The first corporate governance code was enacted in Greece in 1999 (Committee on Corporate Governance, 1999), following the crash of the capital market, in an attempt of the Capital Market Commission to restore public confidence. This early code took the form of a white paper depicting the intent of the legislator to provide time to the companies to comply with the spirit and not merely with the letter of the regulation. A set of these corporate governance practices became obligatory for the listed companies three years later with the issuance of Law 3016/2002. The provisions of Law 3016/2002 are complemented by the consolidated Law 2190/1920 for the limited liability companies with regard to the responsibilities of the board members and the chief executive officer.

Research on corporate tax evasion must take into account the (divergent) incentives of the firms when reporting for tax and financial accounting purposes. When tax and accounting earnings are highly aligned, the underreporting of taxable income will also lead to reporting lower profits to the shareholders. On the other hand, when different accounting regimes apply for taxation and financial reporting purposes, tax aggressive behavior does not affect the reported earnings_[1]. Greek public companies operated in an accounting environment characterized by a high level of book-tax conformity up to 2005 since Greek GAAP were applicable both for  financial  reporting  and  tax purposes. In 2005, the adoption of International Financial Reporting Standards (I.F.R.S.) became obligatory for the Greek public companies. After 2005, the accounting regime can be characterized as a two-book one since the financial statements are published according to IFRS whereas Greek GAAP still applies for tax purposes.

The aim of the present paper is to examine the association between corporate governance practices and the tax behaviour of the Greek public when they operated in an accounting environment characterized by a high level   of   book –   tax   conformity.  Specifically,  this research focus on the period 2000-2004, since in 2005 Greece adopted International Financial Reporting Standards (IFRS) and moved from a one-book to a two-book system. As noted above, the separation between tax and financial reporting alters managers’ reporting incentives. The impact of deferred taxation on corporate tax decisions is left for future research.

The rest of the paper is organized as follows. First is a presentation of the literature review, followed by a description of the methodology as well as sample and data collection. Thereafter, the univariate and multivariate results is reported, and then conclusion.[1] However, the companies face the question of whether to report lower accounting earnings or disclose the book-tax difference (Hanlon and Heitzman, 2010).

One of the most crucial roles of the board of directors is to mitigate the agency problems that arise in large corporations due to the separation of ownership and management. The effectiveness of the board of directors in deterring earnings management and other cases of accounting fraud has been examined in a number of studies. Fama (1980) and Fama and Jensen (1983) contend that corporate boards should   include top-level   managers   who   have a thorough   knowledge of the organization that enables them to take decisions that will potentially maximize its wealth. At the same time, the board should include outside members in order to monitor top-management and protect minority interests. The participation of outside members in the board has been found to be effective in constraining earnings manipulation by a number of studies (Beasley, 1996; Dechow et al., 1996; Fanning and Cogger, 1998; Beasley et al., 1999; Beasley et al., 2000; Klein, 2002; Xie et al., 2003; Saksera, 2003; Carcello and Nagy, 2004). Osma and Noguer (2007) have found that outside members are able to prevent earnings manipulation only in firms that have a nomination committee principally made up of institutional directors.

According to Jensen (1993), the independent members of the board are more effective in their monitoring role when they also hold shares of the company. The research by Beasley (1996) has come to support this notion as he has found that the more shares the independent members hold the less possible it is that the company issues falsified financial statements. Beasley (1996) has also shown that independent outside members with long tenures on the board exercise their duties more efficiently since they have a better understanding of the company and more negotiating power in the board.  However, Xie et al. (2003) have reached contradictory conclusions since outside board members with longer tenures turn out to be less effective monitors. The authors assume that this may happen because the independent members have been co-opted by management. There are also opposite opinions regarding the impact of multiple  directorships. Fama (1980) argues that serving on more boards  is  an  indication  of  great skills and competence while Morck et al. (1988) believe that serving on more boards is very time-consuming and thus prevents directors from focusing on each of the companies. The studies of Beasley (1996) and Crutchley et al. (2007) corroborate Mock’s et al. (1988) argument. The financial sophistication of the board members has been examined by Xie et al. (2003) and has been found to play an important role in constraining the propensity of managers to engage in earnings management.

