Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Corporate Governance and Earnings Management in Latin American markets        Jesus Alfonso Saenz Gonzalez                ...
1. Introduction   In recent years, large accounting fraud uncovered in the stock markets has once againconfirmed the impor...
alignment of ownership and management interests. Thus, several studies documented asignificant relationship between the ch...
among such owners (La Porta et al., 1999; Castañeda, 2000a, 2000b; Rabelo & Coutinho,2001; Santiago et al., 2009). Moreove...
(Dechow & Schrand, 2004). We use the modified version of Jones (1991) proposed byDechow et al. (1995) and which has been u...
discretionary accruals in developed market such as the U.S. (DeFond & Jiambalvo, 1991;Wartfield et al., 1995; Dechow et al...
section 4, we shows the statistical results; in section 5, we discusses the results, the limitationsand future lines of re...
Servaes, 1990; Agrawal & Knoeber, 1996; Demsetz & Villalonga, 2001; De Miguel et al.,2004; Boubraki et al., 2005; Sanchez-...
proportion of their wealth materialized in shares of the company that they lead, at mostdirectly affect their personal wea...
implementation of Code of Best Corporate Practices for a sample of Mexican listedcompanies, found that firms with a fewer ...
In this way, large shareholders play a key role on internal control of companies, becausethe volume of participation encou...
concentration and, second, many firms are directly controlled by one of the industrial orfinancial conglomerates that oper...
concentrated ownership structures, given the incentives and the greater ease with which thebig shareholders can monitor th...
reputation concerns means that family firms do not act opportunistically in reporting earnings,such those actions that are...
usually used to align the interest of the company in benefit of the majority shareholder, indetriment of minority sharehol...
suggests a negative relation between the proportion of shares held by family owners and theabsolute value of discretionary...
that may arise therein.   In Latin America context, Lefort (2005) pointed out that institutional investor have animportant...
John & Senbet, 1998; Daily et al., 2003; Chatterjee et al., 2003). In this way, the board ofdirectors is an essential elem...
management team and a higher quality of corporate decisions (Pearce & Zahra, 1992). In thissense, Chin et al. (2006) for a...
external members in board of directors are greater than the owners, for there to be moreoversight of management and to max...
making boards less effective in monitoring the decisions taken by managers opposed to thoseboards that have a majority of ...
principle, a negative relationship between the boards activity and EM (Monterrey & Sanchez,2008). An opposing view is that...
the Codes of Good Governance, given the high concentration of ownership and control heldby families that produces an effec...
(Gill & Kharas, 2007; Aidt, 2009). Voliotis (2011) looked at different forms of organisationalcorruption at the European U...
effectiveness of government (a high index value) increase the chance that entrepreneurscapture a larger portion of the rev...
study (Arellano & Bover, 1991; Arellano, 1993; Himmelberg et al., 1999; Palia, 2001; Bricket al., 2005). Additionally, sin...
integrate the board of directors; the board independence (Board_IND) is measured by theproportion of external directors in...
more negotiating power with the external auditor and are subject to increased monitoring byinvestors and analysts, so acco...
variable that takes value of 1 if the company has had losses in the last two years and, 0otherwise. Finally, we consider t...
TABLE 2:                 Descriptive Statics of Discretionary Accruals [Abs (DAC)it]                                    Es...
have a higher internal ownership with 7.3%, followed by Argentinean (7.1%), Chilean (6%)and Mexican (4.5%) companies. Fina...
TABLE 3:                      Descriptive Statistics of Quantitative and Dichotomous Variables                            ...
panel data, the absolute value of discretionary accruals [Abs (ADD) it] on the variables ofownership structure, board of d...
management team decreases due to the existence of problems of communication andcoordination that increases the use of disc...
not an attitude, because the absence of sufficient distance from the management of thecompany could concentrate in fact th...
motivated to manipulate the results obtained with the intention to make the company moreattractive (Kothari et al., 2005; ...
TABLE 4:                           Discretionary Accrual Regressions on Corporate Governance and Control Variables        ...
TABLE 4:                                     Discretionary Accrual Regressions on Corporate Governance and Control Variabl...
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Corporate Governance And Earnings Management In Latin American Markets
Upcoming SlideShare
Loading in …5
×

Corporate Governance And Earnings Management In Latin American Markets

1,114 views

Published on

Forthcoming in Journal of Business Ethics

  • Be the first to comment

Corporate Governance And Earnings Management In Latin American Markets

  1. 1. Corporate Governance and Earnings Management in Latin American markets Jesus Alfonso Saenz Gonzalez Emma Garcia Meca Profesor Investigador UACJ Profesora Titular UPTCAbstract: We use panel data to examine the relationship between the internal mechanisms ofCorporate Governance and Earnings Management measured by discretionary accruals incompanies listed on Latin American markets. Our results show how the Board dimensioncreates problems of communication and coordination that reduce the oversight over themanagement team, increasing earnings management. Also, we point out how in the LatinAmerican context the role of external directors is limited and those Boards that meet morefrequently takes a more active position in the monitoring of the insiders, showing a lower useof manipulative practices. In addition, we find a non-linear relationship between insiderownership and discretionary accruals, also pointing out that ownership concentration may bea manipulative practices constrictor mechanism only when the ownership of mainshareholders is moderate. These results support the general opinion that the full applicationof the Anglo-Saxon corporate governance model to a continental institutional setting isinappropriate. In addition, the findings have important policy implications since it is, to thebest of our knowledge, the first study to analyze the relationship between the effectiveness ofthe government and the earnings management behaviour. We document how when a countryimplements controls aimed to reduce corruption, to strengthen the rule of law or to improvethe effectiveness of government, it leads to reduce earnings management. These data wouldprovide useful information for testing complementarities between low-quality financialaccounting regimes and quality control mechanisms to promote economic efficiency.Key Words: Board of Directors; Corporate Governance; Corruption; Discretionary Accruals;Ownership Structure; Stock Markets. 1
  2. 2. 1. Introduction In recent years, large accounting fraud uncovered in the stock markets has once againconfirmed the importance of transparency and reliability of financial information provided tomarkets. It is no coincidence that this lively interest in accounting issues is accompanied by asignificant concern for good governance. The regulatory response to financial scandals hasbeen taking measures to protect information transparency, mitigate conflicts of interest andensure the independence of auditors, all in order to protect the investors interests’ and increasethe confidence of capital markets. Thus, the strong pressure that capital markets exert on managers is an important conditionthat can motivate them to engage in manipulative practices, to the point of being willing, asshown by Graham et al. (2005), to sacrifice the value to meet investor expectations and liveup to forecasts by analysts tagged. For this reason, the accounting function should enjoyadequate protection, and hence the corporate governance (CG) practices stand as a guaranteeof its integrity, improving the quality and transparency of financial statements1 (Johnson etal., 2002; Garcia Osma & Gill de Albornoz, 2005; Biddle & Hilary, 2006; Biddle et al.,2009). Thus, the establishments of internal governance processes are essential to maintain thecredibility of firms’ financial information and safeguarding against earnings manipulations(Dechow et al., 1996). A weak governance structure may provide an opportunity for managersto engage in behaviour that would eventually result in a lower quality of reported earnings.This opportunistic behaviour of managers can be explained by agency theory, given theseparation between ownership and control in a firm, managers may act to maximize their ownwealth at the expense of shareholders’ wealth (Jensen y Meckling, 1976; Fama, 1980) Thus, from the studies published by Jensen & Meckling (1976) and Fama & Jensen(1983), it is assume that both, the role of board of directors and ownership structure, arecrucial in monitoring managerial activity, capable of reducing agency costs resulting from the 2
  3. 3. alignment of ownership and management interests. Thus, several studies documented asignificant relationship between the characteristics of the board of directors and the integrityof accounting information (Klein, 2002; Xie et al., 2003; Anderson et al., 2004; Peasnell etal., 2005; Karamanou & Vefeas, 2005; Saleh et al., 2005; Ahmed et al., 2006; Bradbury et al.,2006; Cheng et al., 2006; Rahman & Ali, 2006; Patelli & Prencipe, 2007; Hashim & Devi,2008). Regarding to the ownership structure, previous studies analyzed the effect of theinternal ownership and shareholding concentration held by major shareholders on the qualityof financial results, either individually as a proxy for ownership structure or along with theparticipation of managers as a distinct dimension of that structure (Wartfield et al., 1995;Short & Keasey, 1999; Demsetz & Villalonga, 2001; Fan & Wong, 2002; Yeo et al., 2002;Han, 2004; Lefort, 2005; Kim & Yi, 2006; Price et al., 2006). All these studies related mainlyto Anglo-Saxon countries, where outsides investors are well-protected by the legal system(e.g. United Stated, United Kingdom) and the level of transparency is high, most listed firm’spresent widely-held ownership structures. In this setting, the main agency conflict stems fromthe divergence of the interests between managers and shareholders (Jensen & Meckling,1976). The results documented by La Porta et al. (1998, 2000), Leuz et al. (2003), Kim & Yi(2006) reveal that the manipulative practices are higher in economies with less developedstock markets, with more concentrated ownership structures and weak investor protectionlaws. In the Latin American context, the ownership structure of listed firms is characterisedby high levels of concentrated ownership where many firms are directly controlled by one ofthe industrial or financial conglomerates that operate in the region (Khanna & Palepu, 2000;Lefort, 2003; Lefort & Walker, 2005, 2007; Lopez & Saona, 2005; Cespedes et al., 2008), bythe use of pyramidal structures that enable controlling shareholders to separate their votingand cash flow rights (Mendes & Mazzer, 2005), and by the notable presence of family groups 3
