Neither Shareholder nor Stakeholder Management:

European Management Journal Vol. 25, No. 2, pp. 146–162, 2007 Ó 2007 Elsevier Ltd. All rights reserved. 0263-2373 $32.00

doi:10.1016/j.emj.2007.01.002

Neither Shareholder nor Stakeholder Management: What Happens When Firms are Run for their Short-term Salient Stakeholder? LAURENT VILANOVA, CoActis University of Lyon, France One of the critical distinctions between shareholder theory and stakeholder theory rests on the role of management in the resolution of the firm’s internal conflicts. Whereas managers are considered as a source of conflicts by agency/shareholder theorists, they are often viewed as useful mediators in the stakeholder approach. This paper proposes an alternative theory on the role of management in corporate governance, the so-called short term salient stakeholder theory, and illustrates it with a longitudinal case study of Eurotunnel, the Channel Tunnel operator. When the firm’s legitimate stakeholders have very different information levels and bargaining strengths, this theory predicts that (i) firms are governed in the interests of a unique stakeholder group (ii) managers have a minor role and are prone to collude with the most powerful interest group (iii) this autocratic type of governance is unstable in the long-term as the legitimate stakeholders expropriated at one period use influence strategies to gain power in the next period (iv) the chronic conflicts associated to short-term salient stakeholder management lead to poor organizational performance. Ó 2007 Elsevier Ltd. All rights reserved.

146

Keywords: Stakeholder Theory, Shareholder Theory, Corporate Governance, Role of Management Finance and strategic management theories often take divergent positions on two fundamental questions: Should firms be run in the interests of their shareholders or for all of their stakeholders? More broadly, what should be the objective of firms? Most of the time, the defenders of each approach rely on normative arguments and make simplifying assumptions. Shareholder theorists argue that managers should be pledged to shareholders. Because managers are self-serving and opportunistic, the governance structure should be designed so as to limit managerial discretion as strictly as possible. The stakeholder management view considers on the contrary that managers are not always opportunistic and that they should retain sufficient autonomy so as to influence corporate decisions in a way satisfying all the key stakeholders of the firm. Beyond this normative debate, little is known about the actual behavior of firms and about the connexion between corporate governance strategies and organizational performance. One of the main problems in designing a

European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

tractable test of instrumental stakeholder/shareholder theories comes from the measure of the stakeholder/shareholder orientation of firms. How to decide that a firm is run for its shareholders or alternatively for its stakeholders? Should we trust managers’ speeches or rather consider the firm’s actual decisions? By considering the case of large project companies, our paper seeks to develop and illustrate an alternative theory of corporate governance, what we call short-term salient stakeholder management. Although the terminology of salient stakeholder is a direct reference to the work of Mitchell et al. (1997), our paper goes beyond the question of who or what constitutes a stakeholder. It also considers the link between the firm’s organizational structure, the number of salient stakeholders and organizational performance. Even if they have many idiosyncratic features, large project companies are attractive research sites for people interested in evaluating the relative performance of shareholder and stakeholder theories. Their organizational structure reflects a shareholder management view of the firm (managers are strictly controlled by financiers) in the mere context where stakeholder theorists argue for the needs to consider the interests of multiple constituencies (these ‘mega’ investments affect numerous groups or individuals and can dramatically change the economic conditions for local citizens). Moreover, empirical evidence shows that this type of firm often exhibits bad performance (Flyvberg et al., 2003). In this paper, we use the case of Eurotunnel (the Channel Tunnel operator and the largest project company in the world) to illustrate some of our propositions. Short-term salient stakeholder management is defined by the following attributes: (i) firms are run in favor of one salient stakeholder group. This group is the one that simultaneously possesses the three attributes defined by Mitchell et al. (1997), that is legitimacy, power and urgency (ii) the identity of the salient stakeholder group can change depending on the firm considered and on the period of time. The salient stakeholder at one period is often the one that suffered the most from corporate decisions in the near past (iii) managers are pledged to the salient stakeholder group and have little discretion concerning corporate decisions (iv) short-term salient stakeholder management increases agency conflicts and may lead to poor performance. Our theory brings insights on three fundamental questions concerning corporate governance and performance. The first question is quite classical: for whom companies are run? The answer is far from being obvious as the balance of power inside the firm can take a less trivial form than the one supposed by shareholder and stakeholder theorists. In accordance with shareholder theory, our theory suggests that firms are run for a unique stakeholder group (the European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

‘‘salient stakeholder group’’ in our terminology) but, contrary to the shareholder theory’s core assumption, this group is not always constituted by shareholders. Obviously, the fact that corporate decisions give more importance to one stakeholder group is susceptible to increase conflicts within the firm. In other words, the autocratic governance structure associated with short-term salient stakeholder management increases the incentives of non salient stakeholders to gain authority. The Eurotunnel case illustrates clearly this dynamic dimension of shortterm salient stakeholder management: since its creation the firm was successively controlled by the construction companies, banks and dispersed shareholders. Each time, the new salient stakeholder group was the one that suffered the most important losses in the preceding years. The second insight of the paper concerns the efficiency of this autocratic type of management. In other words, is it economically efficient to concentrate power in the hands of one type of stakeholder instead of balancing power among the different key stakeholders? The answer is unambiguously yes for agency and transaction costs theorists (see for example Jensen, 2001; Williamson, 1991). For them, the concentration of decisions rights in the hands of the ‘‘natural’’ residual claimant, that is shareholders considered as an homogeneous group, economizes on agency and transaction costs. Even if stakeholder theorists are reluctant to the efficiency notion, they implicitly assume that the dispersion of control rights among all the stakeholders and favorable performance go hand in hand. Although the problem of the optimal allocation of control rights is difficult to answer on a single case basis, the Eurotunnel case suggests that the concentration of power in unique hands does not automatically reduce agency conflicts and can even exacerbate them. Lastly, our paper sheds new light on a fundamental while understated difference between alternative theories of corporate governance, that is the role of management: should organizational structures limit managerial discretion (contractual theories) or instead expand it (see for instance Donaldson, 1990)? Is managerial discretion valuable because of the managers’ ability to take pertinent strategic decisions or because of their ability to limit conflicts between the firm’s multiple constituencies? The Eurotunnel case is a unique opportunity to examine these questions. Eurotunnel is indeed a typical project company, an organizational structure where managers take few strategic decisions and where long-term contracts are supposed to be an efficient tool to resolve the potential conflicts between the firm’s multiple constituencies. These special features explain why shareholder management theorists consider project companies as a kind of ideal firm where agency conflicts between managers and shareholders are minimized. Our analysis shows that the successive management structures of Eurotunnel, that is a 147

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

no-management structure during the first years of the project and a bank-controlled management afterwards, were inefficient: the former led to a ill-defined project and opportunistic behavior from the construction companies at the expense of banks, the latter led to the expropriation of dispersed shareholders by banks. Overall, the Eurotunnel case illustrates the limits of contracts in governing relationships between different claimants and suggests that the resolution of conflicts between various stakeholders necessitates the intervention of an independent powerful third party. Whether a strong management can exercise this arbitrage role is suggested by our theory but requires additional reflection on the ways legitimate stakeholders can prevent managers from colluding with the most powerful claimant. The paper is organized as follows. The first section underlines the conflicting views between shareholder and stakeholder theorists on the role of management. These theories take opposite positions on the optimal level of managerial discretion. The second section briefly presents the characteristics of large project companies and explains why shareholder theorists consider that this type of firm should be efficient. The third section describes the Eurotunnel case. We address first the question of stakeholder identification and salience. We also investigate the consequences of the short-term salient stakeholder management adopted. The fourth section returns to theorizing. This final section presents our theory of short-term salient stakeholder management and stresses the need for stakeholder theorists to consider more closely managerial incentive mechanisms.

