Rule with an iron hand: powerful CEOs, influential shareholders and corporate performance in Russia

Abstract

Purpose

This study examines whether CEO power influences the book-based and market-based performance of Russian companies when it is restricted by the presence of essential shareholders, namely, state and influential businessmen.

Design/methodology/approach

Managerial power is divided into structural, ownership, expert and prestige. The proposed power metrics include not only CEOs but also the board of directors' characteristics that may restrict or enhance CEO power. The empirical analysis is based on the sample of 90 large traded Russian firms, which shares are included in the Moscow Stock Exchange Broad Market Index (MICEX BMI), observed from 2012 to 2019.

Findings

Panel data analysis suggests that higher board ownership and tenure may restrict CEO power, which in turn would be beneficial for corporate performance. the authors also see that in companies owned by influential businessmen, CEO power influence on M/B value is more negative, while state ownership does not moderate it. CEO power metrics, based on political experience and tenure, affect corporate performance differently in companies affiliated with extractive industries.

Originality/value

First, the authors consider two channels through which a company in emerging markets may get additional resources: CEOs and influential owners. Second, the authors develop power metrics based on Finkelstein's managerial power classification (1992) and the idea of relative power proposed by Bebchuk et al. (2011). It allows identifying whether the board of directors' may constrain or enhance CEO power to raise corporate performance. Third, the authors analyze developing Russian markets that represent a good ground for testing the question, whereas empirical research on Russia is relatively scarce (Grosman and Leiponen, 2018). Fourth, the authors pay particular attention to the CEO power in the extractive industry, strategically important for the Russian economy.

研究目的

本研究擬探討行政總裁的權力，若因有不可或缺的股東 - 即國家和具影響力的實業家 - 的存在而受到約束時，其權力會否影響俄羅斯公司以賬簿為基礎和以市場為基礎的表現

研究設計/方法/理念

管理權分為結構性的、所有權的、專家的和聲望的。提出的權力指標不但包括行政總裁，也涵蓋或會限制或增加行政總裁權力的董事會特徵。本研究的實證分析法是基於90間股份被納入莫斯科股票交易廣泛市場指數的大型俄羅斯上市公司樣本，觀察期由2012年至2019年

研究結果

面板數據分析顯示、較高的董事會所有權和較長的任期或會限制行政總裁的權力，這因此有利於提升企業績效。我們亦看到，在具影響力的企業家擁有的公司裡，行政總裁權力對市價淨值的影響是較負面的，而國有制沒有把它減低。就隸屬採掘業的公司而言，基於政治經驗和任期的行政總裁權力指標，會對企業績效帶來不同的影響

研究的原創性/價值

(一) 我們考慮在新興市場公司可取得額外資源的兩個途徑：行政總裁和具影響力的所有者。(二) 我們基於芬克爾斯坦 (Finkelstein, 1992) 的管理權分類，以及 Bebchuk et al. (2011) 所提出相對功率的學說，建立了權力指標。憑著這權力指標，我們可鑑定董事會會限制、抑或增強行政總裁提升企業績效的權力。(三) 我們分析發展中的俄羅斯市場，其為測試我們問題的良好地方，而探討俄羅斯的實證研究較為稀有 (Grosman and Leiponen, 2018) 。(四) 我們特別關注在採掘業的行政總裁權力，而採掘業對俄羅斯經濟來說、是具有重要戰略意義的

Citation

Zavertiaeva, M. and Ershova, T. (2022), "Rule with an iron hand: powerful CEOs, influential shareholders and corporate performance in Russia", European Journal of Management and Business Economics, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/EJMBE-08-2021-0228

1. Introduction

Emerging economies are usually characterized by underdeveloped formal institutions and market (Enikolopov and Stepanov, 2013; McCarthy et al., 2014). Firms have to use informal channels to get resources. According to the resource dependency view, directors' knowledge, expertise and relationships can provide a firm with needed resources (Pfeffer and Salancik, 1978). The literature describes plenty of personal traits of CEOs that may enable corporate cooperation and monitoring (Grosman and Leiponen, 2018), bring resources or commitments outside the firm (Lynall et al., 2003). For example, a director with influential connections may provide access to external financing (Durbach and Parker, 2009) and share the best business practices (Renneboog and Zhao, 2011). More experienced directors better understand a company and industry and create “mutual knowledge” inside the firm (Cramton, 2001). Foreign directors are characterized by a global mindset and foreign expertise that provide high social competence and monitoring capability (Gregorič et al., 2009).

The ability of a CEO to bring additional resources can be regarded from the other side. Those CEOs who know their company well, have essential connections, and are characterized by greater self-confidence can deviate from corporate goals in favor of personal ones more easily. The concept of managerial power can aggregate these factors: the more professional advantages a CEO has, the more decision-making power they get compared to other managers. Power gives a more significant stimulus to maximize personal welfare and more knowledge of how to do it. While power consequences are twofold, the performance of companies managed by a powerful CEO is still unclear and should be studied.

