Abstract
This article explores how both corporate governance and corporate social responsibility (CSR) can be improved by using insights from complexity theory. Complexity theory reveals that decentralized governance architecture is required for firms to absorb competently the increased intricacies, variety of variables, and objectives introduced by CSR. The current predominant form of centralized governance based on command-and-control hierarchies copes with complexities by reducing data inputs. This approach results in firms reducing their objectives, concerns, and insights about CSR. Firms with a decentralized “network” form of governance architecture are used to illustrate how the data inputs of each manager can be reduced through the decomposition of decision-making labor to improve the capability of the firm to intelligently absorb and manage complexity. Network governance also introduces a division of powers with stakeholders to facilitate information flow and strengthen incentives to manage the enterprise to enhance both shareholder value and CSR.
Keywords
Scholars agree that in recent decades the rate of change has accelerated, and that through globalization and technological advances the world is getting more connected and complex (Drayton, 2009; Paine, 2003; Senge, 2010). In these circumstances it is becoming more difficult to manage organizations and specifically firms, because management requires increased awareness of changes in stakeholder preferences as well as the environment at large (Freeman, Martin, & Parmar, 2007). Many business associations such as the World Economic Forum or the Business Round Table offer programs for CEOs and board members to learn how to deal with increased complexity. They discuss how managers should handle social challenges that heretofore were only of marginal interest to directors or C-suite officials including climate change, social inequity, or poverty. 1
Paine (2003) made the case early on that corporations will need to become more socially aware to deal effectively with increased societal expectations. Indeed, the Financial Times reports that in 2014 almost all Fortune 500 companies engage in some form of corporate social responsibility (CSR).
2
As a result, corporate spending has increased vastly (Forbes)
3
because managers feel a need to safeguard their license to operate, increase societal goodwill, and legitimacy while appeasing public demands for more accountability and transparency. Such engagement with CSR, even within the narrowest economic understanding (Garriga & Mele, 2004) most likely increases the complexity of managerial and governance processes. As Ackermann and Eden (2011) stated, Current experiences suggest that, as organizations become more aware of the needs to develop corporate social responsibility and to respond to regulatory frameworks, the numbers of their stakeholders are likely to increase, raising the complexity of stakeholder management still further. So paying attention to the strategic management of stakeholders is likely to become increasingly important. (p. 194)
As complexity is increasing, good corporate governance seems to matter even more as the global financial crisis along with other scandals (such as Bernard Madoff and Allan Stanford) have highlighted (Adams, 2012; Pirson & Turnbull, 2011a). Governance concerns are increasingly influenced by social and environmental dimensions of business conduct and elevate the necessity of CSR to new heights (Margolis & Walsh, 2003; Senge, 2010). As a consequence, the challenges for global managers have become increasingly complex, and observers describe them as a perfect storm (Hart, 2005; Jackson & Nelson, 2004).
Many thought leaders in management argue that business practice requires an entire paradigm shift to deal successfully with the increased complexity (Hart, 2005; Jackson & Nelson, 2004; Senge, 2010). So far, academic research is struggling to provide guidance on managing the challenges of complexity, namely the increased need to receive and process higher levels of information for adequate decision making. The authors use the word “complexity” here to indicate (a) the variety of firm objectives including financial and social responsibilities, (b) the variety of external and internal information inputs available for executives to evaluate risks and opportunities in achieving such firm objectives, and (c) the ensuing intricacies in the relationships between various firm objectives and the operating environment of the firm that need to be understood and managed by firm executives. Whereas research has increased within the realm of corporate governance as well as CSR, the majority of such research has discussed the development, dimensions, and determinants separately (Aguilera & Cuervo-Cazurra, 2009; Aguilera & Jackson, 2003; Carroll, 1999; Garriga & Mele, 2004; Turnbull, 2000b). Furthermore, very little research has examined the foundation of current ills at the intersection of organizational structures and strategic decision making (Jamali, Safieddine, & Rabbath, 2009). More specifically, the interactions between corporate governance structure and CSR practice have been neglected (Jamali et al., 2009; Kolk & Pinkse, 2010). This neglect is surprising because as the global business environment evolves, the question arises whether current centralized corporate governance models best fit the increased complexity compounded by CSR practices. This question guides this article and the authors argue that complexity-inducing stakeholder engagement enhances the need for adequate information gathering and processing and thereby challenges traditional corporate governance systems designed to deal with stable environments. In this article, the authors therefore suggest that CSR-induced complexity can be better governed through decentralized governance systems.
To demonstrate how decentralized governance systems allow for higher levels of stakeholder engagement and support of CSR, the authors will use insights from complexity theory. Based on complexity theory, the authors will showcase in this article the various shortcomings of centralized governance and engage with the question of how decentralized governance architecture can improve operations given the increased complexity leading to the higher volume of data and information inputs required for CSR. We explore the insights of complexity theory that notes that centralization such as in hierarchies relies on the “reduction” of data inputs and internal processing as dictated by the limited span of control of each manager. However, data reduction aggravates transmission errors up and down a hierarchy. Firms made up of a network of decentralized decision-making modules create a basis to “absorb” much greater data inputs, and internally filter and cross-check data to more reliably and comprehensively discover and manage an increased variety of risks and opportunities to which a firm may be exposed. In addition, the authors explore how the decentralization of power can also facilitate meaningful stakeholder engagement to improve business operations, risk management, stakeholder interests, and CSR. In this way, CSR becomes integrated into corporate governance.