The effect of the board size has also been examined in a number of studies which have reached contradictory conclusions. Beasley (1996) has found that the companies that have committed accounting fraud have larger boards, reinforcing the opinion of Jensen (1993) that a large board is not very functional due to bureaucratic problems. On the other hand, Xie et al. (2003) found that companies with larger boards are less likely to engage in earnings management, which may be due to the broader experience of the board members and the increased likelihood that there are more members with a financial background. The effect of the board size on the credibility of the financial statements has also been examined by Dechow et al. (1996) and Carcello and Naggy (2004) but the results were not statistically significant. The number of the board meetings (as a proxy for the level of commitment of the board members) has been found by Xie at al. (2003) to be negatively associated with the incidence of earnings management. However, similar research conducted by Erickson et al. (2006) did not yield any significant results.

A good corporate governance practice is considered to be the separation of the roles of the chairman of the board and the chief executive officer (CEO).When the same individual holds both positions, it is much more difficult for the board of directors to evaluate CEO’s performance and at the same time it is much easier for the CEO to control the board in order to serve his own personal interests (Jensen, 1993). The majority of the studies (Dechow et al., 1996; Abbott et al., 2000; Saksera, 2003; Carcello and Nagy, 2004; Erickson et al., 2006) show that fraudulent financial reporting occurs in organizations where the chairman of the board is also the CEO. However, this assumption was not confirmed in the studies of Beasley (1996) and Xie et al. (2003). Dechow et al. (1996) also found that the role of the board weakens when the chairman of the board is also the firm’s founder, reinforcing Jensen’s theory (1993) that the owner in this case takes decisions without reporting to the board. Beasley (1996), Abbott et al. (2000) and Saksera (2003) also examined whether accounting fraud is associated with the tenure of the CEO, assuming that as his tenure increases, his power increases which could decrease the monitoring provided by the board. However, they did not find any significant results.

Family   relationships   among   board    members   are considered to facilitate the management of earnings (Fanning and Cogger, 1998). Beasley et al. (1999) found close family relationships in the 40% of the fraud cases that they examined. On the other hand, the participation of institutional directors in the board seems to protect minority shareholder rights. Dechow et al. (1996) and Osma and Noguer (2007) have found that financial statement quality is greatly enhanced by the addition of institutional directors to the board. However, similar research conducted by Beasley (1996), Abbott et al. (2000) and Klein (2002) did not give any significant results.

Apart from the role of the board of directors in ensuring the quality of financial reporting, the role of the audit committee has also been under scrutiny. Abbott et al. (2000), Beasley et al. (2000), Klein (2002), Bédard et al. (2004) and Crutchley et al. (2007) have found that the existence of an independent audit committee is negatively related to the likelihood that a firm will engage in earnings management. More specifically, Klein (2002) has found that the audit committee is effective when at least 50% of its members are independent whereas Bédard et al. (2004) suggest that there is a significant reduction in the likelihood of aggressive earnings management  only when the audit committee solely consists of independent members. These findings come in support of the Sarbanes-Oxley Act which requires the independence of all the members of the audit committee. Moreover, audit committees that have members with corporate or financial backgrounds (Beasley et al., 1999; Xie et al., 2003; Bédard et al., 2004) as well as audit committees that meet frequently (Beasley et al,. 1999, 2000; Xie et al,. 2003) have also been found to be important factors in constraining the propensity of managers to engage in earnings management.

