Minority Shareholders’ Rights and Audit Quality: Empirical Evidence from Turkey

The analysis of corporate governance systems has attracted increasing attention in recent years. Many studies have analysed efficient monitoring and disciplining mechanisms that can reduce agency costs associated with the separation of ownership and control. Since the separation of ownership and control is achieved in various ways in different corporate governance systems, the emerging conflicts of interest vary in nature (Mohammed, 2018). Most of the existing studies on corporate finance have focused on public firms with a large number of dispersed shareholders and entrenched managers who control the company. The agency problem exists because managers, who are not under shareholders’ control, can pursue their own goals. Prominent among the examples of managerial discretion are the pursuit of growth and overinvestment (Fama & Jensen, 1983). A possible way to correct this is to have a less dispersed ownership structure. Shareholders with a large stake in the company have greater incentives to monitor and take corrective action, because they partially internalize the benefits from their monitoring effort. Chandren, Ahmad, and Ali (2015) report empirical evidence on the monitoring role of large shareholders. This type of agency problem is mostly prevalent in the US and the UK, where the corporate governance system is characterized by a large number of listed firms with dispersed ownership, a liquid capital market facilitating trade of ownership and control rights and very few inter- corporate shareholdings (Mohammed & Saeed, 2018). Franks and Mayer (1997) call such a market-oriented system an outsider system.

In most European countries, share ownership is much more concentrated than in the US. Most firms are not listed on stock exchanges and even when they are, a single large shareholder or a small group of shareholders retains a controlling share in the firm (Mustafa, Che-Ahmad, & Chandren, 2018). Since the concentrated ownership structure makes firms impervious to takeovers, the controlling stake commonly stays with the founder of the company and his family, even when the company is large and publicly listed. The controlling shareholder generally takes active interest in running the firm, by appointing the management and directly taking executive positions. In this situation, the conflict of interest arises mainly between the controlling shareholders and the minority shareholders, instead of the salaried managers and the dispersed shareholders in the Arab- Saxon countries. This conflict can arise in several ways. Rent extraction in the form of diversion of company earnings to the advantage of the controlling shareholders or the use of the company’s assets to favour other firms owned by them, are two examples that readily come to mind. In the terminology of Franks and Mayer (1997), these features are shared by an insider corporate governance system, which they characterize has having few listed firms, large number of substantial share stakes and large inter-corporate shareholdings. While outside investors can and are encouraged to participate in equity returns through the stock market, they cannot exert much control. Japan, Germany, Italy, Austria and Belgium are prominent examples of this type of corporate governance system. Issues related to corporate governance in the US and Europe is relatively well documented. However, similar studies are very rare for developing countries. During the 2008, many developing countries reformed their financial system. A major new element in the development of the financial system was the establishment and fast expansion of stock markets. Singh (1994) illustrates that large firms in developing countries depend heavily on external funds and new share issues to finance their growth. This research underlines the importance of equity finance as in the 1980s, 25% to 40% of corporate growth of net assets was financed by new share issues. Given the perceived relationship between economic development and corporate financing patterns, we feel that corporate governance has received very little attention in developing countries, particularly Turkey. In 2012, Turkey made substantial amendments to corporate governance mechanisms in order to improve financial reporting quality and to subsequently integrate the Turkish market with that of Europe (Mustafa, Che Ahmad, & Chandren, 2017). Therefore, it is interesting to analyse the effects of 2012 reforms on clients’ demand for high quality audit. The paper describes the Turkish case by studying the ownership data of the top 100 firms listed on the Borsa Istanbul (BIST). The study demonstrates that the Turkish system shows strong similarities to the insider system. The Turkish traded firms exhibit highly concentrated and centralized ownership structures. Families, directly or indirectly, own more than 75% of all firms and have the majority control (Mustafa et al., 2017). The separation of ownership and control is mainly achieved through pyramidal or complex ownership structures and by building coalitions with other families or foreign firms (Mustafa & Che-Ahmad, 2017). In addition, an active market for corporate control does not exist, given the limited openness and concentrated ownership of the typical traded firm. It is almost impossible to acquire a traded firm without the prior willingness of the controlling owner to sell. There are also no signs that a market for large stakes operates in a way that disciplines poor performance.

