Leverage, corporate governance and real earnings management: Evidence from Korean market

Tulcanaza-Prieto, Ana; Lee, Young Hwan; Koo, Jeong-Ho Article Leverage, corporate governance and real earnings management: Evidence from Korean market Global Business & Finance Review (GBFR) Provided in Cooperation with: People & Global Business Association (P&GBA), Seoul Suggested Citation: Tulcanaza-Prieto, Ana; Lee, Young Hwan; Koo, Jeong-Ho (2020) : Leverage, corporate governance and real earnings management: Evidence from Korean market, Global Business & Finance Review (GBFR), ISSN 2384-1648, People & Global Business Association (P&GBA), Seoul, Vol. 25, Iss. 4, pp. 51-72, https://doi.org/10.17549/gbfr.2020.25.4.51 This Version is available at: https://hdl.handle.net/10419/253313 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/ Received: Nov. 24, 2020; Revised: Dec. 22, 2020; Accepted: Dec. 21, 2020 † Younghwan Lee E-mail: [email protected] ⓒ Copyright: The Author(s). This is an Open Access journal distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution , and reproduction in any medium, p rovided t he original work is properly cited. GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 pISSN 1088-6931 / eISSN 2384-1648∣Https://doi.org/10.17549/gbfr.2020.25.4.51 ⓒ 2020 People and Global Business Association GLOBAL BUSINESS & FINANCE REVIEW www.gbfrjournal.org1) Leverage, Corporate Governance and Real Earnings Management: Evidence from Korean Market A na Belen Tulcanaza-Prietoa, Younghwan Leeb†, Jeong-Ho Koob aSchool of Public Economics and Strategic Sectors, Instituto de Altos Estudios Nacionales, and Business School, Universidad de Las Americas, Quito, Ecuador bDepartment of Business Administration, Kumoh National Institute of Technology, Gumi, Republic of Korea A B S T R A C T Purpose: This study examines how corporate governance (CG) and leverage simultaneously influence real earnings management (REM). Methodology: We employed CG score (CGS), total, short-term, and long-term debt ratios as independent variables, and REM metrics as dependent variables. We include ordinary least-squares (OLS) panel data regressions, residual test, and interaction analysis in our study. Findings: While a significant positive relationship existed between leverage and REM, CG had a negative effect on real manipulations. Our results from the interaction analysis and residual test show that CG is a key player in explaining the relationship between leverage and REM. We also find that firms with a low-level of CG were more likely to conduct REM activities than those with a high-level CG. Research limitations/implications: These results imply that the reduction in opportunistic behavior of managers in the presence of strong CG could decrease the leverage of firms and REM activities and improve the quality of their earnings. Moreover, shareholder rights and audit organizations were the prominent CG characteristics influencing REM activities, as they enforce additional monitoring of financial reporting quality and increase audit standards. CG strategies mitigate the corporate corruption scandals through the adoption of high-quality accounting and financial norms in reporting and management. Thus, executives decrease their incentives to conduct REM activities and leverage cannot be used freely as a mechanism to manipulate earnings, given firms’ leverage position is audited and reviewed by the financial committee in firms with strong CG. For future research, the authors suggest including the degree of leverage as a disaggregation sample and adding the degree of cost stickiness. Originality/value: Several studies have investigated either the relationship between (1) REM and leverage, or (2) REM and CG, including leverage as a control variable. Our study, extends this bilateral relationship to simultaneous relationship between leverage, CG, and REM. Moreover, we investigate the predominant CG characteristics that influence the association among leverage and REM. We conduct interaction analysis and residual effect to investigate if CG should be considered as a determinant variable in the recurrence of REM activities of managers. We include firms with high- and lowlevels of CG to show which firms are leveraged. Keywords: Real earnings management, Corporate governance, Leverage GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 52 Ⅰ. Introduction Schipper (Schipper, 1989) defines earnings management (EM) as “a purposeful intervention in the external financial reporting process with the intent of obtaining some private gain”. EM influences on stock market perceptions, increases managers’ rewards, reduce the probability of violating lending agreements, and avoid any regulatory intervention (Healy & Wahlen, 1999). Furthermore, managers engage in various EM practices to avoid earnings volatility and to meet or beat earnings benchmarks because investors prefer firms with consistent profitability and stability, which would affect their bonuses, based on the financial performance of the firm. Hence, they prefer to manage earnings through real earnings management (REM) because of the lesser pressure of debt covenants (Kim & Sohn, 2013; Vakilifard & Mortazavi, 2016), lesser scrutiny by auditors and regulators, and therefore, a lesser probability of being detected (Cohen et al., 2008; Cohen & Zarowin, 2010; Graham et al., 2005). Studies have provided evidence of both positive and negative relationships between REM and leverage. The positive association between debt and REM is grounded in the high default risk of high-leverage firms, which might lead managers to adopt measures such as evading the reporting of annual loses. Roychowdhury (Roychowdhury, 2006) documents that managers from high-leverage firms manipulate real earnings, activities by, (1) providing price discounts to temporarily increase the sales volume, (2) overproducing inventory to decrease the cost of goods sold (COGS), or (3) cutting discretionary expenses to improve the reported margins. Tulcanaza-Prieto, Lee, and Koo (Tulcanaza-Prieto, Lee, et al., 2020) also reports the positive relationship between leverage and REM using quarterly financial data. Therefore, firms might increase their REM activities (1) to show stable results and fewer volatile earnings to their lenders and future investors (Burgstahler & Dichev, 1997; Wijesinghe & Kavinda, 2017), (2) to reduce their debt covenant violations, contraventions, and penalties (Y. Chen et al., 2015; Lemma et al., 2013), and (3) to secure debt refinancing (Huang & Sun, 2017; Roychowdhury, 2006). In contrast, the negative relationship between leverage and REM focuses on reducing the opportunistic behavior of managers, which modifies their astute conduct into discipline, as lenders and institutional investors increase their scrutiny and control on firms (Jelinek, 2007; López-Iturriaga & Hoffman, 2005). Moreover, there is a reduction in the free cash flow when managers engage in REM practices, which also decreases the leverage of the firm (Wasimullah & Abbas, 2010), suggesting that lenders and investors of high-leverage firms are more susceptible to increase their accounting and financial scrutiny and control by adopting corporate governance (CG) policies. The agency costs also develops an