Study sample

The study sample consists of the public companies listed on Athens Stock Exchange (A.S.E.) during the period 2000-2004. Data regarding the corporate governance practices have been manually collected from company annual reports. Data regarding the extent of tax evasion are available since the Greek public companies are obliged to publish the outcomes of the tax audits on the Athens Stock Exchange website, on their corporate website and in any new prospectus that they release (Athens Exchange Rulebook, Article 275). The above-mentioned announcements and prospectuses are used as the main source of information for estimating the extent of tax evasion. Information regarding the outcome of the tax audits was obtained for 96 firms in total and for 225 fiscal years. Specifically, there are 39 observations for the year 2000, 50 observations for 2001, 52 observations for 2002, 41 observations for 2003 and 43 observations for 2004. The mean rate of tax evasion has been estimated at 13.58% for the year 2000, at 22.70% for 2001, at 20.79% for 2002, at 17.25% for 2003 and 20.32% for 2004. Respectively, the mean tax gap has been estimated at €295,237 for 2000, at €350,672 for  2001,  at  €327,719  for  2002,  at  €275,871  for  2003  and  at €247,048 for 2004.

Similar to univariate analysis, each year is examined separately since law obligations have changed over the five-year period. The amount of total assets is used as a proxy for the size of the firms since it cannot be as easily manipulated as the measure of earnings or profits. Regression results are not reported here for brevity but are available upon request.

The extent of tax evasion, measured as the amount of the tax gap, continues to appear significantly lower in the companies where the owner is also the chairman of the 2000. 2001 and 2002).

In accordance to Mann-Whitney board (that is, significant associations found for the years and Kolmogorov-Smirnov tests for the years 2001 and 2002. the rate of tax evasion is found to be lower when board members are compensated through profit distribution. Moreover a strong negative relationship is found between the extent of the tax gap in years 2001 and 2002 and the distribution of profits to the board members. There is limited evidence (only for year 2001) that tax evasion decreases as the percentage of stock held by board members increases. There is  also  limited  evidence  (that is, only  for  year  2000)  that  tax  evasion  is  higher  in companies where the board members are also employees of the company. Both results agree with the outcomes of univariate analysis. Lastly, the extent  of  tax evasion is found to be higher in companies that the board members serve on more boards. As previously, “multiple directorships” is first estimated as the number of board members that serve on other boards. second as the total number of the board positions held by board members and third as the percentage of the board members that serve on other boards.

This paper has examined the association between corporate governance practices and the extent of tax evasion. Although univariate analysis has shown that the amount of the tax gap is significantly lower in companies where the board of director is also the chief executive officer. the results were not corroborated when controlling for firm size. This suggests that larger companies tend to separate the two roles. This is consistent with previous studies (Xanthakis et al., 2004; Florou and Galarniotis, 2007; Grant Thornton, 2008) suggesting that voluntary adoption of corporate governance practices among Greek public companies increases with firm size.

Tax evasion has been found to be significantly lower in the companies where the chairman of the board is also the owner of  the  company. In addition. there is a  strong negative association between tax evasion and a) the percentage of stock held by the owner and its family members and b) the percentage of stock held by board members.

The results contradict international codes on corporate governance practices that call for greater independence of the board. In the case of family-controlled public companies the owner acts in order to ensure the long-term prosperity of the firm and does not behave in an opportunistic manner pursuing shortsighted goals and personal benefits. On this account. Robins (2006) states that “the particular concerns of owner managed companies provide just one illustration of problems likely to arise from across-the-board application of strict. detailed rules”.

The remuneration of the board members through the distribution of profits has been found to significantly decrease the evasion of taxes. This is according to expectations as the tax and accounting earnings during the period of study aligned. A decrease in the reported earnings would also decrease or even prevent the distribution of profits to the board members. Lastly, it has been found that tax evasion is higher when the board members also hold positions on other boards. This may indicate the inability of the directors to manage successfully multiple corporations. Another plausible explanation is that individuals who participate in multiple boards do so in order to gain personal benefits by recording personal expenses in company’s accounts. which later on are rejected by the tax authority.

The present paper focuses on an accounting period characterized by a high level of book – tax conformity. It is of great interest to examine how managers balance the tax incentives to external reporting incentives after the application of International Financial Reporting Standards. This is left for future research.

The authors have not declared any conflict of interests.