discretionary accruals in this regression is estimated using the Kothari et al. (2005) model. From Table 3, 3 out of 7 variables are found to be significantly associated with EM. Free cash flow is positively associated to EM and significant at a 1% level. This result implies that the higher the free cash flow, the higher will be the EM practice among sample firms. This result supports H1: Firms with higher level of free cash flow would have higher earnings management. Dividend ratio shows a negative result to EM which implies firms with higher dividend ratio would have lower EM. However the result shows a non-significant association between dividend ratio and EM. Therefore, the result rejects the second hypothesis H2: Firms with higher level of dividend payment would have lower earnings management. Leverage shows a significant association at 5% level and negative to EM. This result implies that the higher the firms leverage, the lower will be the EM practice. This result therefore supports H3: Firms with higher level of leverage would have lower earnings management. This result is supported by Aini et al. (2006) that found leverage is not a concern to EM. However this result contradicts Beatty and Weber (2003), (Dichev and Skinner, 2002), and Sweeney (1994) 20

studies which suggested that firms engaging in EM are more likely to avoid debt covenant default from the leverage made. On the other hand, among the control variables, trading and services industry shows a positive and significant association to EM. The adjusted R 2 result for the data is 5.2 percent. The adjusted R 2 is acceptable since it is quite consistent with the previous study such as revealed by Aini et al. (2006) which range between 1 to 6 percent while Becker et al. (1998) show only 1 percent of adjusted R 2. The situation of small R 2 also occurs to Chang and Sun (2009) which range between 0.7 to 7.4 percent. 4.4 Robustness To ensure robustness of the results, this study also estimates the absolute value of discretionary accruals using the Kasznik (1999) model. Compare to Kothari et al. (2006), this model added the change in operating cash flows and subtract the changes in receivables from the changes in revenue as follows to derive at the total accruals: TA it = β o (1/T.A it-1 ) + β 1 ( REV it + REC it ) + β 2 PPE it + β 3 CFO it + ε it where: (3.4) TA it = Absolute Discretionary Accruals of firm i in year t based on Kasznik model as (3.4); 21

CFO it = changes in cash flows from operation in year t for firm i obtained from statement of cash flow; Table 4: Regression results of discretionary accruals estimated using Kasznik (1999) model. NKASZNIKDACC = β o (1/T.A it-1 ) + β 1 ( REV it + REC it ) + β 2 PPE it + β 3 CHGCFO it + ε it Coefficients t-stat. VIF Intercept β 0-0.190** -2.656 FCF β 1 0.190* 4.684 1.424 DIVRATIO β 2-0.009-0.232 1.215 LEVERAGE β 3-0.665* -3.016 2.901 CONSTRUCTION β 4 0.080 0.625 1.168 CONSPROD β 5 0.094 1.212 1.544 INSPROD β 6 0.190** 2.656 1.635 TRADINGSER β 7 0.121 1.555 1.416 Adjusted R-Squared 0.058 N=819 Notes: * and ** represents statistical significance at 0.01 and 0.05 levels, respectively (two-tailed test). NKASZNIK is the normal score of discretionary accrual estimated by using Kasznik (1999) model through Van der Waerden s formula. FCF is the normal score of free cash flow using Van der Waerden s formula. DIVRATIO the normal score of dividend ratio using Van der Waerden s formula. LEVERAGE is the total debt divide by total assets. CONSTRUCTION takes the value of 1 if the firm is construction firm; 0 otherwise. CONSPROD takes the value of 1 if the firm is consumer product; 0 otherwise. INSPROD takes the value of 1 if the firm is industrial product; 0 otherwise. TRADINGSER takes the value of 1 if the firm is trading and services; 0 otherwise. Table 4 presents the results of the ordinary least square regression used to test the relationship between earnings management (NKASZNIKDACC) and free cash flow (FCF), dividend (DIVRATIO) and leverage (LEVERAGE). The absolute value of discretionary accruals in this regression is estimated using the Kasznik (1999) model. 22

Similar to Kothari et al (2005) model, the results from discretionary accruals estimated using Kasznik (1999) model also report that free cash flow (FCF) is positively associated to EM and significant at a 1% level. This result therefore supports H1: Firms with higher level of free cash flow would have higher earnings management. However, dividend ratio shows a non-significant association and negative result to EM which implies firms with higher dividend ratio would have lower EM. Therefore, the result rejects the second hypothesis H2: Firms with higher level of dividend payment would have lower earnings management. Leverage also shows a similar result between Kotahri s et al. (2005) model and Kasznik (1999). Leverage is found significant at 1% level and negative to EM. This result implies that the higher the firms leverage, the lower will be the EM practice. This result therefore supports H3: Firms with higher level of leverage would have lower earnings management. Among the control variables, industrial product industry shows a positive and significant association to EM. However, Kasznik (1999) model shows a slightly better adjusted R 2 result for the data which is 5.8 percent. 5.0 CONCLUSION The aims of study is to examine the relationship between EM (discretionary accruals) and free cash flow, dividend ratio and leverage by controlling the construction, consumer product, industrial product and trading & services industries. Based on the study period of 2010-2012, the result of this study revealed that free cash flow is positive and significant to EM while leverage 23