CEOs are not the only informal channel that brings resources to a firm. Influential owners can be even more critical resources providers on emerging markets, especially when they participate in decision-making processes and day-to-day management. Thus, state-owned companies are believed to be protected from bankruptcy and have a competitive advantage (Enikolopov and Stepanov, 2013). Other influential owners are large shareholders who have financial and non-financial resources that develop a business even in a very hostile environment (Guriev and Rachinsky, 2005). Influential owners can diminish the positive and negative outcomes of CEO power: to prevent personal welfare maximization and decrease the director's stimulus to maximize firm value. The importance of state and large shareholders on emerging markets raises the necessity to study relative CEO power: an ability of a CEO to use professional advantages in decision-making that is conditional on the presence of influential owners.

This paper aims at identifying the outcomes of relative CEO power. We study the performance of companies ruled by a powerful CEO considering the influence of state and large private shareholders. In contrast to previous studies that analyze CEO power influence in developed countries, we study the emerging Russian market. The Russian business environment provides a good context for exploring CEOs' personalities and power distribution. First, Russia is still classified as an emerging economy with insufficient economic freedom, weak legal protection of property rights and non-transparent companies (Lazareva et al., 2008; McCarthy et al., 2013). It raises the necessity to use directors and influential owners to get resources. Second, the influence of managerial traits is more substantial in uncertain conditions where directors rely more on their own opinions and experience. Third, Russian corporate culture is hierarchical and authoritarian, raising the importance of managerial power. Finally, Russian business is characterized by a high involvement of the state. According to the Federal Antimonopoly Service of the Russian Federation, by the end of 2016 state controls over 70% of the Russian economy (Mereminskaia, 2016). At the same time, large private owners, so-called oligarchs, still hold a large share of Russian business and are highly involved in day-to-day management.

The empirical part of the study is based on a database of large listed Russian companies included in the Moscow Stock Exchange Broad market index (MICEX BMI). The final sample analyzed consists of 379 firm-year observations between 2012 and 2019. The results show that the Market-to-Book (M/B) value and return on assets (ROA) are lower in those companies ruled by CEOs with ownership power; oligarch ownership reduces CEO expert and ownership power impact. State ownership does not affect the relationship between CEO power and the performance of state-owned companies. The relative indicators of power allow recognizing that raising of CEO ownership and tenure compared to board of directors (BoD) members harm corporate performance. The additional analysis considers the extractive industry, strategically important for the Russian economy. It shows that company affiliation to an extractive industry decreases the expert power impact on both market and book-based performance, while prestige power raises M/B and ROA.

This paper contributes to the literature in three main ways. First, we consider two channels through which a company in emerging markets may get additional resources: CEOs and influential owners. We suppose that the CEO's power should directly influence corporate performance, and influential owners moderate the relationship between a CEO's power and corporate performance. Second, we develop power metrics based on Finkelstein's managerial power classification (1992). We borrow the idea of Bebchuk et al. (2011) to consider board directors' characteristics when calculating CEO power and build metrics of prestige, ownership and expert power. It allows identifying whether the Board of directors' may constrain or enhance CEO power to raise corporate performance. Third, we analyze developing Russian markets that represent a good ground for testing our question, whereas empirical research on Russia is relatively scarce (Grosman and Leiponen, 2018). Fourth, we pay particular attention to the CEO power in the extractive industry, strategically important for the Russian economy. The results show that despite the whole sample results, politically connected CEOs (in other words, having prestige power) can benefit the performance of companies from this industry.

The remainder of this paper proceeds as follows. In the next section, we review the existing literature regarding the upper echelons approach and CEO power research. Then we describe the chosen methodology, particularly the metrics for CEO power and data used. In the empirical part, we report and discuss the results of hypotheses testing. The last section concludes the paper, outlining several implications and suggesting avenues for future research.

2. Theoretical background

The upper echelon theory that was developed by Hambrick and Mason (1984) and subsequently proved by several studies (De Hoogh et al., 2005; Guthrie Datta and Deepak, 1997; Kaplan et al., 2012; Musteen et al., 2006; Patzelt, 2010), suggests that the CEO's characteristics and background affect decisions he or she makes and that this, in turn, affects company performance. However, the extent to which the same trait of a CEO affects corporate outcomes may differ. The reason can be the CEO's level of power; the more powerful a CEO, the stronger the link between his or her personality and a company's outcomes and vice versa. At this point, it is essential to define the word power:

Power – “the potential ability to influence behavior, to change the course of events, to overcome resistance, and to convince people to do things that they would not do otherwise” (Pfeffer, 1992).

As underlined by researchers, CEO power is a key for understanding how strategic decisions are made and implemented (Child, 1972; Tushman, 1977). That is why it is crucial to identify the nature of power. Previous studies recognize different power sources, formal and informal and various indicators describing it (Pfeffer, 1992). Finkelstein (1992) attempted to classify different dimensions of power and variables which could reflect them. Numerous authors have used his approach because of its simplicity and universality (Adams et al., 2005; Lewellyn and Muller-Kahle, 2012). According to Finkelstein (1992), managerial power can be divided into four groups: structural, ownership, expert and prestige power.

Structural power is related to the CEO's formal position in the company. A standard measure of structural power is duality when the CEO is also the chairman of the board of directors. Duality has been widely studied, but there is, as yet, no consensus on how this type of power affects performance (Krause et al., 2014). However, in some countries, duality has been abandoned, which forces researchers to look for new measures of structural power. Informal duality has been suggested (Judge et al., 2003), but it remains so far unobserved and requires research.