  4. 4. among such owners (La Porta et al., 1999; Castañeda, 2000a, 2000b; Rabelo & Coutinho,2001; Santiago et al., 2009). Moreover, the control exerted by these family owners is notusually limited solely to their participation in the firms’ ownership since they usually play anactive role in management (La Porta et al., 1999; Castillo-Ponce, 2007). In that regard, alower separation between ownership and control shift the main agency conflict to the possibleexpropriation of minority shareholders by controlling owners (La Porta et al., 2000; Lefort,2005, 2007; Saona, 2009). According to the approaches set out, this paper have the main objective to analyze therelationship between the internal mechanisms of CG and earnings management (EM) in firmslisted on Latin American stock markets, specifically, in the markets of Argentina, Brazil,Chile and Mexico, during the period 2006 to 2009. These countries have not been strangers tothe initiatives of practically all Western countries since the promulgation in 2000 of theSarbanes-Oxley in the U.S. and it seems appropriate to verify empirically the effects of CGmechanisms such as ownership structure and board of directors on EM in these countries,where both the predominant agency conflict and the institutional environment differ fromthose in the Anglo-Saxon and Continental European markets. In addition, according to Boyd & Hoskisson (2010) the nature of institutional countryeffects in which firms are embedded shapes their governance. Thus, by using a governmentindex proposed by previous literature we will test if those countries that control corruption,have a stronger rule of law and higher effectiveness of their government, reduce the EMbehaviour. We define EM in terms of ‘absence of manipulative practices’. This is because theintentional manipulation of earnings by managers may reduce the usefulness of earnings tothe overall users (Velury & Jenkins, 2006; Dechow et al., 2010; Matis et al., 2010). Earningsthat are persistence and predictable may not be of high quality if this results from EM 4
  5. 5. (Dechow & Schrand, 2004). We use the modified version of Jones (1991) proposed byDechow et al. (1995) and which has been used in other studies such as Teoh et al. (1998), Xieet al. (2003) and Francis et al. (2008) to determine the discretional accruals. Thus, we use theabsolute value of discretionary accruals [Abs (DCA) it] as a measure of the degree of EM.This is consistent with previous studies on EM who pointed out that the study of the quality ofresults does not impose any direction or sign on the expectations of EM (Wartfield et al.,1995; Klein, 2002; Gabrielsen et al., 2002; Bowen et al., 2004, 2008; Van Tendeloo &Vanstraelen, 2005; Wang, 2006; Chen et al., 2007; Barth et al., 2008). This study contributes to the growing body of literature related to CG in the followingways. First, it extends the very limited research on the relationship between CG and EM inLatin America and provides a more comprehensive picture of this association. Second, itprovides further evidence by analyzing the empirical evidence in a Latin American context,where the boards of directors, legal investors’ protection, the presence of reference investors’and the threat of corporate takeover differs substantially from other regions of the world,especially in those countries with developed markets. Third, our study extends the literature toethical aspects that are scarce and have not been tested yet in the relationship between internalmechanisms of CG and EM in Latin America, such as corruption, rule of law and governmenteffectiveness. In this way, we include a proxy that represents the country governability level(government index). This is because corruption is prevalent in emerging countries, affectingthe effective function of governments and economies (Gill & Kharas, 2007; Aidt, 2009). Theimplementation of controls aimed to reduce the corruption, to strengthen the rule of law or toimprove the effectiveness of the government in a country could lead to reduce anopportunistic behaviour and, consequently, could reduce the EM practices in firms. On the other hand, the analysis of the relationship between CG and EM in Latin Americanmarkets is motivated mainly, because several studies have investigated the use of 5
  6. 6. discretionary accruals in developed market such as the U.S. (DeFond & Jiambalvo, 1991;Wartfield et al., 1995; Dechow et al., 1996; Chung et al., 2002; Klein, 2002; Rajgopal et al.,2002; Xie et al., 2003; Bowen et al., 2004; Cornett et al., 2008), the UK (Peasnell et al., 2000,2005) and Australia (Koh, 2003; Davidson et al., 2005; Hsu & Koh, 2005; Wan et al., 2007).However, very little research has looked at the relationship between CG and EM in emergingmarkets, such as Latin American (Lopez & Saona, 2005; Price et al., 2006; Castrillo & SanMartin, 2007; Teitel & Machuga, 2008). It is possible that in an emerging country, themechanisms of CG are not as effective as in a developed country because of the differentinstitutional environment (Gaio, 2010, Gaio & Raposo, 2011) such as a weaker market forcorporate control (Gibson, 2003; Lins, 2003), more concentrated ownership (Khanna &Palepu, 2000; Lefort, 2003; Lefort & Walker, 2005, 2007; Lopez & Saona, 2005; Cespedes etal., 2008), significant family ownership (Castañeda, 2000a, 2000b; Rabelo & Coutinho, 2001;Santiago et al., 2009) and ineffective shareholder right protection (La Porta, 1998, 2000; Leuzet al., 2003; Lefort, 2007). Additionally, boards of directors in Latin American firms are notas independent as those in developed countries, making less effective in monitoring thedecisions taken by managers (Spencer Stuart, 2000; Santiago & Baek, 2003; Lefort, 2005;Helland & Sukuta, 2005). Given these institutional factors, the results of our study may not besimilar to those found for developed countries. In this case, our study fills a gap in the existing literature by examining the effectivenessof how internal mechanisms of CG (ownership structure and board of directors) constrictmanipulative practices in Latin American firms. Also, as this study was conducted on a largesample of firms over a reasonable time frame, we believe that our findings capture a strongpicture of the relationship between internal mechanisms of CG and EM in Latin America. The remainder of the paper has the following organization: in section 2, the studyhypotheses are developed; in section 3, we present the design and research methodology; in 6
  7. 7. section 4, we shows the statistical results; in section 5, we discusses the results, the limitationsand future lines of research and; finally, in section 6 we present the main conclusions of ourstudy.2. Previous literature and development of hypotheses2.1. Ownership structure The ownership structure is an internal control mechanism that focuses on the aspects thatdefine the ownership of the company and refers to the manner in which titles or rights ofrepresentation redistribute the capital of the firm in one or more individuals or legal entities.In this sense, Demsetz (1983) indicates that ownership structure is just a reflection of theexisting balance between preferences that have the owners and top management team. Thus,the monitoring power derived from the ownership structure resulting in a kind of controlexercised over the company and, particularly, over the top management team. The finalcontrol is given by the distribution of ownership and the capability of any owner or group ofthem to influence in the decisions taken. In this way, several studies have shown that ownership structure is a natural monitoringmechanism that exerts great control over the performance, discretionary and remuneration ofthe top management (Short, 1994; Zajac & Westphal, 1994). Thus, previous studies mainlyfocuses on the effect of insider ownership over the EM (Sanchez-Ballesta & Garcia-Meca,2007; Teshima & Shuto, 2008), analyzing how the managerial ownership influence over theinformativeness of earnings (Wartfield et al., 1995; Gabrielsen et al., 2002; Yeo et al., 2002)or, in conjunction with ownership concentration (measured by the fraction of ownership heldby major shareholders or by the proportion of ownership held by the main shareholders of thefirm), showing how that monitoring of owners improves the quality of managerial decisionsand, consequently, the firm value, since the existence of substantial block holders leads to acloser monitoring of management, implying a lower opportunity for EM (McConnell & 7
  8. 8. Servaes, 1990; Agrawal & Knoeber, 1996; Demsetz & Villalonga, 2001; De Miguel et al.,2004; Boubraki et al., 2005; Sanchez-Ballesta & Garcia-Meca, 2007). Thus, Yeo et al. (2002)and De Bos & Donker (2004) shows that increased ownership concentration is an effectiveCG mechanism in monitoring accounting decisions taken by management. However, Demsetz & Villalonga (2001) affirm that in order to treat ownership structureappropriately and to account for the complexity of interest represented in a given ownershipstructure, must be consider different dimensions of ownership structure, i.e. not only focus onthe insider ownership and ownership concentration. Following this suggestion, we analyzeapart of this two commonly dimensions examined by previous literature, two differentdimensions of ownership structure that previous literature also shown that could be aneffective CG mechanism in monitoring management decisions, capable to constrictsmanipulative practices and, consequently, improving the earnings quality: family ownership(Wang, 2006; Ali et al., 2007; Bona et al., 2008) and institutional ownership (Shleifer &Vishny, 1997; Rajgopal et al., 2002; Chung et al., 2002; Balsam et al., 2002; Jiambalvo et al.,2002; Koh, 2003; Han, 2004; Ferreira & Matos, 2008; Ruiz et al., 2009; Ferreira et al., 2010).The next sections describe the development of the hypotheses related to the four ownershipstructure variables examined in our study.2.1.1. Internal ownership The hypotheses about the influence that the ownership structure has on the value of thefirm, justified mainly through the Agency Theory, have been extended to other aspects ofcompany information, such as EM. Therefore, Agency Theory suggests that when managersare not owners of the entity that they lead or have a low equity stake on it, their behaviour isaffected by self-interest that away goals of maximizing corporate value and, therefore, of theinterest of shareholders, which facilitates EM (Jensen & Meckling, 1976; Fama, 1980; Fama& Jensen, 1983; Healy, 1985; Holthausen et al., 1995). In contrast, if managers have a certain 8