Shareholder vs. Stakeholder Management: The Conflicting Views on the Role of Management The debate opposing the pro and cons of shareholder value is often presented in the following way: for whom should corporations be run? Should management take decisions in the interests of a unique stakeholder, that is shareholders considered as an homogeneous group, or rather consider the interests of the multiple constituencies of the firm? These questions about the legitimacy of shareholder management, while interesting, seem to be unanswerable unless referring to philosophical concepts. For example, each theory (shareholder management and stakeholder management) has its own performance measure (shareholders’ surplus vs stakeholders’ surplus) and there is no particular reason to give the primacy to one measure over the other. 1 Our paper takes a different perspective. Our main question is about the importance and the role of management in both theories. More precisely: Are managers considered as key stakeholders? Should governance 148

structures restrict managerial discretion or on the contrary isolate managers from the control of other stakeholders? If managers take decisions on the behalf of other agents, how can stakeholders (shareholders only or multiple constituencies) be sure that managers will consider their interests? The answers of shareholder and stakeholder theorists differ dramatically on these fundamental questions (see Figure 1). Modern corporate finance asserts quite unanimously the need to control managers. The suspicion on managers is clearly expressed by Tirole (2006, p. 15): ‘‘Most observers are now seriously concerned that the best managers may not be selected, and that managers, once selected, are not accountable’’. The finance view of corporate governance is based on the following premises: (i) In order to obtain financing, the providers of funds must be assured that they will get a return on their investment in the firm, (ii) Managers, who take decisions, have their own interests and need not act in the best interests of providers of funds. They can use their expertise and their informational advantage over outsiders to divert funds to pursue their own objectives, (iii) Consequently, ‘‘much of the subject of corporate governance deals with constraints that managers put on themselves, or that investors put on managers to reduce the ex post misallocation and thus to induce investors to provide more funds ex ante’’ (Shleifer and Vishny, 1997). While this approach considers at first sight the interests of all kinds of providers of funds (creditors and shareholders), shareholders benefit from a special status. They are the residual claimants of the firm: contrary to fixed claimants (creditors and employees), they do not negotiate compensation in advance and their remuneration depends greatly from the ex post performance of the firm. This last point makes obvious that shareholders should obtain control rights (make discretionary decisions) as they have the right incentives to maximize the firm’s value. So, why do shareholders hire managers and give them some control rights? Alternatively, why do not shareholders impose a contract specifying what managers should do (and not do) in each circumstance? The first reason comes from the impossibility to write a contract specifying all future contingencies. According to the terminology used by contract theorists, contracts are by essence incomplete. Shareholders have then to decide how to allocate residual control rights- the rights to make decisions in situations not fully foreseen by the contract (Hart and Moore, 1990). The fact that contracts are incomplete does not however explain why shareholders do not retain all the residual control rights and delegate some of them to managers. According to agency European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

Shareholder management Managers should maximize the wealth of shareholders.

Representation of firms

Key idea

Nature of firms Importance of management Role of management

Behavioral assumptions

Stakeholder Management Managers should satisfy the interests of all (legitimate) stakeholders. Nexus of contracts or relationships between groups with conflicting objectives Managers are key actors because they have residual control rights Managers are viewed as “hired” agents of shareholders and are in charge of day-to-day management in an incomplete contract framework. Managers can destroy shareholder value. Self-interested and opportunistic agents.

Optimal governance system

Control should be concentrated in the hands of shareholders.

Good governance practices

Control mechanisms limiting managerial discretion Incentive mechanisms aligning managers’ objectives with those of shareholders LBO and project company

The “perfect” firm

Managers are viewed as referees between groups with conflicting objectives.

Managers can increase the aggregate value of the firm. The assumption of self-interested and opportunistic agents is overly simplistic. Agents can be organizationcentered and altruist. Control should be divided between the different (legitimate) stakeholders. The ones that increase the democratic representation of noncontrolling stakeholders

German codetermination

Figure 1 A Comparison Between Shareholder and Stakeholder Management

theorists (Jensen and Meckling, 1976), it would be too costly for shareholders to run the day-to-day operations of the firm. Shareholders would be better off by delegating everyday decisions to specialized agents that possess the expertise and the information necessary to operate the firm. If managers have great discretion on day-to-day management, they should however refer to shareholders for strategic decisions such as mergers and acquisitions, dividends or new debt and capital issues. In practice, managers can exploit their informational advantage to overrule the formal authority of shareholders. This leads to a distinction between formal and real authority: ‘‘So, while shareholders have formal control over a number of decisions, managers often have real control’’ (Tirole, 2001, p. 17). Managers have then a considerable role in the shareholder value concept of corporate governance. This is not to say that managers are viewed positively by the tenants of shareholder value. On the contrary, the real authority exercised by managers is considered as a source of danger for shareholders. Once the firm has obtained funds, there is a high probability that self-interested managers use their decision rights to favor their own interests at the expense of investors. Much of the corporate finance literature is then about the optimal means to control management. Many event studies confirm that managers take decisions harmful for shareholders (see Shleifer and Vishny, 1997 for a summary): they cause their firms to grow European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

beyond their optimal size, they favor diversification over profitability, and so on. . . This evidence is often interpreted as consistent with the behavioral assumptions of contractual theories of the firm (agency and transaction cost theories), that is the self-interested and opportunistic nature of economic agents (Williamson, 1991; Jensen and Meckling, 1994). While the huge majority of economists and finance academics adhere to the idea that shareholders need managers to operate the firm, they also consider that shareholders suffer from managerial misbehavior. This leads shareholder theorists to stress the need to circumscribe managerial discretion as much as possible. This is not an easy task and opinions diverge on the optimal control mechanisms. The main difficulty comes from the managers’ ability to make formal control mechanisms (e.g., the board of directors) ineffective. Managers can act strategically to protect their position and hold some winning cards: they can conceal information from their opponents in the board, they can interfere in the nomination process of directors and so on. The dominant opinion is then that corporate boards are quite ineffective in controlling managers (Jensen, 1993). So, it is not surprising that the corporate governance debate is more about public-owned companies where ownership and control are separated and where dispersed shareholders are at a disadvantage with managers (Fama and Jensen, 1983). This is the mere situation where managers can exploit the 149

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

passivity of the formal control structures. If the dispersion of formal authority between dispersed shareholders increases de facto the real authority of opportunistic managers, the logical solution seems to call for a more concentrated ownership. A majority shareholder or an influent minority stakeholder has more possibilities to put pressure on management. The concentration of power in unique hands (different from the ones of managers) is then one of the core ideas of the finance view of corporate governance. This concentration of power is advocated on the part of contractual theorists. For example, Williamson (1991) contests the efficiency of the interest group management of the board- a board in which many constituencies are awarded a prorata stake- and advocates for specialized boards in which one stakeholder group is dominant. Similarly, Jensen (1986) presents LBOs as a kind of ideal organizational structure that increases efficiency and reduces agency conflicts. The high level of debt in LBO transactions- a 10 to 1 ratio of debt to equity is not uncommon- is supposed to be the main reason leading to efficiency. The obligation to pay high interest payments reduces the ability of managers to divert cash flows to their own benefit and gives them incentives to make sufficient effort in order to reach a high level of performance. Moreover, this type of firm has concentrated equity ownership by LBO funds. Creditors and funds are here the locus of control and managerial discretion is kept to its minimum. The main message of the so-called shareholder value theory is then extremely clear: managers are opportunistic agents and have important decision rights; shareholders should either reduce managerial discretion (through monitoring) or give managers adequate incentives to maximise shareholder value (stock-based managerial compensation). Managers should be maintained on a short leash and their unique role is to take decisions that enhance shareholder value in situations where contracts are incomplete. The stakeholder theory attributes also a prominent, although less negative, role to managers. Logically, the premise that firms should operate on the behalf of all the stakeholders of the firm (Freeman, 1984) implies the existence of a third party able to identify stakeholders and to take decisions satisfying the interests of the firm’s multiple constituencies. Whereas managers were at best considered as a necessary evil in the shareholder view of corporate governance, some stakeholder theorists confer them the responsibility to make strategic decisions reconciling the conflicting interests of the firm’s different claimants (Hill and Jones, 1992). This is not to say that managers are the only rightful locus of control (Donaldson and Preston, 1995) or that managers are always considered as key stakeholders. Whereas many authors consider that the only role of managers concerns the repartition of the firm’s economic surplus between multiple stakeholders (Aoki, 1984; 150

