INTRODUCTION
Chief Executive Officer (CEO) change is undoubtedly one of the most significant changes in the existence of any firm because of the fundamental importance of the CEOs influence on a company’s actions, performance, and survival (Kesner & Sebora, Reference Kesner and Sebora1994; Finkelstein & Hambrick, Reference Finkelstein and Hambrick1996; O’Shannassy, Reference O’Shannassy2010). Previous research has concentrated on the effects of executives’ succession on organisational outcomes, including financial performance (Karaevli, Reference Karaevli2007), organisational survival (Haveman, Reference Haveman1993), strategic change and reorientation (Tushman & Rosenkopf, Reference Tushman and Rosenkopf1996; Sakano & Lewin, Reference Sakano and Lewin1999; Lin & Liu, Reference Lin and Liu2012; Karaevli & Zajac, Reference Karaevli and Zajac2013; Dominguez-CC & Barroso-Castro, Reference Dominguez-CC and Barrosso-Castro2017), and top management team change (Tushman & Rosenkopf, Reference Tushman and Rosenkopf1996). Although previous research has discussed the theoretical benefits of CEO succession in general and new CEO origin more specifically, weak and contradictory results regarding their impacts on a firms’ post-succession performance have been widely reported (Virany, Tushman, & Romanelli, Reference Virany, Tushman and Romanelli1992; Shen & Cannella, Reference Shen and Cannella2002b, Reference Shen and Cannella2003; Zhang & Rajagopalan, Reference Zhang and Rajagopalan2004; Karaevli, Reference Karaevli2007).
CEOs are fundamental for executive decision-making and have important effects on organisational outcomes and survival. The upper echelons perspective proposes that CEOs are the most valuable sources of intellectual capital and their effect on the firm’s strategies and organisational outcomes are significant (Hambrick & Mason, Reference Hambrick and Mason1984). Consequently, CEO succession will most likely influence a firm’s performance through changes in leadership skills, experience, adaptability to change, and accessibility of resources (Kesner & Sebora, Reference Kesner and Sebora1994; Thornton & Ocasio, Reference Thornton and Ocasio1999; Giambatista, Rowe, & Riaz, Reference Giambatista, Rowe and Riaz2005; Cao, Maruping, & Takeuchi, Reference Cao, Maruping and Takeuchi2006). However, after several decades of empirical research, findings are still inconclusive regarding the relationship between successor’s origin (outsider vs. insider) and company’s post-succession performance (Karaevli, Reference Karaevli2007). This suggests that prior research has largely neglected to investigate how the characteristics of a firm affect the performance implications of CEO succession (Finkelstein, Hambrick, & Cannella, Reference Finkelstein, Hambrick and Cannella2009). Besides, the effects of insider or outsider succession have not been investigated extensively in nondeveloped countries, especially in emerging market contexts, such as in Turkey. There is a scarcity of comprehensive conceptualisation and testing of the effect of contextual factors on the performance consequences of executive succession (Karaevli, Reference Karaevli2007).
This paper intent to remedy these weaknesses in the CEO succession literature by contributing to existing knowledge in three main domains: first, the impact of the origin of a new CEO will be outlined. Instead of an oversimplified dichotomy of the insider or outsider CEO, CEO origin will be conceptualised and defined as a continuum of new CEO outsiderness (Karaevli, Reference Karaevli2007) in order to capture different characteristics of successors at the same time. We propose that new CEO outsiderness, without managerial discretion context influences, will not have a direct impact on firm’s post-succession performance. Second, the connection between the moderating impacts of managerial discretion will be identified. We suggest that the CEO succession consequences are dependent on the managerial discretion level, which stems from external environmental conditions, organisational features, and the top executive’s personal characteristics. Third, the performance implications of the new CEO outsiderness will be examined in Turkey. Studies conducted in emerging economies (i.e., Claessens & Djankov, Reference Claessens and Djankov1999; Peng, Buck, & Filatotchev, Reference Peng, Buck and Filatotchev2003; Kato & Long, Reference Kato and Long2006; Chung & Luo, Reference Chung and Luo2013) suggest that differences in the performance implications of CEO successor’s origin can be associated to the different circumstances presented by these economies. Turkey, a late-industrialising country that becomes more integrated with the world capital system through liberalisation policies implemented in the early 1980s, provides an interesting setting for the study. Despite the liberal transformation, the dominating economic actors are still state-created, family-controlled and highly diversified big business groups that are strategic responses to strategic factor market imperfections and institutional voids developed due to information asymmetries, poor contract enforcement, and imperfect regulatory structures (Buğra, Reference Buğra1994; Yamak & Üsdiken, Reference Yamak and Üsdiken2006; Üsdiken & Yıldırım-Öktem, Reference Üsdiken and Yıldırım-Öktem2008; Colpan, Reference Colpan2010).