Ownership power is primarily associated with the percentage of shares that a CEO possesses. On the one hand, ownership makes agency costs lower (Chikh and Filbien, 2011) and encourages the CEO to make decisions that will maximize shareholders' wealth (Pathan, 2009). On the other hand, it may cause managerial entrenchment and minority shareholders lose their decision-making power (Onali et al., 2016). Researchers have characterized Russia of the late 1990s and early 2000s as a country with comprehensive insiders' ownership in companies (Dolgopyatova, 2015). In terms of performance, investigation shows that poor separation of ownership and control leads to higher profitability (Kuznetsov and Muravyev, 2001). Similar results were noticed in Chinese banks; high CEO ownership power resulted in enhanced performance (Ting et al., 2017).

The third type of power, according to Finkelstein (1992), is expert power. It reflects all CEO experience, skills and social capital, i.e. social ties with other top managers and experts. The most widely-used measure of this type of power is CEO tenure. During a long tenure in the company, the CEO creates solid social ties with the board. It can create more effective communication within the top-management team, which can, in turn, improve corporate governance (Ryan and Wiggins, 2004). However, as some researchers argue, a close connection between the CEO and the board may result in a situation where the CEO becomes involved in the board's decision-making processes yet lobbies his interests (Shivdasani and Yermack, 1999). Long-tenured CEOs tend to be more conservative and risk-averse (Musteen et al., 2006), which leads to success in stable industries. On the contrary, however, in emerging industries, experienced CEOs often fail to improve performance (Henderson et al., 2006).

The last CEO characteristic reflecting power is prestige, which is defined as social status and reputation. As with expert power, it can be measured by social connections, not only with other top managers (Lewellyn and Muller-Kahle, 2012) but also with other elites. Findings from a range of studies demonstrate that prestige is associated with the board of directors' confidence in the CEO's decisions and the (sometimes over-) confidence of the CEO him/herself (Chikh and Filbien, 2011). High confidence levels lead to weaker monitoring of CEO governance and riskier decisions being taken, with little attention being paid to market signals and the board's opinions (Hengartner, 2007). Such decisions can negatively affect a firm's outcomes (Fan et al., 2007). However, it is also possible to find studies that suggest CEO prestige power relates positively to a company's performance (Ting et al., 2017).

Different CEO power components provide CEOs with different incentives to act. It leads to the influence of a CEO's power on a company's outcomes being mixed; according to the agency theory (Jensen and Meckling, 1976), a CEO is likely to be interested in a firm's performance only when he or she is its shareholder, i.e. the CEO has ownership power. Conversely, structural, expert or prestige power can make agency conflict sharper and provoke CEO entrenchment (Ryan and Wiggins, 2004), leading to poor performance. Empirical results gained regarding CEO power investigation indicate no consensus on how different types of power affect performance. In the review of Krause et al. (2014), we can see different results of the relationship between CEO duality (structural power) and performance. CEO tenure, which reflects expert power, enhances performance in stable industries and decreases it in emerging ones (Henderson et al., 2006). As for prestige power, it is associated with higher firm performance (Davis et al., 2010; Ting et al., 2017); however, it might be data specific.

The influence of managerial power is believed to be more assertive in authoritarian corporate cultures, for example, in Russian. Thus, Abe and Iwasaki (2010) conclude that the power distance between a CEO and other top managers is higher in Russia than in, for example, the United States or Japan and Russian corporate governance could thus be described as authoritarian and hierarchical. In addition, Vernikov (2009) asserts that formal corporate governance structures, familiar to foreign investors and based on best Western practices, function entirely differently in Russian companies; in many cases, the board of directors play only a minor role displaying total obedience to the CEO and/or majority shareholders.

Despite the high importance of managerial power in Russian companies, this has not yet received the attention it deserves. That is why the current paper attempts to validate power metrics developed by Finkelstein (1992) using the developing Russian market. We follow the resource-based view and regard managerial power as the mechanism of providing a company with needed resources. That is why the first hypothesis is stated as follows:

H1.

CEO power increases corporate performance in Russian companies.

Research on Russian corporate governance system usually covers country-specific features like ownership concentration (Dolgopyatova, 2010), “state capitalism”, which implies increasing direct and indirect state ownership in key industries (Djankov, 2015), and particular types of agency problems (Enikolopov and Stepanov, 2013). These papers deliver valuable insights about powerful shareholders who may influence decision-making processes in Russian companies. That is why CEO power may be restricted by other parties of the corporate governance system. The literature allows identification of two essential types of shareholders that can moderate CEO power in Russia: oligarchs and the state.

