INTRODUCTION
The notion of institutional complexity has been explicitly recognized since the introduction of institutional logics, which is defined as the taken-for-granted principles of organizing that provide prescription and appropriateness of actions to organizations (Friedland & Alford, Reference Friedland, Alford, Powell and DiMaggio1991; Thornton & Ocasio, Reference Thornton, Ocasio, Greenwood, Oliver, Sahlin-Andersson and Suddaby2008). Extant studies on institutional complexity have shown that organizational field is characterized by multiple, even conflicting, institutional logics and organizations confront incompatible prescriptions (Marquis & Lounsbury, Reference Marquis and Lounsbury2007; Reay & Hinings, Reference Reay and Hinings2005). Recent theoretical discussions paid closer attention to organizational responses to institutional complexity and a few empirical studies demonstrated idiosyncratic organizational responses to conflicting institutional demands (Battilana & Dorado, Reference Battilana and Dorado2010; Reay & Hinings, Reference Reay and Hinings2009). More theoretical pieces outlined the generic response strategies to institutional demands of different natures (Oliver, Reference Oliver1991; Pache & Santos, Reference Pache and Santos2010), but organizations in the theories were treated as homogeneous and synchronic entities (Kraatz, Reference Kraatz, Lawrence, Suddaby and Leca2009). While most recent studies started to explore certain organizational attributes that influenced organizational responses to institutional complexity (Luo, Wang, & Zhang, Reference Luo, Wang and Zhang2016), a more systematic understanding on the organizational attributes formed throughout the developmental history is needed (Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, Reference Greenwood, Raynard, Kodeih, Micelotta and Lounsbury2011).
This study will examine one of the key organizational attributes that mirrors organizational history, the organizational imprint (Stinchcombe, Reference Stinchcombe and March1965). Existing studies on organizational imprinting have shown the persistence of imprint from the founding environment (Kriauciunas & Kale, Reference Kriauciunas and Kale2006; Marquis & Qian, Reference Marquis and Qian2014; Peng, Reference Peng2004; Shinkle & Kriauciunas, Reference Shinkle and Kriauciunas2012), and explored certain components of the founding environment, including the market, public policy, and culture (Johnson, Reference Johnson2007; Marquis & Huang, Reference Marquis and Huang2010). There are other essential components, however, receiving little empirical attention, such as deep-seated beliefs and social structure of the founding environment (Marquis & Tilcsik, Reference Marquis and Tilcsik2013). Building on the extant imprinting and institutional literature, I argue that organizational imprint can be not only from the dominant logic of organizing, which founders rely on during the founding period, but also from the central versus peripheral institutional position that an organization possesses at founding (Battilana, Leca, & Boxembaum, Reference Battilana, Leca, Boxembaum, Walsh and Brief2009). Both components of the founding environment would shape organizational imprints that influence organizational responses to institutional complexity.
Empirically, I examine the board composition of Chinese state-owned firms listed on the Hong Kong Stock Exchange (HKSE) market. Though they are all state-owned, some firms were founded during the socialist era and others were founded in the transition period. Besides, some state-owned firms were founded and controlled by central government agencies while others were founded and controlled by local government agencies. Different founding periods and state owners created different founding conditions that may leave socialist imprint on state-owned firms’ boards to various degrees. When these state-owned firms went public in the HKSE market, the market logic, as the pushing factor, required more outside directors not affiliated with the socialist state because they were believed to be more objective in the evaluation of the firm and helped to defend shareholders’ financial interests. The socialist logic, as the pulling factor, suggests firms’ board members mainly come from people who built their entire careers in the socialist bureaucracy to defend firms’ deeply entrenched structure and vested interests. In this regard, state-owned firms listed on the HKSE market represent an ideal context of institutional complexity to examine how firms’ imprints affect their board composition as responding to the conflicting institutional demands.
I posit that the board composition of state-owned firms would be influenced by firms' imprint from the dominant logic of organizing during their founding period and their institutional position at founding. State-owned firms founded in the market logic dominant period are more receptive to the prescription of market institutions and tend to have more outside directors than the firms founded in the socialist logic dominant period. Moreover, state-owned firms founded by central government agencies tend to have fewer outside directors than the firms founded by local government agencies, since they were born at the center of the socialist system with boards dominated by state directors to accomplish macroeconomic and political goals of the central government and such strong socialist imprint persisted after listing on the HKSE market.
This study promises to make the following contributions. First, it responds to the contemporary theoretical call for more understanding on the influence of organizational attributes on organizational responses to institutional complexity. Particularly, I demonstrate the importance of organizational imprint in shaping organizations’ strategic actions, thus complementing the extant literature in this area. Second, this study helps to improve our understanding of imprinting theory. I extend the theory by examining how the dominant logic of organizing during the founding period, and the institutional position of an organization at founding, influence contemporary organizational structures and behaviors. Third, my study sheds light on comparative corporate governance by providing a more fine-tuned definition of outside directors in state-owned firms from emerging economies. It allows us to see the nature of governance mechanisms in a comparative perspective.
This article proceeds as follows: I first present the theoretical context where my study is positioned, and then elaborate on the empirical setting. Hypotheses regarding organizational imprint that influences state-owned firms’ board composition as the response to both socialist and market logic pressures are developed, followed by the research design and results. Lastly, I discuss the contributions and limitations of this study and outline directions for future research.
ORGANIZATIONAL IMPRINT AND RESPONSES
The recognition of institutional complexity can be traced back to the introduction of institutional logics. Friedland and Alford (Reference Friedland, Alford, Powell and DiMaggio1991) conceptualized Western society as an ‘inter-institutional system’ comprising ‘the capitalist market, bureaucratic state, democracy, nuclear family, and Christian religion’ (1991: 232). Each sector is associated with a distinctive institutional logic, which is ‘the axial principles of organization and action based on cultural discourses and material practices prevalent in different institutional or societal sectors’ (Thornton, Reference Thornton2004: 2). In this sense, organizations confront multiple institutional logics and are subject to multiple or even incompatible demands.