First of all, this paper concentrates on the theoretical foundations of the association between new CEO origin and post-succession performance by relying on upper echelons theory. Second, as previous studies have failed to identify the features that might moderate the link between CEO succession and performance in an emerging market context, current investigation objects to discover these factors. Upper echelon theory proposes that top executives of companies affect firms’ strategies and outcomes (Hambrick & Mason, Reference Hambrick and Mason1984); however, population ecology (Hannan & Freeman, Reference Hannan and Freeman1977) or neo-institutional perspective (DiMaggio & Powell, Reference DiMaggio and Powell1983) have challenged this view, arguing that, as firms are largely constrained by external forces and organisational contingencies, top executives might have limited impact on organisational outcomes. With this in mind, upper echelons theory has put forward the importance of a moderator factor, that is, managerial discretion, to reconcile these conflicting points of view. In this sense, managerial discretion may help us to explain and understand why CEOs matter significantly more in some circumstances and contexts than in others (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987; Hambrick, Reference Hambrick2007). From this perspective, executives have greater power and influence on decisions and outcomes when they have a greater level of discretion (Finkelstein & Hambrick, Reference Finkelstein and Hambrick1990; Finkelstein & Boyd, Reference Finkelstein and Boyd1998); consequently, managerial discretion should be included and tested as an important moderating variable of any association between CEO succession and a firm’s post-succession performance.
Hambrick (Reference Hambrick2007) proposes that managerial discretion occurs when there are less restriction and more ambiguity in a firm or in its environment; in addition, environmental conditions, organisational factors, or a top executive’s attributes influencing managerial discretion should be taken into the consideration to measure the impact of CEOs on strategic decisions and their outcomes. As most previous studies have accentuated environmental and organisational factors as primary determinants of managerial discretion (Finkelstein & Hambrick, Reference Finkelstein and Hambrick1990; Haleblian & Finkelstein, Reference Haleblian and Finkelstein1993; Hambrick, Geletkanycz, & Fredrickson, Reference Hambrick, Geletkanycz and Fredrickson1993), we also considered and included them in our study. In line with the Dess and Beard’s (Reference Dess and Beard1984) study, two external market features directly affecting discretion and indirectly influencing CEO succession were included in the study: market munificence and complexity. In this sense, a firm’s inertia, resource availability, and CEO influence on its board of directors were the organisational level factors included as determinants of discretion.
The third contribution of this paper is in studying CEO succession and managerial discretion in the Turkish context. We also aim to investigate the influence of an additional organisational factor specific to most emerging markets in general, and to Turkey specifically. In this sense, we also question the impact of the level of family ownership on managerial discretion (Mitchell, Hart, Valcea, & Townsend, Reference Mitchell, Hart, Valcea and Townsend2009). We propose that family control and management might constrain CEO discretion in Turkey.
In summary, our research framework considers how environmental and organisational contexts configure the performance implication of successors’ origin in emerging countries. This framework has been analysed by the help a multilevel data from 75 CEO change events between the years 2008 and 2011. This paper aims to bring a contribution to upper echelons theory by studying an essential but relatively under-investigated proposition that managerial discretion may describe contextual circumstances of the impacts of CEO outsiderness and by identifying how such an association may be generalised to an emerging country.
THEORY AND HYPOTHESES
New CEO outsiderness and firm performance
Prior CEO succession studies have acknowledged the origin of a new executive as an essential element in the context of succession, which has important effects for a firm’s post-succession performance (Karaevli, Reference Karaevli2007). There are several possibilities and related processes for choosing a new CEO, comprising a ‘horse race’ among incumbent top executives of the firm, the selection of an ‘heir apparent’, or a comprehensive search of external candidates (Friedman & Olk, Reference Friedman and Olk1995; Finkelstein & Hambrick, Reference Finkelstein and Hambrick1996). However, several studies report that appointing an outsider CEO has become a frequent recruitment practice in recent decades (Favaro, Karlsson, & Neilson, Reference Favaro, Karlsson and Neilson2013). It has been suggested that outsiders would be privileged over insiders when organisations encounter performance problems necessitating a strategic turnaround, or when the need for information processing and external knowledge are increased. On the other hand, executives of firms would be promoted to the position of CEO when organisations need or prefer continuity (Zajac, Reference Zajac1990; Boeker & Goodstein, Reference Boeker and Goodstein1993; Herrmann & Datta, Reference Herrmann and Datta2002; Menon & Pfeffer, Reference Menon and Pfeffer2003; Magnusson & Boggs, Reference Magnusson and Boggs2006). Yet, the findings of studies investigating the performance implications of a new CEO origin are very controversial (Karaevli, Reference Karaevli2007; Finkelstein, Hambrick, & Cannella, Reference Finkelstein, Hambrick and Cannella2009). For example, an organisational adaptation perspective views outsider succession as a mechanism of adaptation that supports the capability of firms to respond to their ever-changing environmental contingencies (Georgakakis & Ruigrok, Reference Georgakakis and Ruigrok2017) and to initiate strategic changes (Lin & Liu, Reference Lin and Liu2012). The advantages of selecting an outsider as CEO have been also emphasised by resource dependence and upper echelon theory (Pfeffer & Salancik, Reference Pfeffer and Salancik1978; Hambrick & Mason, Reference Hambrick and Mason1984). The hiring of an outsider or even a CEO from outside the industry is related to the necessity of accessing fresh information inflows, which are presumed to improve innovation and information processing capabilities (Menon & Pfeffer, Reference Menon and Pfeffer2003) and contribute to organisational learning, adaptation, and the renewal capacity of the firm (Tushman & Rosenkopf, Reference Tushman and Rosenkopf1996). As outsider successors are nominated from a larger group of external nominees, it is expected their contribution to the firm’s intellectual capital will be greater (Bennedsen, Nielsen, Pérez-González, & Wolfenzon, Reference Bennedsen, Nielsen, Pérez-González and Wolfenzon2007). In particular, it is expected that such executives may be more competent in adjusting to continuous variations and transformations, which are more probable in turbulent environments (Haveman, Russo, & Meyer, Reference Haveman, Russo and Meyer2001). In addition, multinational companies that have to deal with relatively high information processing requirements also prefer to appoint outsiders who have more external networks, experience, and perspectives, and can thus deliver resource-based advantages for the firm (Herrmann & Datta, Reference Herrmann and Datta2002).