Guriev and Rachinsky (2005) define a Russian oligarch as “a businessman who controls sufficient resources to influence national politics”. As their research shows, the top 10 wealthiest businessmen or ownership groups owned about 60% of the Russian stock market in 2003. Despite this, oligarchs in the modern Russian Federation are a point of discussion between the government and society. The official position of the Kremlin posits that there are no oligarchs in Russia now [1]. “The phrase ‘Russian oligarchs' is inappropriate” – Dmitry Peskov, Presidential Press Secretary, told reporters in April 2018 [2]. Arkady Dvorkovich, Deputy Prime Minister, expressed a similar point of view for Bloomberg in January 2018. He believes that “oligarchs are the 90s concept” [3]. However, a poll conducted by The Russian Public Opinion Research Centre (VTsIOM), released in April 2018, indicates that “an overwhelming majority of Russians (94%) do not doubt that Russia has oligarchs” [4]. In addition, the term “oligarch” is still widely used in Russian business media such as Forbes, RBC, Kommersant, and it is applied as a synonym for the wealthiest Russian businessmen. For these reasons, in this paper, oligarchs will be considered as powerful shareholders. They are usually characterized by being heavily involved in managing a company, and their companies are more effective (Guriev and Rachinsky, 2004). They can also foster the value-maximizing behavior of a powerful CEO. We formulate the second hypothesis as follows:

H2.

Oligarch ownership positively moderates the influence of powerful CEOs on a company's performance.

As for the Russian government that is highly involved in corporate activity, especially in large companies that produce strategically important goods like gas, oil, metal and chemicals. Sometimes the government acts as a stakeholder and can directly influence, but more often, the control is less direct, through state-controlled pyramids (Enikolopov and Stepanov, 2013). Since the government does not have the maximization of profits as one of its goals (Lazareva et al., 2008), the management of such companies may differ from traditional styles. We believe that state ownership decreases the ability of a CEO to use power, so the third hypothesis is as follows:

H3.

State ownership decreases the influence of CEO power on a company's performance.

The research framework of the paper is represented in Figure 1. Note that the moderation effect of influential shareholders is supposedly different for the state and oligarchs.

3. Empirical design

3.1 Measuring CEO power

We build our metrics of power based on Finkelstein's classification (Finkelstein, 1992). Many empirical studies provide solid verification of the Finkelstein framework on CEO power (Adams et al., 2005; Lewellyn and Muller-Kahle, 2012; Pathan, 2009; Ting et al., 2017). However, they ignore a vital point in the definition of power: the overcoming of resistance. Without taking into consideration the effect of other powerful actors in a management team, it is impossible to determine whether the CEO is powerful or not. Hence, a recent development is the use of relative indicators of a CEO's power that consider board members' power. Thus, Bebchuk et al. (2011) developed the CEO Pay Slice index (CPS). CPS is defined as “the fraction of the aggregate compensation of the top-five executive team captured by the CEO”. Despite receiving the overall support of researchers (Dutta et al., 2011; Jiraporn and Chintrakarn, 2013), the CPS index has some limitations. For example, it cannot be applied to non-transparent companies where managers' remuneration is not disclosed. Additionally, it applies a strong assumption about the ability of compensation to reflect directors' personal traits. In practice, companies may use remuneration schemes to stimulate directors to reach a particular goal, so the amount paid cannot reflect power distribution in a company. That is why the idea of using relative indicators for measuring some types of CEO power should be further developed and tested.

We follow the idea of Bebchuk et al. (2011) and include the board of directors directly or indirectly in the indicators of power:

1. A CEO's duality usually measures structural power. However, according to Russian legislation, a CEO cannot be a board chairman in the same company. So, we propose using a weaker indicator that equals one if a CEO is a member of the BoD of the same company. Such an indicator considers the ability of a CEO to overcome the resistance of directors because he/she participates in meetings and influences decisions.

2. A CEO's tenure reflects expert power. Long-tenured CEOs know their company well, so they are more entrenched and experience less board monitoring (Cook and Burress, 2013). The tenure of board members may decrease the expert power of a CEO. So, we propose to base the calculation method on Bebchuk's CEO Pay Slice.

(1)CEO expert power=CEO tenureCEO tenure+most experienced directors′ tenure

Where CEO tenure – the number of years a CEO has worked in his/her position in a particular company; Most experienced directors' tenure – the sum of tenures of the four most experienced board members.

1. Prestige power reflects the social status and reputation of a CEO. We follow the paper of Ting et al. (2017) and consider CEO political connections as a metric of prestige. It allows to regard the importance of government in the Russian corporate environment. A CEO with a political background is regarded as a valuable source of information and can mediate between a company and the government. The prestige power of a CEO is higher if he/she is the only politically connected manager. So, we correct the indicator by the political background of BoD members:

(2)CEO prestige power=CEO political background− Number of BoD members with a political backgroundTotal number of directors in a Board

Where the CEO's political background – a binary variable equals one if the CEO has working experience in a governmental body.

1. Ownership power is traditionally measured by the number of shares. In order to correct it by BoD ownership, we suggest following the Bebchuk and co-authors' CEO Pay Slice idea:

(3)CEO ownership power=CEO’s % of shares CEO’s % of shares +BoD % of shares

where CEO's percentage of shares – the percentage of this company's shares a CEO own; BoD % of shares – the sum of this company's shares, owned by the four board members with the highest ownership.

Note that all of the metrics are constructed in such a way that their maximum value is 1. It indicates the largest possible relative CEO power. If CEO power is adjusted by a Board power, the indicators' values are less than 1.

3.2 Variables

Given the hypotheses of our research, our dependent variable is corporate performance. We measure it by market-based indicator, M/B, and book-based one, ROA. M/B is calculated as the equity market value and equity book value ratio. ROA is calculated as the net profit and assets ratio. We test the effect of a CEO's power on M/B value and ROA using the four indicators described above: structural, expert, prestige and ownership power.