The notion of institutional complexity has attracted increasing interest from management scholars and generated numerous empirical research studies from the institutional logics perspective (Thornton, Ocasio, & Lounsbury, Reference Thornton, Ocasio and Lounsbury2012). It has shown that organizational fields are usually characterized by multiple (often two) logics, which impose different, and potentially conflicting, demands on organizations and eventually shape organizational behaviors. For example, Reay and Hinings (Reference Reay and Hinings2005) observed the recomposition of the Alberta healthcare field was driven by the medical professionalism and business-like healthcare logics and actors used their power to accomplish the shift from one dominant logic to another. Lounsbury (Reference Lounsbury2007) demonstrated how the trustee and performance logics, rooted in different cities, shaped different mutual funds' behavior to contract with independent professional money management firms. Similarly, Marquis and Lounsbury (Reference Marquis and Lounsbury2007) found the competition between the national bank and community bank logics in the development of the US banking field. Recently Greenwood and his colleagues (2010) contributed to the exiting research by examining the interplay of three institutional logics and showed how the regional state logic and the family logic impacted organizational response to the overarching market logic in Spain.
Recent theoretical development advocated a shift of attention to how organizations strategically respond to and manage multiple logics and institutional complexity (Greenwood et al., Reference Greenwood, Raynard, Kodeih, Micelotta and Lounsbury2011). A few empirical works have demonstrated idiosyncratic organizational responses to conflicting institutional demands (Battilana & Dorado, Reference Battilana and Dorado2010; Reay & Hinings, Reference Reay and Hinings2009). For example, Reay and Hinings (Reference Reay and Hinings2009) identified four mechanisms through which physicians and government could collaborate inside organizations to manage rivalry logics to coexist for a lengthy time. Using comparative inductive methods, Battilana and Dorado (Reference Battilana and Dorado2010) showed that organizations developed a common organizational identity to combine conflicting logics and formulate a hybrid one. These pioneering empirical studies, however, focused on distinctive organizational responses in particular contexts and did not develop theoretical models that can systematically examine organizational responses to institutional complexity.
A more systematic effort was first made by Oliver (Reference Oliver1991), which proposed a repertoire of strategies available to organizations as they faced institutional pressures: acquiescence, compromise, avoidance, defiance, and manipulation. But the outlined response strategies to institutional pressures are rather generic. Building on Oliver's (Reference Oliver1991) work, Pache and Santos (Reference Pache and Santos2010) suggested a more fine-tuned model taking into consideration the nature of institutional demands and the degree to which the demands were represented within organizations. What Oliver (Reference Oliver1991) and Pache and Santos (Reference Pache and Santos2010) have in common is their shared assumption that organizational responses are collectively shaped and homogenized by field-level institutional pressures. Organizations, in their approach, are treated as synchronic entities, rather than diachronic ones (Kraatz, Reference Kraatz, Lawrence, Suddaby and Leca2009). The unique organizational characteristics, which are formed throughout their developmental history and enable organizations to take different strategic actions towards multiple institutional demands, received little attention.
In contrast, Kraatz and Block (Reference Kraatz, Block, Greenwood, Oliver, Suddaby and Sahlin-Andersson2008) emphasize unique organizational characteristics (e.g., identity) and agency facing institutional complexity. They provide a detailed typology of organizational responses enabled by the organizational identity: eliminating the tension of complexity by committing to a preferred identity, balancing multiple demands by increasing cooperativeness among identities, building new identities, or compartmentalizing multiple identities. Hence organizations are not passive recipients of institutional demands, but rather can exercise a certain level of strategic choices enabled by their characteristics. Extending Kraatz and Block's (Reference Kraatz, Block, Greenwood, Oliver, Suddaby and Sahlin-Andersson2008) argument, Greenwood and his colleagues (Reference Greenwood, Raynard, Kodeih, Micelotta and Lounsbury2011) emphasize the important role of organizational attributes in determining organizational actions. It is argued that pressures from multiple institutional demands do not affect all organizations equally because different organizations possess different attributes or characteristics (e.g., identity, legacy of past conditions, position in an institutional field) that may shape organizational strategies and actions. Thus, they call for more systematic theories of organizational responses through examination of various organizational attributes. Most recent research has examined the attributes that increased scrutiny from the conflicting institutional constituencies, such as organizational size and institutional linkages (Luo et al., Reference Luo, Wang and Zhang2016). This study also responds to Greenwood et al. (Reference Greenwood, Raynard, Kodeih, Micelotta and Lounsbury2011) and explores the organizational attribute that mirrors the developmental history of organizations.
One such key organizational attribute is the organizational imprint (Stinchcombe, Reference Stinchcombe and March1965). The concept of imprinting that stemmed from studies on animal behaviors was borrowed into organizational studies by Stinchcombe in his classic 1965 essay, ‘Social Structure and Organizations’. Stinchcombe (Reference Stinchcombe and March1965) suggested that initial organizational structures and features were shaped by the socioeconomic conditions present at their founding and that these characteristics would persist over time. Building on Stinchcombe's seminal essay, Marquis and Tilcsik (Reference Marquis and Tilcsik2013) identified three essential features of the imprinting process. The first is ‘the existence of a temporally restricted sensitive period’ (Reference Marquis and Tilcsik2013: 201) during which focal organizations are highly susceptible to environmental influence. Second, the environment exerts significant influence on these organizations during the sensitive period so that they develop characteristics that reflect prominent features of the environment. Finally, these characteristics continue to persist despite subsequent environmental changes. The imprinting concept has been widely applied in an array of fields such as organizational ecology (Carroll & Hannan, 2004) and institutional theory (Marquis & Huang, Reference Marquis and Huang2010) and become an important lens for understanding organizational attributes.
Most studies on organizational imprinting focus mainly on the persistent effect of imprint from the founding environment (Kriauciunas & Kale, Reference Kriauciunas and Kale2006; Marquis & Qian, Reference Marquis and Qian2014; Peng, Reference Peng2004; Shinkle & Kriauciunas, Reference Shinkle and Kriauciunas2012). Some research has paid attention to the essential components of the founding environment but focuses on market environments (e.g., population density) directly related to organizations’ economic performance (Johnson, Reference Johnson2007). Relatively little research has examined other components like public policy and culture (Marquis & Huang, Reference Marquis and Huang2010). As Lounsbury and Ventresca stated, the institutional environment at founding is a ‘richly textured n-dimensional space in which organizations navigate’ (Reference Lounsbury, Ventresca, Lounsbury and Ventresca2002: 3) such that more essential components, particularly deep-seated beliefs and social structure, are worthy of careful consideration.