is negatively significant to EM based on Kothari et al. (2005) and Kasznik (1999) discretionary accruals model. However, in Kothari et al. (2005) model, trading & services was significant to EM while in Kasznik (1999), industrial product shows a significant result to EM. However this study found insignificant association between dividend ratio and EM, suggesting that dividend ratio does not significantly affect EM in Malaysian companies. This also shows that Malaysian firms with higher level of dividend payment would have lower earnings management. Dividends could act as a credible signal of earnings quality, with companies unengaged with EM being more likely to pay dividends, to have higher dividend yields amongst dividend payers and to increase dividends per share (Farinha and Moreira, 2007). However this study found an insignificant association between dividend ratio and EM, suggesting that dividend ratio does not significantly affect EM based on Malaysian firms. 5.1 Limitations of the study There are some limitations listed in this section which could have been the underlying factors for the rejection of most hypotheses. First, this study used a sample of three years period with 819 firm-years observations, which is considered short as compared to the other studies. Second, the current study only covers 4 sectors which are constructions, industrial products, consumer products and trading and services. Third, this study only used two different accruals models which is performance-matched discretionary-accrual approach by Kothari et al. (2005) and Kasznik (1999). By using more 24

accruals models, the possibility for the findings to be significant could be higher. This is because Jaggi and Lee (2002) argued that the use of different accruals models will reduce errors in calculating discretionary accruals. 5.2 Future Study This study suggests that that there is possibility for future studies to find a significant relationship by extending the sample size. It is strongly encouraged that future studies use a larger sample because it might be able to find more significant results. The time frame could also be extended to 5 or 10 years, compared to just 2 years as done in this study. The latest studies or new models to detect the discretionary accruals (EM) could also be used in comparing which model gives the best possible answer in detecting EM. This study can also be extended to include detailed information for other relevant factors such as non-audit fees, audit tenure, qualified opinion and auditor changes. Future research also needs to be done to examine to which extent EM is harmful or useful to shareholders. 25

REFERENCES Aini, A., Takiah, M.I., Pourjalali, H. and Teruya, J. (2006). Earnings Management in Malaysia: A Study on Effects of Accounting Choices. Malaysian Accounting Review, 5(1), 185-207. Barth, M.E., Cram, D.P. and Nelson, K.K. (2001). Accruals and Prediction of Future Cash Flows. The Accounting Review, 76(1), 27-58. Beatty, A. and Weber, J. (2003). The Effect of Debt Contracting on Voluntary Accounting Method Changes. The Accounting Review, 78(1), 119-142. Becker, C.L., Defond, M.L., Jiambalvo, J. and Subramanyam, K.R. (1998). The Effect of Audit Quality on Earnings Management. Contemporary Accounting Research, 15(1), 1-24. Bukit, R. B. and Mohd Iskandar, T. (2009). Surplus Free Cash Flow, Earnings Management and Audit Committee. International Journal of Economics and Management, 3(1), 204-223. Chang, J.C. and Sun, H.L. (2009). Crossed-Listed Foreign Firms' Earnings Informativeness, EM and Disclosures of Corporate Governance Information under SOX. The International Journal of Accounting, 44, 1 32. Chung, R., Firth, M. and Kim, B.K. (2005). Earnings Management, Surplus Free Cash Flow, and External Monitoring. Journal of Business Research, 58, 766-776. Cooke, T. E., (1998). Regression analysis in accounting disclosure studies. Accounting & Business Research, 28(3), 209. DeAngelo, H., DeAngelo, L. and Wruck, K.H. (2002). Asset Liquidity, Debt Covenants, and Managerial Discretion in Financial Distress: the Collapse of L.A. Gear. Journal of Financial Economics, 64 (1), 3-34. DeFond, M. and Jiambalvo, J. (1994). Debt Covenant Effects and the Manipulation of Accruals. Journal of Accounting and Economics, 17, 145-176. 26