important argument on the literature of REM. First, the agency costs of debt motivate the positive relationship between leverage and REM activities, if REM would be associated with stockholders’ selfish strategies against bond holders such as taking large risks, engaging underinvestment, or milking properties. On the other hand, an increase of leverage may result in heavy monitoring activities on firm’s financial policies by the institutional bond holders, which insists on the negative relationship between leverage and REM. Second, the agency cost of equity raises the two types of disputes in firms. The first disagreement is the conflict of interest between the majority and minority shareholders, grounded in the expropriation risk (Goh et al., 2013), while the second problem arises from different benefits between managers and shareholders inspired by wealth of individuals and of the firm, respectively. For both types of conflicts, REM could be considered as a manager’s expedient of personal extortion from shareholders. Therefore, the agency cost of equity implies the negative relationship between leverage and REM. Appropriate governance structures are created to protect the interests of all stakeholders by increasing independence of the board and committees, raising the access of transparent information and disclosure, and owning high-level audit committees. A strong CG might improve transparency and reduce both agency conflicts of debt and equity, while a weak CG might increase Ana Belen Tulcanaza-Prieto, Younghwan Lee, Jeong-Ho Koo 53 the asymmetry of information between parties and increase the agency costs. Agency problems and asymmetric information often increase the probability of managers engaging in REM activities. Therefore, CG could loosen the relationship between leverage and REM activities. Several studies have investigated either the relationship between (1) REM and leverage (Kaushik & Kumar, 2018; Tulcanaza-Prieto, Lee, et al., 2020; Vakilifard & Mortazavi, 2016; Zamri et al., 2013), or (2) REM and CG, including leverage as a control variable (Kang & Kim, 2012; Kurnia & Pradipta, 2016; Lee et al., 2015). This study, however, extends this bilateral relationship to simultaneous relationship among leverage, CG, and REM. We also analyze which characteristics of CG influence the association between REM and leverage. For this purpose, we incorporate five CG characteristics in our statistical analysis, namely, shareholder rights, board structure, disclosure, audit organization, and management error. We conduct interaction analysis and residual effect to investigate if CG should be considered as a determinant variable in the recurrence of REM activities of managers. Moreover, to reinforce our previous findings, we divide our sample into firms with high- and low-levels of CG based on the median value of their CG score. We compare the standardized beta coefficients of four regressions between both the samples to verify that firms with high-level of CG show lower leverage standardized beta coefficients than those with low-level of CG. Analyzing Korean non-financial firms for the period of 2003-2011, we find a positive relationship between leverage and REM activities and a negative association between CG and REM. Moreover, the interaction term between leverage and CG has a positive relationship with real manipulations. Using residual tests, we show that CG is a key player in explaining the relationship between leverage and REM. These findings imply that CG could act as a corporate finance tool reducing REM activities of the managers. Furthermore, our findings suggest that the reduction in the opportunistic behavior of managers in the presence of strong CG could decrease leverage and REM activities of the firm and improve its quality of earnings. We also find that managers are less likely to manage earnings in high-leverage firms with strong CG, whereas they are more likely to conduct REM activities in high-leverage firms with a low-level of CG. We show that shareholder rights and audit organizations are the prominent CG characteristics that influence REM activities in the Korean market. Finally, we conduct a two-stage least square (2SLS) regression analysis, which provides a robustness check for our results and controls the endogeneity problem. The rest of the paper is composed as follows. Section 2 presents a literature review and describes the development of the hypothesis. Section 3 illustrates the empirical design. Section 4 defines the data collection procedure and presents the empirical findings. Section 5 discusses the results, highlights the conclusions, and offers recommendations for future research. II. Development of Hypothesis Managers show personal motivations to manipulate earnings using their judgment to alter the financial reports by structuring transactions in order to increase the firm performance or influence the contractual outcomes (Healy & Wahlen, 1999; Schipper, 1989). In the accounting literature, there are two methodologies to conduct EM in a firm, namely accrual-based EM and REM. Zang (Zang, 2012) suggests that there is a trade-off between accrual-based EM and REM because both EM categories are substitutes and are influenced by the relative costliness and the timing of EM activities. Previous studies suggest the preference of managers to conduct real manipulations instead of accrual-based modifications due to its low probability of being detected (Graham et al., 2005). Firms might increase their real manipulation activities to show good results to their lenders to reduce their debt covenant violations and secure refinancing of debt (Klein, 2002; Lemma et al., 2013; Othman & Zeghal, GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 54 2006; Sweeney, 1994). Thus, there is a positive relationship between REM and leverage, suggesting that the market participants underreact to REM (Anagnostopoulou & Tsekrekos, 2017; Kuo et al., 2014). In contrast, there is a negative relationship between REM and leverage, motivated by the reduction in the opportunistic behavior of managers to meet (or beat) earnings targets, owing to the scrutiny and control imposed on firms by lenders and institutional investors (Jelinek, 2007). If the free cash flow decreases by their REM practices, managers prioritize their debt payments and repayments, and increase their prudency to invest in non-value maximizing projects, which reduce leverage of the firms (Wasimullah & Abbas, 2010). The presence of a high-level of CG introduces a higher transparency and public information disclosure in a firm, which decrease the agency costs, asymmetric information, and opportunistic behavior of managers. Transparency is aligned with openness and willingness to disclosure financial performance, suggesting that firms provide clear information to shareholders and stakeholders. Transparency guarantees that stakeholders show confidence in the decision-making and management processes of a firm, which increases the quality of CG. Transparency is a characteristic of the high-level of CG, which increases reliability of firms. Black, Jang, and Kim (Black et al., 2006) mentioned that most Korean companies have a controlling shareholder or family, which is considered as a tool of CG, because it