In this study, I identify such two essential components of the founding environment – the logics of organizing and institutional positions – and examine how they shape organizational imprint. The first component of the founding environment is the ‘logics of organizing’ that founders relied on to build their organizations (Baron, Burton, & Hannan, Reference Baron, Burton and Hannan1999). The founding environment could be characterized by multiple logics of organizing, but one logic is dominant over others in a certain period. Organizations tend to be more susceptible to the influence of the dominant logic of organizing in order to achieve isomorphism and relieve uncertainty and legitimacy pressures (DiMaggio & Powell, Reference DiMaggio and Powell1983). The dominant logic of organizing is to be reflected in organizations’ structures, culture, and routines that broadly constitute new organizational capabilities (Winter, Reference Winter2003). These organizational capabilities are institutionalized over time and become the organizational imprint, which later shapes organizational responses to external pressures. Less dominant logics may also affect the organizations, but not as profoundly and persistently as the dominant logic.
The second component is the institutional position an organization possessed in the founding environment. Existing institutional literature has noted that organizations are not equally distributed in an institutional field: some organizations are located at the center, while others at the periphery (Shils, Reference Shils1975). Different institutional positions in a field could affect organizations' access to resources and their behaviors (Battilana, Leca, & Boxembaum, Reference Battilana, Leca, Boxembaum, Walsh and Brief2009). Organizations at the center of a field are secured in their social acceptance, so they can deviate from conventional behaviors (Greenwood & Suddaby, Reference Greenwood and Suddaby2006; Greenwood, Suddaby, & Hinings, Reference Greenwood, Suddaby and Hinings2002). Organizations at the periphery of the field may also initiate changes to seek a more favorable social reception from other available audiences (Garud, Jain, & Kumaraswamy, Reference Garud, Jain and Kumaraswamy2002; Haveman & Rao, Reference Haveman and Rao1997; Leblebici, Salancik, King, & Copay, Reference Leblebici, Salancik, King and Copay1991). The notion of institutional positions is also applicable to the founding environment because some organizations were founded at the center and others on the periphery. Organizations founded at different institutional positions face different contextual features and would develop characteristics that reflect prominent features of their own positions. In other words, different institutional positions may leave organizations different imprints that would affect their responses to multiple institutional pressures.
The Chinese state-owned firms listed in the HKSE market, which were founded in different periods and by different government layers, provides an ideal context to examine how firms’ imprints affect their board composition pressured by two conflicting institutional logics: the socialist and market logics. The next section further elaborates the empirical setting.
EMPIRICAL SETTING
Boards of State-owned Firms
The empirical setting I explore is the publicly traded Chinese state-owned firms listed in the HKSE market. In the late 1970s, China launched the socioeconomic transition from a socialist planned economy to a market-oriented economy. The market transition accelerated after 1992 as the former paramount leader Deng Xiaoping demonstrated the determination to end the ideological debate and reassure the market-oriented reform during his southern tour in the spring 1992 (Zhao, Reference Zhao1993). State-owned firms, the pillar of the socialist economy, underwent a series of ownership reforms as the transition unfolded. State-owned firms founded in the socialist era were transformed into joint-stock corporations, while newly founded state-owned firms were organized around the Western model of corporations (McFarlan, Xu, & Manty, Reference McFarlan, Xu and Manty2009). Besides, firms operating in related industries were consolidated into business groups, and governments used parent firms to control the conglomerates (Keister, Reference Keister2000). The central and local governments owned and managed different sets of business groups through State-Owned Assets Supervision and Administration Commissions (SASACs) – a hierarchical government agency created in 2003 to oversee state assets on behalf of the government at different layers – and such classification of business groups followed the governmental administrative rank of the parent firms at founding (Deng, Morck, Wu, & Yeung, Reference Deng, Morck, Wu and Yeung2011). Most often the strongest performing assets of a business group were reconsolidated into an existing or a new firm ready for going public (Lin & Milhaupt, Reference Lin and Milhaupt2013). Since the mid-1990s, there were a growing number of state-owned firms going public in overseas stock exchanges. The HKSE market was the most favorable one due to its geographic and cultural proximity.
In order to meet listing requirements of the HKSE market, state-owned firms adopted internationally recognized corporate governance structures and built boards by imitating both the US and German board models. They set up shareholders’ meetings, the board of directors, and the supervisory board. The board of directors was responsible for making key decisions on firm operation and have the authority to appoint or dismiss managers. In addition to the state-appointed directors, the board allowed directors who represented the interests of foreign and private investors to participate in firm management for the sake of board independence required by the stock market. The supervisory board, designed to oversee the board and management, was often considered more cosmetic since they had basically no power to question board decisions and acts and became the subordinate of board directors and senior executives (McFarlan et al., Reference McFarlan, Xu and Manty2009). Figure 1 illustrates the board structure of listed state-owned firms.
Figure 1. Board structure of Chinese listed SOEs
Board members of HKSE-listed state-owned firms are officially categorized into three groups: executive directors (EDs), non-executive directors (NEDs), and independent non-executive directors (INEDs) (Clarke, Reference Clarke2006). EDs are management directors directly appointed by governments or the SASACs and serve as the agent of government owners. They are usually party members and implicitly enjoy the equivalent administrative position in governments. Unlike their counterparts in US corporations who defend shareholders’ interests for maximizing returns on investment, EDs instead work closely with management to defend state interests, which are not limited to financial ones (Wong, Opper, & Hu, Reference Wong, Opper and Hu2004). NEDs are non-management directors who have professional relationships with the focal state-owned firm. NEDs include both individual and institutional investors (called ‘legal persons’ by Chinese terminology). INEDs are non-management directors having no professional relationship with the focal state-owned firm.