The remainder of the article following this introduction has six sections. The section “The Nexus of Corporate Governance and CSR,” after this introduction, presents the nexus between corporate governance and CSR and how both corporate governance (structural) and CSR (strategic) are responses to the problem of managerial responsibility. In the section “Complexity Theory, Corporate Governance, and CSR,” we outline why engagement with CSR increases the complexity of the governance process and outline complexity theory as pertinent to the study of organizations. In the section “Complexity Reduction and its Problems for Management Responsibility,” we outline how in a more dynamic world complexity reduction approaches such as the traditional, centralized governance architecture (unitary boards or unitary boards with a supervisory board) fail to adequately support monitoring and strategic advice functions concerning CSR. In the section “Complexity Absorption and Managerial Responsibility,” we consequently explore the insights from complexity absorption approaches leading to a decentralized governance architecture, which can better support CSR-related strategy and monitoring. In the section “Integrating Corporate Governance and CSR,” we outline how a decentralized governance architecture with many boards can allow strategically important stakeholders to be constructively included in the governance processes to enhance CSR. In the final section “Concluding Remarks,” we suggest that organizations learn from the transition from authoritarian toward democratic government systems by installing a division of power and checks and balances to respond to increased environmental complexity.
The Nexus of Corporate Governance and CSR
As firms embrace CSR in a variety of formats, they do so often to increase their legitimacy, enhance their reputation, and maintain their license to operate (Garriga & Mele, 2004; Margolis & Walsh, 2003; Poppo & Schepker, 2010). Whereas the need for corporate responsibility is questioned less and less, the predominant argument for such changes in managerial behavior posits an enhancement of long-term profitability (Paine, 2003). Despite a wave of authors claiming the mutual benefits of corporate responsibility to society at large and responsibility to shareholders in specific (Collins & Porras, 2002; Margolis & Walsh, 2003; Ogden & Watson, 1999; Porter & Kramer, 2011), the relationship between corporate governance and CSR remains under-researched (Jamali et al., 2009). The question of how to design the governance architecture to integrate CSR is crucial because corporate governance determines the relevant measures of control and strategy a board is tasked with (Child & Rodrigues, 2004). Some scholars argue that corporate governance is determined by law and regulation (Bebchuk & Weisback, 2010; Bhimani & Soonawalla, 2005; Dunlop, 1998) whereas others consider CSR a more fungible strategic task (Bhimani & Soonawalla, 2005; Porter & Kramer, 2006). Others suggest that CSR is framed by the complexity of institutional arrangements that are predicated by corporate governance codes (Aguilera & Cuervo-Cazurra, 2009; Jamali et al., 2009; Waddock, 2008). A further perspective views CSR as modes of self-governance (Jamali et al., 2009) and differentiates corporate governance as external governance (Bebchuk & Weisback, 2010; Tricker, 1994). Authors such as Turnbull (2002), however, have long suggested that corporate governance can be viewed as self-governance mechanism itself, reducing the need for both external regulation and CSR codes.
Irrespective of such differences, an increasing number of societal challenges arguably increase the complexity of the governance process as they require not only more information to understand them but also additional partners in forms of cross-sectoral collaboration to address them (Margolis & Walsh, 2003). Responding to increased societal expectations (Paine, 2003), many boards progressively focus more on external stakeholders, driving up the complexity of the decision-making process rather than simplifying it. As a consequence, environmental complexity often increases the complexity of governance-related decision processes. The adoption of CSR policies (Garriga & Mele, 2004; Paine, 2003), in general, increases the number of actors, variety, and quantity of data inputs in the governance and strategy decision-making processes. Such a broadened focus seems to conflict with the apparent goal of the corporation, namely shareholder value maximization (Charreaux & Desbrières, 2001; Paine, 2003). These potential conflicts have been reflected in many debates since the rise of the manager-led corporation (as an example, see K. Davis, 1973; Friedman, 1970). Whereas this conflict seems most prominent in the Anglo-Saxon tradition of corporate governance, it has permeated the discourse in many cultures (Charreaux & Desbrières, 2001).
The common denominator of the discussion concerning corporate governance and CSR is the problem of managerial opportunism (Carroll, 1999; Jensen & Meckling, 1976) and how to deal with it given ever-increasing decision-making complexity (Paine, 2003). One strategy has focused on establishing corporate governance systems and codes (Organisation for Economic Co-Operation and Development [OECD], 2001); the other strategy has been based on adopting responsible products and services (Paine, 2003).
Corporate Governance as Structural Response
With the rise of the manager-run corporation, the problem of managerial responsibility rose to prominence (Khurana, 2007). Corporate governance is often understood as “the system by which companies are directed and controlled” (Cadbury, 1992, p. 14). Many scholars argue that the role of corporate governance is to manage constructively the conflict of interests between the shareholders who are described in the literature as “principals” and the managers who are described as the “agents” to create what is described as “the principal–agent” problem (Berle & Means, 1968; Jensen & Meckling, 1976). As CSR increases, the complexity of management information it exacerbates further information asymmetry to facilitate managerial opportunism at the expense of the owners (Berle & Means, 1968; Jensen & Meckling, 1976). Although there have been other notions of corporate governance, such as the stewardship notion (J. H. Davis, Schoorman, & Donaldson, 1997), the agency perspective has largely dominated the discussion (Blair, 1995). According to the agency perspective, the interest of managers can be structurally aligned with owners by issuing them with shares and/or options (Hambrick & Jackson, 2000; Hawley & Williams, 2000). It is argued, however, that such shared ownership compensation schemes have led to even more managerial opportunism hurting shareholders in the long run (Cadbury, 1992; Kirkpatrick, 2009). After various crises (Enron, global financial crisis), it became obvious that such structural responses were insufficient to ensure managerial responsibility. As a consequence, the effectiveness of internal agency-theory-driven solutions as well as external regulation has been questioned (OECD, 2009; Pirson & Turnbull, 2011b). The authors argue that increasing environmental complexity generally requires corporate directors and managers alike to pay attention to more than just the classical principal–agent relationship.