increases the control of managers and their actions, promoting more efficient operations, increasing overall firm value, or transferring part of this value from majority to minority shareholders (Yoon et al., 2006). Prior studies show that firms with strong CG policies are more likely to avoid REM activities (Byard et al., 2006; Jianga et al., 2008), suggesting a negative relationship between both variables, because the supervisory role of the majority owner decreases the opportunistic behavior of managers, and their motivations to engage in REM. Piosik and Genge (Piosik & Genge, 2019) shows that the optimal level of ownership concentration minimizes the magnitude of REM and increases financial transparency, which is one of the pillars of sustainable firms. Analyzing American firms, He et al. (2009) shows that a strong CG decreases REM practices, increases the accounting-monitoring process, improves the quality of financial reporting, and raises the shareholder confidence (He et al., 2009). Similarly, García-Osma and Noguer (García-Osma & Noguer, 2007), Alves (Alves, 2012), and Hashim and Devi (Hashim & Devi, 2012) found that managers of well-governed firms in Spain, Portugal, and Malaysia are less likely to engage in REM activities because of the high-level institutional investors and independent directors, managerial ownership, and ownership concentration, respectively. These findings suggest that strong CG measures might decrease REM practices, accounting violations, and frauds (Xie et al., 2003). Therefore, effective CG mitigates REM activities. Lee, Kang, and Cho (Lee et al., 2015) showed that the improved CG through adopting IFRS impulses the manager to avoid REM. Waweru and Riro (Waweru & Riro, 2013) also showed that the board composition and ownership structure significantly influence the earnings management respectively in a negative and positive manner. They mention that the higher composition of independent directors reduces real manipulations because boards of directors play an important role in the financial reporting process and increase the reporting quality. However, the increase in ownership concentration motivates the engagement in REM activities, as there is a reduction in external pressure for higher reporting quality (Saona et al., 2020). Byun, Kwak, and Hwang (Byun et al., 2008) showed that shareholders’ rights protection is the most representative practice for the reduction of the cost of equity because it mitigates the agency problems and the information asymmetry, and improves the financial reporting quality. The opportunistic behaviors of managers decrease when the firms adopt CG policies with transparent information and disclosure. According to the agency theory, the improved transparency reduces conflict of interest and the asymmetry of information, as well. There is evidence in the finance and accounting literature that contracts with identification of roles, Ana Belen Tulcanaza-Prieto, Younghwan Lee, Jeong-Ho Koo 55 rights, and obligations of managers provide more transparent information, which influence on the negative relationship between leverage and REM activities (Cohen et al., 2008; López-Iturriaga & Hoffman, 2005). Furthermore, the active supervisory role of regulators and the presence of a strong CG structure in firms might decrease leverage of the firms and reduce the opportunities for managers to engage in REM activities given the high possibility of being discovered. Therefore, CG practices might loosen the relationship between leverage and REM. A couple of the previous studies analyze the relation among CG, leverage, and REM together. Swai (Swai, 2016) argues for a significant negative relationship between REM and CG. He asserts that audit quality decreases the possibility of managers engaging in REM activities, and that the adoption of CG practices reduces the financial leverage of firms. However, the study has not investigated the interaction effect of CG on the relationship between leverage and REM. Fitri et al. (Fitri et al., 2018) demonstrates a significant negative relationship between REM activities and institutional and managerial ownership, and found that influence of leverage on REM was insignificant. Therefore, institutional and managerial ownership act as a monitor of managerial performance, which limits and reduces the opportunistic behavior of managers. Amertha et al. (Amertha et al., 2014) studied the relationship among the management of earnings, leverage, and CG. They showed that CG has a significant negative effect on REM, whereas influence of leverage was insignificant. However, they did not investigate the simultaneous relationships among REM, leverage, and CG. In this study, we simultaneously investigate the relationship among leverage, CG, and REM using descriptive statistical analysis, interaction effect, and residual test. All these tests are performed to evaluate CG as a predominant variable affecting managerial decision to engage in REM activities. We also examine which characteristics of CG play an important role in explaining the relationship between leverage and REM. Finally, the samples are classified into firms with high- and low-levels of CG, based on median value of their CG score to confirm our previous finding of the relationship between REM and CG. Therefore, our hypothesis is: Hypothesis: Effective CG mechanisms decrease the magnitude of relationship between leverage and REM. III. Empirical Design A. Detecting REM We employ Roychowdhury’s (Roychowdhury, 2006) model to measure manipulation in real earnings activities as it is the most frequent and convenient method used in several REM studies (Anagnostopoulou & Tsekrekos, 2017; Cohen et al., 2008; Cohen & Zarowin, 2010; Roychowdhury, 2006; Zamri et al., 2013). We examine patterns in the individual and aggregate values of cash flow from operations (CFO), selling, general, and administrative (SG&A) expenses, and production costs (sum of COGS and change in inventory) for firms close to the zero earnings benchmark, to detect real activities manipulation to avoid losses. All metrics are calculated using the difference between the actual value and the normal value by estimating coefficients from the corresponding industry-year and the firm-year sales and lagged assets. All parameters are calculated using estimation samples (all industries) and then we introduce these parameters in our models using test sample. The abnormal aggregate REM (ABN_REM) is measured by the aggregation of the abnormal CFO (ABN_CFO), abnormal SG&A expenses (ABN_SG&A), and abnormal production costs (ABN_PROD). For understanding purposes, we report the inverted sign for the variables ABN_CFO and ABN_SG&A, as both measurements show negative residuals when firms engage in REM activities. We estimated Equation (1) using annual information; therefore, high residuals correspond to high levels of real earnings manipulation, resulting in positive ABN_REM when a firm manages earnings through REM initiatives. GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 56 ___& _  (1) where _  is the abnormal aggregate REM, _  is the abnormal CFO, _&  is the abnormal SG&A expenses, and _  is the abnormal production costs. The subscripts i and t denote the firm and fiscal year, respectively. The ABN_CFO was estimated as follows: _          ∆,(2) _  is measured by _        , where _ is the CFO,   denotes the total assets of a firm, and ∆  is the change in sales of the firm measured as ∆          , where   and   are the total sales of firm i in the years t and t-1, respectively.   