To test hypotheses 2 and 3, we use two characteristics of a company's ownership structure: oligarch and state ownership. Oligarch ownership is a binary variable equal to 1 if an oligarch owns a share in a company. To identify an oligarch, we used the Forbes recent annual list of Russia's 25 wealthiest businessmen. A person needs to have at least $4.5 billion to be included in the list. All of the persons included there are known not only by their wealth but also by their political “weight”. Thus, we assume the people in this list to be oligarchs. State ownership is a binary variable equal to 1 if the Russian state owns a share in a company.

The control variables include usual corporate finance determinants such as firm age, financial leverage, firm size, the board size and tangibility. Firm age is defined as the number of years since the company was established. Financial leverage is the ratio of debt to equity. Firm size is measured as the log of sales. Board size is the number of board of directors' members listed in the annual report of a particular year. Tangibility is the ratio of fixed to the book value of assets. ROA is used as a control variable in models with M/B as a dependent variable. M/B is used as a control variable in models with ROA as a dependent variable. We also control on the state and oligarch ownership. All variables are described in Appendix.

3.3 Sample

The database used in this research includes companies, whose shares are included in Moscow Stock Exchange Broad Market Index. It consists of 100 shares of 90 large listed Russian companies where shares are selected by liquidity, capitalization, and free float. We consciously limit our sample to traded companies to ensure homogeneity. Listed companies with non-actively traded shares have other goals on the capital market that reflect in their corporate governance system. They also disclose less information about their board members. The initial sample consists of 720 firm-year observations from 90 firms between 2012 and 2019 [5].

The Russian stock market is usually regarded as being underdeveloped, and there are few companies with actively traded shares (Adams et al., 2005; Pathan, 2009; Davis et al., 2010; Lewellyn and Muller-Kahle, 2012). As a result, our sample is very representative of the population of listed firms since it refers to approximately 80% of the total market capitalization of Russian companies. The sample structure reflects the Russian financial market structure, characterized by the predominance of the manufacturing, oil and gas sectors (Lazareva et al., 2008). Thus, the potential for sample selection bias can be ruled out.

Several different sources were used to collect the data. Financial information was gathered using the Ruslana database provided by Bureau van Dijk. Information on CEOs and board compositions was collected from the companies' websites and annual reports. The personal data of directors were hand-collected from annual reports and websites that contain data on Russian top-managers [6]. State and oligarch ownership data were collected using the Ruslana database, and oligarchs were identified using the Forbes' journal rating of the 25 wealthiest Russian businessmen published on the Forbes official website.

3.4 Method

We ran two separate analyses. First, we checked the influence of CEO power on corporate performance. We implemented panel data analysis with fixed effects to test the following equation:

(4)FinPerformanceit=α+β∗CEOPowerit−1+δ∗Controlsit−1+εit

Where FinPerformance – the metric of a company's market-based performance (M/B value) and book-based performance (ROA);

CEOPower – the indicators of a CEO's power: structural, expert, prestige and ownership power.

Controls – control variables: firm age, financial leverage, ROA (in regression with M/B as a dependent variable), M/B (in regression with ROA as a dependent variable), firm size, the board size, tangibility, state ownership and oligarch ownership;

ε – the error term.

Note that serving as a CEO in companies with high value may raise CEO power or more successful companies may hire more powerful CEOs, the endogeneity problem may arise. Also, there could be omitted variables, which we do not include in the regression equations. We use fixed effects and the lagged variables to decrease the endogeneity impact. Thus, fixed effects allow consideration of unobserved variables, which decreases the omitted variables problem. The lagged variables allow decreasing the impact of the reverse causality problem.

Then we analyze CEO power outcomes in firms owned by oligarchs and the state. So, we keep the same dependent variables and test hypotheses 2 and 3 by including the interaction effect between CEO power and ownership variables. Thus, the model to be tested is equation (5), and the method of estimation is panel data analysis with fixed effects:

(5)FinPerformanceit=α+β∗CEOPowerit−1+δ∗CEOPowerit−1∗Ownershipit−1+γ∗Controlsit−1+εit 

Where, Ownership – the binary variable that characterizes the specific types of ownership: oligarch ownership or state ownership.

4. Results

4.1 Descriptive analysis

Descriptive statistics are presented in Table 1. The sample has been reduced significantly due to missing data, especially regarding variables describing power. They are based on CEO and boards' personal information, which some companies do not publish. Moreover, the ownership structure of Russian companies is complicated; it is difficult to find the ultimate beneficiary and detect whether an oligarch owns it or not. In such cases we deleted the observation from the sample.

Minimum and maximum M/B values show that the sample also includes outperforming and underperforming firms. The average ROA value indicates that companies get approximately six rubles of profit on every 100 rubles of assets invested in a company. Some companies are characterized by a negative ROA.

An overwhelming majority of CEOs from the sample are board members, so they have structural power. As for expert power, on average, this has a value of 0.211. According to the design of expert power measure, in the case of equal tenure of the CEO and the four most experienced board members, this indicator equals to 0.2. Thus, CEOs and boards have roughly the same amount of power in the working sample. It may potentially lead to the insignificance of CEO expert power. The average prestige power is positive and equals to 0.090. It indicates that CEOs, as well as board members, tend to have a political background. The average value of ownership power is 0.212, which means that CEOs usually have less percentage of a company shares than board members.