CSR as Strategic Response
Another, complementary response to the problem of managerial responsibility has been developed on the strategic level, namely CSR. Although there has been a long ongoing theoretical debate about the role of social responsibility in theory, business practice has long adopted CSR in some way shape and form. Although it is debatable which form of CSR companies pursue, it is almost always a form of increasing legitimacy with societal stakeholders that heretofore were not considered critical to business success. CSR as a concept equally has its roots at the beginning of the 20th century with the rise of the corporation (Hoffman, 2007). It is also a result of increased corporate influence and power over time. With decreased regulations and more open markets post–Cold War, CSR has become an umbrella term for the need to establish increased managerial responsibility, largely through enlightened strategy (Carroll, 1999; Paine, 2003). CSR gained increasing visibility after various corporate scandals that increased the public’s concern with corporate behavior overall (Paine, 2003). Different approaches to CSR have emerged. Garriga and Mele (2004) classified them as instrumental, political, integrative, or ethical approaches. At its core, CSR is expressed by the strategic choices of a company with regard to social, political, legal, and normative matters in a given location of operation (Garriga & Mele, 2004; McWilliams & Siegel, 2001). As such, CSR strategy complements the structural quest to establish managerial responsibility (Garriga & Mele, 2004).
Complexity Theory, Corporate Governance, and CSR
Whereas there is an established literature on corporate governance and a burgeoning literature on CSR, there is less research examining the connection between the two concepts (Jamali et al., 2009). As environmental complexity is increasing, we argue that both concepts can be connected via the lens of complexity theory. Complexity theory has developed over the last half century, inspired by the complexity of biological structures and process and is now applied in a variety of fields. Complexity theory according to Thrift (1999) is an economy of concepts based on an “emergent or self-organizing impulse, usually involving . . . non-linearity, self-organization, emergent order and complex adaptive systems” (Jencks, 1995, p. 34). Complexity theory aims to predict, explain, and control such complex systems (Chu, Strand, & Fjelland, 2003).
Complexity theory has roots dating back to the early 20th century (Ashby, 1956; Weiner, 1948). Complexity arises from systems that according to Chu et al. (2003) possess radical openness and contextuality. Scholars convincingly argue that organizations including businesses display openness and contextuality, especially in a globalizing world (Anderson, Meyer, Eisenhardt, Carley, & Pettigrew, 1999; Chu et al., 2003; Fontana & Balati, 1999). From the perspective of complexity theory, organizations are treated as instances of adaptive systems (Boisot & Child, 1999), that is, systems that have to match in a non-trivial way the complexity of their environment. This matching cannot be achieved by a centralized control system that by its nature cannot provide a matching richness of variety to cope with complex data inputs and decision making.
Complexity theory applied to organizations allows further specific insights. As Boisot and Child (1999) argued in line with Giddens (1984) and Weick (1969), organizations can be viewed as interpretative systems that first create and then objectify the world through structuration. However, opportunities for varying the structure of hierarchies are by their definition limited. Decentralized modular organizations can be conceived as loosely coupled systems with many combinatorial possibilities. Members of hierarchical organizations are forced to make sense of complexity by reducing both the number of agents and the variety of data inputs that influence their behavior (complexity reduction). The shareholder value maximization paradigm also further achieves complexity reduction by having only a single firm objective. As a result, potential risks and opportunities, as well as other major interests and concerns, could be factored out.
Another way members of an organization can cope with complexity according to Boisot and Child (1999) is to adopt cognitive structures that widely distribute data inputs and so simplify the inputs of each member and so the organization as a whole can absorb complexity through structural design (complexity absorption); see Figure 1. Boisot and Child (1999) further suggested that in different cultures, different approaches to dealing with complexity have emerged. They posit that the Chinese have historically dealt with complexity by absorbing it through dense interpersonal linkages within tightly bounded communities (Anderson et al., 1999). Western companies in contrast have usually decided to reduce complexity through, heuristics, rule systems, codification, and legal institutions.

Schematic differences of complexity reduction and complexity absorption approaches.
Extending complexity theory to the notion of corporate governance is timely as increased complexity affects not only managers but also board members who need to advise and oversee management as well. Its theoretical proscriptions are especially relevant when treating the question of how to design governance systems for enhanced CSR. Concerns for CSR tend to make governance decisions more complex as they introduce more variables and input factors, which in turn create more uncertainty in the decision-making process. Complexity theorists describe two basic ways an organization can decide to deal with increased environmental complexity. One approach is labeled complexity reduction through a centralized hierarchical structure that leads to simplification of management, with a very linear focus on one goal such as shareholder value maximization. A second approach is labeled complexity absorption and allows organizations to absorb information in non-linear ways through network relationships. We argue that what drives the organization from the first approach to the second approach is the governance structure.
Complexity Reducing Approaches of Corporate Governance and CSR
The prevailing conceptual basis for corporate governance and CSR is driven by the attempt to reduce complexity of (a) human behavior, (b) organizational structure, and (c) the interaction of both in the decision-making process (Dierksmeier & Pirson, 2010). Human behavior is modeled on assumptions from economics (“homo oeconomicus”), organizational structures are command-and-control-based hierarchies, and decision making is geared to fulfill shareholder value imperatives. The dominating approach to corporate governance, agency-theory, is based on a reductionist perspective of a self-interested, amoral, utility maximizing individual that needs to be kept in check and monitored to enact managerial responsibilities toward the benefit of shareholders (Dierksmeier & Pirson, 2010; Jensen & Meckling, 1976; Pirson & Lawrence, 2010). The most cost-efficient organizational structure then is command-and-control based and hierarchical. The resulting governance system is equally hierarchical and centralized, reflected by a centralized board structure of a single board or an executive board with a supervisory board. This approach oftentimes emphasizes strong monitoring activities, relies on tying executive remuneration on quantifiable performance measures, and focuses on shareholder value creation as the central prerogative.
CSR within this complexity reducing approach must reduce its scope and concerns with management driven by shareholder value creation. As such, CSR is reduced to the economic aspect of the interactions between business and society, such as by ways of branding (Garriga & Mele, 2004). Along this approach, the corporation’s purpose is to maximize shareholder value, which can occur in an enlightened manner that includes any advantageous CSR activity (Friedman, 1970; Garriga & Mele, 2004; Jensen, 2002).