is the intercept term and   is the error term. We estimated the ABN_SG&A expenses as in Equation (3): _&                       ,(3) where _& is measured by  &   & , and &  is the SG&A expenses of firm i in year t. The ABN_PROD was estimated as: _                           ∆         ∆      (4) where _  is measured by _   .        is the production cost measured by     ∆  , where   is the COGS of firm i in year t and ∆  is the change in inventory measured by ∆          , where   and   are the total inventories for firm i in the years t and t-1, respectively. ∆  is the change in sales measured by ∆          . B. Corporate Governance Metrics Standard and Poor’s (S&P) calculates the CG score (CGS) using the CG principles established by the OECD in 1999. The CGS is calculated by aggregating the scores of five CG characteristics. A higher CGS suggests better CG implementation and higher transparency. The scoring process uses both public and private information. The maximum value of the CGS is 300 and its characteristics are presented in Figure 1. The Korean Commercial Code is the primary source of law relating to CG, which applies to both listed and unlisted firms. All listed companies need to show public disclosure, establish an audit committee, elect outside directors, and issue securities. The applicability and practice of CG scoring is challenging because most of the components are assessed as a qualitative exercise, which is the main difference with the financial analysis where quantitative measures are the components of benchmarks. The CGS in Korea assesses the CG practices in a firm and extends to the firms’ financial stakeholders, emphasizing on shareholders’ interests. Furthermore, CGS allow the comparison of individual companies within a national context, as well as comparisons of companies in different jurisdictions (Standard & Poor’s Governance Services, 2004). C. Research Model We used ordinary least-squares (OLS) panel data regression models with fixed effects to investigate the relationship between leverage, CG, and REM in the Korean non-financial firms because the results of the Hausman test revealed that error terms are not correlated with the constant, which captures the individual characteristics (Nwakuya & Ijomah, 2017). We adopted ABN_REM, ABN_CFO, ABN_SG&A, and ABN_PROD as the dependent variables, with Ana Belen Tulcanaza-Prieto, Younghwan Lee, Jeong-Ho Koo 57 CGS and leverage as the independent variables. The CGS is calculated as the sum of five measurements, which are (1) shareholder rights (CG1), (2) board structure (CG2), (3) disclosure (CG3), (4) audit organization (CG4), and (5) management error (CG5), while leverage from debt ratios (Lev) comprises total (TLev), short-term (StLev), and long-term (LtLev) debt ratios. Our models included the most frequent control variables from previous studies, which are asset tangibility, profitability, size, and firm liquidity (Tulcanaza-Prieto & Lee, 2019). Profitability includes return on assets (ROA) in the lagged form following methodologies in previous studies (Vakilifard & Mortazavi, 2016; Wijesinghe & Kavinda, 2017; Zamri et al., 2013). Furthermore, we incorporated net interest payment and foreign investor ownership rate as new control variables because an increase in leverage might raise interest expense, which lowers net income, while foreign investor ownership rate might contribute transparent information (Jelinek, 2007; Jensen, 1986; Vakilifard & Mortazavi, 2016). In Equation (5), the coefficient   measures the relationship between leverage and REM. If the   is positive, leverage will positively influence the real earning manipulation activities. Therefore, we expect a positive value for   . _                                ∑       ∑          (5) where _  is the abnormal aggregate REM and is composed of the abnormal CFO _  , abnormal SG&A expenses _&  , and abnormal production cost _  of firm i for year t.   is the debt ratio of firm i in year t, composed of a total debt ratio   , short-term debt ratio   , and long-term debt ratio   , estimated as              © , and            , respectively.             is the assets tangibility,            is the proxy for firm profitability,       is the size of the firm represented by natural logarithm of total assets,          is the firm liquidity,             is the net interest payment, and   is the foreign investor ownership rate. The dummy terms   and   represent industry of a firm (there are eleven Figure 1. Characteristics of corporate governance score (CGS). GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 58 non-financial industries listed on KOSPI) and the year of information, respectively, and   is the error term. The subscripts i and t denote the firm and fiscal year, respectively. In Equation (6), the coefficient   measures the relationship between CG and REM. If   is negative, CG will negatively influence the real earning manipulation activities. We expect a positive value for   . _    ∑  ∑  (6) where   is the CG score of firm i in year t. It is composed of shareholder rights    , board structure    , disclosure    , audit organization    , and management error    .           for firm i in year t. In Equation (7), the coefficients   and   measure the relationship between (1) leverage and REM, and (2) CG and REM, respectively. We expect a positive value for   and negative value for   based on previous findings from Equations (5) and (6). _                                    ∑       ∑          (7) In Equation (8), the coefficient   measures the relationship between leverage and REM while   measures the relationship between CG and REM. We introduce an interaction term  in our model. The coefficient   is the result of our hypothesis, which shows the interaction effect of CG on the relationship between leverage and REM. If   is positive, CG provides more incremental information than the information provided by the individual relationship between leverage and REM, and CG and REM. We expect positive values for the coefficients   and   , and a negative value for the coefficient   . _       ∑   ∑  (8) where    is the interaction term between debt ratio and CGS of firm i in year t. To reinforce our findings, we included the residual test, which shows the effect of CG on the relationship between leverage and REM. The residual value in a regression model shows the lack of fit resulting from the deviation in the linear relationship between independent variables (Gujarati, 1988). The steps for the residual examination are detailed as follows: (1) Regress CGS towards Lev (Equation (9)), (2) calculate the absolute value of residuals, and (3) regress the absolute value of residuals with ABN_REM as the independent variable (Equation (10)). In Equation (10), the coefficient   reinforces the result of our hypothesis because it can explain the abnormal volatility of CGS given firms’ leverage. If   is negative, then CG is a key variable on the relationship between leverage and REM.           ,(9)       _  ,(10) where    is the absolute value of residuals of firm i in year t. Residuals from Equation (9) collects the abnormal CGS that leverage does not explain. In Equation (10), we use the absolute value of estimated residual. Therefore, the absolute value of estimated residual contains the abnormal volatility of CGS that total leverage cannot explain. IV. Empirical Results The initial sample consists of non-financial firms listed on the Korean Composite Stock Price Index (KOSPI). The financial sector was excluded, as those firms are considered financially different from the industrial companies. Therefore, a high leverage for Ana Belen Tulcanaza-Prieto, Younghwan Lee, Jeong-Ho Koo 65 REM, in firms listed in East Africa and Indonesia, respectively. They suggested that the signaling theory and the increase in reporting quality would reduce the agency costs, limiting the managerial opportunistic Variables ABN_REM ABN_CFO (-1) ABN_SG&A (-1) ABN_PROD TLev -0.138 -0.621 -0.372 -0.215 -0.367 (-1.563) (-0.749) (-1.592) (-1.221) (-0.729) CGS -0.794*** -0.176*** -0.290*** -0.271*** (-4.304) (-4.163) (-5.591) (-2.994) CG1 -0.404*** (-2.978) CG2 -0.116 (-1.514) CG3 -0.104 (-1.072) CG4 -0.139** (-2.568) CG5 -0.020 (-0.585) TLev * CGS 0.199*** 0.086*** 0.077*** 0.069*** 0.064*** (2.646) (2.751) (3.075) (3.312) (2.886) Tang 0.056 0.024 -0.077*** 0.002 0.132*** (1.427) (0.439) (-7.828) (0.207) (4.536) ROA -0.176** -0.397** -0.152*** -0.031 0.060 (-2.248) (-2.338) (-8.417) (-1.405) (1.051) Size 0.016** 0.037*** 0.004*** 0.011*** -0.001 (2.449) (3.760) (2.626) (5.312) (-0.179) Liq 0.018*** 0.026*** 0.003** 0.003*0.010** (3.107) (3.186) (2.165) (1.926) (2.425) NIP -0.177** 0.179 0.084*** 0.016 -0.276*** (-2.198) (1.423) (4.417) (0.680) (-4.648) FIOR -0.016** -0.024** -0.005** -0.013*** 0.000 (-1.977) (-2.225) (-2.337) (-5.390) (0.065) Intercept 0.102*** -0.025 0.250*** 0.343*** 0.470* (2.944) (-0.062) (2.976) (3.325) (1.704) Year-fixed effects Yes Yes Yes Yes Yes Industry-fixed effects Yes Yes Yes Yes Yes Adj. R20.300 0.341 0.130 0.255 0.211 F-Stat. 50.138*** 36.384*** 20.391*** 72.525*** 31.647*** DW 1.898 1.853 1.945 1.821 1.860 Note: The results indicate (1) a significant negative relationship between CGS, CG1, and CG4, and all REM metrics, (2) a significant positive relationship between all REM metrics and the interaction term (TLev * CGS), and (3) an insignificant negative relationship between total borrowings and REM metrics. Beta corresponds to unstandardized coefficients. The numbers inside the parentheses are t-statistics. ***, **, and * indicate statistical significance at 1%, 5%, and 10% levels, respectively. Table 6. Relationship between REM, leverage, CG, and interaction term GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 66 behavior to engage in REM. Jin et al. (Jin et al., 2018) found that CG introduces more control in the accounting processes of firms. Managers do not freely access the cash flow and the financial committees supervise and approve the debt position of firms. Therefore, motivation to engage in REM activities in the managerial level decreases. Finally, F-test results from Table 5 were higher than those results from Table 6. We concluded that CG metrics are key determinants that influence REM activities, while leverage’s importance decreases when the we included the interaction term, implying that the joint effect of CG metrics (CG1-CG5) over REM activities are significantly higher than leverage effect. 4. Residual effect of CG on relationship between REM and leverage Table 7 shows the results of the residual test conducted to explain the residual effect of CG on the relationship between total borrowings and REM. Therefore, if the result of the residual regression is negative, and the independent variable is considered as a key variable as it affects and breaks the relationship between the variables used in the first step of regression. We found significant negative beta coefficients for all REM metrics, suggesting that CG would weaken the relationship between leverage and REM. We showed that the increase in monitoring and controlling mechanisms protects all the stakeholder interests, which would align the managerial goals with those of corporates. The existence of CG in a firm increases the transparency and reliability of a company by information disclosure, which also reduces company risk and expropriation of minority shareholders. As a consequence, managers reduce their opportunistic behavior and do not engage frequently in REM activities. Therefore, our hypothesis was true for all REM metrics, implying a significant residual effect of CG on the relationship between leverage and REM. 5. Robustness test To confirm our previous results, we conducted a robustness test. We classified firms with high- and low-level of CG according to the median value of their CGS. The median value of the CGS logarithm was 1.996 for all samples. Therefore, firms with firm-year observations having a CGS higher than 1.996 were classified as firms with a high-level of CG, whereas those with firm-year observations having a CGS lower than 1.996 were classified as firms with a low-level of CG. Our sample was classified into 1,724 and 1,273 firm-year observations with high- and low-levels of CG, respectively. We decided to classify our sample into two groups to identify the degree of preference of managers to engage in Model Unstandardized Coefficients Standardized Coefficients t Beta Std. Error Beta 1(Constant) 0.020 0.002 11.012*** ABN_REM -0.015 0.005 -0.056 -3.078*** 2(Constant) 0.006 0.002 1.003 ABN_CFO (-1) -0.005 0.020 -0.004 -2.056** 3(Constant) 0.036 0.002 1.004 ABN_SG&A (-1) -0.079 0.015 -0.087 -5.253*** 4(Constant) 0.012 0.002 10.910*** ABN_PROD -0.007 0.007 -0.018 2.326** Note: The result indicates that CG is a key variable on the relationship between leverage and all REM metrics. Dependent variable: the absolute value of leverage residuals. Steps: (1) regress CGS towards TLev, (2) calculate the absolute value of residuals (    ), and (3)    regressed by all REM metrics. *** and ** indicate statistical significance at 1% and 5% levels, respectively. Table 7. Residual test Ana Belen Tulcanaza-Prieto, Younghwan Lee, Jeong-Ho Koo 67 REM activities depending on the CG level. Therefore, we expect lower REM incurrence in firms with high-level of CG than its value in firms with low-level of CG, because the adoption of CG increases the control and supervision in firms, and thus, managers might not have free access to the internal and external financing, which decreases the frequency of REM engagement. Table 8 shows that the mean of REM activities was higher in suspicious firms (whose net income scaled by total assets was ≥ 0.0 but < 0.005) than that in non-suspicious firms, which was approximately zero. Similarly, the mean of the leverage ratio in suspicious firms was higher than that in non-suspicious firms. Moreover, we show that firms with high-level of CG are less likely to engage in REM activities (31.03%), while firms with low-level of CG frequently manage their earnings by real activities (52.16%), suggesting that one-third of the firms with high-level of CG engage in REM activities compared to half of the firms with low-level of CG that engage in