The ownership structure of companies included in the sample could be described as follows. Oligarchs own approximately 22% of sample companies, and more than half of companies are involved in direct or indirect state ownership. It reflects the tendency of the Russian economy to increase governmental presence in large, strategically important companies.

4.2 Explanatory analysis

We applied panel data analysis with fixed effects and lagged variables to test our hypotheses. Table 2 reports the models (4) and (5) results, which are estimated to test the relationship between CEO power and market-based and book-based performance. Note that the number of observations decreases compared to reported in Table 1 because of the use of lags.

The regressions (1) and (2) represent the results of Equation (4) testing. It shows the effect of CEO power on book-based and market-based corporate performance. We identify that only one type of CEO power, namely, ownership power, negatively impacts both types of company results. An increase of a relative ownership power by 1% leads to a decrease in M/B value of 0.17%. While the average M/B value for the sample is 0.770, it will be diminished by 0.0013. At the same time, a growth of relative ownership power by 1% leads to a decline in ROA value by 0.0002%. Taking the mean value of ROA which equals to 0.061, the absolute value of a decrease is 0.000012. Note that the growth of ownership power relates to the raise of a CEO's shares holding compared with the board members.

Next, regressions (3) and (4), reported in Table 2, are estimated to test hypothesis 3, suggesting that state ownership might interact with CEO power and decrease its impact on performance. We cannot find any direct or indirect effect of state ownership on the corporate value of Russian companies. As a side result, we observe that the CEO expert power has a significant and negative impact on ROA, while ownership power negatively affects M/B value. It indicates that the raise of CEO tenure and shares ownership in comparison to the board members harms book-based corporate results.

The last reported models (5) and (6) test the interaction effect between state ownership and CEO power. Despite the initial idea that oligarch ownership should positively moderate the impact of CEO power on corporate performance, we observe the opposite. In firms, owned by influential businessmen, the expert power of a CEO decreases M/B value: the growth of expert power by 1% leads to a decline of M/B value by 1.01%. The absolute value of M/B change is 0.0078. Also, oligarch ownership strengthens the negative effect of the ownership CEO power. Each increase in ownership power by 1% leads to a decrease of M/B value of oligarch-owned companies by 0.456%, with the absolute value of a decrease of 0.0035.

4.3 Additional analysis

In this section, we introduce additional non-hypothesized CEO power issues which might be related to corporate performance on the Russian market. We propose to observe the CEO power effect on corporate performance in the extractive industry, which represents a large share of a Russian economy and is regarded as strategically important. According to the results of descriptive analysis, 40% of the sample companies belong to it. Because of the strategic importance of such companies, CEO power may be particularly important to raise their performance.

We include in the regressions dummy variable on being belonged to the extractive industry and calculate the interactions of the CEO power variables with it. Since the industry is not changed in years, we implement ordinary least squares (OLS) analysis with control variables on year and industry. We also keep the use of lagged variables to reduce the reverse causality problem impact. The results are presented in Table 3.

We observe that both the M/B value and ROA of companies from the extractive industry are higher in comparison to the rest of the sample. It highlights the important role of this industry in the Russian economy. We also see that the CEO power of companies from extractive industries impacts corporate performance differently. Thus, company affiliation to an extractive industry decreases the expert power impact on both market and book-based performance. The size of the effect differs. The raise of the expert power by 1% leads to a 0.0004% increase of M/B value, while companies from the other industries get 0.004% growth in M/B. As for ROA, the increase of the expert power by 1% decreases it by 0.00075%.

Also, affiliation to the extractive industry changes the effect of the prestige power on corporate performance. A 1% increase in prestige power leads to an increase in M/B value of 0.0023%. The same growth is reflected in the 0.0011% growth of ROA.

5. Conclusion

In this paper, we employ the concept of CEO power in the specific case of Russian companies. This topic is new in Russian corporate governance research even though Russian firms are often described as fully hierarchical, authoritarian structures with solid leaders. Using Finkelstein's managerial power framework and Bebchuck's CEO Pay Slice idea, we constructed our relative expert, prestige and ownership power metrics that consider board members' power. Then we analyze whether the CEO, powerful according to Finkelstein, influences the performance in the presence of powerful shareholders. From a Russia-specific viewpoint, oligarchs and the state were defined as powerful shareholders who could reduce managerial power.

The resource-based view, which is the theoretical background of the current study, supposes that CEO power provides a firm with needed resources. So, we expect it to increase market-based and book-based performance. Also, we suggest that the presence of powerful shareholders reduces CEO power, yet only a tiny part of our results supports this. M/B value and ROA are lower in those companies ruled by CEOs with ownership power. Oligarch ownership makes CEO expert and ownership power impact more negative, while state ownership does not cause any effect. In other words, in companies owned by influential shareholders, the raise of CEO power is not beneficial for corporate performance: more experienced CEOs and those who have more company's shares compared to board members, harm market-based performance. Thus, our findings provide new evidence, not only about Russian corporate governance specifically reflected by power distribution between the CEO, board and shareholders but also about differences of CEO power performance between state-owned and private-owned firms.