Freeman and Evan, 1990), others expand the role of management to the creation and the repartition of the economic surplus (Donaldson and Preston, 1995; Hill and Jones, 1992; Mitchell et al., 1997). One may obviously argue that the negative vs. positive evaluation of management is more representative of the opposition between finance and strategic management than of the one between shareholder and stakeholder theories. In our opinion, it is however not coincidental that the strategic management field was greatly influenced by the stakeholder perspective whereas finance theory considers nearly exclusively the shareholder approach. 2 We have found no criticism against managerial misbehavior in the stakeholder literature. Rather than restricting managerial actions, corporate governance structures should give managers sufficient discretion in order to satisfy multiple stakeholders simultaneously. The proponents of the stakeholder theory of the firm argue that the influence of managers should be reinforced and that a strict control of managerial decisions can impede value creation (Rappaport, 1990; Charreaux and Desbrie`res, 2001). Even if this point is rarely underlined, stakeholder theorists usually consider that the assumption of self-interested and opportunistic agents is overly simplistic (Jones and Wicks, 1999). Many of them implicitly acknowledge the propensity of managers to act in the interests of their organization rather than in their sole interests. Managers are then considered more like stewards than like opportunistic agents. These diverging behavioral assumptions appear clearly in Davis et al. (1997, p. 25): ‘‘Stewards in loosely coupled, heterogeneous organizations with competing stakeholders and competing shareholder objectives are motivated to make decisions that they perceive are in the best interests of the group. . .A steward who successfully improves the performance of the organization generally satisfies most groups because most stakeholder groups have interests that are well served by increasing organizational wealth’’. In short, the stewardship theory assumes that managers and the other firm’s constituencies are willing to act cooperatively and that managers will naturally give consideration to the interests of multiple stakeholders. The organization-centered behavior postulated by the stewardship theory can be then considered as a sufficient condition for implementing the kind of democratic governance structure described by the tenants of the stakeholder society. But, is it a necessary condition? In other words, is stakeholder management implementable in a world of self-interested and opportunistic agents? The answer is clearly negative for the defenders of shareholder management. For them, stakeholder management would dilute managerial accountability and would give managers the opportunity to invoke multiple and hard-to-measure missions to hide their selfserving behavior (Jensen, 2001; Tirole, 2006). European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

If it seems utopian to satisfy simultaneously all the stakeholders of the firm, it does not mean that a ‘‘moderate’’ form of stakeholder management is impossible. Mitchell et al. (1997) propose in this spirit a very interesting theory of stakeholder identification and salience departing from the broad view of Freeman (1984) – the firm should satisfy the interests of all the groups or individuals who can affect or be affected by the achievement of the firm’s objectivesas well as from the narrow view of shareholder management – the firm should only consider the interests of shareholders. They start with a typology of stakeholders based on the possession of three attributes: power (the stakeholder’s ability to influence the firm), legitimacy (the legitimacy of the stakeholder’s relationship with the firm based upon contract or legal title) and urgency (the degree to which managerial delay in attending to the claim is unacceptable to the stakeholder). These three attributes can be used to identify all relevant stakeholders (a stakeholder must have at least one attribute) but also to determine the salience of each stakeholder for managers. The idea is quite simple: managers pay more attention to a ‘‘definitive’’ stakeholder, who holds simultaneously the three attributes, than to a latent one (with only one attribute). This theory has many interesting features. It attributes a central role to managers who are in charge to identify and to organize stakeholders into a hierarchy. It does not stipulate a priori the identity nor the number of ‘‘definitive’’ stakeholders. Lastly, the theory is dynamic as it considers that stakeholders can shift over time from one class to another. More broadly, this approach denies that all the stakeholders of the firm have convergent objectives and considers rather conflict between stakeholders as an important while understated premise of stakeholder theory in the same way that conflict between managers and shareholders is a key premise of shareholder theory. This idea, opposite to the one advocated by stewardship theory, is also clearly expressed by Frooman (1999, p. 193): ‘‘I suggest, then, that if the potential for conflict did not exist – that is, if the firm and all its stakeholders were largely in agreement – managers would have no need to concern themselves with stakeholders or stakeholder theory’’. Although both theories drastically diverge on the role of managers, few attempts have been made to test the connections between the practice of management and the achievement of corporate performance. For many scholars, the main difficulty comes from the explanatory variable, that is the measure of corporate performance. Tirole (2006, p. 62) asserts along this line that there exists no reliable measure of the impact on managerial decisions upon stakeholders’ welfare. In our opinion, the measure of the dependent variable (the stakeholder or shareholder orientation of the firm) is even more problematic: how to measure stakeholder salience? Should we rely on CEOs’ speeches? Agle et al.’s (1999) work illustrates some of these methodological issues. Their empirical European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

test of stakeholder salience uses survey data provided by CEOs. They find a strong relationship between stakeholder attributes (power, legitimacy and urgency) and stakeholder salience but they identify no effects of salience on performance. In our opinion, this last result asks two questions: are CEOs’ speeches reliable measures of actual stakeholder salience? To what extent do managers have sufficient discretion to give the primacy to one type of stakeholder over the others? Looking at archetypal forms of stakeholder or shareholder firms might be a better approach to test the relative performance of the two dominant models of corporate governance (Figure 1).

Large Project Companies and the Performance of ‘‘Pure’’ Shareholder Firms The finance view of corporate governance has received substantial attention from the media throughout the world with the release of the socalled ‘‘codes of best practice for boards of directors’’. 3 These codes made recommendations on incentive mechanisms (managerial compensation) and control mechanisms (composition of the board) and were essentially concerned with the resolution of conflicts between managers and shareholders. However, shareholder theorists don’t restrict their attention to the question of board structure. They also identify special forms of organizational structures allegedly supposed to resolve agency conflicts in a more efficient way than traditional structures. We present here one type of ‘‘perfect’’ firm for shareholder theorists, i.e. a project company. According to Esty (2003), a project company is a legally independent entity financed with equity from one or more sponsoring firms and nonrecourse debt for the purpose of investing in a capital asset, usually with a single purpose and a limited life. This type of structure is today one of the leading finance vehicles for investments in natural resources and infrastructure sectors such as power plants, toll roads, mines, pipelines and telecommunication systems. The financing and organizational structures differ drastically from the ones of other companies. Project companies typically employ very high leverage compared to public companies. Debt-to-capitalization ratios in the range 70%–90% are usual. Debt ownership is most of the time concentrated in the hands of several banks (syndicated bank loans) and creditors have no possibility to recourse to sponsoring firms (shareholders) if any problem of debt repayment occurs. Equity ownership is also far more concentrated than in public companies as the typical company is privately held and has only one to three sponsors. The board of directors is supposed to control very closely managers as more than 80% of the 151