Complexity Absorbing Approaches to Corporate Governance and CSR
Complexity absorbing approaches to corporate governance and CSR accept the complexity of (a) human behavior, (b) organizational structure, and (c) the interaction of both in the decision-making process. This alternative approach rejects the simplistic economic view of human nature as plainly wrong (Mintzberg, Simons, & Basu, 2002), and instead views the individual as a “zoon politicon”—a relational (wo)man, who materializes freedom through value-based social interactions (Mele, 2008). People, accordingly, do not have fixed preconceived utility functions; rather, their interests, needs, and wants vary and take shape through discourse and continuous exchange with the outside world (Pirson & Turnbull, 2011b). Organizations in which they work are organically shaped toward the needs of all stakeholders required to provide the necessary support for organizational success (Freeman, Wicks, & Parmar, 2004). The decision-making process is supported by a structural design based on complexity absorption. This complexity absorption is achieved through the decomposition of decision-making labor, which mimics the distributed intelligence architecture of the human brain (Lawrence, 2007). In organizations, such complexity absorption designs provide a checks-and-balances structure to allow stakeholders to influence the decision-making process (see also Gratton, 2004). Managerial responsibility is thus facilitated by a governance structure that systematically reduces managerial opportunism.
CSR can similarly be viewed as being dependent upon complexity absorption via strategy, rooted in that enlarged vision of human nature that does not support command and control only, but where responsibility, including managerial responsibility, is part of a successful human interaction (Dierksmeier & Pirson, 2010). The political, integrative, and ethical approaches suggested by Garriga and Mele (2004) embrace that view and allow for CSR beyond the business case transcending the narrow, economic rationale:
Representatives of the political approaches to CSR argue that strategies should account for higher levels of complexity. They posit that the corporation by sheer facticity of power has become a political actor, which leads to certain responsibilities beyond the economic imperative (Scherer & Palazzo, 2007). This perspective leads to the notions of corporate citizenship. In line with this perspective, the political approach bestows the corporation with political right as well as social duties that need to materialize in its organizational architecture. In line with this perspective then, managerial responsibility is a result of managerial personhood and reflected in stewardship perspectives of corporate governance (J. H. Davis et al., 1997; Muth & Donaldson, 1998).
Representatives of the integrative approaches imply that environmental complexity needs to be mirrored in CSR approaches. They argue that the corporation depends on society and, therefore, managerial action needs to create social legitimacy (Carroll, 1991; J. H. Davis et al., 1997; Garriga & Mele, 2004; Werther & Chandler, 2011). As such, managers should take into account social demands and integrate them in a way that reflects social values in business operations. Such an approach is contingent on the societal values and could vary according to culture and stakeholders. Stakeholder management is often viewed as a way to integrate such values in actual managerial decision making (Donaldson & Preston, 1995; Freeman, 1984). Managerial responsibility then includes the ability to understand the various stakeholder interests and their integration that allows business to operate. As such, the corporate governance perspectives based on a stakeholder view fit well with this perspective on CSR (Spitzeck & Hansen, 2010; Turnbull, 2000c).
Proponents of the ethical approaches similarly argue against the complexity reduction approach, and posit that the relationship between corporations and society rests on ethical values (Dierksmeier & Pirson, 2009; Sison, 2008). These values are universal in nature and do cut across societies. As such, managerial responsibilities require an understanding of such universal values and the ability to implement them within varying societal contexts that may or may not cherish such an approach. This approach thus views managerial responsibilities first and foremost as ethical responsibilities not operational tasks (Garriga & Mele, 2004). Such perspectives are represented by the UN Global Compact and the reference to universal human rights, and the common good approach promoted by catholic social teaching, as well as the sustainable business perspective adopted by the United Nations (Garriga & Mele, 2004).
Ultimately, the three latter approaches to CSR acknowledge an increase in environmental complexity for firms; the authors argue that these approaches cannot adequately be used following a centralized complexity reduction approach.
Complexity Reduction and Its Problems for Management Responsibility
At the nexus of corporate governance and CSR rests the concern for managerial responsibility and the potential neglect of stakeholder interests, including shareholder interests, in the corporate decision-making process. The notion of managerial responsibility implies a notion of managerial freedom to choose an option from a variety of options (Dierksmeier & Pirson, 2010). It also implies semantically the ability to respond (response-ability) toward various stakeholders including shareholders (Pirson & Kimakowitz, 2014; von Kimakowitz, Pirson, Spitzeck, Dierksmeier, & Amann, 2010). Consequently, the ability to respond to various stakeholder claims (including normative questions) rests upon the ability of the board and management to deal with higher levels of complexity (Galbraith, 1974; Pirson & Turnbull, 2011a; Premkumar, Ramamurthy, & Saunders, 2005; Rost & Osterloh, 2010).
Scholars have consistently argued that complexity reduction leads to decreased performance in more complex environments (Hambrick, Werder, & Zajac, 2008; Jamali et al., 2009; Tricker, 1994). In addition, such complexity reduction in the area of corporate governance and CSR may very well lead to inadequate execution of managerial responsibility. Such failures of managerial responsibility can be facilitated by (a) a reductionist governance structure and (b) a reductionist CSR strategy.
Complexity Reduction, Structure, and Corporate Governance
Complexity reduction with centralized governance can lead to loss of important information in complex environments. Managerial responsibility in environments where multiple stakeholders demand accountability requires matching complexity in governance support. Such support requires (a) monitoring and advice functions or (b) external accountability to stakeholders and internal supervision as well as (c) long-term strategic thinking, (d) setting internal corporate policy as well as (e) appointing and remunerating the chief executive (Tricker, 1994). A single board that is adequate in stable environments will provide only limited CSR support to management with regard to strategic advice and monitoring, as it will have difficulty to predict or control the vast variety of stakeholder actions.
An example that highlights the challenges for a unitary board advising and monitoring managerial action in a more complex environment (such as increased media attention) is provided by Nike when accused of using child labor (Beder, 2002). For a consumer product firm such as Nike, brand equity is key to success. Board members are faced with the task of managing risks to the brand from poor CSR. Nike was criticized in the 1990s for the lax labor standards in its suppliers’ factories, of which there were more than 650 (Beder, 2002). In response to the ongoing criticism, Nike formulated a code of conduct for its contractors, which affected more than 500,000 workers around the globe. The code, first formulated in 1992 and amended in 1997 and 1998, was supposed to apply in all factories producing Nike products. It included recommendations for minimum wages (as set in the host country), maximum mandatory working hours of 60 per week, a minimum age for workers of 16 years old, a ban on forced labor, and minimum safety and environmental standards (Beder, 2002).