Studies on corporate governance of Chinese firms followed agency theory and applied the US definition of outside directors into the Chinese context (Child, Reference Child1994; Peng, Reference Peng2004). For example, Peng (Reference Peng2004) examined the effects of outside directors on publicly traded firm performance in China and the factors that affected the appointment of outside directors. He defined outside directors as directors representing institutional and individual interests, and insiders as those currently employed by the focal firm and its subsidiaries. For Chinese state-owned firms, however, the distinction between insiders and outsiders is different from the US firms. Some NEDs of state-owned firms, which are considered as outsiders in US firms, come from other state-owned firms of the same business group. Moreover, some INEDs are not truly independent since they are former leaders of the same state-owned firms or former government officials who later serve in government-affiliated entities (e.g., industrial associations) with interests in the state-owned firms. These people are an integral part of the state running the firms and hardly able to evaluate management impartially. In this study, I define insiders in state-owned firms as state directors who are directly appointed by the state and those from business-related state-owned firms, government agencies, and government-affiliated entities. Outsiders are non-state directors who are not affiliated with any state-owned firm, government agency or government-affiliated entity. They represent non-state shareholders who have less similarity with the state shareholders so that they can more effectively monitor the state-owned firms to defend minority shareholders’ interests and are much more valued by firms’ stakeholders. Board independence in state-owned firms is more precisely captured by the weight of non-state directors on the board. Such categorization also reflects the major conflict between the state as the controlling shareholder and non-state investors as minority shareholders in state-owned firms.
For publicly traded state-owned firms, their board independence is pressured by the HKSE market that epitomizes the market logic and the state owners that epitomize the socialist logic. In the next part, I analyze and compare the differences between the socialist and market logics that create the institutional complexity.
Socialist versus Market Logics
Under socialist economic institutions, the state controlled most critical resources and allocated them to state-owned firms based on central planning. State-owned firms could not determine what product they should make, how many they should output, and what price they should set, but had to comply with the state directives. The state procured the output of state-owned firms for further distribution (Kriauciunas & Kale, Reference Kriauciunas and Kale2006). Since the state was the sole owner of these firms, leadership and personnel were ultimately determined by the state to execute the state will. Although there was no board of directors in the state-owned firms founded in the socialist era, its counterpart governance body was constituted by nearly all government officials and party members.
After the market transition, the state still remains the controlling shareholder (Fligstein & Zhang, Reference Fligstein and Zhang2009). State-owned firms are considered extended arms of the state in the market in order to act in the best interests of the state. The state formally exercises its rights as the controlling shareholder to appoint directors who build their entire careers in the socialist state bureaucracy, so that the state can exert influence over nearly all decisions of state-owned firms to achieve goals beyond financial performance (McFarlan et al., Reference McFarlan, Xu and Manty2009). Non-state directors who focus more narrowly on financial goals are thus inconsistent with the socialist logic, and state-owned firms are inclined to minimize the number of non-state directors. In short, the socialist logic prescribes internal structure and practices that reflect the persistent commitment of state-owned firms to the socialist system.
In contrast, market institutions are characterized by private ownership of means of production. Firms have autonomy and control over their own production and pricing. In the US, firms are characterized by relatively dispersed ownership and top executives could take control of firms since a controlling shareholder rarely exists (Hall & Soskice, Reference Hall and Soskice2001). This separation of ownership and control between shareholders and management leads to the principle-agent problem where top executives may not act in the best interests of shareholders. Agency theory suggests governance mechanisms must be established to mitigate the principal-agent problem, and the board of directors, among others, is considered essential to monitor self-interested managers’ actions and maximize shareholder value (Fama & Jensen, Reference Fama and Jensen1983). In particular, a more independent board that consists primarily of outside directors is believed to be better able to monitor management actions because outside directors tend to be more objective in the evaluation of management performance given their arm's length relationships to firms’ top executives (Dalton, Daily, Ellstrand, & Johnson, Reference Dalton, Daily, Ellstrand and Johnson1998).
A plethora of corporate governance research has shown that the market logic with respect to the agency theory has risen to its dominance and firms’ stakeholders are heavily influenced by the prevailing market logic to evaluate firms by placing greater importance on board independence (Davis, Reference Davis2009; Westphal & Graebner, Reference Westphal and Graebner2010). This can be seen in the listing requirements of the HKSE market that underscore the increasing weight of outside directors on the board and the media's more favorable coverage of firms and their chief executive officers (CEOs), following increased board independence (Bednar, Reference Bednar2012). State-owned firms are also pressured to build more independent boards to meet listing requirements and stakeholders' expectations. Those founded in the market-oriented environment would be more susceptible to the influence of the prevailing market logic.
Table 1 summarizes the differences between the socialist and the market logic. For HKSE listed state-owned firms, the institutional complexity was created in a way that the market logic, as the pushing factor, demanded more non-state directors, while the socialist logic, as the pulling factor, mandated fewer non-state directors.
Table 1. Socialist versus market logics
HYPOTHESES
Imprinting from the Logics of Organizing and Board Composition
Recent imprinting research has paid attention to the socioeconomic transition from socialist to market-oriented economies throughout the world in the past few decades. These studies showed that organizations founded in the socialist economic system were stamped with a strong socialist imprint, rendering them difficult to adapt to market environments during the transition. Studies on Central and Eastern Europe, for instance, demonstrated that firm-specific capabilities and knowledge sets developed in the Communist system were era-specific and persistent so that these socialist imprints adversely impacted firms’ operating knowledge and competitive aspirations in market-oriented environments (Kriauciunas & Kale, Reference Kriauciunas and Kale2006; Shinkle & Kriauciunas, Reference Shinkle and Kriauciunas2012). Research on China has found that older firms were less likely to adopt new practices, since they had greater socialist imprints that were incompatible with market environments (Marquis & Qian, Reference Marquis and Qian2014). The key difference between the socialist and transition period is that they are characterized by different dominant logics of organizing that founders rely on. Hence, state-owned firms founded in different periods would be stamped with a socialist imprint to various degrees, which shaped firms’ subsequent behaviors.
Chinese state-owned firms founded after 1992 faced declining socialist institutions as the market-oriented reform accelerated. Organizational structures and practices consistent with the socialist logic were found to be ill-suited to the market-oriented environment and state-owned firms had to develop structures and practices prescribed by the ascending market logic (e.g., more dispersed ownership structure and representative governance body) to establish legitimacy and look for a fit with the market-oriented environment (Meyer & Rowan, Reference Meyer and Rowan1977; Peng, Reference Peng2004). Listing in the HKSE market enforced the market logic such that state-owned firms founded in the market logic dominant period were more oriented by the demands and perception of various domestic and international shareholders to have more non-state directors on their board for legitimacy purposes. Domestically, these firms were more visible and scrutinized by the public and media when they were created more recently. As an important part of the people's wealth, they were deemed to perform well for the people by strengthening internal control, including the functioning of the boards. Moreover, they were listed to access the global market, and international investors had rigorous demands for more outsiders to increase transparency and board effectiveness and protect their interests. Poorly monitored firms would expose themselves to the criticism of the domestic public for squandering people's wealth and aggravate international investors’ concerns about their investment. Even under the pressure of the socialist state to keep the dominance of state directors on the board, the firms were inclined to have more non-state directors who could send the market the signal that these firms may have better internal control mechanisms and minority shareholders’ interests could be protected.