On the contrary, according to organisational disruption view, hiring an outsider CEO might bring some inconvenience for performance for several reasons (Vancil, Reference Vancil1987). Mainly, an outsider might lack firm-specific competencies and knowledge (Bailey & Helfat, Reference Bailey and Helfat2003; Zhang & Rajagopalan, Reference Zhang and Rajagopalan2004, Reference Zhang and Rajagopalan2010), and therefore necessitate more effort and time to familiarise themselves with, and integrate themselves into, the new organisational settings (Fondas & Wiersema, Reference Fondas and Wiersema1997; Dai, De Meuse, & Gaeddert, Reference Dai, De Meuse and Gaeddert2011). This absence of organisational know-how and familiarity with operations and people might often force a CEO to take some premature strategic decisions, which consequently might impact a firm’s outcomes negatively (Zhang & Rajagopalan, Reference Zhang and Rajagopalan2004). Second, some studies have also documented new outsider CEOs as being less likely to find internal support from incumbent top executives of the firm when compared with internal successors (Friedman & Saul, Reference Friedman and Saul1991). It is very likely that outsiders will be seen as contenders for power, thus posing more of a threat to existing internal and external social alliances (Shen & Cannella, Reference Shen and Cannella2002a). Once an outsider is appointed, it is also assumed that this choice will most likely create a tension within the executive team (Shen & Cannella, Reference Shen and Cannella2002a). It is also presumed that the lack of a shared past and intra-firm experience between incumbent executives and the outside successor might create communication issues at the early phase of a post-succession period, which promotes organisational disruption and lowers performance levels even if the executives accept the new CEO as the new leader. Following this line of argument, several researchers have reported that insider successions could result with a better post-succession performance than outsider successions (Zajac, Reference Zajac1990; Cannella & Lubatkin, Reference Cannella and Lubatkin1993; Shen & Cannella, Reference Shen and Cannella2003; Zhang & Rajagopalan, Reference Zhang and Rajagopalan2004).
Several researchers propose that inconsistencies in the definition and scope of the outsider and insider CEO dichotomy may also have significantly added to the contradictory results regarding the performance implications of successor origin (Karaevli, Reference Karaevli2007; Finkelstein, Hambrick, & Cannella, Reference Finkelstein, Hambrick and Cannella2009). New CEO origin is still defined and studied as a binary variable in most studies; however, in this study, we conceptualise the concept of new CEO outsiderness (Karaevli, Reference Karaevli2007) to improve the constructs used by prior research in measuring successors’ origin. Instead of dichotomous variables such as insider and outsider, we adopt the outsiderness continuum suggested first by Finkelstein and Hambrick (Reference Finkelstein and Hambrick1996) so as to capture different managerial knowledge, skills, and styles that CEOs bring to a company related to their prior professional experiences from other companies and industries and their effect on organisational performance. Finkelstein, Hambrick, and Cannella (Reference Finkelstein, Hambrick and Cannella2009) propose that firms’ boards prefer outsider succession over insider succession to achieve different objectives; consequently, different degrees of outsiderness might help to realise different amounts of these desired objectives. As defined by Finkelstein, Hambrick, and Cannella (Reference Finkelstein, Hambrick and Cannella2009), our CEO outsiderness continuum captures the concept from extreme insider to extreme outsider and also takes into consideration CEO transfers within business groups, between affiliated companies, or from the headquarters of multinational enterprises. In emerging countries, the family business group structure is a dominant form for organising (Buğra, 1994; Yurtoğlu, Reference Yurtoğlu2000; Yamak & Üsdiken, Reference Yamak and Üsdiken2006; Khanna & Yafeh, Reference Khanna and Yafeh2007; Üsdiken & Yıldırım-Öktem, Reference Üsdiken and Yıldırım-Öktem2008); in addition, foreign investments and multinational companies also play an significant role (Tatoğlu & Glaister, Reference Tatoğlu and Glaister2000; Demirbağ, Tatoğlu, & Glaister, Reference Demirbağ, Tatoğlu and Glaister2009). Therefore, these structural characteristics might provide different alternatives in the selection of a successor and should be taken into consideration in the CEO outsiderness continuum. A recent study conducted in Turkey reveals that the CEO succession rate reached an accumulated level of 57 per cent between the years 2005 and 2011 in publicly listed firms in Turkey, and 60 per cent of these companies assigned an outsider as a successor (Ataay, Reference Ataay2016). However, when the origin of outsiders is analysed in detail, it can be seen that 47.5 per cent of these new CEOs are related outsiders; in other words, they are internal transfers between business groups or affiliated multinational enterprises companies.