What is particularly important, we test relative indicators of CEO power. So, our analysis regards not the absolute value of power, but its value, compared to the board of directors. Thus, we observe that raising CEO's shares and tenure in comparison to BoD lead to lower performance. It indicates that higher board ownership and tenure may restrict CEO power, which in turn would be beneficial for corporate performance. Raising board ownership leads to lower board independence. Despite the common knowledge that independent boards provide for better monitoring of CEO and increase corporate results, previous studies on the Russian market show the insignificance of this factor (Berezinets et al., 2017). So probably the specific of the Russian companies lead to the necessity to control CEO by board members, own corporate shares. It may be an avenue for future research.

We also provide additional analysis of the extractive industry, which is strategically important for the Russian economy and characterized by higher M/B value and ROA in comparison to the rest of our sample. Extractive industry affiliation decreases the effect of expert power, which is based on CEO and board experience. At the same time, it raises the effect of prestige power, built on the CEO and board political connections. It may indicate that in this industry the role of political connections is more important than the experience. Nevertheless, this supposition requires further testing.

It is important to note that there are no conclusions from previous research which are similar to the results gained in this paper since new metrics of CEO power were used and that the investigation was built with the consideration of specifically Russian ownership structures. Thus, only the following results could be partially compared with previous results on the investigation of CEO power. First, despite the previous papers, particularly, the study of Judge et al. (2003), which is based on Russian companies' data, the metric of CEO structural power is not associated with performance. A similar effect of CEO duality on performance was got by Daily and Jonathan (1997). Note that we analyzed only a weak metric of duality because of the Russian legislature, which forbids CEO duality. Second, based on the agency theory (Jensen and Meckling, 1976), we expected that CEO will benefit the company when he owns a certain number of shares. This assumption has found its empirical support in previous research (Fischer and Pollock, 2004; Ting et al., 2017). However, we find the opposite result. Higher ownership power of CEOs in Russian companies harms both market- and book-based performance. Third, the CEO's expert power enhances a company's outcomes in some of the tested specifications, which was confirmed by Ting et al. (2017) and Henderson et al. (2006) for stable industries. Nevertheless, despite the previous research, we use metrics corrected for the board members' power.

The limitations of our research include the problem of the latent nature of CEO power. We develop new metrics of relative expert, prestige and ownership CEO power based on the Bebchuk et al. (2011) idea of CEO Pay Slice and find relations between them and corporate performance. Nevertheless, they need further validation. What is more, the concern about possible endogeneity remains open, although we have addressed it using fixed-effects analysis and lagged variables. We also lack a theoretical framework for some of our additional analyses.

Our findings suggest new directions for future research and discussion. Developing indicators describing CEO power while considering other powerful stakeholders is essential to research. Bebchuk's CEO Pay Slice considers only the most powerful board members, and this design could be inappropriate for countries with collectivist-oriented cultures. Perhaps, it would be more appropriate to use the average power of the whole board to correct CEO power in Russia. Next, as was revealed in our paper, CEO power measure has a different impact on corporate performance. Therefore, each component of CEO power should be investigated in further detail separately from each other. In addition, as far as managerial power is an agency theory problem, it would be interesting to explore ways in which shareholders could take CEO power under control, for example, by designing contracts.

Figures

Figure 1

Research framework

Table 1

Descriptive statistics of the sample

Variable	Obs	Mean	Std. Dev	Min	Max
ROA	379	0.061	0.071	−0.099	0.352
M/B	379	0.770	0.609	0.006	3.434
Structural power	379	0.879	0.327	0.000	1.000
Expert power	379	0.211	0.225	0.000	1.000
Prestige power	379	0.090	0.458	−0.857	1.000
Ownership power	379	0.212	0.353	0.000	1.000
Financial leverage	379	1.090	1.396	0.017	9.767
Log of firm size	379	14.970	1.682	6.855	18.894
Tangibility	379	0.715	0.148	0.173	0.999
Board size	379	10.340	2.261	5.000	19.000
Log of firm age	379	3.179	0.885	1.609	7.610
State ownership	379	0.599	–	0.000	1.000
Oligarch ownership	379	0.219	–	0.000	1.000
Extractive industry	379	0.400	–	0.000	1.000