On the other hand, state-owned firms that were founded before 1992 possessed greater socialist imprint since their founding period was dominated by the socialist logic. Such firms were organized to rely on the state planning for input and output and instituted the governance body filled with directors who were civil servants accountable to the state and responsible for achieving the goals of the state (Newman, Reference Newman2005). The decision-making process was controlled in the boardroom and the results went in favor of government owners. When listed in the HKSE market, these firms had to adjust board composition to meet listing requirements, but were inclined to preserve the deeply entrenched governance structure by keeping as many state directors on the board to ensure continuing reliance on the state. Given their long-standing role in the socialist system, these firms were less concerned with the legitimacy issue and less oriented by other shareholders’ demand for more outside directors who would pose a challenge to the firms’ governance structure and vested interests. Thus, state-owned firms founded in the socialist logic dominant period were disposed not to increase the number of non-state directors so long as the board composition could meet the listing requirements and the expectation of government owners. In other words, the socialist imprint of these firms persisted and rendered the partial and gradual increase of non-state directors. Thus, I hypothesize:
Hypothesis 1: State-owned firms founded in the market logic dominant period will have more non-state directors on the board than state-owned firms founded in the socialist logic dominant period.
Imprinting from Institutional Positions and Board Composition
Studies on state-owned firms usually consider the state as a monolithic entity and focus on performance and behavior differences between state-owned and non-state-owned firms (Marquis & Qian, Reference Marquis and Qian2014; Nee, Opper, & Wong, Reference Nee, Opper and Wong2007; Ralston, Terpstra-Tong, Terpstra, Wang, & Egri, Reference Ralston, Terpstra-Tong, Terpstra, Wang and Egri2006). However, the state is a hierarchically institutional field comprising a multi-layer government administration (Weber, Reference Weber, Roth and Wittich1978). A typical nation-state consists of the central or federal government and multi-layer local governments acting within the power delegated by legislation or directives of higher levels of government. The central and local governments may own and manage different sets of firms, and this is particularly true for emerging economies, which are characterized by the dominance of the state sector (Li, Cui, & Lu, Reference Li, Cui and Lu2014). In India, for example, there were less than 200 firms owned by the central government while local governments managed about 800 to 1,000 firms by 2005 (Mishra, Reference Mishra2009). As of 2011, the Chinese central government agencies owned and managed 121 central state-owned firms and business groups, and the rest of the state-owned firms were owned and managed by a total of 54 provincial and prefecture city government agencies.
The Chinese state-owned firms listed in the HKSE market include both the firms founded by central government agencies and the firms founded by local government agencies. They were later managed by the SASACs that represented government owners at the central and local level, respectively. In the socialist system, state-owned firms founded by central government agencies were born at the center while others founded by local government agencies were born on the periphery. Their different positions in the founding environment would leave them different imprints in that the central government and local governments may put different goals and governance structures in place for their own firms at founding. Subsequent structures and personnel would bear a strong resemblance to their founding models (Burton & Beckman, Reference Burton and Beckman2007).
State-owned firms founded by the central government agencies were positioned to serve as policy instruments to support the macroeconomic and political goals of the central government (e.g., secure strategic assets and resources for national security and competitiveness) (Li et al., Reference Li, Cui and Lu2014). When founded, the firms were tightly controlled by the central government through the board dominated by state directors to mirror and accomplish the goals. The central positioning shaped the blueprint for state-owned firms founded by central government agencies, such that the firms continued to pursue those strategic goals rather than purely commercial interests when listed in the HKSE market. Their boards, dominated by state directors, were maintained and reinforced over time to guarantee that the strategic purposes were fulfilled under pressure from the rest of the stakeholders (Beckman & Burton, Reference Beckman and Burton2008). Non-state directors that represented the interests of private and global investors usually focused narrowly on maximizing investment return, and would have divergent views on firms’ goals and decisions made by the state directors. Having more non-state directors would disrupt decision-making routines and not necessarily help these firms achieve the strategic goals. In this regard, the central position left state-owned firms founded by central government agencies a stronger socialist imprint, which rendered them less likely to pursue board independence.
In contrast, state-owned firms founded by local government agencies were created on the periphery of the socialist system and less tightly controlled by the socialist state. Local governments did not share the broader national goals with the central government and their firms were not obligated to pursue similar strategic goals as the central state-owned firms had at founding. Instead, these firms were positioned to generate rent and support budgets for local governments (Li et al., Reference Li, Cui and Lu2014). Such positioning of local state-owned firms was heightened as the administrative and fiscal decentralization in the transition period entailed local governments to be more responsible for their own budget and fiscal health (Wang & Hu, Reference Wang and Hu2001). Being less privileged by the socialist system, state-owned firms founded by local government agencies had to seek for a more favorable social perception in the market-oriented environment to gain more financial support, including going public and raising capital in the HKSE market. Having more non-state directors helped alleviate investor concerns about firms’ corporate governance and accomplish the financial goals of local governments. In addition, adding more non-state directors provided these firms with necessary knowledge, skills, and competences to evaluate management and assess business strategies and operations, and eventually strengthen their performances and competitiveness in the global market (Hillman, Cannella, & Paetzold, Reference Hillman, Cannella and Paetzold2000; Hillman & Dalziel, Reference Hillman and Dalziel2003). Therefore, the peripheral position left a weaker socialist imprint on the state-owned firms founded by local government agencies and these firms were more responsive to the market pressure to have more non-state directors than the firms founded by the central government agencies. I therefore hypothesize:
Hypothesis 2: State-owned firms founded by the central government agencies will have fewer non-state directors on the board than state-owned firms founded by local government agencies.