REM. These findings are consistent with our previous results and with our hypothesis that CG could be considered as an effective variable to demotivate managers to engage in REM activities. Therefore, the interaction term of CG and leverage might weaken the relationship between leverage and REM. Table 9 shows the standardized beta coefficients of the independent variables to explain the relationship between REM and both CG and leverage. We confirmed the significant negative relationship between CGS and ABN_REM in firms with high- and low-levels Disaggregation ABN_REM TLev High-level of CG Low-level of CG Mean No. % No. % Non-Suspicious firms -0.004 0.401 1,189 68.97% 609 47.84% Suspicious firms 0.053 0.526 535 31.03% 664 52.16% Total 0.017 0.454 1,724 100.00% 1,273 100.00% Note: “Suspicious” is defined as firms whose net income scaled by total assets is ≥ 0.0 but < 0.005. Table 8. REM activities by level of CG Model ABN_REM Independent Variables Firms with high-level of CG (N = 1,724) Firms with low-level of CG (N = 1,273) (1) TLev 0.179*** 0.269*** CGS -0.121*** -0.097*** (2) StLev 0.094*** 0.151*** CGS -0.120*** -0.103*** (3) LtLev 0.042** 0.083** CGS -0.118*** -0.086** (4) TLev 0.117*** 0.225*** CG1 -0.069*** -0.075* CG2 -0.028 -0.076 CG3 -0.064** -0.040 CG4 -0.094*** -0.052 CG5 0.001 -0.033 Note: The magnitudes of standardized CG and standardized leverage coefficients are higher and lower in firms with high-level of CG than their standardized value in firms with a low-level of CG, respectively. Beta corresponds to standardized coefficients. ***, **, and * indicate statistical significance at 1%, 5%, and 10% levels, respectively. Table 9. Standardized beta coefficients GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 68 of CG, whereas the effects of all Lev on ABN_REM were significantly positive. The effects of shareholder rights and audit organization were significantly negative on REM in firms with a high-level of CG, whereas shareholder rights in firms with a low-level of CG showed a significant negative effect on REM. The standardized beta coefficients of different regressions and samples could be compared because the beta coefficients are expressed in units of standard deviations (Gujarati, 1988). The higher and lower magnitudes of CGS and leverage regression coefficients of firms with both high-and low-levels of CG are presented. For model 1, if CGS increases by one standard deviation, ABN_REM, on average, will decrease by 0.121 and 0.097 units of standard deviation, in firms with high- and low-level of CG, respectively. Meanwhile, if total debt rises by one standard deviation, ABN_REM, on average, will increase by 0.179 and 0.269 units of standard deviation, in firms with high- and low-level of CG, respectively. Similar findings and interpretation of the remaining models. Therefore, the magnitude of standardized CGS and standardized leverage coefficients was higher and lower in firms with a high-level of CG than its standardized value in firms with a low-level of CG, respectively. 6. Two-Stage Least Square Regression Analysis In the literature of finance and accounting, there is concern about the endogenous relationship between REM and CG. To address this concern, we employed Equations (11) and (12) to control the endogeneity problem by a 2SLS regression analysis. The possible endogenous problem arises because our dependent variable is created using lagged assets while our independent variables are expressed using assets in term t (Equations 1-8). To increase the econometric specification, we recur to 2SLS regression procedure. Black, Jang, and Kim (Black et al., 2006) implemented simultaneous equations using 2SLS. According to their study, CG can be influenced by firm size, long term profitability, and industry factors. In the first stage, we ran ABN_REM and CGS, where we controlled for “ Size_Dummy ” and “ Size ”. In the second stage, we ran our model using results from the first stage. First stage:       _    _         (11) Second stage: _                                    ∑       ∑          (12) where _ is an indicator variable with a value of 1, if total assets are equal to or above 2 trillion Korean Won (firms that have assets of over 2 trillion Korean Won are required by law to have an internal audit committee, which provides an internal control mechanism for monitoring management’s activity), an 0 otherwise. (Table 10) shows the results of 2SLS regression that are consistent with our previous results, even though each coefficient of CGS,   , is larger than the coefficients presented in Table 5 and Table 6. These results suggest that there is a negative relationship between CG and REM activities, and it is aligned with the hypothesis that effective CG mechanisms might decrease the magnitude of the relationship between leverage and REM using the interaction variable between CG and leverage. V. Conclusions This study analyzes the simultaneous relationship between REM, leverage, and CG using a sample of 2,997 firm-year observations of non-financial firms listed on KOSPI for the years 2003-2011. By introducing ABN_REM, ABN_CFO, ABN_SG&A, and ABN_PROD as proxies for REM, we find that all of long-term and short-term leverage variables are significantly positive on REM activities and that Ana Belen Tulcanaza-Prieto, Younghwan Lee, Jeong-Ho Koo 69 most of CG variables are significantly negative on REM metrics. These results imply that the managerial discretional behavior motivates managers to engage in REM activities by providing price discounts, tolerant credit terms, and overproduction. We also find that CG acts as a powerful barrier to conduct REM initiatives as it increases the accounting transparency by active supervision, which reduces the probability to engage in “masked” everyday transactions. Moreover, introducing an interaction term between leverage and CG and performing a residual test, we find that CG plays a significant role to explain the positive relationship between leverage and REM. Unlike previous studies, leverage is no longer effective in preventing the firms from engaging in REM because it acts only in an interactive role with CG, suggesting that the CG influences the relationship between leverage and REM. These findings imply that CG could act as a corporate finance tool, providing credible and sustainable financial information to make decisions in firms, and reducing the possibility of managers engaging in REM activities. For robustness test, our results show that (a) 31% of firms with a high-level of CG engage in REM while 69% of them do not manage earnings while (b) 52% of firms with low-level of CG engage in REM while the remaining 48% do not manipulate real earnings. We also find that most of firms with high-level CG are characterized with low-level leverage while most of firm with low-level CG are characterized with high-level leverage. These results imply that the reduction in opportunistic behavior of managers in the presence of strong CG could decrease the leverage of firms and REM activities and improve the quality of their earnings. Our findings are consistent with those of Wijesinghe and Kavinda (Wijesinghe & Kavinda, 2017), He et al. (He et al., 2009), and García-Osma and Noguer (García-Osma & Noguer, 2007) which show