In conclusion, based on the above discussions regarding the several benefits and inconveniences of a new CEO outsiderness, we can conclude that the impact of a new CEO is dependent to the managerial discretion context of the firm (Karaevli, Reference Karaevli2007). CEOs have greater power over outcomes when they have a greater level of discretion. Organisational and environmental aspects sometimes work for or sometimes against potential benefits that successor outsiderness brings to a firm. As it is very challenging to envisage and isolate the main effect of CEO outsiderness on firms’ post-succession performance without controlling for the managerial discretion level, we propose that:
Hypothesis 1: New CEO outsiderness is not significantly associated with post-succession firm performance when controlling for the managerial discretion level.
The moderating role of managerial discretion
To remedy the contradictory conclusions of previous studies on the impact of new CEO outsiderness (Giambatista, Rowe, & Riaz, Reference Giambatista, Rowe and Riaz2005; Karaevli, Reference Karaevli2007), we propose, based on upper echelons theory, that managerial discretion could be a critical moderating variable in the association between chief officer succession and post-succession performance. Most of the research on the effects of CEOs on performance is conducted in the United States where the CEOs have an ultimate power on decisions (Taniman & O’Shannassy, Reference Taniman and O’Shannassy2015). However, previous theory also proposes that managerial discretion effects the extent to which CEOs are essential for firms’ strategic intents and performance (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987; Hambrick, Reference Hambrick2007), and empirical findings has evidenced that top executives’ influence on the company’s processes and outcomes are stronger when they have greater discretion (Finkelstein & Hambrick, Reference Finkelstein and Hambrick1990; Finkelstein & Boyd, Reference Finkelstein and Boyd1998; Crossland & Hambrick, Reference Crossland and Hambrick2007). Drawing on these propositions and findings, this study intents to analyse the organisational and environmental aspects that impact the scope of managerial discretion (Hambrick, Reference Hambrick2007).
Environmental determinants of managerial discretion
Market munificence. Market munificence can be defined as an external environment’s capacity to backing sustainable organisational growth (Dess & Beard, Reference Dess and Beard1984; Keats & Hitt, Reference Keats and Hitt1988). It is expected that a CEO’s discretion will be enriched dependent on the market munificence. As more opportunities and resources can be provided by a munificent market for firms, we can assume that the CEO of such companies will be more likely to have more freedom to take some strategic decisions (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987). However, as the competition in munificent markets is fierce (Wiseman & Bromiley, Reference Wiseman and Bromiley1996; Palmer & Wiseman, Reference Palmer and Wiseman1999), there is a necessity for more dynamic and strategic decision-making (Hambrick & Abrahamson, Reference Hambrick and Abrahamson1995). In other words, the CEOs can have a wider scope for making strategic choices in munificent environments (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987). As enhanced market discretion may intensify the effect of the new CEO on the performance we propose that,
Hypothesis 2a: Market munificence strengthens the positive association between new CEO outsiderness and post-succession firm performance.
Market complexity. An industry’s several structural characteristics might have an impact on the managerial discretion level of the organisations, and one of these characteristics, market complexity, might enhance managerial discretion level (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987). Market complexity can be defined as the extent to which a firm’s external environment is heterogeneous and competitive. Market complexity is likely to increase as industry concentration rate decreases (Palmer & Wiseman, Reference Palmer and Wiseman1999). Hambrick and Finkelstein (Reference Hambrick and Finkelstein1987) proposed that CEO’s discretion will be limited in markets where there are fewer challenges created with a limited number of competitors as this markets tend to function simpler way and highly regulated by developed norms or rules. However, CEOs tend to have more discretion in complex markets where there are fewer restrictions for organisations (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987). We propose that,
Hypothesis 2b: Market complexity strengthens the positive association between new CEO outsiderness and post-succession firm performance.
Organisation-level determinants of managerial discretion
Organisational inertia. As organisational inertia may limit the CEO’s discretionary choices, the direction and fate of a firm are driven by internal inertial forces (Tushman & Romanelli, Reference Tushman and Romanelli1985). Hambrick and Finkelstein (Reference Hambrick and Finkelstein1990) suggested firm’s age as one of the most critical elements of organisational inertia. When firms become older, CEOs prefer to follow established routines, tending to focus more on exploitation rather than exploration (Hannan & Freeman, Reference Hannan and Freeman1984). Several scholars have suggested that firms may grow more weaknesses regarding the effectiveness of their actions, and develop more fossilised communication configurations as they age (Barron, West, & Hannan, Reference Barron, West and Hannan1994; Guillen, Reference Guillén2002). Consequently, as firms get older, a CEO’s discretion can be constrained by the increasing level of inertia. Therefore,
Hypothesis 3a: A firm’s age weakens the positive association between new CEO outsiderness and post-succession firm performance.