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

directors come from the sponsoring firms. Lastly, the relationships with the company’s multiple stakeholders are essentially managed by way of contracts as ‘‘the typical project has 40 or more contracts uniting 15 parties in a vertical chain from input supplier to output purchaser’’ (Esty, 2004, p. 2). In short, project companies are financier-dominated firms where contracts are preferred to managers for governing relationships with related parties. These special features are often justified by the efficiency gains associated with the reduction in transaction costs or agency conflicts. Said differently, the typical project structure would promote the interests of shareholders. Consider for example the project companies’ modes of financing. In reference to Jensen’s theory of free cash flows, high leverage is supposed to increase the managers’ incentives to reach a high-level of performance and to limit managerial discretion over cash flows. The fact that debt comes from a few banks reduces also managerial discretion as bank credit contracts include many restrictive covenants and involve more monitoring than arm’s length debt. According to this view, the banks’ dominance should benefit shareholders. Everything takes place as if the banks and the shareholders shared the same objective of reducing managerial discretion. In the same perspective, the concentration of equity ownership strengthens the control of shareholders over managers. Finally, the use of long-term contracts is thought to be an ideal governance tool to prevent any opportunistic behavior of related parties (input suppliers, output buyers) given highly specific assets. All these structural attributes of project companies contribute to reducing potential managerial opportunism. The banks and the shareholders take the strategic decisions before the project begins and the role of project managers is restricted to the management of day-to-day operations. The special purpose of project companies- dedicated to the management of only one asset-, the predominance of long-term contracts and the strict monitoring exerted by financiers also restrict ex post managerial opportunism. In summary, this type of organizational structure seems to be a pure illustration of the ‘‘shareholder view’’ of the firm that posits the strict control of managers as a necessary condition for achieving high (shareholder) performance. Despite the supposed qualities of project companies, the debate on the efficiency of project companies is still open. For Esty (2003), the typical ‘‘flat’’ payfor-performance compensation schemes of project managers demonstrates the efficiency of project governance systems in resolving managerial agency conflicts. Despite that project finance loans are made to riskier borrowers, they have lower spreads than other types of loan (Kleimeier and Megginson, 2000). A recent survey initiated by Standard & Poor’s Risk solutions and involving 30 project finance lenders from around the world also shows that project loans have lower default rates and higher recovery 152

rates than corporate loans. Globally, this empirical evidence tends to confirm the ability of project companies to mitigate agency conflicts. The few studies that have focused on the performance of large projects are however far more pessimistic. Large projects seem to perform very badly and experience systematic cost overruns (Flyvberg et al., 2002, 2003). This suggests that asset size is a key determinant of the performance of project companies. The reasons underlying this empirical observation are still unknown. Why do large project companies fail so often while other project companies outperform more traditional organizational structures? One potential explanation lies in the discrepancy between the project company’s simplified organizational structure and the multiplicity of stakeholders. To illustrate this idea, consider large infrastructure projects. The usual scheme, the so-called BOT (Build, Operate and Transfer), involves a private consortium agreeing to build the project facility, operate it for a given length of time (the concession period) and then transfer it back to the public sector. Many agents intervene at the successive stages of the project and are parties in the project’s main contracts : (i) the concession contract is signed between the host government and the project company (a consortium of private sponsors) (ii) the debt contract links the project company with banks (iii) the construction contract is negotiated between the project company, represented by its sponsors and lenders, and construction companies (iv) the contracts that govern the relationship between the project company and the related parties that supply critical inputs (a fuel producer for a thermal power plant) or that buy primary outputs (a company that resells electricity to final consumers for a power plant). Each of these relationships gives rise to potential opportunism from related parties. The firm’s sponsors can hide information to the host government in order to win the invitation to tender and can overestimate future cash flows in order to convince lenders to bring sufficient funds. The construction companies can voluntarily underestimate costs or potential delays in order to obtain the construction contract or provide insufficient effort to complete the construction on time after the contract has been signed (Flyvberg et al., 2002). Output suppliers can renege on their promise to buy the outputs at the pre-specified price and try to impose ex post a lower price. Host governments can also behave opportunistically by seizing the company’s asset or by increasing taxes. Obviously, the parties involved in formal contracts are not the only ones susceptible to influence a project company. For very large projects, other stakeholders such as politicians, the media, residents or associations of individual shareholders can interfere in the decision-making process and can increase the difficulties to take value-maximizing decisions. Of course, each constituency may pursue its own objective and use its informational advantage at European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

the different steps of the project to expropriate other stakeholders. The multiplicity of stakeholders involved in large project finance as well as the frequent failure of large project companies raises the question of the optimal governance structure: Is it efficient to concentrate power in the hands of investors (banks and sponsors) in large project companies that often involve many constituencies? According to shareholder theorists the high concentration of power in large project companies should lead to high performance. Even if numerous influential parties enter the decision-making calculus, the need to raise external funds and to convince risk-averse banks makes it unlikely that the decisions of large project companies do not reflect careful and deliberate attempts to increase shareholder value (Esty, 2004, p. 219). The answer of stakeholder instrumental theorists would be different: if managers don’t view the interests of stakeholders as having intrinsic worth and pursue only the interests of a single group, the firm they manage will achieve lower performance than had they pursued the interests of multiple stakeholders (Donaldson, 1999, p. 238). Large project companies are thus attractive research sites to compare these two approaches of corporate governance. Indeed, their organizational structure reflects a shareholder management view of the firm in the mere context where stakeholder theorists underline the need to adopt a more democratic governance structure.

The Failure of the Shareholder Management View in Large Project Companies: The Case of Eurotunnel The decision to build a fixed link between France and England was made official on April 2, 1985 when the French and the UK governments launched the invita-

tion to tender. The competing offers were receipted seven months later and the selection was effective in January 1986. The retained offer was the one submitted jointly by France Manche (FM) and the Channel Tunnel Group (CTG), which formed twin holding companies (Eurotunnel SA and Eurotunnel PLC), chartered in France and Great Britain, respectively. The winning project consisted in two single track rail tunnels and a service tunnel, including a shuttle service for cars and trucks. At the same time, the main legal instrument governing the construction and operation (The Treaty of Canterbury) was signed by both governments in February 1986 (later ratified in July 1987) and the concession was formally awarded to Eurotunnel (a 50/50 joint venture between FM and CTG) on March 14, 1986 for 55 years. As usual in project companies, a fairly complex set of contracts was implemented during the first months between multiple stakeholders (see Figure 2). Many features of Eurotunnel’s initial organizational structure fit well with the so-called efficiencyenhancing features of project companies. First, investors controlled quite closely the firm as the main provider of funds, i.e. the banks, controlled closely managers through its 50% equity ownership. The banks also had 9 out of 18 seats on the board of directors. Construction companies were also heavily implicated through a 50% equity ownership and through their representation in the board (2 seats out of 18). The fact that constructors were at the same time Eurotunnel’s main contractor and one of its leading shareholder was supposed to decrease the propensity of construction companies to behave opportunistically. Each cost overrun or delay in the construction phase would run counter to the interests of the construction companies due to the adversarial impact on Eurotunnel’s share value. Furthermore, the potential opportunism of construc-

Constructors Lenders (Banks)

Shareholders Credit agreement (November 1987)

Railways

Railway usage contract (July 1987)

TML Construction contract (August 1986)

Eurotunnel (FM/CTG) Concession (March 1986)

UK Government

Treaty (February 1986)

French Government

Figure 2 The Channel Tunnel’s Initial Contractual Structure

European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

153

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

tors seemed at the time all the more unrealistic since the construction contract included an incentives/ penalties scheme supposed to deter any voluntary delay or cost overrun. The problem of the ex post monopolistic position of the output buyer, i.e. railways, was also considered through the railway usage contract that defined a minimum payment to Eurotunnel during the first years of operations.