  9. 9. proportion of their wealth materialized in shares of the company that they lead, at mostdirectly affect their personal wealth on the decisions taken will tend to align, to a greaterextent, their interests with other shareholders (convergence of interests’ hypothesis) and showless discretionary behaviour (Chaganti & Damanpour, 1991; Mehran, 1995; Alonso & DeAndres, 2002; Minguez & Martin, 2003). However, excessive internal ownership may alsohave an adverse effect on the company, because the higher power of the managers could leadthem to take accounting decisions that reflect personal reasons, affecting the goal ofmaximizing the value of the company (Yermack, 1997; Aboody & Kaznik, 2000). In line withthis, Weisbach (1988) and Fernandez et al. (1998) point out that managers could use thehigher power contained by their shares to avoid be removal in case of inefficient behaviour. Inthis way, arise the entrenchment hypothesis pointed out by Fama y Jensen (1983), whichestablished that high levels of internal ownership could lead to a greater EM by managers(Cornett et al., 2008). Therefore, Yeo et al. (2002) concludes that the informativeness ofaccounting results increases with low levels of internal ownership, while for high levels, theinternal ownership is not sufficient as alignment interest’s mechanism (Mork, et al., 1988;Wartfield et al., 1995; Sanchez-Ballesta & Garcia-Meca, 2007). In Latin American context, Santiago et al. (2009) for a sample of listed companies inBrazil, Chile and Mexico, found that a small number of insiders owns shares of companies,making this group unable to exert a great influence on decisions and strategies taken by theboard of directors. This suggests an agency problem, because the internal directors interestsseem not aligned with the interests of the other shareholders (convergence of interests’hypothesis), that could increase the likelihood of managers to use manipulative practices fortheir own benefit (Koh, 2003; Bowen et al., 2004; LaFond & Roychowdhury, 2006), reducingthe quality of the information issued by companies in Latin America (Lopez & Saona, 2005).However, Machuga & Teitel (2009) who analyzing earnings quality surrounding the 9
  10. 10. implementation of Code of Best Corporate Practices for a sample of Mexican listedcompanies, found that firms with a fewer internal ownership shows a greater earnings qualitycompared to those firms that not have managerial ownership, i.e. shows less manipulativepractices by managers. Therefore, the argument that insider ownership constrains the opportunistic interest ofmanagers suggests a negative relation between the proportion of shares held by insiders andthe absolute value of discretionary accruals. Nevertheless, the argument that high levels ofinsider ownership can become ineffective in aligning insiders to take decisions thatmaximizes the firms value, suggests a positive relation. Similarly, although the convergenceof interests’ hypothesis suggests a positive association between the informativeness ofearnings and insider ownership, the entrenchment hypothesis leads to a negative relationship,suggesting that when accountings numbers are less informative in measuring the firmperformance, high managerial ownership is likely organizational response. In this way, weaddress the competing views by testing the following unsigned hypothesis: H1: The insider shareholding affects earnings management.2.1.2. Ownership Concentration The degree of ownership concentration is an important factor because it helps to overcomethe problem of the lack of incentives for monitoring. Several studies shows the importance ofconcentrated ownership structures in which one or a few major shareholders exert ahighlighted level of control on listed companies (Demsetz, 1983; Shleifer & Vishny, 1986; LaPorta et al., 1998; Lefort & Walker, 2000; Han, 2004; Lefort, 2005; Sanchez-Ballesta &Garcia-Meca, 2007; Cespedes, 2008). The basic idea lies in that large shareholders hasincentives to take responsibility and cost involved in the monitoring of managers, unlike smallshareholders who tend to adopt a passive attitude in defence of their interest. 10
  11. 11. In this way, large shareholders play a key role on internal control of companies, becausethe volume of participation encouraged them to monitoring and influence in the strategy ofthe firm where they have invested (Fernandez et al., 1998; Yeo et al., 2002; Gabrielsen et al.,2002). This means that a greater ownership concentration should conduct, in accordance withthe efficient monitoring hypothesis (Jensen & Meckling, 1976), to a less opportunisticbehaviour and a greater tendency to maximization the value of the firm (Fama, 1980; Fama &Jensen, 1983), having a positive impact on the informativeness of accounting earnings, sincewhen increasing the participation of the controlling shareholder reduce the incentives of thisowner to expropriate the wealth of minority shareholders (McConnell & Servaes, 1990;Agrawal & Knoeber, 1996; Demsetz & Villalonga, 2001; De Miguel et al., 2004; Boubraki etal., 2005). In this sense, De Bos & Donker (2004) point out that increased ownership is aneffective CG mechanism in monitoring accounting decisions taken by management thatimplies a higher earnings quality and a strong positive effect on earnings informativeness(Yeo et al., 2002). However, when the level of ownership concentration is too high it can lead to agencyproblems due to the expropriation of the minority shareholders’ interests (La Porta et al..1998, 2000, 2002, Leuz et al., 2003; Boubraki et al., 2005; Lefort, 2007). In this way, arisethe entrenchment effect formulated by Morck et al. (1988), based on the influence of thecontrolling shareholder over the information provided by company to the market (Zingales,1994; Shleifer & Vishny, 1997; Sanchez-Ballesta & Garcia-Meca, 2007), where the externalinvestor will put scant attention to financial data, because he expects it to remain objective, toa greater extent, the particular interests of the majority owner rather than a true reflection ofthe economic consequences of transactions made by the company (Fang & Wong, 2002). In Latin America there are two important aspects that characterize the structures ofownership and corporate control: first, companies shows a high degree of ownership 11
  12. 12. concentration and, second, many firms are directly controlled by one of the industrial orfinancial conglomerates that operate in the region (Khanna & Palepu, 2000; Lefort, 2003;Lefort & Walker, 2005, 2007; Lopez & Saona, 2005; Cespedes et al., 2008). In their studies,Santiago et al. (2009) and Saona (2009) for a samples of companies in Chile, Brazil andMexico, found that 96% of Chilean companies are affiliated to a conglomerate, followed by72% of Brazilian companies and 57% of Mexican firms. Usually, the majority shareholderholds the power of the conglomerates through complex structures of ownership, calledpyramids2. The problem with pyramidal structures is that they allow a phenomenon known astunneling3, whereby the last shareholder can divert resources among different companies ofthe pyramid, of course in their own benefit, hiding the true structure of the company anddistorting the relationship between ownership and control (Shleifer & Vishny, 1997; Bertrand& Mullainathan, 2002; Mendes & Mazzer, 2005). This type of governance structure can inferon the quality of the financial results of companies in Latin America, increasing EM. Therefore, the argument that a greater ownership concentration should conduct, inaccordance with the efficient monitoring hypothesis, to a less opportunistic behaviour and agreater tendency to maximization the value on the firms’ informative quality of accountingearnings, suggests a negative relation between the proportion of shares held by largeshareholders and the absolute value of discretionary accruals. However, the argument thathigh levels of ownership concentration can lead to agency problems due to the expropriationof the minority shareholders’ interests, in accordance with the entrenchment effect, suggests apositive relationship. In this way, we address the competing views by testing the followingunsigned hypothesis: H2: The ownership concentration affects earnings management.2.1.3. Family Ownership The Agency theory assumes that moral hazard problems are lower in companies with 12
  13. 13. concentrated ownership structures, given the incentives and the greater ease with which thebig shareholders can monitor the management team. If the companies management is in handsof the owners, as usually occurs in family firms, will tend to eliminate the agency problemthat arises from the separation of ownership and control, achieving greater alignment ofinterests between shareholders and management, resulting in a higher value creation in thecompany through the benefits arising from the increased monitoring (Shleifer & Vishny,1986). In this sense, several studies have shown how certain distinctive characteristics of familyfirm have a positive impact on corporate behaviour. More specifically, in their study of asample of 500 US firms between 1992 and 1999, Anderson et al. (2003) reveal that familyfirms achieve higher levels of performance than non-family firms. This result would bejustified by certain characteristics associated with family nature of the firm, such as it long-investment horizons and its reputation concerns. In this sense, compared with other types ofowners, families are interested in remaining in the firm over a long period time, so they aremore prone to make investments that maximize value in the long term (Anderson & Reeb,2003; Jaggi et al., 2009). Thus, a family owner would tend to have incentives to followmarket rules when making decisions, since the firm is not considered a resource to beconsumed during the owner’s lifetime but rather an asset to be transferred to his/her heirs inthe future (Dyer, 2003). Therefore, the firm’s survival becomes a “family matter” in this typeof enterprise. Furthermore, Anderson et al. (2003) suggest that the long-term ties typical ofthe family owner mean that external agents, such a supplier or lenders, develop their businesswith the controlling family over a long period of time. This leads to those external agentsperceiving a “family reputation” that has economic consequences that last not only forowners’ lifetime, but throughout the lives of his/her heirs. On the same lines, Wang (2006) and Ali et al. (2007) states that long-term orientation and 13