The distinctive features of the Turkish case are its financial system and the presence of business groups (BGs). Almost every private bank is under the control of families who typically control a large number of other financial and industrial firms (Che-Ahmad & Mustafa, 2017). Hence, the monitoring function of banks works in a way that reinforces the interests of family owners. There are substantial inter-corporate shareholdings organized around holding firms, similar to those observed in Korea, Mexico and Singapore (Granovetter, 1995).

In the next section, the authors lay out a review of the relevant studies, followed by a brief description of the rules concerning ownership disclosure and the rights of shareholders. Section 1.5 displays the hypothesis development for the study. The research design is discussed in section 1.6. Section 1.7 provides the discussion and results of the findings. Finally, section 1.8 displays the conclusion of the study.

Ownership structure is considered as a significant mechanism of corporate governance. Concentrated ownership refers to the degree of distribution of power between agent and principle. Large shareholders have the ability to directly monitor management actions (Desender et al., 2013; Jensen & Meckling, 1976). Academic reports have shown that high levels of concentrated ownership influence shareholders’ motivation, tendency and the ability to monitor managers, and this may lead to minimizing agency cost (Desender et al., 2013).

Ownership structure is considered as an essential driver of agency problems, particularly that present between insider controllers and outside investors; it has a significant influence on the value of the firm (Lemmon & Lins, 2003). For example, outside investors’ resources may be utilized in a proper way to maximize inside controllers' interests or may be invested in unprofitable projects that give extra benefit for the insider-controllers'. Consequently, in both cases, this issue impacts negatively on the firm’s value (Amran & Che-Ahmad, 2010).

It has been proposed that ownership structure has a significant influence on the extent of the audit process and scope. Chan et al. (1993) find that audit is an important instrument to oversee management behavior, particularly for firms with widely dispersed ownership. Furthermore, the shareholders’ direct monitoring process becomes more challenging with widely dispersed ownership, because it is difficult to coordinate their monitoring efforts, as opposed to concentrated ownership (Laallam, Alom, & Mohamad, 2017). Thus, dispersed ownership is more likely to rely on the audit as an instrument of corporate governance to monitor management activities. In contrast, Jensen and Meckling (1976) suggest that managers get less opportunity to maximize their interests at the expense of shareholders with concentrated ownership. Therefore, firms with concentrated ownership are less likely to demand more extensive (costly) audit function.

Turkish firms are categorized as having a highly concentrated ownership structure. The pyramidal ownership structure is the most dominant structure of Turkish firms listed on the BIST, in the form of family-controlled firms. Groups of firms arrange some BGs and a family or group of families controls these firms. One of the most important reasons why firms adopt this kind of ownership is to support their capital through debt and retain their control rights simultaneously (Morck & Yeung, 2003). Also, the pyramidal ownership structure is used in order to control additional firms and contribute less stakes of capital (Akdogan & Boyacioglu, 2010).

Gürsoy and Aydoğan (2002) examined the effects of ownership structure on both the firm’s performance and risk taking of Turkish firms listed on the BIST. They define ownership structure in terms of two dimensions, i.e., concentrated ownership and mixed ownership. Concentrated ownership refers to shared stakes owned by majority shareholders, such as a certain number of individuals, institutions and families; whereas mixed ownership refers to the identity of the majority shareholders. Yurtoglu (2003) finds that the Turkish business environment shares weak features of a corporate governance regime, particularly those related to ownership structure, such as concentrated ownership, family ownership and wedge (deviation between control rights and cash flow rights) that are used through pyramidal BGs and dual-class shares. Wedge is used as a method to expropriate minority shareholders’ property rights to the interests of majority shareholders (La Porta, Lopez-de- Silanes, Shleifer, & Vishny, 1999).

The Capital Market Board (CMB) requires all investors, individuals, legal entities and coalitions, acting in concert, to disclose their share stakes when they acquire 10% or more of the total voting rights or of the total shares outstanding; or when the fraction falls below 10% of the total voting rights or of the total shares outstanding. There is a compulsory disclosure requirement when the total purchases or sales of the outstanding shares of a traded company by one insider (Chairman, members of the board of directors, general manager, his or her deputies or those shareholders who own 10% or more of the total number of shares outstanding) reach 1% of the total number of shares outstanding. The investors must notify the CMB and BIST of the commencement of the sale of shares in circulation (i.e., not freely floated) on the exchange. The disclosed information should constitute information on stocks to be sold, its nominal amount and the time period of sale.