that the CG mechanisms increase the supervisory and controlling role of owners, raise the quality of financial statements, and reduce the opportunistic behavior of managers. We also provide evidence that shareholder rights and audit organizations are the most prominent CG characteristics which influence REM activities. Both of these CG characteristics enforce additional monitoring on the quality of financial reporting and increase the audit standards First-Stage Regression Results Variable CGS ABN_REM -0.043 *** (3.601) Intercept 0.489*** (6.708) Adj. R2 0.014 F-Stat. 12.966*** Covariance ABN_REM 0.016 Second-Stage Regression Results Variables ABN_REM TLev 0.494*** 0.549 (3.431) (1.553) CGS -0.840*** -0.817*** (-7.777) (-6.882) TLev * CGS 0.493 *** (2.924) Tang -0.449*** -0.448*** (4.726) (-4.725) ROA -0.124*** -0.109*** (-6.413) (-6.328) Size 0.093*** 0.091*** (5.821) (5.710) Liq 0.039*** 0.038*** (2.877) (2.827) NIP -0.520*** -0.603*** (-2.859) (-3.278) FIOR -0.110*** -0.108*** (-5.680) (-5.589) Intercept 0.582*** -0.576*** (8.108) (-6.865) Year-fixed effects Yes Yes Industry-fixed effects Yes Yes Adj. R20.318 0.320 F-Stat. 64.310*** 62.224*** DW 1.230 1.237 Note: The results indicate a significant negative (positive) relationship between CGS (interaction term) and REM, using 2SLS regression to control endogeneity. Beta corresponds to unstandardized coefficients. Numbers inside the parentheses are t-statistics. *** and ** indicate statistical significance at the 1% and 5% level, respectively. Table 10. Regression results of the 2SLS model GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 70 of firms. These results suggest that the frequency of managerial engagement in REM activities depends on the degree of CG because it is an effective mechanism to control the opportunistic behavior of managers and it decreases the possibility to freely access the internal and external financing (Tulcanaza- Prieto, Shin, et al., 2020). Specifically, improving shareholder rights or establishing secure audit organizations would help to avoid management’s incentive to engage in REM. Finally, we conducted a 2SLS regression analysis to control endogeneity problem in our regression model. In the second-stage regression, we find a negative relationship between CG and REM and a positive association between the interaction term and REM. These results are consistent with our previous results. Acknowledgments and Legal Responsibility We reveal special remarks to the committee of 2020 Joint Conference organized by the Korean Finance Association for their valuable comments for our paper presentation. This research was funded by KUMOH NATIONAL INSTITUTE OF TECHNOLOGY, grant number 2020-0229-0001. References Alves, S. (2012). Ownership Structure and Earnings Management: Evidence from Portugal. Australian Accounting, Business, and Finance Journal, 6(1), 57-74. Amertha, I., Ulupui, I., & Putri, I. (2014). Analysis of Firm Size, Leverage, Corporate Governance on Earnings Management Practices (Indonesian Evidence). Journal Of Economics, Business and Accountancy Ventura, 17(2), 259-268. https://doi.org/10.14414/jebav.14.1702009 Anagnostopoulou, S., & Tsekrekos, A. (2017). The Effect of Financial Leverage on Real and Accrual-Based Earnings Management. Accounting and Business Research, 47(2), 191-236. https://doi.org/10.1080/00014788.2016.1204217 Black, B., Jang, H., & Kim, W. (2006). Does Corporate Governance Predict Firm’s Market Values? Evidence from Korea. The Journal of Law, Economics, & Organization, 22(2), 366-413. Burgstahler, D., & Dichev, I. (1997). Earnings Management to Avoid Earnings Decreases and Losses. Journal of Accounting and Economic, 24, 99-126. https://doi.org/10.1 016/S0165-4101(97)00017-7 Byard, D., Li, Y., & Weintrop, J. (2006). Corporate Governance and the Quality of Financial Analysts’ Information. Journal of Accounting and Public Policy, 32(4), 609-625. Byun, H., Kwak, S., & Hwang, L. (2008). The Implied Cost of Equity Capital and Corporate Governance Practices. Asia-Pacific Journal of Financial Studies, 37(1), 139-184. Chen, E., & Dixon, W. (1972). Estimates of Parameters of a Censored Regression Sample. Journal of the American Statistical Association, 67(339), 664-671. https://doi.org/1 0.2307/2284463 Chen, Y., Lee, C., & Chou, P. (2015). Stock-Based Compensation and Earnings Management Behavior. Review of Pacific Basin Financial Markets and Policies, 18(2), 1-33. https://doi.org/10.1142/S0219091515500083 Cohen, D., Dey, A., & Lys, T. (2008). Real and Accrual-Based Earnings Management in the Pre- and Post-Sarbanes-Oxley Periods. The Accounting Review, 83(3), 757-787. https://do i.org/10.2139/ssrn.813088 Cohen, D., & Zarowin, P. (2010). Accrual-Based and Real Earnings Management Activities around Seasoned Equity Offerings. Journal of Accounting and Economics, 50, 2-19. https://doi.org/10.1016/j.jacceco.2010.01.002 Fama, E., & French, K. (1992). The Cross-Section of Expected Stock Returns. The Journal of Finance, 47(2), 427. https://doi.org/10.1111/j.1540-6261.1992.tb04398.x Fitri, A., Muda, I., & Badaruddin. (2018). The Influence of Good Corporate Governance, Leverage, and Profitability on Earnings Management with Firm Size as Moderating Variable in the Banking Companies Listed in Indonesia Stock Exchange in the Period of 2012-2016. International Journal of Research and Review, 5(9), 49-66. García-Osma, B., & Noguer, B. (2007). The Effect of the Board Composition and its Monitoring Committees on Earnings Management: Evidence from Spain. Corporate Governance: An International Review, 15(6), 1413-1428. Geiger, M., & North, D. (2013). Do Strong Shareholder Rights Mitigate Earnings Management? Journal of Accounting, Ethics and Public Policy, 14(2), 289-322. Goh, J., Lee, H., & Lee, J. (2013). Majority Shareholder Ownership and Real Earnings Management: Evidence from Korea. Journal of International Financial Management & Accounting, 24(1), 26-61. https://doi.org/10.1111/jifm.12 006 Graham, J., Harvey, C., & Rajgopal, S. (2005). The Economic Implications of Corporate Financial Reporting. Journal of Accounting and Economics, 40, 3-73. https://doi.org/10. 1016/j.jacceco.2005.01.002 Gujarati, D. (1988). Basic Econometrics (Second). McGraw Ana Belen Tulcanaza-Prieto, Younghwan Lee, Jeong-Ho Koo 71 Hill. Hashim, H., & Devi, S. (2012). Institutional Monitoring and Earnings Quality in Malaysia. Procedia - Social and Behavioral Sciences, 65, 419-426. He, L., Labelle, R., Piot, C., & Thornton, D. (2009). Board Monitoring, Audit Committee Effectiveness, and Financial Reporting Quality: Review and Synthesis of Empirical Evidence. Journal of Forensic & Investigative Accounting, 1(2), 1-14. Healy, P., & Wahlen, J. (1999). A Review of the Earnings Management Literature and its Implications for Standard Setting. Accounting Horizons, 13(4), 365-383. Huang, X., & Sun, L. (2017). Managerial Ability and Real Earnings Management. Advances in Accounting, 39, 91-104. https://doi.org/10.1016/j.adiac.2017.08.003 Jelinek, K. (2007). The Effect of Leverage Increases on Earnings Management. Journal of Business and Economic Studies, 13(2), 24-46. Jensen, M. (1986). Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review, 76, 323-339. https://doi.org/10.2139/ssrn.99580 Jianga, W., Lee, P., & Anandarajan, A. (2008). The Association between Corporate Governance and Earnings Quality: Further Evidence using the GOV-Score. Advances in Accounting, 24(2), 191-201. Jin, S., Hwang, I., & Kang, S. (2018). Improving Sustainability through a Dual Audit System. Sustainability, 10(1), 1-15. https://doi.org/10.3390/su10010137 Kang, S., & Kim, Y. (2012). Effect of Corporate Governance on Real Activity-Based Earnings Management: Evidence from Korea. Journal of Business Economics and Management, 13(1), 29-52. Kaushik, N., & Kumar, S. (2018). Leverage Effect on Earnings Management: Evidence from India. The IUP Journal of Accounting Research & Audit Practices, 17(4), 21-34. Kim, J., & Sohn, B. (2013). Real Earnings Management and Cost of Capital. Journal of Accounting and Public Policy, 32(6), 518-543. https://doi.org/10.1016/j.jaccpubp ol.2013.08.002 Klein, A. (2002). Audit Committee, Board of Director Characteristics and Earnings Management. Journal of Accounting and Economics, 33(3), 375-400. https://doi.org /10.1016/S0165-4101(02)00059-9 Kuo, J., Ning, L., & Song, X. (2014). The Real and Accrual-Based Earnings Management Behaviors: Evidence from the Split Share Structure Reform in China. The International Journal of Accounting, 49(101-136). https://doi.org/10.1016/j.inta cc.2014.01.001 Kurnia, Y., & Pradipta, A. (2016). Corporate Governance and Real Earnings Management. International Journal of Business, Economics and Law, 9(1), 17-23. Lasfer, M. (2006). The Interrelationship between Managerial Ownership and Board Structure. Journal of Business Finance and Accounting, 33(7-8), 1006-1033. Lee, Y., Kang, S., & Cho, S. (2015). The Effect of Voluntary IFRS Adoption by Unlisted Firms on Earnings Quality and the Cost of Debt: Empirical Evidenced from Korea. Journal of Business Economices and Management, 16(5), 931-948. Lemma, T., Negash, M., & Mlilo, M. (2013). Determinants of Earnings Management: Evidence from around the World. SSRN Electronic Journal, 1-56. https://doi.org/10.2139/ssr n.2370926 Lopes, A. (2018). Audit Quality and Earnings Management: Evidence from Portugal. Athens Journal of Business and Economics, 4(2), 179-192. López-Iturriaga, F., & Hoffman, P. (2005). Earnings Management and Internal Mechanisms of Corporate Governance: Empirical Evidence from Chilean Firms. Corporate Ownership and Control, 3(1), 17-29. https://doi.org/10.22 495/cocv3i1p2 Luoma, P., & Goodstein, J. (1999). Stakeholders and Corporate Boards: Institutional Influences on Board Composition and Structure. Academy of Management Journal, 42(5), 553-563. Nwakuya, M., & Ijomah, M. (2017). Fixed Effect versus Random Effects Modeling in a Panel Data Analysis: A Consideration of Economic and Political Indicators in Six African Countries. International Journal of Statistics and Applications, 7(6), 275-279. https://doi.org/10.5923/j. statistics.20170706.01 Othman, H., & Zeghal, D. (2006). A Study of Earnings- Management Motives in the Anglo-American and Euro- Continental Accounting Models: The Canadian and French Cases. The International Journal of Accounting, 41(4), 406-435. https://doi.org/10.1016/j.intacc.2006.09.004 Piosik, A., & Genge, E. (2019). The Influence of a Company’s Ownership Structure on Upward Real Earnings Management. Sustainability, 12(152), 1-24. https://doi.org/10.3390/su12 010152 Prawitt, D., Smith, J., & Wood, D. (2009). Internal Audit Quality and Earnings Management. The Accounting Review, 84(4), 1255-1280. Roychowdhury, S. (2006). Earnings Management through Real Activities Manipulation. Journal of Accounting and Economics, 42(3), 335-370. https://doi.org/10.1016/j.jacce co.2006.01.002 Saona, P., Muro, L., & Alvarado, M. (2020). How Do the Ownership Structure and Board of Directors’ Features Impact Earnings Management? The Spanish case. Journal of International Financial Management & Accounting, 31(1). https://doi.org/doi.org/10.1111/jifm.12114 Schipper, K. (1989). Commentary of Earnings Management. Accounting Horizons, 3(1), 91-102. Standard & Poor’s Governance Services. (2004). Criteria, Methodology and Definitions. In Standard & Poor’s Corporate Governance Scores and Evaluations (pp. 1-23). McGraw-Hill Companies Inc. Swai, J. (2016). The Impact of Corporate Governance and Firm-Specific Characteristics on Earnings Management: Evidence from East Africa. Research Journal of Finance GLOBAL BUSINESS & FINANCE REVIEW, Volume. 25 Issue. 4 (WINTER 2020), 51-72 72 and Accounting2, 7(8), 139-156. Sweeney, A. (1994). Debt-Covenant Violations and Managers Accounting Responses. Journal of Accounting and Economics, 17, 281-308. https://doi.org/10.1016/0165-41 01(94)90030-2 Tulcanaza-Prieto, A., Koo, J., & Lee, Y. (2019). Does Cost Stickiness Affect Capital Structure? Evidence from Korea. Korean Journal of Management Accounting Research, 19(2), 27-57. Tulcanaza-Prieto, A., Lee, Y., & Koo, J. (2020). Effect of Leverage on Real Earnings Management : Evidence from Korea. Sustainability, 12(2232), 1-20. https://doi.org/10.3 390/su12062232 Tulcanaza-Prieto, A., & Lee, Y. (2019). Internal and External Determinants of Capital Structure in Large Korean Firms. Global Business & Finance Review, 24(3), 79-96. https://doi.org/10.17549/gbfr.2019.24.3.79 Tulcanaza-Prieto, A., Shin, H., Lee, Y., & Lee, C. (2020). Relationship among CSR Initiatives and Financial and Non - Financial Corporate Performance in the Ecuadorian Banking Environment. Sustainability, 12(1621), 1-17. https://doi.org/10.3390/su12041621 Vakilifard, H., & Mortazavi, M. (2016). The Impact of Financial Leverage on Accrual-Based and Real Earnings Management. International Journal of Academic Research in Accounting, Finance and Management Sciences, 6(2), 53-60. https://doi.org/10.6007/IJARAFMS/v6-i2/2039 Wasimullah, T., & Abbas, Z. (2010). Can High Leverage Control the Opportunistic Behavior of Managers: Case Analysis of Textile Sector of Pakistan. International Research Journal of Finance and Economics, 47, 134-142. https://www.researchgate.net/publication/287397802_Ca n_high_leverage_control_the_opportunistic_behavior_of _managers_Case_analysis_of_textile_sector_of_Pakistan Waweru, N., & Riro, G. (2013). Corporate Governance, Firm Characteristics and Earnings Management in an Emerging Economy. Journal of Applied Management Accounting Research, 11(1), 43-64. Wijesinghe, R., & Kavinda, C. (2017). The Impact of Leverage on Real Earnings Management: Evidence from Listed Manufacturing Companies in Colombo Stock Exchange. Kelaniya Journal of Management, 7, 63-82. https://doi.org/ 10.4038/kjm.v6i1.7527 Xie, B., Davidson, W., & DaDalt, P. (2003). Earnings Management and Corporate Governance: The Role of the Board and the Audit Committee. Journal of Corporate Finance, 9(3), 295-316. Yoon, S., Miller, G., & Jiraporn, P. (2006). Earnings Management Vehicles for Korean Firms. Journal of International Financial Management & Accounting, 17(2), 85-109. https://doi.org/ 10.1111/j.1467-646X.2006.00122.x Zamri, N., Rahman, R., & Isa, N. (2013). The Impact of Leverage on Real Earnings Management. Procedia Economics and Finance, 86-95. https://doi.org/10.1016/S 2212-5671(13)00222-0 Zang, A. (2012). Evidence on the Trade-Off between Real Activities Manipulation and Accrual-Based Earnings Management. The Accounting Review, 87(2), 675-703. https://doi.org/10.2308/accr-10196