  14. 14. reputation concerns means that family firms do not act opportunistically in reporting earnings,such those actions that are more in line with a short-term orientation. These authors uses thesearguments to offer possible explanations for the result obtained in his study using a sample ofU.S. firms and concludes that family firms provide better earnings quality than non-familyfirms, leading to reduced managerial discretion that results in better performance of thecompanies (Anderson et al., 2004). At this point, it could be concluded that compared with non-family firms, controllingfamily firms would tend to maximize the firm’s wealth in the long term. Thus, there would befewer incentives to obtain private benefits at the expense of minority shareholders, which inturn could result in a higher earnings quality (Bona et al., 2008). However, Wang (2006) andAli et al. (2007) also point out that one of the main limitations that have their studies is thedifficulty to extend their results to other settings where there is a lower protection of minorityshareholders, and consequently, more concentrated ownership structures such as LatinAmerican context. This is because the presence of concentrated ownership structures and thepresence of family groups may trigger other problems of CG. In this sense, when there arelarge shareholders on firms it is more likely to arise conflicts of interests between these partiesand the minority shareholders. In family firms, given their greater information asymmetries,the likelihood of expropriation of corporate resources is high, including the likelihood ofentrenchment of unskilled family management team (Mcvey et al., 2005; Sacristan & Gomez,2007). According with this argument, Castrillo & San Martin (2007) for a sample of Mexicancompanies, studying the relationship between ownership structure and the board of directorswith managerial discretion, finding that family ownership and the level of corporate leverageexplain the degree of discretion that managers have to manipulate accounting numbers inMexico. They also point out that the high concentration of family ownership is a mechanism 14
  15. 15. usually used to align the interest of the company in benefit of the majority shareholder, indetriment of minority shareholders. Other studies conducted on Latin American context suchas Castañeda (2000a, 2000b) and Rabelo & Coutinho (2001) shows that a high familyparticipation exerts a decisive influence on the control of companies, where the ownersusually issued non-voting shares and develop pyramidal ownership structures to obtain fundswithout dispersing their capacity to control the companies. According to previous arguments,it could be argued that the greater concentration of voting rights could entail greaterincentives for controlling shareholders to obtain private benefits, i.e. increasing EM (Bona etal., 2008). In this respect, some studies have provided evidence on the expropriation actions carriedout by family groups. DeAngelo & DeAngelo (2000) shows how the controlling family of alarge North American firm cut dividends to minority shareholders while paying itself a specialdividend. Similarly, drawing on data based on a entire population of Spanish newspapersGomez-Mejia et al. (2001) analyze the role that family relations play in agency contracts andprovide evidence of the entrenchment of the chairman of the board when he/she has a familyties with the controlling shareholders. In those circumstances, controlling shareholders wouldhave incentives to manipulate the accounting information in order to avoid the cost associatedwith the detection of this kind of behaviour (Fan & Wong, 2002; Haw et al., 2004; Francis etal., 2005; Santana et al., 2007). In this way, Fan & Wong (2002) states that, when an ownereffectively controls a firm, she/he also controls the production of the firms’ accountinginformation and reporting policies. Therefore, the argument that a greater family ownership should conduct to a positiveimpact on corporate behaviour, justified by certain characteristics associated with familynature of the firm such as it long-term investment horizons and its reputation concerns,leading to reduced managerial discretion that results in a better performance of the firms, 15
  16. 16. suggests a negative relation between the proportion of shares held by family owners and theabsolute value of discretionary accruals. However, the argument that high levels of familyownership can lead to agency problems due to the expropriation of the minority shareholders’interests, suggests a positive relationship. In this way, we address the competing views bytesting the following unsigned hypothesis: H3: The family ownership affects earnings management.2.1.4. Institutional Ownership The literature review carried out related to the influence of institutional ownership on EM,reveals the existence of two conflicting views regarding the general role of institutionalinvestors in the companies. On one hand, Porter (1992) and Bhide (1993) argue thefragmented ownership that usually have these institutional investors and the frequency oftrading investments, does not actively engage in CG of companies that are part of theirportfolios. Furthermore, Bushee (1998) and El-Gazzar (1998) argue that institutional investorsplays an active role in controlling managerial discretion and improve the efficiency ofinformation in capital markets, being sophisticated investors with advantages to acquire andprocess information (Balsam et al., 2002; Jiambalvo et al., 2002; Koh, 2003; Han, 2004;Ferreira & Matos, 2008; Ruiz et al., 2009; Ferreira et al., 2010), limiting the opportunism andpromoting the reduction of agency costs (Shleifer & Vishny, 1997; Rajgopal et al., 2002;Chung et al., 2002). In this way, Koh (2003) and Hsu & Koh (2005) proposed that the role ofinstitutional investors in firms can be approximated considering the level of participation ofthe institutional shareholders in them, i.e., that institutional ownership may act as agovernance mechanism that affects the EM based on the level of their participation. Inconcrete, low levels of investor participation is assimilated to temporary or short-term view,whereas when the level of participation increases, the institutional investor is assimilate as aninvestor more engaged with the company, and hence, involved in the resolution of conflicts 16
  17. 17. that may arise therein. In Latin America context, Lefort (2005) pointed out that institutional investor have animportant role in CG of companies. The early reform of the pension funds in Chile, followedlater by Argentina, Colombia, Peru and Mexico, gave to institutional investors an importantrole as providers of capital and prompted several changes to the laws of capital markets in theregion, helped to substantially improve the protection of minority shareholders (Iglesias,2000), given the nature of funds administered and their political influence. In this way,Walker & Lefort (2001) shows that the participation of institutional investors create a moredynamic legal framework in capital markets using good CG practices that improvestransparency of accounting information, reducing the cost of capital in firms. Therefore, the argument that a higher institutional ownership should conduct to a positiveimpact on corporate behaviour, because the managers would be discouraged to make EM dueto the pressure from institutional investors to focus in long term, suggests a negative relationbetween the proportion of shares held by institutional owners and the absolute value ofdiscretionary accruals. However, the argument that low levels of institutional ownership couldlead to managers to manipulate accountings numbers, due to the short-term vision and thepreference for short-term gains, suggests a positive relationship. In this way, we address thecompeting views by testing the following unsigned hypothesis: H4: The institutional investors affect earnings management.2.2. Board of Directors The board of directors is the governance body in which shareholders delegate theresponsibility to oversee, compensate and substitute managers, as well as to approve majorstrategic projects, and therefore plays a key role in the overall oversight of the company andthe monitoring of top management, in particular (Jensen & Meckling, 1976; Jensen, 1993; 17
  18. 18. John & Senbet, 1998; Daily et al., 2003; Chatterjee et al., 2003). In this way, the board ofdirectors is an essential element of CG and is considered the main internal mechanism toreduce agency conflicts, either between managers and shareholders or between majority andminority shareholders (LaFond & Roychowdhury, 2006; De Andrade et al., 2009). The CG literature shows different characteristics that may influence in the effectivenesswith which the boards monitor the performance of managers in firms (John & Senbet, 1998;Rahman & Ali, 2006). In this sense, this influence depends on the ability of control that exertthe board over the top management (Brick et al., 2005), which can be affected by aspects sucha number and type of directors (Karamanou & Vafeas, 2005), the ownership structure of thefirm (Kim & Yi, 2006), the quality of CG system (Gibson, 2003), the activism of institutionalinvestors (Koh, 2003), the CEO duality or concentration of power (Saleh et al., 2005; Hashim& Devi, 2008), the existence of specific committee (Klein, 2002; Goodwin & Kent, 2006), thepresence of a hostile takeover market or the existence of a competitive labour market (Brick etal., 2006; De Andrade et al., 2009). Although, according to Fernandez et al. (1997), most ofthe previous CG literature discusses two mainly characteristics or variables that influence themonitoring capabilities of boards: its independence and size. In this way, a part of these twocharacteristic, we analyze its activity and the CEO duality or concentration o power. The nextsections describe the development of the hypotheses related to the four board characteristicsexamined in our study.2.2.1. Board Size Studies such as Davila & Watkins (2009) in Mexico and Ferraz et al. (2011) in Brazil,found that if the size of Board is very small the monitoring of the management team is smallertoo, so they tend to a greater discretion to receive a higher remuneration, a greater chance ofEM and a more likely to information asymmetry (Fernandez et al., 1998; Azofra et al., 2005;Brick, et al., 2006). Thus, a larger size of board assumes a better supervision of the 18