Even if Nike engaged only in whitewashing as its many critics allege, the board could not have monitored the CSR standards in the numerous supplier factories Nike had contracts with. Despite the availability of information regarding CSR violations in the supply chain, the board could also not adequately ensure standard compliance. Ashby (1956), a pioneer of complexity theory, has shown the impossibility of many variables being controlled centrally (see also Pirson & Turnbull, 2011a). Ashby (1956) argued that centralized monitoring and/or control systems cannot possess the capability to reliably and comprehensively either identify or control code violations. In particular, when considering the capabilities of a unitary board, Mitchell 4 states, “boards are overwhelmed, overscheduled, undereducated and often uncoordinated in addressing key concerns of the enterprise and its stakeholders.” 5 Carter and Lorsch (2003) pointed out in their synopsis “that boards are being pressed to perform unrealistic duties given their traditional structure” (p. 1). 6 Unsurprisingly, many more violations to Nike’s code were detected and led CEO and Chairman Phil Knight to concede that “The Nike product has become synonymous with slave wages, forced overtime and arbitrary abuse” (Herbert, 1998).
Complexity Reduction, Strategy, and CSR
In addition to the inherent problems of centralized governance, reductionist approaches to strategy can add to the troubles of managerial responsibility. Strategy-based complexity reduction that relegates CSR to branding or a means for short-term shareholder value maximization can also lead to compromising outcomes.
British Petroleum (BP) had embraced CSR as a strategic imperative in the 1990s under the leadership of Lord Browne (Arena, 2004). A rebranding of BP to “beyond petroleum” was intended to highlight higher levels of CSR. According to scholars, this move paid off handsomely, but was also often criticized as green-washing (Cherry & Sneirson, 2011). Arguably BP had intended to complexify its strategic aims by serving a variety of stakeholders. Many observers, however, attribute the failures leading up to the Deepwater Horizon explosion in 2010 to the prevailing reductionist approach of cost cutting and short-term shareholder value creation (Chazan, Faucon, & Casselman, 2010).
The branding of the company was not reflected in its strategy, which resulted in ongoing failures of managerial responsibility. According to an internal BP report in 2007, there were 10 near-miss incidents in the Gulf of Mexico attributable to shareholder primacy concerns. As stated in the report, a “common theme . . . was a failure to follow BP’s own procedures and an unwillingness to stop work when something was wrong” (quoted in Chazan et al., 2010, p A1). In 2007 alone, BP paid US$373 million in settling lawsuits that arose from the Texas City refinery explosion, oil spills along BP’s pipeline in Alaska, and the allegation that BP traders were manipulating the market for propane (Cherry & Sneirson, 2011). Safety concerns were continually discarded despite the high risk any incident would cause for the brand image.
The reductionist strategic approach to dealing with the many stakeholder concerns lead to the near demise of the company as a whole. The overriding concern with cost reduction led to neglect of employee safety and the resulting environmental impacts. During the Deepwater Horizon explosion, 11 employees were killed, and the resulting oil spill impacted entire communities along the Gulf coastlines, whose livelihoods depended on fishery. As a consequence, even shareholders were severely affected because the oil spill resulted in a drop in BP’s credit rating, downgraded by rating agencies such as Moodys, Standard and Poor, and Fitch. BP’s shares fell significantly (by up to 45%) and BP had become a target for takeovers (Logendran, 2010). This result provides evidence that a reductionist CSR strategy in a complex environment, however well intentioned, does not reliably enhance managerial control and responsibility.
Complexity Absorption and Managerial Responsibility
The insights of complexity theory and the science of corporate governance (Turnbull, 2002) explain why the reductionist approach as used in centrally governed organizations fails to reliably and comprehensively cope with complexity. As a result, the authors argue to consider the complexity absorption approach as a way to establish managerial responsibilities to shareholders as well as stakeholders in a broader sense. Such complexity absorption can be achieved through either structure or strategy, ideally as a combination of both. In the following discussion, we present generic ways in which structures and strategies can help absorb complexity to improve managerial responsibility and both CSR and corporate governance together.
Complexity Absorption via Complexified Governance Structures
Historically, Western societies have replaced centralized national governance with decentralized governance that introduces a division of powers with democratic checks and balances to ensure responsible governance. Democratic, decentralized structures were deemed superior in dealing with the increased complexity of citizen demands. Similarly, to absorb increased complexity of the business world in the early 20th century, corporations complexified managerial structures by introducing divisional structures (Williamson, 1985). A global survey by Bernstein (1980) revealed without exception that sustainable stakeholder controlled firms had all introduced a division of board powers to decompose decision making into different boards. The sustainable competitive advantage obtained by firms governed by many boards has been demonstrated by VISA International Inc. in the United States, the John Lewis Partnership in the United Kingdom, and the Mondragón Corporacion Cooperativa (MCC) in Spain (Turnbull, 1995).
Complexity Absorption via Division of Labor
A number of organizations have experienced troubles due to a reductionist governance approach and have shifted toward a more complex governance structure to better absorb the environmental uncertainties.