Firm size is another essential determinant of organisational inertia. Since the larger organisations are more likely to have more hierarchical structures and established routines (Audia & Greve, Reference Audia and Greve2006), they encounter some difficulty for initiating strategic and organisational changes such as expanding their operations (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987). The inertia level of an organisation increases with its size and consequently, this lead organisation to have less and less scope for managerial discretion. For example, some scholars have reported a positive and strong link between the tenure of top executives and firm performance in small-sized firms (Finkelstein & Hambrick, Reference Finkelstein and Hambrick1990). Therefore we propose that,
Hypothesis 3b: A firm’s size weakens the positive association between new CEO outsiderness and post-succession firm performance.
A firm’s intangible resources can be categorised as another crucial determinant of managerial discretion since a relevant level of resources offer firms with the latitude to indulge into some exploratory activities (Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987). Intangible resources, among all the other different types of resources a firm might have, has been suggested to be especially crucial for achieving sustainable competitive advantage (Hall, Reference Hall1992). For example, in research-intensive industries, research and development (R&D) intensity of a firm, an important indicator of intangible resources, is influenced by the CEO’s strategic tactics regarding resource allocations (Burgelman, Reference Burgelman2002; Burgelman & Grove, Reference Burgelman and Grove2007). Moreover, as the research and development intensity of a company increases, the information asymmetry between outsiders and insiders increases and this leads to an increase in the CEOs’ power and discretion. Therefore,
Hypothesis 3c: A firm’s research and development intensity strengthen the positive association between new CEO outsiderness and post-succession firm performance.
CEO duality is a crucial corporate governance feature of a firm where a single executive undertakes the responsibilities of both the CEO of the company and the chairman of the board concurrently. Many researchers support the view that a dual leadership role fundamentally compromises the effectiveness and independence level of the board (Filatotchev & Nakajima, Reference Filatotchev and Nakajima2010). Managerial power and agency theory both suggest that the centralisation of authority in one person will result in management domination of the board, thus negatively affect performance (Finkelstein & D’Aveni, Reference Finkelstein and D’aveni1994; Shleifer & Vishny, Reference Shleifer and Vishny1997; Core, Holthausen, & Larcker, Reference Core, Holthausen and Larcker1999; Goyal & Park, Reference Goyal and Park2002). Therefore, when the same person assumes both the chairman and the CEO roles, the CEO may have more discretion to influence organisational change and performance as the supervisory power of the board of directors will be weaker (Cadbury, Reference Cadbury2002).
As a common characteristic of late industrialised countries, the founder rules as the chairman and CEO in the majority of Turkish family business group where the first generation is still in power till the late 1990s (Colpan, Reference Colpan2010). However, some institutional reforms such as the adoption of corporate governance principles in 2003, very much along the lines promulgated by the OECD (Uğur & Ararat, Reference Uğur and Ararat2006) brought some important changes to the conventional integration of direction and execution in Turkish companies. The Capital Markets Board of Turkey, the securities regulator in Turkey, initially followed a soft law approach; some provisions of the corporate governance principle were recommended but not mandatory for listed firms. The Capital Markets Board of Turkey suggests that firms should separate their chairperson and CEO. In Turkey, corporate boards are one-tier structures (Yurtoğlu, Reference Yurtoğlu2000) and CEO role duality is now relatively low in listed Turkish companies (Yamak, Ertuna, & Bolak, Reference Yamak, Ertuna and Bolak2006; Ararat, Black, & Yurtoğlu, Reference Ararat, Black and Yurtoğlu2017). We propose that,
Hypothesis 3d: CEO duality strengthens the positive association between new CEO outsiderness and post-succession firm performance.
Family ownership concentration. Percentage of share hold, as well as the identity of the owners shape also the managerial discretion. The shareholder’s amount of control is related to their stake in the firm’s ownership. If firm ownership is highly concentrated, the owners will tend to have more concentrated power and impact on the strategic processes and their discretion will be high (Crossland & Hambrick, Reference Crossland and Hambrick2007).
Turkey has a peculiar corporate governance system compared with its European and North American counterparts. Volatile and unreliable financial and capital markets, insufficient liquidity to provide external control, and weak legal protection for minority shareholders shapes external corporate governance environment in Turkey (Nilsson, Reference Nilsson2007; Oba, Tiğrel, & Şener, Reference Oba, Tiğrel and Şener2014; Yamak & Ertuna, Reference Yamak and Ertuna2017). On the other hand, internal governance system in Turkey features the main characteristics of family-based systems. Similar to other emerging markets where large block-holders dominate the companies (LaPorta, Lopez-de-Silanes, Shleifer, & Vishny, Reference LaPorta, Lopez-de-Silanes, Shleifer and Vishny1999; Claessens, Djankov, & Lang, Reference Claessens, Djankov and Lang2000), Turkish companies’ ownership structures are highly concentrated (Demirağ & Serter, Reference Demirağ and Serter2003; Orbay & Yurtoğlu, Reference Orbay and Yurtoğlu2006; Gugler, Mueller, & Yurtoğlu, Reference Gugler, Mueller and Yurtoğlu2008), and block-holders are generally families (Yurtoğlu, Reference Yurtoğlu2000; Gündüz & Tatoğlu, Reference Gündüz and Tatoğlu2003). Companies are characterised by highly concentrated ownership, pyramidal-type ownership structures and dominance of business groups (Colpan, Reference Colpan2010). In this sense, families control more than 75 per cent of publicly listed organisations (Yurtoğlu, Reference Yurtoğlu2000; Ararat & Uğur, Reference Ararat and Uğur2003), and family business groups are among the largest economic units in Turkey (Buğra, Reference Buğra1994; Yurtoğlu, Reference Yurtoğlu2000; Gökşen & Üsdiken, Reference Gökşen and Üsdiken2001; Khanna & Yafeh, Reference Khanna and Yafeh2007; Colpan, Reference Colpan2010). A family firm’s business operations are normally constrained by family power and goals (Miller & Le Breton-Miller, Reference Miller and Le Breton-Miller2011) therefore the family ownership may also have implications on how much discretion a CEO might have in the Turkish context. As this ownership structure enables families to exercise extensive and, sometimes, unrestricted control and intervention (Anderson & Reeb, Reference Anderson and Reeb2003), CEOs decision-making power in family-owned firms can be constrained, and they may experience less discretion. We propose that,
Hypothesis 4: Family ownership weakens the positive association between new CEO outsiderness and post-succession firm performance.