  19. 19. management team and a higher quality of corporate decisions (Pearce & Zahra, 1992). In thissense, Chin et al. (2006) for a sample of 313 firms from Hong Kong, found a negativerelationship between the size of the board and EM, concluding that a larger board fewer arethe manipulative practices made by the management of companies. However, excessive size can be an obstacle for quick and efficient making decisions, dueto problems of coordination and communication. In this sense, Santiago & Brown (2009) in asample of 97 companies in Brazil, Chile and Mexico, found a positive relationship betweenthe size of the board and EM. This indicates that the low separation between ownership andcontrol that exists in Latin American companies assumes that a larger size of the board thelevels of monitoring over the management team decreases, so it increases the risk ofexpropriation by controlling shareholders and the propensity to the discretion of the boardmembers to establish a higher level of remuneration and manipulate the results of companiesfor their own benefit (Jensen, 1993; Yermack, 1996; Fernandez et al., 1997; Eisenberg et al.,1998; Sanders & Carpenter, 1998; Core et al., 1999; Thomsen, 2008). In line with this, Xie etal. (2003) for a sample of 110 U.S. companies, shows a positive relationship between size ofthe board and EM, concluding that a larger size of board greater are the manipulativepractices made by the management of companies. Finally, Bradbury et al. (2006) in a sampleof companies in Malaysia and Singapore found no relationship that links the size of the boardwith the EM. Since we do not can say that there is a conclusive position about the effects thatboard size might have on the quality of results in companies, we formulate the followingunsigned hypothesis: H5: The board size affects earnings management.2.2.2. Board Independence Because previous CG literature shows that independence is often considered as asubstitute for transparency and disclosure of annual reports, recommending that a number of 19
  20. 20. external members in board of directors are greater than the owners, for there to be moreoversight of management and to maximize the value of the organization (Zattoni & Cuomo,2010; Ferraz et al., 2011). This suggests that the degree of board independence is directlyrelated to the quality of information that firms issues (Cheng & Courtenay, 2006). Also, CGliterature affirmed that a greater degree of board independence provides more control over thedevelopment of company activities and a better defence of the issue of information as amechanism to carry out processes of accountability to different groups of business interest,because the external directors are not linked to the management of the entity (Willekens et al.,2005; Karamanou & Vafeas, 2005; Cheng & Courtenay, 2006). Thereby, seeks fairness in thestrategic decisions taken by the board and effective monitoring of the decisions and activitiesof managers, thus ensuring transparency of information and proper image on the outside oforganizations (Chen & Jaggi, 2000; Patelli & Prencipe, 2007). Furthermore, several studiesprovide empirical evidence relating to the role of external directors on the constriction of EM,documenting that a higher proportion of external directors, greater and better will be thequality of financial information that is issued by firms, reducing the chances of EM (Klein,2002; Xie et al., 2003; Peasnell, 2005; Davidson et al., 2005; Garcia-Osma & Gill deAlbornoz, 2005, 2007; Bradbury et al., 2006; Jaggi et al., 2009). However, Park & Shin (2004) argues that transparency of information and effectiveness incontaining manipulative practices may be seriously compromised by the type of CG and,particularly, by the high ownership concentration that could neutralize performance of theexternal directors. In this way, the Securities and Exchange Commission (SEC) of the UnitedStates noted that in Latin America, given the widespread use of pyramidal structures and thehigh degree of participation that have the controlling shareholders in the daily activities ofLatin American firms, generally the boards of directors are mainly integrated by internaldirectors, who tends to be associated with the majority shareholders and/or control groups, 20
  21. 21. making boards less effective in monitoring the decisions taken by managers opposed to thoseboards that have a majority of external directors4 (Santiago & Baek, 2003; Lefort, 2005;Helland & Sykuta, 2005). In this sense, external directors have a very limited participationthat facilitates the EM and managerial discretion (Silveira et al., 2003; Schiehll & Santos,2004). Nevertheless, most recent studies such as Price et al. (2006, 2007), Teitel & Machuga(2008), Chong et al. (2009), Davila & Watkins (2009) and Ferraz et al. (2011) shows thatlegal framework in capital markets (such a Code of Best Corporate Practices) forced LatinAmerican firms to include more external directors, allowing to improve the way that firmsdisclosed their financial information, showing a greater transparency in their reports anddecreasing the chances of EM. By the above, we formulate the following hypothesis in thesense that it could be expecting a possible negative association between the degree of boardindependence and EM: H6: The boards independence affects negatively on earnings management.2.2.3. Board Activity Another characteristic related to the board of directors is its activity, measured by thenumber of meetings, since its size and independence are necessary but not sufficient. Thus,Adams (2003) and Garcia Lara et al. (2009) suggest that the number of meetings is a goodproxy for the directors’ monitoring effort. As Menon & Williams (1994) notes, boards that donot meet, or meet only a few numbers of times, are unlikely to be effective monitors. In thisway, Eguidazu (1999) argues that it is also essential that the boards being active andunderstand its task as a continuous process, and empirically Vafeas (1999) has demonstratedthe existence of a direct relationship between the board activity and the profitability of thefirm. In consequence, is possible that boards with more engaged on their duties take a moreactive stance in order to safeguard the quality of accounting information, so hopefully, in 21
  22. 22. principle, a negative relationship between the boards activity and EM (Monterrey & Sanchez,2008). An opposing view is that board meetings are not necessarily useful because routinetask absorb much of limited time that directors and CEO’s spend together to set the agendafor board meetings (Lorca et al., 2011). By the above, we formulate the following hypothesisin the sense that it could be expecting a possible negative association between the boardsactivity and EM: H7: A greater number of board meetings influences negatively on earnings management.2.2.4. CEO Duality It is understood that there is concentration of power in a company when the same persontakes charge of chief executive and president of the board. This assumption of powerincreases, from the perspective of the Theory of Agency, the risk that the chief executive candevelop strategies that promote their personal interests (Jensen & Meckling, 1976; Jensen,1986) encouraging the management team to adopt decisions for their own benefit withoutconsidering the interests of owners and even develop actions against them (Cole et al., 2001;Jensen & Zajac, 2004), which could create information asymmetries and conflicts of interest,thus leading to inefficiency in CG (Fama & Jensen, 1983; Jensen, 1993). Furthermore, theexistence of duplicity of functions arises the illogical that the manager is controlled byhimself, i.e., has a greater power to influence members of the board of directors, which altersthe functioning of the company and the handling of the ownership structure of companies(Cyert, 2002; Dahya & Travlos, 2002; Brick et al., 2006; Faleye, 2007; Ganga & Vera, 2008).Also, CEO duality may damage the transparency of information of the company, raising thepossibility of the development and disclosure of fraudulent financial statements (Forker,1992; Pi & Timme, 1993; Hashim & Devi, 2008). Some empirical studies developed in Latin America, shows that in practice is not fulfilledwith the separation of roles between the president and CEO, despite the recommendations of 22
  23. 23. the Codes of Good Governance, given the high concentration of ownership and control heldby families that produces an effect of entrenchment by the chairman of board of directorswhen it maintains family ties with the major shareholders. In this sense, in Mexico, Castañeda(2000b) found that in 85% of Mexican companies listed on the Stock Exchange in New York,the majority owners presides the board of directors and also exerted the role of CEO.However, Husted & Serrano (2002) argues that while in Mexican firms, the family retainedboth functions, a group of them showed that the majority owner delegated the role of generalmanager to a family member, which responds to succession process and the need to provide aresource management of the business trust (Hoshino, 2004; Ruiz-Porras & Steinwascher,2007). Also, Leal & Carvalhal (2005) in Brazil, through the application of surveys on a sample of400 listed companies, documented that 36% of companies have concentrated power in thesame person. In Argentina, Chisari & Ferro (2009) for a sample of 100 listed firms, found that75% of the corporations the chairman and CEO are the same person. This situation is not verydifferent in Chile, Lefort and Walker (2005) obtained similar results in a sample of 120 listedcompanies, pointing out that only in 21% of corporations Chairman of the board isindependent, that is, not have duplication of functions between President-CEO, a situationthat is widespread throughout Latin America. By the above, we formulate the followinghypothesis in the sense that it could be expecting a possible positive association between theCEO duality and EM: H8: The existence of concentration of power (CEO duality) increases earnings management.2.3. Government Index While corruption is prevalent in emerging countries, there is increasing focus on thedegree of its predictability to affect the effective functioning of governments and economies 23