METHOD
Sample and Data Collection
The sample firms of this study are all P. R. China-registered state-owned firms listed in the HKSE main board between 2005 and 2010 and a total of 107 firms are included. Choosing publicly listed state-owned firms at the HKSE market is mainly because the HKSE market is commonly considered as the most efficient and transparent capital market worldwide. It has more stringent and exhaustive corporate governance and disclosure requirements for listed firms compared to Mainland China's capital markets. In this regard, the HKSE market can better represent a true market environment. Besides, the data from the HKSE market is believed to be more reliable than the data obtained from domestic capital markets in China because it serves investors across the globe and emphasizes information accuracy.
I chose this timeframe between 2005 and 2010 because since March 31st of 2004, HKSE revised the Main Board Listing Rules and required all listed firms to increase the minimum number of independent directors to three, and recommended firms to increase the proportion of outside directors (HKEx Mainboard Listing Rules, 2005). Such revision was a mandate and sent a signal to all listed firms about appropriate governance practices. Thus, choosing 2005 as the starting point, we can better observe the variation in state-owned firms’ board composition when responding to both the market and socialist logics.
My first data source is HKExnews, which is the official dataset of Hong Kong Exchange and Clearing Limited solely for the issuer information. This dataset provides sample firms' annual reports, which contain the information of directors, boards, and supervisory boards. I manually collected each director's biography and coded their government affiliation. The size and change of boards and supervisory boards were also obtained from firms' annual reports. My second data source includes public official records of the sample firms as well as the SASACs (e.g., official websites), where I collected each sample firm's founding dates and identified whether the firm was founded and subsequently managed by central government or local government agencies. Additional information of firms’ founding history and directors’ background has been obtained from media coverage to complement and cross validate the coding of founding owners and directors’ government affiliations. The last one is the Chinese Overseas Listed Company Research Dataset, a sub-dataset of China Stock Market and Accounting Research Dataset (CSMAR). It is compiled by Guo Tai An (GTA), an economics and finance research institute in China. The dataset covers company profiles, share structures, income statements, balance sheets, and trading volume for all Chinese firms listed in overseas stock exchanges including the HKSE market. I mainly used this dataset to obtain the data of control variables.
Variables
Dependent variable
The dependent variable is the board composition of a focal firm in a given year. It is a count variable measured as the total number of non-state directors in a sample firm's board. As discussed earlier, non-state directors, rather than the independent directors in US firms, are the outsiders in state-owned firms. I define state directors as directors who are affiliated with state-owned firms, government and legislative agencies, or government-affiliated entities. The can be EDs, NEDS, or INEDs in the focal state-owned firm. Thus, non-state directors are directors having no such affiliation.
Using the number of non-state directors in the board, instead of the percent, is because the number can better capture the structural variation of the board and the effect of the organizational imprint. Two state-owned firms may have the same percent of non-state directors, but could differ in the number. More non-state directors are able to be involved in a wide range of firm management and have more power to influence the board's decisions in line with the market demands; and this is more likely to be allowed in firms that had less socialist imprint in the first place. This variation in the board structure and imprint effect are reflected by the number of non-state directors, however, cannot be seen from the same percent. To further corroborate the imprint argument, the percent of non-state directors is included in robustness checks.
Independent variables
To test Hypothesis 1, I treat founding period as a dummy variable which is equal to 1 if the focal firm was founded after 1992 when the market logic became dominant, and is 0 if the focal firm was founded in and before 1992 when the socialist logic dominated. The cutting year is 1992 because the former paramount leader Deng Xiaoping made the southern tour of China in 1992 to galvanize the market-oriented reform and the economy rapidly embraced the market logic afterwards. Right after Deng's southern tour, the Company Law was promulgated in 1993 for the purpose of corporatization of traditional state-owned firms and establishment of Western-like ownership structure and corporate governance practices. Following was the large-scale reorganization of the state-owned firms and the issuance of their initial public offerings (IPOs) in the HKSE and other overseas capital markets.
To test Hypothesis 2, position is measured in two ways. To be consistent with prior research that classified state-owned firms into a few categories (Guthrie, Reference Guthrie1997), I code position with three dummy categories. Those founded by central government agencies or the business groups wholly owned by the central SASAC are classified as central government ownership, those founded by provincial government agencies or the business groups wholly owned by the provincial SASACs as provincial government ownership, and those founded by the government agencies at or below prefecture city level or the business groups wholly owned by the prefecture city SASACs as city government ownership. I include central government ownership and city government ownership in the regression analyses, leaving provincial government ownership as the omitted category. Second-level subsidiaries of the business groups wholly owned by the central SASAC are classified as provincial government ownership, because such subsidiaries were not directly controlled by the central government at founding and usually started operation locally. Alternatively, I recode provincial and city government ownership into a new category, local government ownership and code position with two dummy categories. It is equal to 1 if the focal firm is classified as central government ownership, and 0 if it is classified as local government ownership.
It is noted that some state-owned firms underwent a series of ownership restructuring during the transition period and their government owners may change accordingly. To identify the extent of ownership change in the sample, I also collected sample firms’ government owner data between 2005 and 2010 and compared it with firms’ government owners at founding. Three sample firms experienced the change of government ownership since their founding and government owners of the rest remain unchanged. All three firms were devolved from central government ownership into local government ownership, and this allows regression analysis to provide a more conservative test. In this vein, firms' government owners between 2005 and 2010 are not included in the analyses.
Control variables
A number of control variables are included to account for alternative explanations. Firm age is measured by the number of years since the founding of a state-owned firm. It is used to capture firms' socialist imprint in prior studies (Marquis & Qian, Reference Marquis and Qian2014). In this study, some state-owned firms founded in the socialist logic dominant period were restructured in the market logic dominant period. Their governance structures with socialist imprint were disrupted as they became aged. Firm age may not precisely capture the persistent effect of firms' socialist imprint. Board size, measured as the total number of directors sitting on a board in a given year, is considered to affect the monitoring capability of the board. Larger boards may endanger participation and cohesiveness, and are less likely to be independent (Firstenberg & Malkiel, Reference Firstenberg and Malkiel1994). I also controlled for supervisory board size, measured as the count of the number of supervisory directors in a given year, since the supervisory board, though more often symbolic, is designed to oversee the boards of directors and the management in the two-tier board system and may affect the board composition. CEO duality is a dummy variable, equal to 1 if the CEO is not the chairman of the board of directors and 0 otherwise. A CEO holding both positions is seen to have greater formal authority, often hampering board independence (Cannella & Lubatkin, Reference Cannella and Lubatkin1993). Firm performance should influence the likelihood of appointing more outside directors since high-performing firms tend not to change their current board composition. Performance is measured as ROA – return on assets – which represents the efficiency of firm operation. Firm size, measured as the logarithm of annual sales, is controlled in that larger firms are less likely to change their existing governance structures. Industry is controlled using four-digit industry code of the HKSE classification, and I include 19 industry dummies representing the 20 industry categories. To control for a potential time effect, I include year dummy variables in the analyses.