Dichev, I.D. and Skinner, D.J. (2002). Large-Sample Evidence on the Debt Covenant Hypothesis. Journal of Accounting Research, 40(4), 1091-1123. Duncan, J. R. (2001). Twenty Pressures to Manage Earnings. The CPA Journal, 71, 32-37. Edelstein, R., Liu, P. and Tsang, D. (2009). Real Earnings Management and Dividend Payout Signals: A Study for U.S. Real Estate Investment Trusts. Retrieved May 15, 2010 from http://www.ires.nus.edu.sg/researchpapers/visitors/remreit%20(dr%20desmond%20 Tsang).pdf Fama, E. and French, K. (1997). Industry Costs of Equity. Journal of Financial Economics, 43, 153-193. Farinha, J. and Moreira, J.A. (2007). Dividends and Earnings Quality: The Missing Link. CETE / Faculty of Economics / University of Porto. Retrieved June 10, 2010 from http://www.clsbe.lisboa.ucp.pt/resources/documents/professores/seminarios/paper_ 21Jan2008.pdf. Fey, S.L., Mohd Lukman, A.M., Aidil Rizal, S. and Mohd Said, O. (2008). Dividend Policy: Evidence from Public Listed Companies in Malaysia. International Review of Business Research Papers, 4(4), 208-222. Gu, Z., Lee, C. and Rosset J. (2005). What Determines the Variability of Accounting Accruals? Review of Quantitative Finance & Accounting, 24(3), 313-334. Hart, O. and Moore, J. (1995). Debt and Seniority: An Analysis of the Role of Hard Claims in Constraining Management. American Economic Review. 85(3), 567 85. Healy, P.M. and Wahlen, J.M. (1999), A Review of the Earnings Management Literature and Its Implications for Standard Setting. Accounting Horizons, 13(4), 365-383. Jelinek, K. (2007). The Effect of Leverage Increases on Earnings Management, Journal of Business & Economic Studies, 13(2), 24-26. 27

Jensen, M. (1986). Agency Cost and Free Cashflow, Corporate Finance, and Takeovers. American Economic Review, 76, 323-326. Kasznik, R., (1999). On the Association between Voluntary Disclosure and Earnings Management. Journal of Accounting Research, 37(1), 57. Kim, H.J. and Yoon, S.S (2008) The Impact of Corporate Governance on EMin Korea. Malaysian Accounting Review, 7(1). Kothari, S.P., Leone, A.J and Wasley, C.E. (2005). Performance Matched Discretionary Accrual Measures. Journal of Accounting and Economics, 39, 163-197. Lehn, K. and Poulsen, A. (1989). Free Cash Flow and Stakeholder Gains in Going Private Transactons, The Journal of Finance, 44(3.), 771-787. Leuz, C., Nanda, D. and Wysocki, P.D. (2003). Earnings Management and Investor Protection: An International Comparison. Journal of Financial Economics, 69, 505 527. Linacre, J. (2008). The Expected Value of a Point-Biserial (or Similar) Correlation. Rasch Measurement Transactions 22(1), 1154-1163. Mashayekhi, B., Bagheri, B. and Tahriri, A. (2010). Free cash flows, financial leverage and earnings management: the case of Iran, International Journal of Accounting and Information Management, 18 (1), 5-13. Mohammed, A.A. (2009). Corporate Governance and Earnings Management: An Empirical Study of the Saudi Market. The Journal of American Academy of Business, Cambridge, 15(1). Mohd. Ali, S., Mohd. Salleh, N. and Hassan, M.S. (2008). Ownership Structure and Earnings Management in Malaysian Listed Companies: The Size Effect. Asian Journal of Business and Accounting, 1(2), 89-116. Abdul Rahman, R. and Wan Abdullah, F.H. (2006). Board, Audit Committee, Culture and Earnings Management: Malaysian Evidence. Managerial Auditing Journal, 21(7), 783-804. 28

Schooley, D.K. and Barney, L.D. (1994). Using Dividend Policy and Managerial Ownership to Reduce Agency Costs. Journal of Financial Research, 17, 363-373. Skinner, D. J. (1993). The Investment Opportunity Set and Accounting Procedure Choice: Preliminary Evidence. Journal of Accounting Economics, 18, 407-445. Spohr, J. (2005). Essays on Earnings Management. (Doctoral dissertation, Swedish School of Economics and Business Administration, 2005). Retrieved April 16, 2010 from http://dhanken.shh.fi/dspace/bitstream/10227/125/2/153-951-555-898-0.pdf Stulz, R. (1990). Managerial Discretion and Optimal Financing Policies. Journal of Financial Economics, 26, 3-27. Sun, L. and Rath, S. (2008). Fundamental Determinants, Opportunistic Behaviour and Signaling Mechanism: An Integration of Earnings Management Perspective. International Review of Business Research Papers, 4(4), 406-420. Sweeney, A.P. (1994). Debt Covenant Violations and Managers Accounting Responses. Journal of Accounting and Economics, 17, 281-308. Syed Zulfiqar, A.S., Hui, Y. and Nousheen, Z., (2010). Earnings Management and Dividend Policy: An Empirical Comparison between Pakistani Listed Companies and Chinese Listed Companies. International Research Journal of Finance and Economics, 35, 51-60. Watts, R.L. and Zimmerman, J.L. (1990). Positive Accounting Theory: A Ten Year Perspective. The Accounting Review, 65, 131 156. Zunaidah, S. and Fauzias, M.N. (2008). Dividends, Ownership Structure and Board Governance on Firm Value: Empirical Evidence From Malaysian Listed Firms. Malaysian Accounting Review, 7(2), 55-94. 29