For Whom Eurotunnel was Run? At first sight one may consider that the banks and the construction companies have shared the control of Eurotunnel during the first years of the project: they were both the main contractors and the main sponsors of the project company. This shared-control was also reflected in the 50–50 partition of equity ownership. However, in spite of appearances, real authority was concentrated in the hands of the construction companies. The dominance of constructors was primarily due to their informational advantage in the evaluation of construction costs. Faced to TML, the construction consortium, the funding banks were unable to elaborate their own projections. 4 The fact that banks relied almost exclusively on the information provided by constructors was all the more disastrous since both agents had very diverging objectives and interests. TML had short-term interest in the project and was only concerned in the construction phase. In contrast, the banks had a longer-term interest as debt repayment was scheduled to commence after the commercial opening of the Tunnel. The discrepancy between these two key stakeholders’ investment-term horizon was confirmed ex post. Whereas the banks retained their shares and the control of the Board of Directors for many years, the constructors sold their shares and left the Board very quickly (the last one left the Board at the end of 1989). Because of their lack of technical expertise, the banks were however unable to circumscribe the opportunistic behavior of constructors. For example, the banks were convinced at the outset of the project that the construction costs computed by TML were overestimated. In reality, TML communicated artificially low costs in order to win the tender offer. For many observers, the banks were incredibly naı¨ve as deliberate cost underestimation is a usual practice in tender offers. 5 This seminal mistake greatly influenced the relationship between Eurotunnel main financiers (the banks) and constructors. Conflicts appeared

Table 1

from the very beginning of the construction schedule and intensified at the same time as delays and cost overruns increased. Table 1 shows the evolution of projected construction costs over time at the different steps of the seven years construction phase and illustrates the problem of cost escalation. 6 The banks were clearly the first victims of their inability to control the opportunistic behavior of TML. The escalation of construction costs increased the need for financing and the banks that had initially planned to bring £5 billion were obliged to increase their investment to £7.8 billion. While the bank syndicate agreed to provide additional funds, it set strict conditions. The banks asked for a major change of ownership structure and risk allocation. This led Eurotunnel to become a public company in 1987 and to realize two additional equity issues in 1990 and 1994. As emphasized by Stonham (1995), the lending banks were satisfied to transfer part of their risk to individual shareholders as equity offerings allowed the Tunnel to be completed while decreasing the leverage to capital ratio. At the time of the IPO, the bank-dominated management of Eurotunnel made very optimistic speeches to encourage individual investors to subscribe to new equities: ‘‘This is an ideal investment to finance the studies of your children; you receive nothing at the beginning but in turn the value of your shares will treble in seven years’’ (Jean-Louis Dherse, Eurotunnel’s managing director, reported in Les Echos, October 29, 1996). In reality, Eurotunnel stock was issued at 350 pence in November 1987, was priced around 260 pence at the time of the commercial opening of the Tunnel (in June 1994) and has constantly decreased since then. During this period, the projected traffic and revenues were systematically updated downward after each equity offering. This led many observers to question the fairness and independence of Eurotunnel’s management. 7 More precisely, the managers were suspected of favoring banks by communicating over-optimistic projections to small investors. During the following years, the financial situation of Eurotunnel deteriorated. The firm suspended interest payments on its junior debt in September 1995. Complex negotiations with banks took place and lasted more than two years before an agreement was reached. During this period, the position of management was extremely ambiguous. Patrick Ponsolle, at the time CEO of Eurotunnel, declared at the outset of the negotiations its intention to seek concessions

Evolution of Projected Construction Costs (in £ million – 1985 prices)

Total construction costs

November 1987

November 1990

May 1994

% change 1987–1994

2710

3969

4844

78.8%

Data are Drawn From Eurotunnel prospectus (1987 offer for sale, 1990 and 1994 rights issues).

154

European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

from banks and to defend the interests of individual shareholders. He advocated for reduced interest margins on bank debt and rejected a debt-for-equity swap that would dilute the claims of individual shareholders. Ponsolle’s position was radically different a few months later when he tried to convince shareholders of accepting the financing restructuring proposals negotiated with the representatives of the bank syndicate (Les Echos, June 24, 1997). The individual shareholders had no other choice but to trust the CEO and to approve the plan. They had only partial information on the precise terms of the revised credit agreement as none of their representatives was invited to the negotiations. Moreover, the plan was incredibly complex and most of the individual shareholders were unable to evaluate its implications for their own welfare. Despite the fact that the banks declared having made huge concessions to Eurotunnel, the situation of the firm didn’t improve significantly after the implementation of the restructuring. While the financial pressure was delayed through debt rescheduling, actual revenues were still disappointing. Two groups came in open conflict. The first group (presented as ‘‘the rebels’’ in the press) was constituted by several associations of individual shareholders and pressed simultaneously the banks for a debt write-off and the British and French governments for a bail-out of the troubled group. The second group included the banks and the management in place. It advocated for less radical measures including negotiations with banks to restructure Eurotunnel’s £6 billion debt and a commercial plan intended to boost traffic and revenues. The conflict grew by the end of 2003 as the future of Eurotunnel became more uncertain. The MUC (Minimum Usage Charge) that guaranteed Eurotunnel a minimum payment from the railways was planned to end in November 2006. This was a clear threat on Eurotunnel’s revenues. The minimum user fee arrangements guaranteed Eurotunnel a revenue equivalent to about 10 million Eurostar passen-

Launching phase

Formal authority Real authority / Salient SH

Constructors

Expropriated SH

Banks

Management

No management

Each group tried to improve its position through influence strategies. Several associations of French individual shareholders were active in the media to denounce the ‘‘scandal of the century’’ and seek to bring the matter into the political arena. The dispute became a major topic of political debate in France with more than 620,000 French shareholders embittered by the collapse in the share price from a high of 800 pence in 1987 to 40 pence in March 2004. A cross-party working group was formed at the French Parliament and several deputies took positions in favor of the rebels. On the other side, the French finance minister intervened in the dispute by asserting that the strategy proposed by the existing board was ‘‘responsible and serious’’ (Financial Times, March 12, 2004). Eurotunnel’s management also warned shareholders against the rebels’ plan: ‘‘Based on [the rebels’] public pronouncements, shareholders could face losing their entire investment.’’ (Richard Shirrefs, Eurotunnel’s CEO in Financial Times, March 12, 2004). Finally, the revolt succeeded and rebel shareholders ousted Eurotunnel’s Board of Directors during a stormy annual meeting in April 2004. Since then Eurotunnel’s situation has not significantly changed. Once elected, the new management team has roughly adopted the same strategy than the former one and is still negotiating a financial restructuring with its creditors since April 2005. No agreement has been reached yet.

Operating phase

Construction phase

Banks

Banks and constructors

gers annually and 5 million tonnes of freight. After the end of the MUC, Eurotunnel would suffer a huge decline of its revenues given its 6.3 million Eurotunnel passengers and 1.7 million tonnes of freight in 2002. The second threat was due to the end of the stabilization period in December 2005. According to the 1998 financial restructuring, Eurotunnel had no repayment obligation on its junior debt and had the possibility to issue new debt bearing no interest to cover interest payment shortfalls until the end of 2005. The common opinion was then that a new debt restructuring was unavoidable.

Banks and individual shareholders

Banks

Banks and individual shareholders Individual shareholders

Bank-dominated

Affiliated to individual shareholders

Figure 3 Stakeholder Identification and Salience in the Eurotunnel Case

European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

155

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

As a whole, the history of Eurotunnel is characterised by huge conflicts between the firm’s key constituencies. The salient stakeholders were successively the constructors, the banks and the individual shareholders (Figure 3). The management, if any, paid only attention to the short-term salient stakeholder, the one with the more urgent claim, and ignored the others. The other stakeholders were constantly ignored (for example, individual shareholders until 2004). The expropriated stakeholders in one period adopted in turn strategies to gain power and became salient in the subsequent one.