The rights attached to shares of a firm give investors the power to extract from managers the returns on their investment. These rights become critical as a protection against managers or majority owners who act in their own interests. Because shareholders exercise their power by voting for directors, La Porta, Shleifer, and Vishny (1998) focus on voting procedures in evaluating shareholders’ rights. A main finding of La Porta et al. (1998) is that, in a cross-country comparison, shareholder protection measures are associated with lower concentration of ownership, suggesting that concentrated ownership is a response to poor investor protection.

To explore this possibility, we begin by looking at voting rights attached to shares and rights that support the voting mechanism against interference by the insiders. Investors are better protected when dividend rights and voting rights are closely linked. There are at least three ways around the one-share-one-vote principle in Turkey. First, firms can issue shares with different cash-flow and collateral rights in liquidation. Non-voting shares and shares that assign an arbitrarily high number of votes to one class of shares outstanding can be issued. The second way in which owners can reduce their ownership below their control rights is to organize the ownership structure of the firm in a pyramid. Finally, the owners can enforce their control via cross-shareholdings, i.e., having the firm own shares through its corporateshareholders.

The next set of rights is referred to as anti-director rights (La Porta et al., 1998). These rights measure how strongly the legal system favors minority shareholders against managers or dominant shareholders in the corporate decision-making process as follows:

1-General meetings are mandatory for firms. A shareholder cannot attend and vote unless he or she has lodged his or her shares with the company one week before the meeting. This prevents shareholders from selling their shares for several days around the time of the meeting and keeps shareholders who do not bother to go through this exercise from voting.

2-Shareholders must show up in-person or send an authorized representative to the general meeting to be able to vote. They are not allowed to mail their proxy vote directly to the firm.

3-Shareholders can call an extraordinary shareholders’ meeting to challenge the management. The percentage of share capital necessary for this action is 10% and can be reduced by the firm’s charter.

4-The shareholders are entitled to the pre-emptive right to buy new issues of stock. This right is intended to protect shareholders from dilution, whereby shares are issued to favored investors at below market prices.

5-Cumulative voting for directors or any other mechanism of proportional representation on the board, by which minority shareholders may name a proportional number of directors, is not allowed in Turkey.

6-Minority shareholders are entitled to some legal mechanisms against perceived oppression by directors. These mechanisms allow minority shareholders to demand a special investigation of the firm’s records by the audit committee or to go to the Court if the necessary investigation in not undertaken. The percentage of share capital necessary for this action is again 10%.

The research findings on this list of rights for shareholders’ protection in Turkey show mixed results. While the minority shareholders have in practice no chance to exert direct influence on the long-term goals or strategies of the firm, they have some devices, though difficult to use, to protect themselves from obvious oppression by the directors. Mandatory dividends are often observed in countries with poor investor protection. Nonetheless, this remedial measure is not used in Turkey where managers are free to pay or not to pay dividends.

The agency theory is the most dominant theory that has been used extensively in auditing research. This research is explained using the agency theory as the underpinning theory. Based on the agency theory, divergence between ownership and control leads to misalignment of interests of both agent and principle (Jensen & Meckling, 1976). Thus, monitoring management activities is necessary to protect investors’ wealth (Fama & Jensen, 1983). Hence, the dominant agency problem in emerging markets is not that which exists between agent and principle as Berle & Means (1932) report, but the conflicts of interests between majority shareholders and minority shareholders (Shleifer & Vishny, 1997).

Controlling shareholders have strong incentives to jeopardize minority shareholders’ wealth, by issuing unreliable accounting information, selecting second best investment opportunity, misusing firm’s fund and earnings manipulation. This indicates the misalignment of interests that gives rise to the need for corporate governance mechanisms, for instance, external audit. Dispersed ownership in firms allows managers to control and direct the firm’s activities. The managers might be incentivized to maximize their interests at the expense of minority shareholders’ wealth. Although dispersed owners participate in the process of selecting the board of directors, they face difficulty in combining their power in order to choose a board that will have the interests of minority shareholders at heart (Desender et al., 2013). Thus, in order to mitigate agency conflicts and monitor management activities, monitoring costs should be incurred via engagement with the Big4 auditors. Therefore, this study is motivated to investigate whether or not minority shareholders with 10 % of control rights, have the power to enhance clients’ ability to engage a high quality auditor in order to mitigate agency conflicts.