Table 2

Results of the regression analysis

	(1)	(2)	(3)	(4)	(5)	(6)
	MB	ROA	MB	ROA	MB	ROA
Structural power	−0.016 (0.102)	−0.021 (0.016)	0.084 (0.103)	−0.019 (0.016)	0.130 (0.100)	−0.019 (0.016)
Expert power	0.122 (0.155)	−0.028 (0.024)	0.069 (0.166)	−0.053* (0.027)	−0.015 (0.152)	−0.040 (0.025)
Prestige power	−0.024 (0.078)	−0.006 (0.012)	0.025 (0.078)	−0.013 (0.012)	−0.011 (0.075)	−0.011 (0.012)
Ownership power	−0.170** (0.078)	−0.020* (0.012)	−0.146* (0.080)	−0.014 (0.013)	−0.132* (0.076)	−0.013 (0.012)
State ownership	−0.174 (0.115)	0.024 (0.018)	−0.166 (0.115)	0.021 (0.018)	−0.103 (0.111)	0.019 (0.018)
Oligarch ownership	0.056 (0.132)	0.014 (0.020)	0.038 (0.126)	0.014 (0.020)	0.178 (0.139)	0.025 (0.023)
Structural power*State			0.111 (0.099)	−0.012 (0.016)
Expert power*State			0.095 (0.194)	0.038 (0.031)
Prestige power*State			−0.113 (0.097)	0.008 (0.016)
Ownership power*State			−0.005 (0.110)	0.010 (0.018)
Structural power*Oligarch					−0.013 (0.139)	−0.018 (0.023)
Expert power*Oligarch					−1.011* (0.600)	−0.008 (0.098)
Prestige power*Oligarch					0.219 (0.179)	0.036 (0.029)
Ownership power*Oligarch					−0.324** (0.155)	0.019 (0.025)
Constant	5.886*** (0.995)	0.359** (0.155)	5.362*** (1.025)	0.419** (0.167)	5.487*** (0.972)	0.379** (0.160)
Observations	325	325	315	315	315	315
R2	0.159	0.087	0.179	0.106	0.203	0.108
Control variables	Included	Included	Included	Included	Included	Included

Note(s): Standard errors are in parentheses

***p < 0.01, **p < 0.05, *p < 0.1

Table 3

Results of the regression analysis (ordinary least squares, interaction effects with extractive industry)

	(1)	(2)
	MB	ROA
Structural power	0.111 (0.090)	−0.010 (0.012)
Expert power	0.409*** (0.141)	0.026 (0.019)
Prestige power	0.077 (0.072)	−0.019* (0.010)
Ownership power	0.095 (0.089)	0.015 (0.012)
State ownership	−0.088 (0.062)	0.013 (0.009)
Oligarch ownership	0.247*** (0.073)	0.009 (0.010)
Structural power*Extractive industry	−0.207 (0.169)	−0.026 (0.023)
Expert power*Extractive industry	−0.797*** (0.236)	−0.075** (0.032)
Prestige power*Extractive industry	−0.199* (0.116)	0.077*** (0.016)
Ownership power*Extractive industry	−0.089 (0.150)	−0.032 (0.020)
Extractive industry	0.426** (0.177)	0.055** (0.024)
Constant	0.100 (0.360)	0.019 (0.049)
Observations	325	325
R2	0.535	0.298
Control variables	Included	Included
Year dummies	Included	Included
Industry dummies	Included	Included

Note(s): Standard errors are in parentheses

***p < 0.01, **p < 0.05, *p < 0.1

Variable name	Explanatory notes on the variable
ROA	Return on assets calculated as net income divided by book value of total assets
M/B	The ratio of the equity market value and equity book value
Structural power	A binary metric that equals one if a CEO is a member of a BoD in the same company
Expert power	CEO tenure divided by the sum of CEO tenure and tenures of the four most experienced board membersa
Prestige power	CEO prestige is calculated as the difference between the binary metric equals to one if a CEO has a political background and the share of board members with a political backgroundb. We recognize that a CEO/director has a political background if he/she has working experience in a governmental body
Ownership power	CEO % of shares divided by the sum of CEO % of shares and % of shares of the four board membersc
Oligarch ownership	A binary variable equal to 1 if an oligarch owns a share in a companyd
State ownership	A binary variable equal to 1 if the Russian state owns a share in a company
Extractive industry	A binary variable equal to 1 if a company represents the extractive industry (oil, gas, metals, etc.)
Firm age	Log of the number of years since a company's establishment
Firm size	Company size indicator calculated as the natural logarithm of company total sales
Leverage	Financial leverage is defined as the ratio of total debts to shareholders' funds
Board size	Number of directors on a board
Tangibility	A ratio of fixed assets to the book value of total assets

Note(s): ^aCEO Tenure Slice=CEO tenureCEO tenure+Most experienced directors tenure, where CEO tenure – the number of years a CEO has worked in his position in a particular company; Most experienced directors' tenure – the sum of tenures of the four most experienced board members

^bCEO Prestige=CEO political background− Number of BoD members with a political backgroundTotal number of directors in a board, where CEO (board member's) political background equals one if a CEO (board member) has working experience in a governmental body

^cCEO Shares Slice=CEO′s % of shares CEO′s % of shares +BoD % of shares, where CEO's % of shares – the percentage of this company's shares a CEO own; BoD % of shares – the sum of this company's shares, owned by the four board members with the highest ownership

^dAn oligarch is assumed to be a person included in the Forbes' annual list of Russia's 25 wealthiest businessmen

Notes

1. Article from The Moscow Times: https://themoscowtimes.com/news/there-are-no-oligarchs-in-russia-kremlin-claims-amid-reports-new-us-sanctions-61062

2. Article from the Russian news agency TASS: https://tass.ru/ekonomika/5097768

3. Article from the Russian news agency TASS: http://tass.ru/ekonomika/4901002

4. A press release of WTsIOM –located at: https://wciom.com/index.php?id=61&uid=1526

5. Some observations were excluded because the company was founded after 2012 or delisted before 2019.

6. We used the following websites: http://whoiswho.dp.ru; https://finparty.ru; http://www.forbes.ru