Shell Oil, for example, was seen as leading proponent of CSR by the mid-1990s and had arguably taken a lead role in developing corporate environmental relationships (Ledgerwood, 1998). However, that did not help Shell’s board to prevent a variety of CSR-related scandals. With regard to the disposal of an oil rig, the Brent Spar, the board of Shell UK commissioned a scientific, economic, and environmental study pertaining to the eventual disposal of the Spar, and concluded that deep sea disposal was the best option (Zyglidopoulos, 2002). Unfortunately, the board of Shell UK forgot to take into consideration the social, political, and public effects of the planned disposal leading to major boycotts and decreases in revenue. In the Brent Spar case, Shell’s board was not aware of the complex ramifications of its decision despite its willingness to support responsible decision making overall. What Shell Oil learned was that it had to complexify its governance structure in addition to its strategy, by including stakeholders in the decision-making process. As a consequence, Shell installed advisory councils of different stakeholder groups to increase the variety of data inputs for decision making and obtain feedback from its stakeholders affected by its decisions. This type of decomposition of decision-making labor is fundamentally different from establishing sub-committees within a command and control hierarchy. The difference is an increase in the sources and variety of data inputs with increased decision-making centers and cross-checking communication and control channels as required by the insights of Ashby (1956) and Shannon (1948). Sub-committees that can often be found on centralized boards do not necessarily increase the variety of sources of data inputs or introduce feedback from stakeholders who are affected by the decisions. Also, sub-committees do not introduce additional decision makers to reduce the complexity of data processing required by the individuals concerned.
Complexity of the various governance tasks can further be absorbed by introducing multiple boards, specific to the various governance tasks (Jones, Hesterly, & Borgatti, 1997). Tricker (1994) specified five functions for unitary boards, being (a) external accountability to stakeholders, (b) internal supervision, (c) long-term strategic thinking, (d) internal corporate policy, and (e) to appoint and remunerate the chief executive.
An analysis of the MCC architecture by Turnbull (1995) revealed that a separate board carried out each of the five functions identified by Tricker in each firm of the MCC network. This approach is illustrated for one firm in Table 1 that shows how all members of the firm are involved in making up the membership of the five boards that are collectively referred to as a “Compound Board.” The MCC is a network of network-governed employee and other stakeholder controlled firms with total workers in more than 200 firms exceeding 80,000.
Mondragón Compound Board Compared With a Unitary Board.
Note. Degrees of decomposition of data processing labor indicated by allocations of “X” in the table.
Omits the General Assembly, which elects Watchdog Council and Supervisory board.
Descriptions follow typology of Tricker (1994, pp. 244, 287) with typical number of people involved in each board.
Each component of the compound board in a Mondragón firm is constituted in a different way to undertake different functions and activities indicated in Table 1. The supervisory board is elected by employees and their strategic stakeholders. It determines long-term strategic direction with engagement and feedback to and from stakeholders of the enterprise. It also appoints the management board, whereas the social council determines working conditions, welfare, and profit-sharing policies. Work-units of 10 to 20 workers determine relative wage rates, and appoint a delegate to the Social Council. A watchdog board of three people provides oversight on the efficacy and integrity of governance processes including the control of the auditor. Finally, the management board orchestrates the efficient allocation of resources (Turnbull, 1995).
Whether board tasks are split up according to function or geography (see examples of VISA International, Raiffeisen Banks, and Red Cross International), in both ways decision-making labor is decomposed, strategic stakeholders are involved, and greater complexity is absorbed by the firm while reducing the complexity for the individuals involved through the reduction of function and activities, and thus data input and processing. Mondragón and Raiffeisen Banks have long been viewed as exemplary in their CSR practice while generating billions in revenue and remaining profitable over decades. Eckart (2005) and Moellner (2010) suggested, therefore, that any business aiming to improve its CSR can learn from their governance architecture (see also Turnbull, 1995).
Complexity Absorption via Division of Power
As a consequence of the Gulf Oil Spill, BP tried to complexify its structure by installing a central safety and operational risk organization directly reporting to the CEO (Dudley, 2011). Even if that new organization obtains the capacity to provide better safety, the board will not be able to directly monitor, as reporting will still go through the CEO. Nike similarly tried to complexify its structure by installing an additional committee dealing with matters of CSR on the board level. The CSR committee is tasked to review Nike’s activities and make recommendations to the board of directors. 7 Even though there is arguably more division of labor, the power to act remains centralized.
The problem of BP and Nike was not that internal violations were unknown. Rather executives and the board did not do anything about them and were not incentivized to do so. BP and Nike complexified their structures within a centralized governance system. Such structural changes are comparable with a dictator changing or adding some advisors. Centralized governance gives rise to self-serving decision making, and without external feedback data leads to poor risk management (see Turnbull, 2009). In most advanced, democratic societies, the executive branch is checked and controlled by a legislative and judicative body informed by independent press that may communicate contrary viewpoints. No such division of power with communications independent of the power of a board is systematically found in corporations with centralized governance.
Executives are commonly only subjected to being checked by other executives and their friends (Monks & Sykes, 2002). Centralized governance can be efficient and effective when objectives and concerns are also simple as promoted by Jensen (2002). However, the complication introduced by the many objectives and concerns of CSR requires a complex governance architecture to reduce the complexity that individuals need to manage within their operating limits. For example, when CSR becomes a strategic goal as, for instance, stated and symbolized by BP’s rebranding, there needs to be governance structure in place that can ensure managerial responsibility to all stakeholders.
If political history is any guide, a separation of powers can help to ensure effective CSR. Political scientists have long explored how stakeholder-oriented governance can be supported through a division of power. Most prominently, the political philosopher Montesquieu proposed systems of checks and balances based on three pillars (executive, legislative, and judicial branches) to ensure responsible governance. Corporate governance arrangement could emulate such governance architecture by including systemic checks and balances in its governance architecture. The John Lewis Partnership for example, has given itself a constitution and emulates the checks-and-balances system of a democratic governance system (John Lewis Partnership, 2010b). The John Lewis Partnership is one of the largest U.K. retailers owned and controlled by 69,000 employees operating major department stores and super markets (John Lewis Partnership, 2010a). It has “Committees of communication” and “Registrars” for any employee to provide feedback to “Branch Heads,” “Branch Councils,” “Central Council,” “Executive Committee,” “Board of Directors,” and the “Trustees.” All such officers are directly or indirectly elected or appointed by the employees. Letters from any employee can be published anonymously in the internal Gazette or Chronicle with the power to petition a response. Given this complexified structure that allows for a division of power, it is interesting to note that the John Lewis Partnership has long been held up as role model in terms of CSR as well as sustainable economic growth (Idowu & Louche, 2010; Yudelson, 2009).