Previous literature has documented that the agency theory is an appropriate theory to illustrate the relationship between audit quality and agency conflicts, which in this study, is represented by the Type II Agency Problem. Under the agency theory, the external auditor can play a greater role in monitoring management activities and reducing the incident of minority shareholders’ expropriation. This study employs minority shareholders’ rights as a type of ownership to study their relationship with audit quality, measured by international audit firms. Claessens, Djankov, and Lang (2000) report that ownership patterns of firms listed in emerging markets have two countervailing affects. Controlling shareholders’ interests might be entrenched or aligned with minority shareholders’ interests. In the case of entrenchment effects, controlling shareholders jeopardize minority shareholders’ wealth more than increase shareholders’ value. Thus, they are less likely to engage with a high quality auditor. Claessens et al. (2002) document that a firm’s value decreases with strong entrenchment effects of controlling shareholders. Demirag and Serter (2003) analyzed Turkish listed firms and measured the entrenchment effects as the wedge between control rights and cash flow rights. They found the entrenchment effects are dominant in Turkish family firms. This is in contrast to the alignment of interests between majority shareholders and minority shareholders. In this case, majority shareholders are more likely to improve firm value and monitoring mechanism by hiring a high quality auditor. This study uses minority control rights at the 10% rate to measure minority shareholders’ rights to enhance the clients’ ability to demand high audit quality in an environment of controlling shareholders’ entrenchment. When we study the arguments of the extent of the minority shareholders rights’ in enhancing clients’ demand for audit quality, we might conclude that minority shareholders are more likely to demand high audit quality in order to mitigate agency problems. In other words, it is more likely that a high quality auditor will be engaged in order to reduce the agency conflicts between majority and minority shareholders. Therefore, this study hypothesizes as follows:

H1: There is a positive relationship between minority shareholders’ rights and audit quality.

According to our theoretical framework and the hypothesis, the conceptual model of audit quality introduced by DeAngelo (1981) is adopted in this study. We utilize the panel data regression model to investigate the relationship between ownership structure and audit quality. Henderson and Kaplan (2000) report that panel data regression offers more useful benefits than cross-sectional or time-series data analysis as follows: (1) it assists to create a model that is more relevant than time-series and cross-sectional models to overcome the problems inherent in repeated measures of a study sample; and (2) it assists to handle the issues of omitted variables and heterogeneity bias. Therefore, the following panel regression model is developed and according to the previous study of DeAngelo (1981), the binary measurement of audit quality is measured by (1) if the firms hire international audit firms, otherwise, (0).

This study’s independent variable is shareholders who control about 10% of voting rights. The presence of minority shares is a dummy variable. Shareholders who own equal to or more than 10% of voting rights are coded (1), otherwise (0).

The audit quality model in previous studies has found various firm-specific variables to control for cross-sectional variance among listed companies on the stock exchange. Audit quality is related to firms size measured by (total assets), return on assets (ROA) and leverage is measured by (total debt% total assets). Previous studies have documented that these variables have a significant role and robust power across countries and at different points of time. Similar to previous studies, this study adopts and assumes total assets, ROA and leverage possess robust predictive power. Following the previous study by Adeyemi & Fagbemi, (2010), this study controls for the influence of firm size using the natural log of transformation of total assets.

This study also includes profitability to control for the firm’s ROA and incentive to engage with international audit firms. We follow previous studies to find out the value of profitability by dividing net income by total assets (Johnson & Lys, 1990; Lin et al., 2009; Srinidhi et al., 2011; Uang et al., 2006). Finally, we include leverage to control for auditee’s financial health and incentive to hire high quality auditor. This study defines leverage as a combination of short-term debt and long-term debt divided by total assets (Engel et al., 2010; Zanani et al., 2008). To meet the objective of this study, the following panel equation model is used to analyze the relationship between minority shareholders’ rights, control variables and auditquality.

where,

For each firm (i) and each year (t)

INAU_it= Audit quality measured by international audit firms. A dichotomous variable is used to examine the hypotheses variables related to the brand name auditor

MIRI= minority shareholders’ rights

FSIZE = Total assets ROA= return on assets

LEVE= Total debt divided by total assets

ε_it = Error term supposed to be normally scattered with constant differences.

The descriptive statistics for audit quality, minority shareholders’ rights and control variables (FSIZE, ROA and LEVE) are as disclosed in Table 1. The average value for firms which hire international audit firms is 47%; while firms which hire non-international audit firms is 53%. The minority shareholders who possess more than 10% of voting rights is 55% of firms included in this study sample. The average size of firms is 5.246 and ranges from 3.253 to 7.420, with standard deviation of 0.882.