I also controlled for two context-specific variables: controlling shares and H shares. HKSE-listed Chinese state-owned firms offer shares classified as non-tradable state shares, non-tradable legal person shares, and tradable public shares (Haveman & Wang, Reference Haveman and Wang2013). State shares are held by firms' government owners. The SASACs under the jurisdiction of government owners hold state institutional shares, which are categorized into legal person shares, to control their firms.[ Footnote 1 ] Tradable public shares, also known as as H-shares[ Footnote 2 ], are open to foreign individual and institutional investors. This system of split ownership (partly state, partly non-state) enabled state-owned firms to raise capital while maintaining state control (Fligstein & Zhang, Reference Fligstein and Zhang2009). Controlling shares are measured by the percentage of non-tradable state and state institutional shares in total shares. It is included because contemporary state control may affect board composition and the conversion of non-tradable shares to tradable shares in the 2000s may result in reduced controlling shares. H shares, measured by the percentage of publicly traded number of H-shares in the total shares issued by the focal firm, is controlled for temporary market volatility.
Analyses
The unit of analysis of this study is the state-owned firm, and the unit of observation is the firm-year. I used generalized estimating equations (GEE) to conduct the regression analysis in the reported models, given that the panel data has repeated measurements for each sample firm from 2005 to 2010. GEE models are well suited for the analysis of longitudinal data because this estimation technique can increase the efficiency and consistency of estimators while controlling for potential autocorrelation and heteroskedasticity in the data, and allows the dependent variable to take on a wide range of forms (Liang & Zeger, Reference Liang and Zeger1986; Zorn, Reference Zorn2001). For GEE models, a distribution of the dependent variable, a link function, and a correlation structure need to be correctly specified. Since the models predicted the count of the number of non-state directors, I specified a log link function and a Poisson distribution for the dependent variable. All models presented below used an independent correlation structure.
RESULTS
Table 2 displays descriptive statistics and a correlation matrix for all variables. Table 3 presents the results of GEE analyses predicting the number of non-state directors on the boards of listed Chinese state-owned firms. Model 1 in Table 3 includes all control variables as the baseline model.
Table 2. Descriptive statistics and Pearson correlations
Table 3. GEE results of predicting the number of non-state directors
Notes: p-values in parentheses.
*** p < 0.01; ** p < 0.05; * p < 0.1.
Hypothesis 1 predicts that state-owned firms founded in the market logic dominant period tend to have more non-state directors on the board. Model 2 includes firms’ founding period to test Hypothesis 1. State-owned firms founded after 1992 has a significantly positive coefficient (p < 0.01). The expected number of non-state directors will increase by a factor of 1.817 with a 95% confidence interval (CI) of (1.546, 2.136) when state-owned firms are founded after 1992. It suggests that state-owned firms founded after 1992 follow the market logic and tend to have more non-state directors than those firms founded in and before 1992. The results provide strong support to Hypothesis 1.
Hypothesis 2 predicts that state-owned firms founded by central government agencies tend to have fewer non-state directors on the board than state-owned firms founded by local government agencies. Model 3 includes the position to test Hypothesis 2. It is operationalized by the three dummy categories with city government ownership as the omitted category. The coefficient of city government ownership is not significant, but the central government ownership has a significantly negative coefficient (p < 0.01). The results suggest that the number of non-state directors of state-owned firms founded by city government agencies are not significantly different from that of state-owned firms founded by provincial government agencies, but state-owned firms founded by central government agencies tend to have fewer non-state directors than state-owned firms founded by provincial government agencies. The expected number of non-state directors will decrease by a factor of 0.647 with a 95% CI of (0.547, 0.766) when state-owned firms are founded by central government agencies. Hypothesis 2 is corroborated. The results are the same in Model 4, when both independent variables are included.
Model 5 uses an alternative measure of position with two dummy categories to test Hypothesis 2. Central government ownership still has a significantly negative coefficient (p < 0.01). The expected number of non-state directors will decrease by a factor of 0.651 with a 95% CI of (0.551, 0.770) when state-owned firms are founded by central government agencies. It indicates that state-owned firms founded at the center of the economy have stronger socialist imprint and tend to have fewer non-state directors than those founded at the periphery. The results also lend strong support to Hypothesis 2. The results are the same in Model 6, when both independent variables are included.
It is noted that firm age has a positive effect on the number of non-state directors in Model 2, 4, and 6. The results confirm the confounding effect of using years to capture the persistent effect of socialist imprint, because firm age covers the market logic dominant period when some firms' governance structures were disrupted. Besides, the size of a supervisory board has a significant negative effect on the number of non-state directors in three models. It is shown that larger supervisory boards tend to perform their duties less effectively and allow fewer non-state directors to sit on boards. The result is consistent with the findings of the extant literature that supervisory boards of state-owned firms are always considered as symbolic. The negative effect of firm size also confirms that larger firms are less likely to adopt new board structure. In addition, CEO duality has no significant effect on the number of non-state directors since most firms in the sample have separate chairman and CEO. ROA and the percent of controlling shares and H shares also do not have significant effects, and a potential explanation is that board independence may be partially symbolic.
Robustness Checks
I conducted several robustness checks following the analyses. To reduce concerns about multicollinearity, I obtained variance inflation factors (VIFs) for each model in Table 3. These VIFs were all less than 10 and I am fairly confident that multicollinearity does not threaten the validity of my findings. Regarding concerns that compound symmetry might be present in the data, I re-tested the hypotheses specifying an exchangeable error correlation structure in the GEE models, and the results did not change. Besides, to corroborate the imprinting effect, I re-tested the hypotheses using the percent of non-state directors on the board as the alternative dependent variable. For this continuous variable, I specified an identity link function and a Gaussian distribution for GEE modeling. Table 4 presents the results. Still, state-owned firms founded after 1992 have a significantly positive coefficient and those classified as central government ownership have a significantly negative coefficient. It suggests state-owned firms founded in the market logic dominant period tend to have a higher percent of non-state directors on the board and those founded by the central government agencies tend to have lower percent of non-state-directors on the board. The results also corroborate the imprint argument.