Complexity Absorption via Complexified Strategy Development
In addition to structural complexification, managerial responsibility can be ensured by a complexity absorbing strategy development process. CSR, by definition, supports the development of stakeholder-oriented strategies. To absorb complexity of the environment, it is possible to include strategic stakeholders in the development process of organizational strategy as is illustrated by the MCC that include suppliers, customers, and community representatives in their compound board. In fact, several companies, including Hewlett-Packard and Shell, are already using stakeholder councils in strategic advisory functions (Spitzeck & Hansen, 2010). Furthermore, ESA, the electric safety agency in Canada, uses stakeholder councils to help the board in the strategic decision making (Electrical Safety Authority, 2009).
However, although advisory councils enrich the variety of data, issues, and concerns to enhance the data inputs, the result is an increase rather than reduction of complexity for decision makers. Complexity absorption only becomes possible when decision-making power becomes shared to create distributed intelligence to reduce the data inputs, concerns, and roles of individual decision makers. Complexity absorption becomes dependent upon the constitution of organizations introducing a network of boards to avoid the corruption of power and by introducing checks and balances on managers to force them and/or incentivize them to become socially responsible. It is by amending the corporate constitution of organizations to introduce an appropriate architecture of network governance that CSR can be integrated into corporate governance as is next considered.
Integrating Corporate Governance and CSR
The integration of CSR into the governance architecture of a firm is not only dependent upon changing the constitution of a firm but also to do so in a way that promotes and sustains its operations in its environment. This relationship means constitutional changes need to be custom designed. Figure 2 can only provide a generic illustration.

Generic integrated governance structure.
Figure 2 indicates how the complexity of director data processing can be reduced by separating the power to manage the business from the powers to govern. This separation also removes serious conflicts of interests identified by Turnbull (2009) while introducing a division of power to create checks and balances to facilitate the participation of a wide variety of stakeholders. In this way, increased data inputs can be absorbed, processed, and acted upon by being distributed among as many additional people that may be required to manage the complexity as indicated in Table 1. Just as importantly, this approach allows the integration of management with governance activities and thus the integration of CSR with corporate governance.
As indicated in Figure 2, this integration can be achieved through each strategic stakeholder constituency establishing a separate board. This structure minimizes differences within each type of stakeholders to focus the common interests of their respective constituencies. In this way, conflicts of interest between different types of stakeholders can arise within the firm to promote negotiations for continuous improvements to facilitate win–win solutions. As argued by Pound (1992, 1993), internal competition for power status and influence within firms provides a more general and less costly way to enhance firm performance than competition for control through a stock market.
Management Board and Board of Governors
Figure 2 illustrates one of many ways to separate the power to govern a business from the power to manage the business. The configuration should not be confused with a European supervisory board that appoints the management board. This European approach maintains a centralized form of governance. In some European jurisdictions, such as Italy, shareholders appoint an audit board as well as a supervisory board that then in turn appoints the executive board. However, in the case of Parmalat, this structure did not create sufficient decentralization of power to protect the company from its CEO. Parmalat failed in December 2000 because the CEO was the major shareholder with power to control the supervisory board, executive board, and audit board (Melis, 2005). This concentration of power is avoided in Figure 2 by the governance board being appointed with one vote per shareholder. This democratic method of voting allows minority shareholders to protect themselves from exploitation by a major shareholder voting on a plutocratic basis of one vote per share. Additional protection for minority shareholders is provided by adopting cumulative voting in appointing supervisory board members. In cumulative voting, each share obtains as many votes as there are board vacancies to allow voters to allocate all their votes to one or more directors. In this way, minority interests can appoint their own directors to advise the governance board privately of any conflicts that could attract their involvement.
There are different ways that the governance board could become involved. One approach adopted in some European jurisdictions is for the governance or “Watchdog” board to take over the powers related to director nomination, director remuneration, and audit (Hatherly, 1994). Another, more nuanced approach adopted by Turnbull (2000a) was to limit the power of the governance board to veto matters in which any director had a conflict of interest. Figure 2 adopts the veto option for conflicts with direct authority to control and remunerate the auditor and other advisers to the firm appointed to inform shareholders independently of management. To exercise their powers, governors have the right to access board papers, minutes, and any other information available to directors. Whether or not the governance board possesses executive or veto power, it still advises shareholders on the appointment, remuneration, and retirement of directors as well as matters affecting the auditor and/or other advisers. In this way, the governance board ensures that management takes note of its concerns.
In addition, the governance board has access to information, independently of management of operations from not only employees but also strategic stakeholder in the supply and distribution chains as well as from stakeholders in the host community from their respective stakeholder boards shown in Figure 2. Although not shown in Figure 2, management would also have access to these data sources. But the governors can now check management information for omissions and identify any biases, distortions, errors, and spin reported to the directors. More important, governors can report to the directors the other side of the story that management may report.
Stakeholder networks, boards, and stakeholder congress provide a systemic basis for governors and directors to evaluate executives independently of the information they may report. So-called independent directors on a unitary board have no such systemic basis to evaluate and remunerate management with data obtained independently of management. This absence is a crucial flaw of the existing system as it means non-executive directors have no creditable basis to perform their fiduciary duties in hiring, remunerating, monitoring, and/or dismissing the CEO.
Stakeholder reporting also institutionalizes a safe, reliable system for constructive whistle blowers. Because stakeholder boards are self-appointed without the grace and favor of management and elected on a direct democratic basis, they have the authority, knowledge, and motivation to counter radical, irresponsible, and/or rat-bag activists. Strategic stakeholders are likely to have a longer time horizon than investors and so more concerned about the long-term future of the enterprise.