The Link Between Governance Structure and Performance Eurotunnel has been in financial distress for nearly twenty years. The first difficulties appear a few months after the beginning of the construction (breach of financial covenants). Since then, and despite several debt restructurings, Eurotunnel is still close to bankruptcy. The creditors and the individual shareholders have experienced negative present value for their investment in the project. For example, the stock price was only 25 pence in April 2006 in comparison with prices between 265 pence and 350 pence for the successive equity offerings during the 1987-1994 period. The overall performance of the Channel Tunnel project is more difficult to evaluate. Anguera (2006) considers however that the producers’ losses (both ferry operators and Eurotunnel) have been much higher than the user’s gains (mainly due to the price cuts). He concludes that the British economy would have been better off had the Tunnel never been constructed. The causes of Eurotunnel’s bad performance are still under question. To what extent was this bad performance due to Eurotunnel’s wrong organizational structure or to bad luck associated to a highly innovative and uncertain project? The interaction between bad performance and the conflictual relationships between stakeholders also deserves attention: Did bad performance exacerbate conflicts or alternatively did conflicts lead to bad performance? The non-existence of Eurotunnel at the birth of the project is the most popular ex post explanation of Eurotunnel’s chronic financial distress. This argument has been regularly invoked by the management staff since 1987 and is also popular among academics (Genus, 1997; Flyvberg et al., 2003). It starts from the premise that the bid submitted to governments in 1986 was prepared by a consortium of banks and construction companies that were primarily concerned in winning the financing and the construction contracts rather than in ensuring the project’s longterm viability. The concessionaire (Eurotunnel) did not exist at that time and was then a missing party in the bidding offer. Consequently, ‘‘traffic and revenue forecasts had, of course, been prepared, but the 156

means by which those figures were to be turned into reality had not been defined’’ (Kirkland, 1995; Eurotunnel Technical Director from 1985 through 1991). The autonomy of Eurotunnel remained potential after the concession had been awarded as its embryonic management was closely affiliated to the banks and the construction companies. Hence, there was no strong representative of the concessionaire to negotiate the two main contracts at arm’s length with the constructors and the banks (Grant, 1997). These contracts were extremely unfair for Eurotunnel and were quite impossible to renegotiate once the operating team was assembled. Interest margins on the debt contract were originally high and were regularly increased between 1987 and 1994. Above all, Eurotunnel had to support most of the construction risk and cost overruns. This point was made very clear by Kirkland (1995, p. 5): ‘‘Any alterations to the construction schedule to facilitate operation. . .were bound to lead to a claim for additional payment (by the construction consortium) with a consequent adverse effect on investor confidence’’. Obviously, the builders have always denied the unfairness of the contract. For them, cost overruns and delays were mainly due to unanticipated ground conditions and major safety changes to the designs imposed by the Intergovernmental Commission. On the other side, the banks and the management recognized the technology risk inherent in such an innovative project but contested the allocation of risk shaped by the construction contract. Quite obviously, this contract contained loose penalties against builders and most of the actions filed by Eurotunnel against TML were dismissed by judges or by the Disputes Panel (a committee that was set up to resolve contractual disagreements between both parties). Finally, Eurotunnel had to endorse most of the cost overruns. The fact that Eurotunnel was an ‘‘empty shell’’ at the birth of the project may be therefore one of the key determinant of its subsequent poor performance. The initial ‘‘no-management’’ structure led to the signature of long-term contracts that did not consider the interests of the firm. While partially true, this line of reasoning fails to capture the full complexity of the interaction between Eurotunnel’s initial governance structure and performance. Opposing banks and constructors on one side, and Eurotunnel on the other side misses two important points. First, the banks and the constructors were not only Eurotunnel main contractors, but also Eurotunnel initial shareholders. In theory, the key contractors’ sponsorship should have precluded any opportunistic behavior against the concessionnaire, the gains obtained as contractors being offset by the losses as shareholders. The reason why this mechanism failed is quite simple. As already argued, the original proEuropean Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

moters sold most of their shares during the construction phase and became minority shareholders in October 1986 just a few months after they won the tender offer and only two months after the signature of the construction contract. This short-term implication in equity ownership has been often interpreted as evidence that the original promoters’ priority was to derive profits from the construction and the financing contract at the expense of the operator. For some observers, it also explains why the banks and the builders had no special interest in creating a strong management staff during the first years of the project. If there is no doubt that short-term sponsorship had a negative effect on Eurotunnel performance, note however that this incentive problem could have been addressed more efficiently had the governments imposed a concession contract prohibiting original equity holders from selling their equity within a certain period of time. 8 Considering that the banks and the builders had, in essence, converging interests is also misleading. Regardless of their equity stake in the firm, the banks had a long-term interest in Eurotunnel as interest and principal payments were scheduled to commence during the operating phase. In theory, the fact that the banks negotiated the construction contract with the builders as equals should have guaranteed the signature of a contract preserving the longterm viability of the project. According to this view, Eurotunnel’s distress would not be attributable to the banks’ opportunism but rather to their technical and information deficit at the outset of the project. The appointment of new management and staff after the 1986 Initial Public Offering (IPO) didn’t change significantly the balance of power and had little positive effect on the performance of the project. The banks kept the control of the Board even if they became at that time minority shareholders. Confident in the project’s success the new dispersed shareholders were passive investors and didn’t try to impose a change of Eurotunnel governance structure. With the complicity of the affiliated management, the banks tried unsuccessfully to challenge the construction contract. Globally, the results of the new governance structure on the long-term viability of Eurotunnel were ambiguous. From 1987 to 1994, the banks regularly increased interest margins on the financing contract, which contributed to deteriorating the financial situation of Eurotunnel. Simultaneously, they helped Eurotunnel to avoid bankruptcy by rescheduling debt repayments and by encouraging small investors to bring new funds. The managers’ behavior was also ambiguous: they allied with the banks to ask for a revision of the construction contract and to communicate overoptimistic projected revenues at the time of equity offerings while denouncing the high interest margins on the debt contract. In fact, the management team had to arbitrate conflicts between two legitimate stakeholders. On one side, the managers European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

were under the pressure of the numerous French individual shareholders that asked for the adoption of a strong line with banks. The managers had therefore strong incentives to declare acting on behalf of small investors. On the opposite side, the managers were submitted to the pressure of banks. As expressed by a former member of the Board, the managers had little room to manoeuvre vis-a`-vis the bank syndicate: ‘‘Members of the bank syndicate received each day the traffic figures. . ..They were the true directors of Eurotunnel’’ (Eurotunnel director, taped interview, 2002). 9 Faced to these two conflicting interest groups, Eurotunnel managers favored the most sophisticated one, i.e the banks. The fact that the small investors’ equity stake was highly diluted and that Eurotunnel was still overindebted after the 1998 financial restructuring seems to illustrate this point. This opinion contrasts with the positive role of management assumed by some observers. Grant (1997) asserts for example that an independent staff could have represented future individual shareholders during the negotiations of the two key contracts in 1986. Similarly, Flyvberg et al. (2003) think that the new management saved the project by adopting a strong line with contractors. If partially true, this view may however overestimate the independence of managers vis-a`-vis the banks and their positive impact on the project’s performance. If the ill-defined governance structure was without any doubt at the origin of the chronic conflicts between builders, banks and dispersed shareholders, in what extent had these conflicts a negative impact on Eurotunnel performance? We have already discussed the long-term implications of the builders’ unresolved incentive problem. More generally, managers have been since the first years more implicated in the ways to resolve major conflicts between stakeholders than in the search for strategic and operational moves to improve the firm’s performance. For example, Eurotunnel had virtually no marketing policy until 1996 despite the fierce competition from ferry operators and low-cost airlines on the crossChannel market (Les Echos, March 14, 1996). In short, the Eurotunnel case shows that the mere elements supposed to improve performance, that is long-term contracts, centralized decision-making and the tight control of managerial discretion, were at the origins of bad performance.