  24. 24. (Gill & Kharas, 2007; Aidt, 2009). Voliotis (2011) looked at different forms of organisationalcorruption at the European Union; Galang´s (2011) study reviewed the corruption literature inleading management journals while Dela Rama (2011) looks at how the CG of family-ownedbusiness groups, deals with different forms of corruption in Asia. However, literatureregarding ethical aspects on Latin American countries is scarce and the effects on discretionalbehaviour have not been tested yet. Thus, we use the Government Index (GOV_Index) taken from the research project“Worldwide Governance Indicators” (WGI)6 proposed by Kaufmann et al. (2010) andpublished by the world Bank7 between the periods 2006-2009. In this way, we integrated thisindex using three mainly indicators that previous literature have shown as more importantfactors to measure the way in which the governability of a country helps to reduce or evenincrease the opportunistic behaviour in firms: Control of corruption, rule of law andgovernment effectiveness (Aidt, 2009; Voliotis, 2011; Galang, 2011). In this sense, low levelsof governability (a low index value) imply, generally, behaviours that affect the trust placed inpublic officials and, therefore, undermine the basis of government trust (Shleifer & Vishny,1993). The presence of corruption, the lack of confidence and respect of the agents in thequality of contract enforcement, property rights, courts, as well as the ineffectiveness ofgovernments about the implementation and formulation of policies, increase the risk of theentrepreneur, because people from outside value chain may have opportunistic behaviour andtake advantage of their profits, situation that is feasible due to the relatively high levels ofasymmetry information that characterize the economic activity (Anokhin & Schulze, 2008). Inaddition, the corruption, the inefficiency of governments and a weak rule of law as well asother weaknesses in the country infrastructure, increase transaction and agency costs, limitingthe income of the firms (Manzetti & Wilson, 2007) and, in consequence, increase theopportunistic behaviour of firms. By contrast, control of corruption, a strong rule of law and a 24
  25. 25. effectiveness of government (a high index value) increase the chance that entrepreneurscapture a larger portion of the revenues that they generate by increasing the reliability of cashflows (Rose-Ackerman, 2001) and, consequently, reduce the opportunistic behaviour in firms.Furthermore, in recent years Latin American countries have been made reforms to their legalframeworks, modifying laws and established harder punishments to those persons who aredemonstrated a corruption practice. By the above, we formulate the following hypothesis inthe sense that it could be expecting a possible negative association between the governmentindex and EM: H9: A country with higher levels of governability shows a lower opportunistic behaviour, i.e. the firm’s shows lower levels on earnings management practices.3. Design and Research Methodology3.1. Sample and Data The sample is obtained from companies listed on the Mexican Stock Exchange (BolsaMexicana de Valores), Santiago Stock Exchange (Bolsa de Comercio de Santiago), StockMarket of Buenos Aires (Mercado de Valores de Buenos Aires) and the Sao Paulo StockExchange (Bolsa de Valores de Sao Paulo) during the period 2006 to 2009. Financialinstitutions are excluded, as is common in this type of studies because their particularaccounting practices. The accounting data on financial statements obtained throughEconomatica database, while data on CG and ownership structure are obtained directly fromannual reports submitted by companies to the different regulatory agencies8 and which areavailable in their Web site. In this way, we get information for 435 firms and a total of 1,740observations by the period from 2006 to 2009. The composition of the sample allows thecombination of time series and cross sections with an adequate opportunity to take advantagein the creation of a panel data, especially in the control of unobserved heterogeneity, i.e. theindividual characteristics of each entity that are not observable but affects the variables under 25
  26. 26. study (Arellano & Bover, 1991; Arellano, 1993; Himmelberg et al., 1999; Palia, 2001; Bricket al., 2005). Additionally, since at present is widely accepted the idea of using unbalancedpanels with total observations is discarded the option of analyzing balanced panels with fewercompanies, because it may be conditioned by the survival bias (Baltagi & Chang, 1994).3.2. Models and Variables Definition Because the intention of investigating the influence that CG mechanisms have on EM,measured by discretionary accruals, we regress the absolute value of discretionary accruals[Abs (DCA)it] on the variables of ownership structure, board of directors and control used inprevious literature, according to the following model:Abs(DAC)it= Β0 + β1 (Int_OW ) + β2 (OW _Con) + β3 (Fam_OW ) + β4 (Inst_OW ) + β5 (Board_SIZE) + β6 (Board_I D) + β7 (Board_ACT) + β8 (CEO_Dual) + β9 (GOV_Index) + β10 (Control) + ηi + λt + υit The unobserved heterogeneity is controlled in the two models through individual effectsof companies (ηi). Also, we included dummy variables to control the temporal effects (λit)and the error term (υit). As a proxy for internal property (Int_OWN) we use the proportion ofshares ≥ 1% owned by members of board of directors and managers of the firms; theownership concentration (OWN_Con) is measured by the proportion of shares owned by themajor shareholder of the company, because many firms in Latin America are directlycontrolled by one of the industrial or financial conglomerates that operate in the region(Lefort & Walker, 2005, 2007; Lopez & Saona, 2005; Cespedes et al., 2008); the familyownership (Fam_OWN) is measured by the proportion of shares held by family members, i.e.the percentage of capital that is directly or indirectly in their hands ≥ 5% and; the institutionalownership (Inst_OWN) through the proportion of shares held by institutional investors.Moreover, the board size (Board_SIZE) is measured by the total number of directors that 26
  27. 27. integrate the board of directors; the board independence (Board_IND) is measured by theproportion of external directors inside the board (external directors / total directors) and with adummy variable (Board_IND50) that takes the value of one when the board comprises with amajority of external directors; the board activity (Board_ACT) is measured by the number ofmeetings held during the year; the President-CEO duality (CEO_Dual) is measured through adummy variable that considers the value of 1 if there is duality of roles between the chairmanand CEO of the firm and, 0 otherwise. Finally, the government index (Gov_Index) thatmeasure the governability level of the country (control of corruption, rule of law andgovernment effectiveness). On the other hand, we control the effect of various factors through the inclusion ofvariables to our model, which have been used in previous studies and have been associatedwith EM and CG. Thus, the variable quality and reputation of the external auditor (Big_4)measured by a dummy variable that takes the value 1 if the company is audited by one of thebig four audit firms, 0 otherwise. In this way, several studies indicate that quality ofaccounting information will be linked to the prestige and quality of the external auditor(DeFond & Jiambalvo, 1991, 1994, DeFond, 1992; Teoh & Wong, 1993; Dechow et al.,1996; Lennox, 1999a, 1999b; Jara & Lopez, 2007), because most reputable auditors limit thepossibility of EM (Becker et al., 1998; Francis et al., 1999; Kim et al., 2003; Van Tendeloo &Vanstraelen, 2005) and therefore, the financial statements audited by these firms have greatercredibility (DeFond & Subramayan, 1998; Teoh et al., 1998). Another control variable is the firm size (Log_ASSET) measured by the natural logarithmof total assets at the end of year (Sanchez & Sierra, 2001; Navarro & Martinez, 2004),controlling with it the effects of company size on accounting choice. Authors generallyexpect, and often proves, a negative relationship between firm size and EM, given that inlarger companies are expected to have more sophisticated control systems, skilled advisers, 27
  28. 28. more negotiating power with the external auditor and are subject to increased monitoring byinvestors and analysts, so accounting fraud is less probable than in smaller firms, where themanagers of these companies have more opportunities to manipulate the information (Beasleyet al., 1999; Bartov et al., 2000; Reynolds & Francis, 2000; Sanchez & Sierra, 2001; Klein,2002; Richardson et al., 2002; Lee & Choi, 2002; Navarro & Martinez, 2004; Garcia Osma &Gill de Albornoz, 2005; Cahan & Zhang, 2006; Goodwin & Kent, 2006; Prior et al., 2008).Moreover, we includes the indebtedness level variable (Debt), calculated as the ratio of totaldebt and total assets. Thus, a high indebtedness is associated with the risk of excessiveleverage (Press & Weintrop, 1990), which motivates the EM to conceal inconvenientinformation and display a greater capacity to generate resources (DeFond & Jiambalvo, 1994;Sweeney, 1994; DeAngelo et al., 1994; Dechow et al., 1995; Krishnan et al., 1996; Frankel etal., 2002; Balsam et al., 2003). Additionally, following the study of Francis & Wang (2004) we include two controlvariables on firm performance. The first is the growth variable (GROWTH), measured interms of the relation of the difference in sales and sales of the previous period for firm i inyear t, which indicates that firms with high growth rates are more likely to use discretionaryaccruals (McNichols, 2000). The second is the variable (ROA), calculated by the ratiobetween earnings before extraordinaire, interest and taxes of year t and the total net assets atbeginning of year t, and positively related to the use of discretionary accruals. This suggeststhat managers are motivated to manipulate the results upward, i.e., increase the benefitsobtained with the intention to make the company more attractive (Kothari et al., 2005;Machuga & Teitel, 2007). Also, because a poor financial situation of the company could increase agency costs andencourage the management to manipulate the accounting numbers (Nurul et al., 2010; Sierraet al., 2010) we includes the control variable loss (Loss) which is measured through a dummy 28