Table 4. GEE results of predicting the percent of non-state directors
Notes: p-values in parentheses.
*** p < 0.01; ** p < 0.05; * p < 0.1.
DISCUSSION
In this study, I explore how organizational imprint influenced state-owned firms’ response to both socialist and market prescription for board composition. I found that state-owned firms founded in the market logic dominant period tend to be more responsive to the stock market prescription and have more non-state directors because they were organized around the market logic and attuned to shareholders’ demands for legitimacy. On the other hand, state-owned firms founded in the socialist logic dominant period tend to be less responsive to the stock market prescription since they were born to be governed by state directors and such strong socialist imprint persisted against the market demand for more outside directors. It is also found that state-owned firms founded by central government agencies tend to have fewer non-state directors because they were positioned at the center of the socialist system at founding to accomplish the strategic goals of the central government and had to be tightly controlled through the boards dominated by state directors accountable to central government agencies. State-owned firms founded by local government agencies were created at the periphery of the socialist system to generate rent for local governments and more receptive to non-state directors who could provide critical financial and management support.
This study contributes to the institutional literature by examining one of the key organizational attributes – organizational imprint – that influences the organizational response to multiple institutional pressures. Extant institutional research treated organizations as synchronic entities and homogeneous passive recipients of multiple institutional pressures. While recent studies explored organizational size and institutional linkages that influenced organizational responses to incompatible institutional demands, other key attributes that mirror the developmental history of organizations have not been fully understood. This study chooses to examine organizational imprint as the enabling or constraining factor of organizational actions in the face of institutional pressures and helps improve our understanding of the attribute formed throughout organizational history.
This study also contributes to the imprinting literature by elaborating the founding environment. Firstly, this study shows that organizations are susceptible to the influence of the dominant logic of organizing for certainty and legitimacy and different founding periods are characterized by different dominant logics of organizing that founders rely on. The dominant logic is to be reflected in organizations’ structures and practices, which are institutionalized over time and become the organizational imprint. Thus, organizations founded in the market logic dominant period are more receptive to the governance practices prescribed by the market logic, while organizations founded in the socialist logic dominant period are inclined to preserve entrenched structures and resist change. Moreover, this study also extends existing works on imprinting by examining the structure of the founding environment – organizational positions in the institutional field – and demonstrates the influence of organizational imprint left by the central and peripheral positions. It is found the central position may leave organizations with stronger imprints than the peripheral position in the founding environment.
Finally, this study contributes to extant works of comparative corporate governance and enhances our understanding of state-owned firms from emerging economies by providing a more fine-tuned definition of outsider directors. Different from the definition of outside directors in the US context, this study defines outside directors in the state-owned firms as directors without any affiliation with state-owned firms, government agencies, and government affiliated agencies. It reflects the major conflict between the state as the controlling shareholder and non-state investors as minority shareholders in state-owned firms. This is also consistent with the core notion that corporate governance systems are embedded in larger institutionalized context, and that effective practices are contingent on the firms’ stakeholders in the local institutionalized context (Davis & Useem, Reference Davis, Useem, Pettigrew, Thomas and Whittington2002).
Limitations and Future Research Directions
Despite this study extending the organizational imprinting argument, there are certain limitations under consideration. Firstly, my argument has received support from the sample firms, which include all state-owned firms listed in HKSE market, but the sample size could be enlarged. Further studies should consider including more state-owned firms listed in other overseas stock exchange markets, such as the New York Stock Exchange and the London Stock Exchange, to offer a better understanding of the imprint effect. Methodologically, the GEE modelling that uses the Huber-White sandwich estimator can provide better variance estimates of coefficients in large samples than small samples (Henderson, Miller, & Hambrick, Reference Henderson, Miller and Hambrick2006).
Secondly, this study has focused on the effect of organizational imprint, but other organizational attributes may also influence organizational response in the face of multiple institutional demands, such as organizational identity, organizational structure, etc. For example, extant studies on organizational identity have shown the competing identity claims and resulting tensions inside organizations (Pratt & Corley, Reference Pratt, Corley, Bartel, Blader and Wrzesniewski2007) and the institutional identity that organizations claim for gaining legitimacy (Glynn, Reference Glynn, Sahlin-Andersson, Greenwood, Oliver and Suddaby2008). But less attention has been given to the link between organizational and institutional identity. How institutional identity will interact with organizational identity and jointly influence organizations’ perception of and responses to institutional complexity remains unclear. Future research is expected to explore more organizational attributes in this regard.
Finally, recent discussion about ‘state capitalism’ remains controversial (Bremmer, Reference Bremmer2010; Musacchio & Lazzarini, Reference Musacchio and Lazzarini2012). With the growing number of state-owned firms from emerging economies adopting the market logic and associated practices, we may observe more conflicts and tensions inside state-owned firms because the socialist logic encoded in state-owned firms’ structures and routines at founding is incompatible with the newly adopted practices prescribed by the market logic. Would there be a new hybrid logic called ‘state capitalism’? If so, how would firms manage multiple institutional logics or form a hybrid one? Whether and how the presence of non-state directors actually affect board decisions? This may be another attractive area of future research to investigate.
CONCLUSION
Research on institutional complexity has shown that organizations may confront multiple, even conflicting, demands prescribed by different institutional logics in the field, and start to explore the organizational attributes that may affect their responses to the institutional complexity. In this study, I suggest that organizational imprint, which mirrors organizational history, is one of the key attributes and argue that organizational imprint can not only be from the dominant logic of organizing, which founders rely on during the founding period, but also from the central versus peripheral institutional position that an organization possesses at founding. My analysis of publicly traded Chinese state-owned firms in the Hong Kong Stock Exchange market, facing pressures on board independence from both the market and the socialist state, indeed shows that state-owned firms founded in the period characterized by the dominance of the market logic tend to have more outside directors than the firms founded in the socialist logic dominant period. Moreover, state-owned firms founded by central government agencies tend to have fewer outside directors than the firms founded by local government agencies. This article contributes to the literature by showing the key role of organizational imprint in shaping organizations’ strategic actions and extends the imprinting theory by considering the dominant logic of organizing during the founding period and the institutional position an organization possesses at founding.