Stakeholder Boards and Stakeholder Congress
Stakeholder representation in the governance and strategy development process is also crucial for enhanced CSR. There are many illustrations of stakeholders volunteering their time, personal resources, and even cash to further not only their interests but also the common good. There are numerous non-profit activist organizations in communities that seek to protect and further a great variety of public interest that given the opportunity would marshal participation. The Citizen Utility Boards (CUBs) established by Ralph Nader provide an example of where utility customers volunteered cash to lobby regulators not to accept management requests to increase prices (Givens, 1991).
Strategic stakeholders have many incentives to participate. For example, supply forums would obtain a formal framework to establish just in time (JIT) delivery processes whereas customer councils would provide a basis for establishing total quality control (TQC) and processes for enhancing after sales services. It would also provide a systemic framework for firms to obtain competitive intelligence on innovations. von Hippel (1986) reported that customers can be responsible for contributing up to 90% of product and/or service innovations.
Stakeholder boards can take on the role of co-regulators to protect and further their own interests in a much more direct, immediate, and nuance way than any arms-length regulator without the information, knowledge, incentive, or even the capability to act, especially when the concern of stakeholders is not actual harms but the risk of harms. A corollary of Ashby’s law of requisite variety that direct amplification of regulation is impossible like it is impossible to amplify the power of TV transmissions to drive a TV set (Ashby, 1956). Amplification is only possible with a supplementary source of power from the local electricity supplier. Stakeholders take on the role of a supplementary source of co-regulation. The variety of stakeholders can provide the requisite variety required to reliably regulate the complexity of a firm’s operations. In this way, the variety of stakeholder interests can generate the requisite variety necessary for either or both the CEO and the government regulator to control a firm to protect and further stakeholders interests.
One indirect but powerful way stakeholder boards can make management accountable in a way that does not necessarily directly conflict with shareholder interests is in the establishment of key performance indicators (KPIs) for evaluating the performance of managers. The board of governors who represent shareholder interests could then use appropriate stakeholder KPIs in remunerating executives. The ability of stakeholders to influence how the firm is managed creates an incentive for stakeholders to participate in the governance of the firm to protect and further their own interests in ways government regulators could not. In this way, stakeholder governance has the potential to reduce the need for government laws, regulations, and manifold compliance standards concerning employment, fair trading, pollution, environmental protection, gender equity, human rights including any other elements of CSR. In other words, decentralized governance provides a basis for firms to internalize the costs of public sector regulation by facilitating self-regulation and self-governance.
Figure 2 only presents a simplistic architecture of decentralized governance. In practice, there could be a variety of each type of stakeholder boards that may in turn represent stakeholder boards in a variety of locations in the case of large or global enterprises. The stakeholder congress shown in Figure 2 negotiates competition for corporate influence and control from the diversity of stakeholder interests. By this means, win–win solutions can be negotiated to enhance performance for all stakeholders including shareholders consistent with the proposals of Pound (1992, 1993).
Because stakeholders possess direct intimate knowledge on how a firm exposes them to harms and risks, the need for firms to provide detailed accounts and reports on their social and environment impacts can be very much attenuated. The need for detailed reporting should also be reduced by problems being avoided, mitigated, and/or managed to the satisfaction of stakeholders. In this way, the need for external entities such as the United Nations and governments to set standards of CSR reporting should be minimized. The stakeholder congress rather than the company could then undertake the reporting of any exceptional matters. This shift would remove the need for firms to produce social and environmental accounts with their associated auditing. The stakeholder congress and its constituent stakeholder bodies would become the auditors on a much more expert, motivated, and no-cost basis.
The stakeholder congress has another role, which is to appoint a chair, without a vote, for shareholder Annual General Meetings (AGMs). There is a need for a stakeholder who is neither a director nor a governor to chair an AGM because the purpose of the meeting is to hold directors and governors accountable, elect them, retire them, and approve their remuneration. For this reason, it would not be ethical for either a director or a governor to control the conduct of the meeting or the counting of votes. The arrangement also provides a formal basis for stakeholders to report on CSR to shareholders and even offer advice on the nomination and remuneration of directors and governors to provide a basis for them to become sympathetic to the firm enhancing its CSR.
Concluding Remarks
Complexity theory provides compelling arguments to illuminate our research question on whether centralized or decentralized governance structures can better support CSR. The authors find that decentralized governance architectures provide operating advantages in managing complexities created by CSR. Empirical evidence of the operating advantages of decentralized, network governance is provided by the success of over generations of CEO’s through a number of business cycles firms such as VISA International Inc. in the United States, the Mondragón Cooperatives in Spain, and the John Lewis Partnership in the United Kingdom. Two key elements help in structuring a complexity absorption rather than complexity reduction-based approach to integrating corporate governance and CSR: (a) division of labor and (2) division of power in both structure and strategy development.
A common response from laypersons and scholars not familiar with the natural laws of governance is to reject the idea of network governance because it increases the complexity of the organizational architecture. But this necessary increase is the point. The laws of requisite variety state that complexity can only be managed with a requisite variety of matching complexity. What may not be intuitive is that the complexity of tasks for individuals within the organization can be substantially reduced as indicated by the number “X” for each board in Table 1. This reduction is the result of the division of labor.
The division of power inherent in decentralized governance architecture provides a basis to introduce checks and balances on managerial power, operating discretions, and the extent that they can act in self-interest ways inconsistent with either shareholder or other stakeholder interests. The division of power created by decentralized/network governance creates a basis for sharing power and influence among stakeholders to discover win–win ways for improving by negotiation outcomes for both shareholders and other stakeholders.
It is conceivable that the many CSR-related scandals would not have occurred if the respective corporations had not adopted a complexity reduction structure but had adopted network governance to provide the ability for a firm to both access and absorb complexity. However, the idea of absorbing complexity via such architecture is rarely considered as can be seen in ‘The State of Corporate Governance Research’ (Bebchuk & Weisback, 2010). The authors hope to provide a stepping-stone in the emerging discussion on how corporate governance and CSR as simultaneous quests to increase managerial responsibility can be integrated.
Footnotes
Acknowledgements
The authors acknowledge the three anonymous reviewers and editor Duane Windsor for helpful guidance on prior versions of the article.
This article was accepted during the editorship of Duane Windsor.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