Short-term Salient Stakeholder Theory While Eurotunnel has a number of unique features, we believe that this study can bring new insights on the way large firms actually work. In this section, we present the premises of an alternative view of corporate governance, what we call the short-term salient stakeholder theory (see Figure 4). 157

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

The attributes of a short-term salient stakeholder firm are clearly different from either a shareholder or a stakeholder firm. At first sight, the main difference with a shareholder firm lies in the identity of the salient stakeholder. The power of shareholders is especially weak in public companies where individual shareholders are numerous and lack sufficient information and expertise to evaluate highly-innovative projects. In this case, they act as ‘‘sleeping’’ stakeholders and are poorly represented by management that brings more importance to more informed and homogeneous stakeholder groups. This departure from a ‘‘pure’’ shareholder firm is however quite secondary in project companies. Remember that in this case, creditors are often the ‘‘quasi-owners’’ of the firm and carry most of the residual risk. Consequently, the shareholder view of the firm is not incompatible with a creditor-dominated governance structure. Our short-term salient stakeholder theory however clearly departs from shareholder theory on the link between governance structure and corporate perfor-

mance. In large firms where many stakeholders have conflicting objectives, the centralization of decision rights in unique hands, the limitation of managerial autonomy, and the use of long-term contracts can increase agency conflicts and limit the firm’s performance. Short-term salient stakeholder firms also differ from stakeholder firms. The governance structure gives the primacy to the interests of one stakeholder and management brings poor attention to other legitimate stakeholders. Said differently, managers do not act as referees and have here a secondary role. The reasons underlying the minor role of management are however ambiguous: Is it due to the fact that managers have limited discretion because of the firm’s governance structure or to the fact that managers rationally choose to act in the interests of the most powerful stakeholder? Both answers seem to be true in the Eurotunnel case. During the first years of the project, managers had no possibility to interfere with the promoters of the project (governments, constructors and banks) as

LEARNINGS FROM EUROTUNNEL GENERAL PROPOSITIONS For whom are firms run? Firms are run in favor of one unique stakeholder Since its creation, Eurotunnel was successively run in favor of the constructors, the banks and the group, i.e the short-term salient stakeholder. individual shareholders.

Identity of the short-term salient stakeholder Among the legitimate stakeholders, the one that possesses the more power and urgency. The group that possesses an informational advantage over other stakeholders has more power and more chance to be salient. The identity of the short-term salient stakeholder can change over time. The salient stakeholder at one period is often the one that suffered the most from corporate decisions in the near past. Role of management Management is pledged to the salient stakeholder and has little discretion.

The construction consortium had an informational advantage in the evaluation of construction costs and was Eurotunnel’s initial salient stakeholder. A few years later, the banks that suffered from the builders’ opportunism became the salient stakeholder group. They transferred part of the financial risk to illinformed individual shareholders with the complicity of affiliated managers. In 2004, the individual shareholders took the control of the Board and gained influence after being expropriated by banks.

Eurotunnel had virtually no management during the first years because the constructors had no special interest in the long-term viability of the project. Once reinforced, the staff has constantly favored banks over dispersed shareholders until the 2004 revolt.

Corporate governance and the firm’s performance Long-term contracts are not an efficient The banks negotiated the terms of the long-term governance mechanism if one party has a construction contract with the builders. Because of significant informational advantage at the their lack of technical expertise, they were unable to shape a contract reducing the constructors’ outset. opportunism.

The concentration of decision rights in the hands of one interest group increases agency conflicts between legitimate stakeholders and leads to poor corporate performance.

Expropriated stakeholders have since the beginning contested the primacy of the salient stakeholder as attested by the successive banks-constructors and shareholders-banks conflicts. Managers have spent most of their time to arbitrate these conflicts instead of elaborating value-enhancing strategies.

Conditions favoring the emergence of short-term salient stakeholder management Important information asymmetries between Builders were the best informed parties during the launching phase. After the 1986 IPO, the banks had legitimate stakeholders. access to private information not available to

Huge differences of bargaining power between individual shareholders. During this period, the banks benefited from the fact that individual legitimate stakeholders. shareholders lacked expertise to evaluate the project and did not act as an homogeneous group.

Figure 4 Key Features of the Short-term Salient Stakeholder Theory

158

European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

Eurotunnel had virtually no existence. After Eurotunnel became a public company, managers clearly chose to give the primacy to the banks over individual shareholders as attested by the constant overoptimistic projections issued at the time of equity offerings. Despite its preliminary form, our theory brings new insight on the role of management in corporate governance. Neither the strict limitation of managerial discretion advocated by shareholder theorists, nor the large managerial autonomy advocated by many stakeholder theorists can be considered as efficient tools to limit conflicts and to increase corporate performance. In the first case, managers are unable to arbitrate the conflicts between the firm’s different claimants. This situation leads to the predominance of the most sophisticated stakeholder, the one that possesses an informational or a bargaining advantage over the others. In the long-term, the resulting autocratic governance structure may however threaten the cohesion of the coalition of social actors who contribute to the firm’s survival. On the opposite side, increasing managerial autonomy is not by itself a guarantee of a more democratic governance structure: managers can be self-interested and are naturally prone to favor the most powerful interest group. The troubled role of Eurotunnel management between 1994 and 1998 clearly illustrates this point. While declaring acting in the interests of individual shareholders, the management voluntarily communicated overestimated projected revenues in order to transfer a part of the financial risk from banks to dispersed shareholders. In short, if our theory shows the limits of an autocratic type of governance it also asks the question of the feasibility of a democratic governance structure organized around a management team acting as mediator between various stakeholders. If our theory fits well with the case of Eurotunnel, one may however wonder about its scope. Is shortterm salient stakeholder management an idiosyncratic type of management or a quite well-spread one? What are the conditions prevailing to the emergence of this type of governance structure? On a general standpoint, two conditions must exist for short-term salient stakeholder management to emerge. The first one is the presence of important information asymmetries between claimants. Among legitimate stakeholders, the one that possesses more information or expertise about the firm’s perspectives has undoubtedly more power than the others. The urgency attribute defined by Mitchell et al. (1997) is also correlated to the stakeholder’s level of information: an ill-informed stakeholder is prone to stay passive and puts less pressure on managers. Short-term salient stakeholder management is also more likely when one stakeholder possesses a strong bargaining advantage when negotiating with other claimants. This privileged status could be due to European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

the retention of an informational advantage over the firm’s perspectives but also to the fact that this particular stakeholder provides a crucial resource to the organization (Pfeffer and Salancik, 1978). The conjunction of these two conditions in large project companies explains why short-term salient stakeholder theory may be especially useful to explain the failure of this type of organization. According to Flyvberg et al. (2003), the poor performance of megaprojects is partly due to risk negligence and lack of accountability in the decision-making process. Project promoters gain primarily from the construction of projects and have a self-serving interest in underestimating costs and overestimating revenues. This rent-seeking behavior is all the more likely that these promoters are powerful in the early stages of project development and do not carry the risks involved in highly risky projects. This analysis fits well with our theory that underlines the dangers associated with the concentration of decision rights in the presence of many legitimate stakeholders. In the case of large project companies, many stakeholders bear some risk and have a claim whose value depends on the project’s performance. This is the case of shareholders but also of creditors that are often considered as ‘quasi-owners’ of project companies, of employees that develop specific skills, of governments that aim to increase the quality of their country’s infrastructures, and so on. Paradoxically, the status of residual claimant is not always adapted to builders, although they are often the main promoters of large infrastructure projects. In such an environment, economic or shareholder theorists would recommend to allocate decision rights to shareholders and to sign incentive contracts with other stakeholders. The construction contract should for example include some rewards/penalties mechanisms that give construction companies the incentive to internalise possible externalities on the operating phase of the infrastructure (Dewatripont and Legros, 2005). The fact that most of large projects perform poorly whereas their governance structure corresponds quite closely to the shareholder approach (Flyvberg et al., 2003; Esty, 2004) may however give substance to our alternative theory. The use of detailed long-term contracts and the concentration of formal decision rights in the hands of shareholders are not enough to deter opportunistic behavior when shareholders lack sufficient information and expertise to shape stringent construction contracts and are at a disadvantage in negotiating with builders during the construction phase. In such a case, the presence of a third party representing the interests of ill-informed stakeholders against powerful builders would be necessary. Our theory is also well adapted to explain stakeholder relationships in a business crisis context. First, information asymmetries and agency conflicts are particularly acute when a firm is financially distressed. Second, because the fact that management 159