METHODS
Data
Common characteristics of the most of the emerging economies such as government interventions, policy ambiguity, and institutional transitions cause some turbulence and complexity for the business environment in Turkey (Wang, Tsui, & Xin, Reference Wang, Tsui and Xin2011). Market turbulence is also set off by financial crises. Although our data set encompassed a longer period, we tested our hypotheses for the years 2008–2011 to be able to control for the effects of the financial crisis of 2007–2008. Our sample was constituted of the publicly traded nonfinancial companies on the Borsa Istanbul (Istanbul Stock Exchange-BIST). All data were manually collected for each year for the period 2006–2013. Financial institutions (i.e., banks, investment trusts, insurance companies, leasing companies, funds, and factoring firms) were omitted as they are regulated according to different accounting rules as well as having different registration systems; they could, therefore, provide misleading evidence regarding the calculation of the performance variables. In total, 824 company-year observations were examined in total. Since some of these companies delisted from BIST during the study period or started to be traded on the custody market, we could not obtain complete data about their financial figures. Also, only firms that had experienced a CEO change were included in our sample.
We collected data about other firm characteristics and annual financial performance from companies’ financial reports and the KAP (Public Disclosure Platform of Turkey) internet-based database. We embraced a multi-source policy to pursuit all pertinent information for CEO succession cases. Data were collected for the years before and after successions to provide a better understanding of causality. We first identified CEO (or general manager) changes from the KAP database. Next, we excluded data with missing information, and we only included 75 cases of succession during our period of observation into our final sample. Of these changes, 16 occurred in 2008, 19 occurred in 2009, and 20 occurred in 2010 and again in 2011. In 14 cases, more than one CEO change occurred during the observation period. We then searched various databases, firms’ financial reports, newspaper and magazine articles, and other documents to find information on both the incumbent CEO and the origin of the successor.
Measures
Dependent variable. We use the average return on assets for 2 years after the CEO change as a proxy for the firms’ post-succession performance in this study. As our aim is to study the association between CEO succession and firm performance, we use an accounting measure of performance instead of market-based measures of performance, which are in general influenced by factors beyond the management’s control (Karaevli, Reference Karaevli2007; Georgakakis & Ruigrok, Reference Georgakakis and Ruigrok2017).
Independent and moderator variables. The new CEO outsiderness variable is operationalised as a continuum (as presented in Table 1) varying from extreme insider to extreme outsider as our sample largely composed of family firms and BGs (Business Group).
Table 1 New Chief Executive Officer (CEO) outsiderness continuum
We differentiate the insiders based on their company tenure, and we also classified new outsider CEOs into two further categories. As BGs fill institutional voids by generating their internal markets for financial capital as well as managerial talent, we defined related outsiders who were employees of other affiliates in the same business group and nonbusiness group affiliated (unrelated) outsiders. Out of the 75 CEO successions in our data set, 43 were outsider successors and 32 were insider successors.
Managerial discretion. The variables such as market complexity and munificence, firm size and age, R&D intensity, chief executive officer duality, and family ownership were used as indicators of managerial discretion. We used the data provided by the Turkey Statistics Institute (www.tuik.gov.tr) to measure market munificence and complexity. These variables were operationalised based on a firm’s primary two-digit NACE Rev.2 (European industrial activity classification) code. Market munificence was measured as the average growth in industry sales over the prior 5 years of succession. Herfindahl index was used to measure market complexity, and since market complexity is likely to increase as industry concentration decreases, this variable was reverse-coded (Keats & Hitt, Reference Keats and Hitt1988). Firm age was calculated by subtracting the data year from the foundation year of the firm. Firm size was measured by the logged values of the company’s total assets. R&D intensity was measured as the ratio of R&D expenditure to total sales. We used a dummy variable for CEO duality which takes the value of 1 when the chairman of the board and the CEO was the same executive. Finally, family ownership was measured as the percentage of stock shares owned by the controlling family. In this sense, the controlling family refers to the family with the largest shareholding. We considered pyramidal structures to calculate family shareholdings and the controlling families own, on average, 29.63 per cent of the total shares in our sample.