The mean percentage of profitability (ROA) is 0.741, with a minimum of -1.154 and maximum of 1.676 and standard deviation of 0.386. The mean percentage of leverage is 1.181 and ranges from -2 to 1.99, with standard deviation of 0.624.

The correlation between this study’s variables is explained in Table 2. The correlation values of this study’s variables are less than the threshold value of 80%. This illustrates that no multicollinearity problem exists between this study’s variables.

MIRI is found to be positively correlated with international audit firms (INAU), while the control variables (FSIZE, ROA and LEVE) are negatively correlated with INAU. The positive correlation means that MIRI positively influences firms’ strategic decisions, particularly with regards to hiring external auditors. Further, Turkish firms are more likely to protect minority shareholders in Turkey.

This paper’s explanatory variable is a categorical variable, i.e., international audit firms and non-international audit firms. Logistic regression is used to examine the influence of a set of predictors on the dependent variable (Pallant, 2007). Pallant (2007) reports that there are three assumptions applied in logistic regression. The first assumption is sample size. In logistic regression, the sample size should be large if the model has a large number of predictor variables. In this paper, the sample size is 100 firms with an average of 25 firms for each variable. The proportion is considered adequate to run the test as least 10 observations are required for each independent variable as suggested by Pallant (2007). The second assumption is multicollinearity problems which deal with intercorrelation problem between independent variables. The Collinearity Diagnostic Test is performed to investigate this problem. The existence of multicollinearity problem is detected if the tolerance value is less than 0.01 and Variance Inflation Factor (VIF) is more than 10 (Hair, Black, Babin, Anderson, & Tatham, 2006). Table 3 shows there is no evidence of multicollinearity problem as VIF is less than 10 and tolerance value is more than 0.01.

The third assumption is outliers. Hair et al. (2006) refer to outliers as having unusually high or low values on a variable or a unique combination of values across several variables that will distort statistics. Pallant (2007) classifies cases with standardized residual of more than 3.3 or less than -3.3 as outliers. In this paper, the maximum standard residual is 2.464 and the minimum is -1.610, which indicates that no outlier is found in this study.

Hosmer Jr., Lemeshow, and Sturdivant (2013) report that linearity is not an assumption in terms of logistic regression. However, the odd ratio needs to be linear with the logit value. As such, linearity tests on continuous variables are required to check for linearity violations. A specific procedure under STATA, the lincheck procedure, exhibits that continuous variables in this study have add ratios that are linear with the logit value. Thus, linearity is not obvious.

The R² using logistic regression is 0.16. This indicates that minority shareholders’ rights explain about 16% of the variation in audit quality. The logistic regression results in Table 5 indicate that minority shareholders’ rights positively influence audit quality (10% level of significance) with P-value 0.061 (t = 1.87). This is consistent with the agency theory’s proposition that external auditors improve clients’ monitoring function of management activities and reduce the incidence of minority shareholders’ expropriation. This indicates that minority shareholders’ rights can impact on clients’ demand for high quality auditor. In other words, the Turkish legal system improves minority shareholders’ rights and power to minimize management’s expropriation with a strong monitoring mechanism. For example, minority shareholders might call an extraordinary shareholders’ meeting to challenge management or request a special investigation of the firm’s records by the audit committee or go to Court if the necessary investigation is not undertaken.

Moreover, with regards to control variables, FSIZE and LEVE possess a significantly negative relationship with clients’ demand for audit quality. The level of significance is at the 1% level for FSIZE (t= -4.42) and 5% level for LEVE (t= -2.81). Besides, the degree of impact on audit quality is 1.47 and 0.78, for FSIZE and LEVE, respectively. This finding displays that big firms with high level of leverage are more likely to be involved with low audit quality. Table 5 displays that profitability (ROA) has positive influence on clients’ demand for high quality auditors at the 10% level of significance (t= 2.68; p= 0.007).

The discussion above reveals that minority shareholders have a positive influence on clients’ demand for high audit quality of Turkish top 100 listed firms. This indicates that minority control rights at 10% enhance clients’ incentive to demand high audit quality in an environment of controlling shareholders’ entrenchment. The research, therefore, recommends policymakers to issue new rules and regulations. These rules and regulations should enhance minority shareholders’ power to mitigate agency conflicts by involving high quality auditors. The study also recommends future studies to include more data and other corporate governance mechanisms to compare client’s demand before and after the regulatory changes in 2012.