  29. 29. variable that takes value of 1 if the company has had losses in the last two years and, 0otherwise. Finally, we consider the sector of activity variables (IND) and year (YEAR), beingthese important factors of measurement, because in sectors and specific years could have abetter results to identifying discretionary accruals (McNichols et al., 1988; Roychowdhury,2006).4. Analysis and Results4.1. Descriptive Analysis Table 1 displays the descriptive statistics and t values of discretionary accruals forestimated [Abs (DAC) it], showing that mean values of discretionary accruals are, in all cases,statistically different from zero. This does not allow us to reject the null hypothesis and,therefore, provides evidence that Latin American companies manipulate their results, eitherby increasing the benefit to denote a better and higher profitability of the company or, on thecontrary, reducing the benefit as fiscal strategy aims to pay fewer taxes and contributions. TABLE 1: Descriptive Statics of Discretionary Accruals [Abs (DAC)it] Estimations by Year Year N Mean Median Std. Dev. T Adjusted R2 DAC-2006 435 0.224 0.119 0.441 0.619 0.536 DAC-2007 435 0.278 0.112 0.704 -1.070 0.285 DAC-2008 435 0.198 0.122 0.293 1.567 0.118 DAC-2009 435 0.249 0.142 0.420 -0.243 0.808 Global 1,740 0.237 0.121 0.489 1.459 0.447 For its part, Table 2 shows the mean, median, standard deviation and the associated t-Value of the estimated coefficients of the absolute value of discretionary accruals [Abs(DAC)it] per country. In it, can be seen that the model significantly explained variations in thecoefficients of discretionary accruals, as its explanatory power shows Adjusted R2 values(significance level) above 40% for all the countries. 29
  30. 30. TABLE 2: Descriptive Statics of Discretionary Accruals [Abs (DAC)it] Estimations by Country Country N Mean Median Std. Dev. T Adjusted R2 Argentina 308 0.247 0.137 0.425 0.412 0.744 Brazil 480 0.293 0.152 0.481 1.749 0.476 Chile 532 0.236 0.101 0.651 1.852 0.542 Mexico 420 0.167 0.110 0.198 1.493 0.408 Global 1,740 0.237 0.121 0.489 1.459 0.447 Table 3 shows the main descriptive statistics of quantitative and dichotomous variables.Thus, with respect the board characteristics variables it can be seen that in the four countriesanalyzed, generally, companies boards meets on average 5 times a year. It can also be seenthat boards are composed with a mean of 11 members, of whom 38.5% are external directors,a fact that clearly indicates that the composition of this organ of government is a majority ofinternal members, thus demonstrating control and domain that have families on this governingbody (Santiago & Brown, 2009; Santiago et al., 2009). Our result contradicts therecommendations highlighted in previous studies that recommend an integration of boards bya majority of external directors (Hermelin & Weisbach, 2003; Sanchez & Guilarte, 2006,2008; Zattoni & Cuomo, 2010); and is in line with results of previous studies conducted inLatin America which document, likewise, a composition of boards by a majority of internaldirectors (Silveira et al., 2003; Schiehll & Santos, 2004; Lefort, 2005, Ferraz et al., 2011). Regarding the ownership structure, Table 3 shows that Mexican companies reveal to havea higher family engagement with the 37.1%, followed by Argentinean (35%), Chilean(26.2%) and Brazilian (24%) companies. The ownership concentration (major shareholder)reflects an average of 29.4% of the social capital of firms. In this way, Chilean companies arethose that revealed have a higher shareholding concentration with 32.2%, followed byBrazilian (29.3%), Mexican (28.6%) and Argentinean (27.5%) firms. Moreover, regarding tothe internal ownership (top management), it can be seen that manager and directors holds, onaverage, 6.1% of the social capital of companies. Thus, Brazilian firms are those that revealed 30
  31. 31. have a higher internal ownership with 7.3%, followed by Argentinean (7.1%), Chilean (6%)and Mexican (4.5%) companies. Finally, the institutional ownership indicates an averagevalue of 22.8% of social capital held by institutional investors. Thus, Brazilian companies arethose that have revealed a higher participation of institutional investors with 23.9%, followedby the Chilean (23.8%), Argentinean (21%) and Mexican (20.6%) firms. TABLE 3: Descriptive Statistics of Quantitative and Dichotomous Variables Observations by Country Variable Statistics Argentina Brazil Chile Mexico Globala) Quantitative Variables 1 N 308 480 532 420 1.740 Int_OWN Mean 0.071 0.073 0.060 0.045 0.061 Std. Dev. 0.046 0.048 0.047 0.038 0.046 Mean 0.275 0.293 0.322 0.286 0.294 OWN_Con Std. Dev. 0.106 0.102 0.118 0.101 0.107 Fam_OWN Mean 0.350 0.240 0.262 0.371 0.305 Std. Dev. 0.163 0.177 0.181 0.181 0.179 Inst_OWN Mean 0.210 0.239 0.238 0.206 0.228 Std. Dev. 0.145 0.136 0.135 0.135 0.137 Mean 11.49 11.38 11.54 11.47 11.47 Board_SIZE Std. Dev. 3.82 3.69 3.60 3.66 3.67 Mean 0.400 0.394 0.366 0.375 0.385 Board_IND Std. Dev 0.730 0.891 0.077 0.862 0.083 Mean 5.42 5.37 5.33 5.15 5.31 Board_ACT Std. Dev. 2.49 2.50 2.47 2.38 2.46 Mean 41.13 51.12 88.05 48.78 49.73 GOV_Index Std. Dev. 3.25 2.51 0.68 2.06 18.78 Mean 13.32 13.39 18.39 16.09 15.58 Log_ASSET Std. Dev. 1.91 1.75 2.32 1.70 2.93 Mean 0.396 0.504 0.280 0.227 0.350 Debt Std. Dev. 1.371 0.691 0.924 0.158 0.862 Mean 4.47 4.58 9.18 6.96 6.56 ROA Std. Dev. 1.84 1.67 2.36 1.70 2.79 Mean 0.236 0.531 0.103 0.124 0.249 GROWTH Std. Dev. 0.617 2.221 0.299 0.484 1.241 b) Dichotomous Variables 2 Variable Statistics Argentina Brazil Chile Mexico Global N 308 480 532 420 1.740Board_IND50 0 148 48.1% 240 50.0% 316 59.4% 224 53.3% 928 53.3% 1 160 51.9% 240 50.0% 216 40.6% 196 46.7% 812 46.7% 0 99 32.1% 193 40.2% 209 39.3% 172 40.9% 673 38.7% CEO_Dual 1 209 67.9% 287 59.8% 323 60.7% 248 59.1% 1,067 61.3% 0 144 46.8% 199 41.5% 180 33.8% 120 28.6% 643 36.9% Big_4 1 164 53.2% 281 58.5% 352 66.2% 300 71.4% 1,097 63.1% 31
  32. 32. TABLE 3: Descriptive Statistics of Quantitative and Dichotomous Variables Observations by Country Variable Statistics Argentina Brazil Chile Mexico Global 0 234 76.0% 373 77.7% 436 81.9% 319 75.9% 1,362 78.3% Loss 1 74 24.0% 107 22.3% 96 18.1% 101 24.1% 378 21.7%(1) Quantitative Variables: Int_OWN = Internal ownership, measured by the proportion of shares owned by managers and members of Boards (≥ 1%); OWN_Con = Ownership Concentration, measured by the ratio of shares held by the major shareholder of the company (≥ 5%); Fam_OWN = Family Ownership, measured by the proportion of shares held by family members (≥ 5%), as a percentage of capital that is directly or indirectly in his possession; Inst_OWN= Institutional Ownership, measured by the proportion of shares held by institutional investors; Board_SIZE= Size of boards of directors, measured by the total number of members of Boards; Board_IND= independence of the Board, measured by the proportion of independent members (independent directors / total directors); Board_ACT= Activity of Boards, measured by the number of meetings; GOV_Index= The degree of law enforcement of each country analyzed, taken from the research project “Worldwide Governance Indicators” (WGI) proposed by Kaufmann et al., (2010); Log_ASSET= Firm size, measured by the natural logarithm of total assets of the companies; Debt= Level of indebtedness, measured by the quotient resulting from gross debt to total assets, ROA= Economic Return, measured by the ratio of the relationship between the result before special items, interest and taxes of year t and the total net assets at the beginning of year t; GROWTH= Growth of the Companies, calculated in terms of the ratio of the difference in sales and sales of the previous period of firm i in year t.(2) Dichotomous Variables: Boad_IND50= Measured through a dummy variable that takes value of 1 if boards has a majority of independent directors and, 0 otherwise; CEO_Dual= Measured through a dummy variable that considers the value of 1 if there is duality of roles between the chairman and CEO of the companies and, 0 otherwise; Big_4= Measured by a dummy variable that takes the value 1 if the firms are audited by one of the big four firms, 0 otherwise; Loss= Measured through a dummy variable that takes value of 1 if the companies have had losses in the last two years and, 0 otherwise.4.2. Regression Results After analyzing the variables descriptively, it is necessary to apply tests to help measurethe linear relationship between the dependent variable “absolute value of discretionaryaccruals [Abs (DAC)it]” and the independent and control variables of the firms. Theexplanatory development is based mainly, on determining the level of influence that CGmechanisms has on discretionary accruals. In order to determine which model is better suitedto our data, either the fixed effects based on groups estimator or random effects based ongeneralized least squares (GLS) we perform the Hausman test (1978), which determineswhether the differences are systematic and significant between the two models. In all cases,the result of this test does not reject the null hypothesis of no systematic differences betweenthe regressors’ and unobserved heterogeneity, therefore assuming the random effects as themost appropriate for our analysis. Thus, in Table 4 the model 1 shows the results obtained from the linear regression of the 32
  33. 33. panel data, the absolute value of discretionary accruals [Abs (ADD) it] on the variables ofownership structure, board of directors and control. With regard to the internal ownership, isobserved that the stake held by managers and directors in Latin American firms have asignificant negative relationship at level of 1% with the absolute value of discretionaryaccruals, suggesting that the low insider’s ownership reduce the EM practices, i.e. reduce theuse of discretionary accruals (convergence of interest’s hypothesis). Our result is in line withthose results obtained by Machuga & Teitel (2009) with a sample of Mexican firms, whoshows that firms with a fewer internal ownership shows a greater earnings quality comparedto those firms that do not have managerial ownership, i.e. shows less manipulative practicesby managers, because the implementation of good CG practices contained in Codes of BestPractices. In similar terms, other studies such as Morck et al. (1988) in Canada, Wartfield etal. (1995) in the U.S., Yeo et al. (2002) in Singapore and Sanchez-Ballesta & Garcia-Meca(2007) in Spain, also point out that the informativeness of accounting results increases withlow levels of internal ownership, while for a high levels, the internal ownership is notsufficient as alignment interest’s mechanism. In relation to the ownership concentration, shows a significant negative relationship atlevel of 1% with the absolute value of discretionary accruals, suggesting that when the mainshareholders have a high percentage of ownership or when a conglomerate directly controlsthe firm, the absolute value of discretionary accruals is reduced, due to the efficientmonitoring hypothesis indicated by the Agency theory (Jensen & Meckling, 1976; Fama,1980; Fama & Jensen, 1983). Thus, Fernandez et al. (1998), Yeo et al. (2002), Gabrielsen etal. (2002) and De Bos & Donker (2004) results are also consistent with the monitoring role ofexternal block holders, and their strong positive effects on earnings informativeness. Moreover, respect to the board size, this indicates a positive relationship significant atlevel of 5%, showing our result that the greater board size the level of monitoring over the 33