Control variables
A number of control variables were incorporated into the model to rule out any alternative explanations. We used the variables Chief Executive Officer age and educational background to control for the effects of differences on new CEOs human capital. Several studies have reported that managers’ demographic characteristics might also affect organisational performance (Finkelstein, Hambrick, & Cannella, Reference Finkelstein, Hambrick and Cannella2009) and therefore might mistake the impact of CEO origin in isolation. CEO age was used as a proxy for chief executive officer’s tenure-in-career (Finkelstein & Hambrick, Reference Finkelstein and Hambrick1996) and was measured as subtracting the year of birth of the CEO from the year of succession. Chief Executive Officer educational background was operationalised by a categorical variable which takes the value from 1 to 6, with each number indicating an ascending level of CEO’s formal education (Finkelstein & Hambrick, Reference Finkelstein and Hambrick1996). Organisational variables that were included as controls, as they might affect a firms’ performance, were: Firm leverage was measured as the ratio of total debt to equity. Since a higher level of debt lowers a firm’s borrowing capacity, this variable is reversely coded. Several studies have pointed out that a poor performance might trigger an executive change and, consequently, the appointment an outsider (Boeker & Goodstein, Reference Boeker and Goodstein1993; Datta & Guthrie, Reference Datta and Guthrie1994). Finally, we also included the variable pre-succession firm performance into the model to control for this effect. It was measured as the average return on assets for 2 years before the change occurs.
RESULTS
The means, standard deviations, and correlations among the variables are shown in Table 2. Although we note significance levels, they should be interpreted with caution. There were numerous significant correlations among the independent variables, the magnitude of which was not very large, thus indicating that multicollinearity did not present a significant problem. A further examination of the correlations does not reveal any serious multicollinearity, showing a mean-variance inflation factor of 1.143 and a maximum of 1.631.
Table 2 Means, standard deviations and correlations
Note: N =75; two-tailed; **p<.01; ***p<.001.
Our hypothesis proposed that managerial discretion will be a moderating factor in the association between CEO succession and performance. There are two techniques proposed to test for contingency relationships within a model. With the inclusion of interaction variables into the equations, the moderating effects can be tested (Aiken & West, Reference Aiken and West1991), and the OLS hierarchical regression estimates for our models are presented in Table 3. The results of the control and moderating variables are included in Model 1 and the key influence of the main independent variable, which is new CEO outsiderness, was included and analysed in Model 2. The inclusion of the independent variable increased significantly the explanatory power of this regression model (Model 2) over the previous model (Model 1). Although the new CEO outsiderness impact on performance was found to be positive, the relationship was not statistically significant. As predicted in the first hypothesis, the relationship between the new CEO outsiderness and a firms’ post-succession performance is not significant if we hold managerial discretion variables constant in the Model (β=0.202, n.s.). This finding supports Hypothesis 1.
Table 3 The regression analysis results
Notes: Standardized coefficients are reported.
N=75; + p<.10; *p<.05; **p<.01; ***p<.001.
The moderating impacts of the managerial discretion dimensions on new CEO outsiderness – post-succession firm performance association were tested in Models 3, 4, and 5. The interaction between new CEO outsiderness and the two environmental characteristics of managerial discretion is tested in Model 3. The addition of the moderating influence of new CEO outsiderness as well as market munificence and market complexity significantly improved the explanatory power of Model 3 (adjusted R 2 change=0.422, p<.001). Hypothesis 2a, predicting that market munificence strengthens the positive association between new CEO outsiderness and post-succession performance, is not supported (β=−0.149, n.s.). However, as predicted in Hypothesis 2b, the moderating impact of new CEO outsiderness and market complexity is negative, with an approaching significance level (β=−0.088, p<.10). However, the influence of this interaction becomes more significant in the Model 5 (β=−0.031, p<.05), and therefore Hypothesis 2b is supported.
The moderating impacts of organisational characteristics which effect managerial discretion on post-succession performance were analysed in Model 4, and their addition to the model significantly augmented the explanatory power of Model 4 (adjusted R 2 change=0.502, p<.001). The moderating effect of organisational inertia was analysed with the variables firm age and firm size. Hypothesis 3a, which predicted that the firm’s age weakens the positive relationship between new CEO outsiderness and performance is supported (β=−0.355, p<.05). However, the moderating impact of firm size on the association between CEO outsiderness and firm performance is not in the proposed direction and has as approaching statistical significance level (β=0.346, p<.10). The moderating effect of the variables of organisational inertia stayed in the same direction throughout all the models. Hypotheses 3c and 3d stated that a firm’s intangible resources and board vigilance strengthen new CEO outsiderness – firm performance relationship. The moderating effect of R&D intensity (used as a proxy for a firm’s intangible resources) on CEO outsiderness – firm performance relationship is not significant (β=−0.143, n.s.). Hence, Hypothesis 3c is not supported. Yet, as predicted in Hypothesis 3d, CEO duality (used as a proxy for the board of directors’ vigilance) strengthens the relationship between new CEO outsiderness and performance (β=0.003, p<.05). The interaction effects of all of the organisational characteristics also stayed equally significant in Model 5.
Finally, the interaction of CEO outsiderness with a variable specific to Turkey is analysed in Model 5. The moderating impact of family ownership concentration on the performance implications of CEO outsiderness was tested, and the results illustrated that the addition of this variable into the equation significantly improved the explanatory power of Model 5. However, the effect of family ownership is not in the estimated direction and is not statistically significant. Thus, Hypothesis 4 is not supported.