  34. 34. management team decreases due to the existence of problems of communication andcoordination that increases the use of discretionary accruals, in line with previous studies asXie et al. (2003), Thomsen (2008) and Santiago & Brown (2009). Also, respect to the board independence, it shows a weak negative relationship significantat level of 10%. Our result contrasts with the prominent role that literature, theoretical andempirical, assigned to this attribute of the board to safeguard the quality and transparency ofresults but, for the case of Latin American countries analyzed, does not seem to be soeffective. In this regard, Price et al. (2006, 2007), Teitel & Machuga (2008), Chong et al.(2009) and Davila & Watkins (2009) in Mexico; Silveira et al. (2003), Schiehll & Santos(2004) and Ferraz et al. (2011) in Brazil; Majluf et al. (1998), Iglesias (1999), Lefort &Walker (2000, 2005) in Chile; suggests that this is due to boards are mainly composed ofmajor shareholders and managers of the companies, having external directors a very limitedparticipation which facilitates the EM and the managerial discretion. It is probably that thisevidence is derived, as stated by Yermack (2004)9, by the presence of grey directors, lack ofrotation of the directors or the two causes simultaneously. Regarding the grey directors, they are those that maintain some kind of family orprofessional relationship (present or past) with the company or its top management, the factthat in the annual reports of CG are designated as external and almost in no way disclose anypossible conflicts of interest, could severely limit the board independence. Regards thesecond, its slow or almost non-existent rotation makes them permanent external, and thus thereport of the First Latin American Corporate Governance Survey, conducted by PriceWaterhouse Cooper (PWC) in 2010 (published in 2011) indicates that on average only12,35% of companies listed on Latin American stock markets put time limits for externaldirectors. In short, according to Monterrey & Sanchez (2008), both groups might fall into thecategory that Eguidazu (1999) calls “the label”, in which independence is an appearance and 34
  35. 35. not an attitude, because the absence of sufficient distance from the management of thecompany could concentrate in fact the power inside the board, thereby facilitating EM (GarciaOsma & Gill de Albornoz, 2005). In addition, the model 1 shows that board activity results to have a negative relationshipsignificant at level of 5% showing that the greater number of meetings held by the boardsdecreases the use of discretionary accruals, i.e., the higher board activity reduces the EM. Wedo not find any statistically significant relationship between family ownership (Fam_OWN),institutional ownership (Inst_OWN), CEO duality (CEO_Dual) and the absolute value ofdiscretional accruals. On the other hand, there is a significant negative relationship at level of 1% betweenGovernment Index (GOV_Index) and discretionary accruals, suggesting that when a countryimplements controls aimed to reduce the corruption, to strengthen the rule of law or toimprove the effectiveness of government seems to influence on EM negatively, i.e., it showsan increase on the quality and transparency of the financial information issued by companies,showing a reduction of discretionary accruals (La Porta et al., 1998, 2002, 2006; Leuz et al.,2003; Bushman et al., 2004; Ball & Shivakumar, 2005). Finally, in the remaining control variables it can be seen that they maintain their level ofsignificance and expected sign: A significant negative relationship at level of 5% betweenfirm size and discretionary accruals, because the largest companies are subjected to a greatermonitoring than smaller firms (Garcia & Gill, 2005; Cahan & Zhang, 2006; Goodwin & Kent,2006; Prior et al., 2008); a significant positive relationship at level of 1% betweendiscretionary accruals and level of debt, due to the companies with more leverage usedmanipulative practices to exhibit a greater capacity to generate resources (Krishnan et al.,1996; Frankel et al., 2002; Balsam et al., 2003); a significant positive relationship at level of1% between economic profitability and discretionary accruals, suggesting that managers are 35
  36. 36. motivated to manipulate the results obtained with the intention to make the company moreattractive (Kothari et al., 2005; Machuga & Teitel, 2007); a significant positive relationship atlevel of 1% between growth and discretionary accruals, indicating that companies thatobserved a high growth are more likely to use a discretionary accruals adjustments(McNichols, 2000), because they experiment better opportunities to attract investment(Young, 1999). Additionally, in column 11 of Table 4 we use a different proxy for board independence,replacing the proportion of external directors on boards (Board_IND) by a dummy variablethat takes the value of 1 if board has a majority of external directors, and 0 otherwise(Board_IND50). The conclusions are the same than model 1, i.e., the board independence alsoshows a weak negative relationship significant at level of 10% with the dependent variable[Abs (ADD) it]. 36
  37. 37. TABLE 4: Discretionary Accrual Regressions on Corporate Governance and Control Variables Random Effects Estimation (GLS)Model 1: Abs(ADD)it= β0 + β1(Int_OW ) + β2(OW _Con) + β3(Fam_OW ) + β4(Inst_OW ) + β5(Board_SIZE) + β6(Board_I D) + β7(Board_ACT) + β8(CEO_Dual) + β9(Big_4) + β10(Log_ASSET) + β11(Debt) + β12(ROA) + β13(GROWTH) + β14(Loss) + β15(GOV_Index) +ηi +λt +υit Expected Model 1 Model 2 Variable Sign (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) -0.433** -0.481*** -0.491*** Int_OWN ¿? (-1.84) (-2.01) (-2.05) -2.340** -2.174*** -2.354*** OWN_Con ¿? (-1.56) (-1.36) (-1.52) -0.089 -0.055 -0.079 Fam_OWN ¿? (-0.91) (-0.55) (-0.80) -0.073 -0.080 -0.083 Inst_OWN ¿? (-0.92) (-1.00) (-1.04) 0.032* 0.053** 0.035** Board_SIZE ¿? (0.73) (1.19) (0.80) -3.201* -3.513* Board_IND - (-1.67) (-1.79) -0.024* -0.027* Board_IND50 - (-1.10) (-1.22) -0.097* -0.127** -0.114* Board_ACT - (-1.61) (-2.07) (-1.87) 0.009 0.005 0.007 CEO_Dual + (0.41) (0.21) (0.29) -0.167*** -0.169*** -0.174*** -0.173*** -0.173*** -0.171*** -0.178*** -0.173*** -0.174*** -0.156*** -0.163*** GOV_Index - (-3.51) (-3.54) (-3.66) (-3.65) (-3.65) (-3.60) (-3.73) (-3.65) (-3.67) (-3.27) (-3.41) -0.027 -0.027 -0.028 -0.028 -0.027 -0.026 -0.027 -0.029 -0.027 -0.029 -0.029 Big_4 - (-1.18) (-1.20) (-1.23) (-1.23) (-1.19) (-1.15) (-1.17) (-1.27) (-1.17) (-1.26) (-1.28) -0.030** -0.032** -0.030** -0.028** -0.028** -0.022** -0.031** -0.033** -0.029** -0.024** -0.035** Log_ASSET - (-1.75) (-1.87) (-1.80) (-1.65) (-1.66) (-1.24) (-1.82) (-1.93) (-1.75) (-1.37) (-2.05) 0.314*** 0.315*** 0.314*** 0.314*** 0.316*** 0.317*** 0.315*** 0.315*** 0.315*** 0.316*** 0.314*** Debt + (2.37) (2.38) (2.37) (2.37) (2.38) (2.39) (2.38) (2.38) (2.38) (2.37) (2.36) 0.045*** 0.045*** 0.047*** 0.044*** 0.046*** 0.043** 0.047*** 0.047*** 0.046*** 0.043*** 0.047*** ROA + (2.58) (2.62) (2.71) (2.54) (2.65) (2.49) (2.72) (2.71) (2.65) (2.45) (2.70) 37
  38. 38. TABLE 4: Discretionary Accrual Regressions on Corporate Governance and Control Variables Random Effects Estimation (GLS)Model 1: Abs(ADD)it= β0 + β1(Int_OW ) + β2(OW _Con) + β3(Fam_OW ) + β4(Inst_OW ) + β5(Board_SIZE) + β6(Board_I D) + β7(Board_ACT) + β8(CEO_Dual) + β9(Big_4) + β10(Log_ASSET) + β11(Debt) + β12(ROA) + β13(GROWTH) + β14(Loss) + β15(GOV_Index) +ηi +λt +υit Expected Model 1 Model 2 Variable Sign (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) 0.031*** 0.031*** 0.031*** 0.031*** 0.031*** 0.031*** 0.031*** 0.030*** 0.031*** 0.029*** 0.029*** GROWTH + (3.49) (3.48) (3.48) (3.52) (3.49) (3.50) (3.50) (3.43) (3.50) (3.27) (3.28) 0.001 0.003 0.005 0.005 0.005 0.005 0.005 0.008 0.004 0.006 0.006 Loss + (0.04) (0.10) (0.19) (0.19) (0.18) (0.18) (0.19) (0.29) (0.13) (0.22) (0.20) Significance 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Adjusted R2 0.2729 0.2726 0.2719 0.2719 0.2717 0.2727 0.2720 0.2726 0.2716 0.2777 0.2770 Number of Observations 1.740 1.740 1.740 1.740 1.740 1.740 1.740 1.740 1.740 1.740 1.740(***) Significant at level 1%; (**) Significant at level 5%; (*) Significant at level 10%.Note 1. The model includes industry sectors and time controls, but they are not reported.Note 2. Z statistics in parentheses.Quantitative Variables: Int_OWN = Internal ownership, measured by the proportion of shares owned by managers and members of Boards (≥ 1%); OWN_Con = OwnershipConcentration, measured by the ratio of shares held by the major shareholder of the company (≥ 5%); Fam_OWN = Family Ownership, measured by the proportion of shares held byfamily members (≥ 5%), as a percentage of capital that is directly or indirectly in his possession; Inst_OWN= Institutional Ownership, measured by the proportion of shares held byinstitutional investors; Board_SIZE= Size of boards of directors, measured by the total number of members of Boards; Board_IND= independence of the Board, measured by theproportion of independent members (independent directors / total directors); Board_ACT= Activity of Boards, measured by the number of meetings; Log_ASSET= Firm size, measured bythe natural logarithm of total assets of the companies; Debt= Level of indebtedness, measured by the quotient resulting from gross debt to total assets, ROA= Economic Return, measuredby the ratio of the relationship between the result before special items, interest and taxes of year t and the total net assets at the beginning of year t; GROWTH= Growth of the Companies,calculated in terms of the ratio of the difference in sales and sales of the previous period of firm i in year t; GOV_Index= The degree of law enforcement of each country analyzed, takenfrom the research project “Worldwide Governance Indicators” (WGI) proposed by Kaufmann et al., (2010).Dichotomous Variables: Boad_IND50= Measured through a dummy variable that takes value of 1 if boards has a majority of independent directors and, 0 otherwise; CEO_Dual=Measured through a dummy variable that considers the value of 1 if there is duality of roles between the chairman and CEO of the companies and, 0 otherwise; Big_4= Measured by adummy variable that takes the value 1 if the firms are audited by one of the big four firms, 0 otherwise; Loss= Measured through a dummy variable that takes value of 1 if the companieshave had losses in the last two years and, 0 otherwise. 38

×