DISCUSSION
The present study extends previous research that dedicated on investigating the role of CEOs characteristics (such as demographic characteristics, functional backgrounds, tenures, educational credentials and affiliations, age, socioeconomic backgrounds) on firms’ performance in developed countries, to consider the association between CEO origin and firm performance in emerging economies. Based on managerial discretion research and upper echelons theory (Hambrick & Mason, Reference Hambrick and Mason1984; Hambrick & Finkelstein, Reference Hambrick and Finkelstein1987), empirical evidence provided in the present study shows that new CEO outsiderness, in isolation, has no direct influence on a firm’s post-succession performance and however its positive impact on performance is contingent on level of managerial discretion. We contribute to the literature by addressing the application of the upper echelons theory to emerging economies.
This study is one of the few research which has explicitly investigated the performance implications of new CEO outsiderness by taking into consideration its multilevel boundary conditions as well as the context of an emerging economy. Although several studies have shown that a CEO’s origin influences a firm’s outcomes (for a review, see Karaevli, Reference Karaevli2007), a very limited number of studies have examined the moderating impacts of various external and internal factors and contexts. All such works have been conducted in developed countries and have not focused on the potential boundary conditions of the effects of CEO outsiderness. This investigation, therefore, aims to prolong the understanding of the impacts of new CEO origin on a firm’s outcomes by filling these empirical gaps in the literature.
Our findings emphasise the moderating impact of managerial discretion, stemming from features in a company’s external and internal environment, which either strengthen or weaken the association between new CEO outsiderness and post-succession performance. We found that if a great deal of discretion is present, then the new CEO outsiderness yield stronger explanations of organisational performance. However, the factors affecting the level of managerial discretion in Turkey differs. The results show that market complexity, but not munificence, provides CEOs with more discretion in the Turkish context, hence strengthening the positive association between CEO origin and firm performance. Once companies encounter with more inertia, as in the case of older firms, managerial discretion is weakened, as is the association between CEO outsiderness and firm performance. The results also show that internal corporate governance mechanisms play a significant role in the models. Finally, when a CEO assumes the dual role of both chairman and CEO, the association between CEO outsiderness and firm performance develops stronger, ostensibly because the CEO has more discretion and board controlling power is weaker.
This analysis used a multi-industry data set in an emerging economy namely in Turkey. The majority of previous research on managerial discretion and CEO succession has, with very few exceptions (i.e., Claessens & Djankov, Reference Claessens and Djankov1999; Peng, Buck, & Filatotchev, Reference Peng, Buck and Filatotchev2003; Kato & Long, Reference Kato and Long2006; Chung & Luo, Reference Chung and Luo2013), generally been conducted in developed countries. As suggested by Crossland and Hambrick (Reference Crossland and Hambrick2007) the impact of managerial discretion would be expected to be weaker in a collective context. Consequently, Turkey’s collectivist culture and family ownership concentration may provide a particularly challenging location in which to find support for the hypothesis tested here. Yet, present investigation nevertheless discovered that the level of managerial discretion significantly moderates the positive impact of new CEO outsiderness on firm performance.
We believe that the results provided in this research contribute to current empirical work aiming, as suggested by Georgakakis and Ruigrok (Reference Georgakakis and Ruigrok2017), to move the research question from ‘whether’ to ‘how and under which conditions’ outside CEO succession is a more appropriate strategy for organisations. The insights gained from this study potentially provide practitioners with parameters that may be useful at the time of CEO selection. Results clearly suggest that, in the new CEO selection process, firms should focus on not only the individual attributes of the contenders but also the different organisational and environmental factors that characterise the companies’ discretional context since the current research become increasingly attentive of the contingency factors that enrich the performance outcomes of an outsider CEO. Results advocate that the performance of a new CEO is contingent to the company-specific context of the firm.
However, several limitations of the study deserve discussion. First, our findings may not be generalisable to small- and medium-sized and financial sector companies since our data was composed of large and nonfinancial firms. For example, managerial discretion may be stronger in small firms because they are less constrained by organisational inertia. Second, we investigated only a limited number of environmental and organisation-level characteristics of managerial discretion, and we did not include individual, in other word, CEO-level elements of managerial discretion. A fruitful extension of the present paper would be then to examine whether other external environmental and organisational factors, as well as CEO attributes, affect the association between CEO succession origin and firm outcomes. As we relied only on archival data, it was not possible to examine in-depth the CEO’s personal attributes and organisational processes through which outsider CEOs control organisations. Therefore, this paper offers a future research agenda in this area. We suggest that future studies employing survey data to identify CEOs’ attributes and organisational processes which shape CEOs’ personalised interpretation of the strategic situations they face would be useful and also add to our empirical understanding of these issues. Third, our discussion of CEO outsiderness and managerial discretion focused on their impacts on firm performance. However, we did not explicitly consider the effects of changes in corporate strategy and top management team composition following a CEO succession, which may affect outcomes. Finally, in order to examine implications of CEO succession more effectively, data on a greater number of successful events would be required.
ACKNOWLEDGEMENTS
This study was supported by Galatasaray University, Scientific Research Projects Fund (Grant No: 12.102.006).