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

pays more attention to one specific stakeholder group, that is creditors, is usual in this particular context (Pajunen, 2006). Nevertheless, our theory goes one step further than current theories of corporate crisis. First, it shows that the failure to implement an initial governance structure that considers the interests of all legitimate stakeholders may increase the ex post agency conflicts and the probability of subsequent distress. Eurotunnel illustrates clearly this path-dependency effect. Once conflicts arise, a mechanism facilitating negotiations between conflicting stakeholders is necessary. However, the long-term resolution of conflicts will be achieved if and only if this mechanism is not biased in favor of one specific group of claimants and is perceived as fair and legitimate by the various stakeholders. A path-dependency effect may also exist at that time. To illustrate this point, consider the Eurotunnel management’s inability to resolve conflicts between the banks and the individual shareholders during the late 1990s. This failure was not surprising as the management lacked legitimacy vis-a`-vis individual shareholders. On the basis of previous managerial decisions (e.g. the optimistic projections at the time of the successive equity offerings), these shareholders suspected the management of acting on the banks’ behalf. In other words, short-term salient stakeholder management may not only increase the probability of economic and financial distress but also hinder the probability of recovery once distress arises. More broadly, short-term salient stakeholder management may be observed in other contexts where the two conditions cited above are met. This could be for example the case in firms where a large shareholder or a large debtor dominates the governance structure and exploits its informational advantage at the expense of other key stakeholders. While the focal position of managers makes them natural candidates to arbitrate conflicts between stakeholders, incentive and control mechanisms are however necessary to make sure that they really act in the interests of the firm’s multiple constituencies. In other words, legitimate stakeholders can’t trust managers blindly and must counter the natural inclination of managers to collude with the short-term salient stakeholder. This raises the problem of the feasibility of a democratic governance structure organized around a management team acting as mediator. Stakeholder theorists often pay little attention to this issue. This could be due to the fact that the stewardship theory is often used as the motivational basis for the stakeholder model of the corporation (Preston, 1998). According to this view, managers are motivated to maximize organizational performance rather than their self-interest and spontaneously try to satisfy the competing interests of stakeholders (Davis et al., 1997). While having moral appeal, this line of reasoning is essentially normative and posits how managers should behave rather than 160

how they actually behave. In fact, our theory stresses the need for the leaders to be held accountable to the interests of the different stakeholders. Dissuading managers to act in the sole interests of the short-term salient stakeholder is not however an easy task. As argued by shareholder theorists, the implementation of efficient managerial incentive mechanisms comes up against the difficulties of defining a clear managerial performance measure in the case of multiple constituencies (Jensen, 2001). While partially true, the argument used by Jensen to discredit stakeholder theory, that is ‘‘stakeholder theory directs corporate managers to serve ‘many masters’’’ (Jensen, 2001, p. 9), is over-simplistic. The problem is not to contest the fact that managers give the primacy to one particular stakeholder, but rather to impose strict rules preventing managers to favor the hold-up of powerless stakeholders. As noted by Kochan and Rubinstein (2000, p. 380), ‘‘any single enterprise is embedded in a larger normative and institutional environment where interest group compete for legitimacy and power’’. When powerless stakeholders have no voice to impose internal reforms of inequitable governance rules, laws protecting these non-salient stakeholders should be implemented. In this spirit, the short-term salient stakeholder management described in the case of Eurotunnel illustrates the failure of French laws protecting individual shareholders and suggests more broadly the need to promote an extended fiduciary duty of managers toward all the firm’s legitimate stakeholders.

Conclusion The relationship between managers and stakeholders is at the core of the corporate governance debate. This debate often comes down to the confrontation between two corporate ‘‘ideal types’’. On one side, the shareholder firm where internal conflicts are resolved through the concentration of power in the hands of shareholders and where managers have little autonomy and, on the other side, the stakeholder firm where managers have more discretion and act as mediators between the firm’s multiple constituencies. Our article contributes to this debate by examining the case of Eurotunnel, a company whose formal organizational structure reflects a shareholder management view of the firm (managers are strictly controlled by financiers) in the mere context where stakeholder theorists argue for the need to consider the interests of multiple constituencies (these ‘mega’ investments affect numerous groups or individuals). In this context, our analysis brings two important insights. First, the traditional mechanisms used by a shareholder firm to resolve conflicts (i.e. bilateral contracts and the strict limitation of managerial discretion) can be counterproductive and can result in increased conflicts. Second, managers don’t always consider the interests of the sole shareholders nor the interests of all legitimate stakeholders but rather European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

give the primacy to the most powerful party, what we call the short-term salient stakeholder. We propose then a new descriptive and instrumental stakeholder theory especially adapted to the situations where the firm’s legitimate stakeholders have very different information levels and bargaining strengths. If both of these conditions are met, our theory predicts that: (i) firms are governed in the interests of one unique interest group (ii) managers are prone to collude with the most powerful group (ii) this autocratic type of governance is unstable in the long-term as the legitimate stakeholders expropriated at one period use influence strategies to gain power in the next period. In its instrumental aspect, our theory states that the chronic conflicts associated to short-term salient stakeholder management lead to poor performance. Additional research is obviously needed to test and refine our theory. Scholars should in particular study the conditions necessary to shift from a short-term salient stakeholder management to a ‘‘true’’ stakeholder management.

Acknowledgement I am grateful to Emmanuelle Reynaud and Franck Tannery for helpful comments and suggestions.

Notes 1. Donaldson and Preston (1995) consider for example that stakeholder theory is based on the idea that the interests of all the stakeholders are of intrinsic value. No details are given on how to measure this variable. Quite obviously, this core assumption of stakeholder theory rests on ethical claims (see Donaldson, 1999 and Jones and Wicks, 1999 for a development on this point). 2. Some finance theorists contest the narrow view of corporate governance which focuses exclusively on the conflict between outside investors and top managers. Berglo¨f and Von Thadden (1999) assert in this way: ‘‘. . .we think that it is important to recognize at the outset that control over a firm’s course involves more than these two groups of actors’’. 3. For example, the Cadbury Report in England (1992) or the Vie´not Report in France (1995). 4. Winch (1998, p. 10) observes that ‘‘the only source of cost information and an estimated outturn cost the banks had was TML’’. 5. Robert Bain, an associate infrastructure analyst at Standard and Poor’s, asserts in this spirit that ‘‘Rewarding the lowest-cost bidder in isolation [from other factors] seems to be a rather dangerous strategy. It leads to the submission of frankly unsupportable bids and it’s these unsupportable bids that lead to trouble later’’ (Financial Times, April 10, 2004). 6. All the figures are expressed in £. This choice can be justified in two ways: (a) From the launching of the project to 2002, Eurotunnel documents used indifferently £ and French francs because of Eurotunnel’s dual nationality (b) Among these two currencies, we chose the one that still exists after the passage to Euro. 7. At the time of the 1994 equity offering, many financial analysts expressed their confusion about Eurotunnel’s projections. An analyst even declared: ‘‘Eurotunnel’s projections are overoptimistic. A new equity issue will be necessary in two or three years’’ (Les Echos, May 27, 1994).

European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007

8. Flyvberg et al. (2003) note the development of this type of mechanism in the UK. 9. The fact that the banks received each day the traffic figures may appear surprising at first sight. It is however consistent with the banks’ dominant position in the Board. This could also be due to the fact that the banks had the authority to put the firm into bankruptcy as Eurotunnel was in financial distress since 1989. In order to avoid bankruptcy, managers had no choice but to communicate their private information to the bank syndicate.

References Agle, B., Mitchell, R. and Sonnenfeld, J. (1999) Who matters to CEOs? An investigation of stakeholder attributes and salience, corporate performance, and CEO values. Academy of Management Journal 42(5), 507–525. Anguera, R. (2006) The Channel Tunnel – an ex post economic evaluation. Transportation Research Part A 40, 291–315. Aoki, M. (1984) The Economic Analysis of the Japanese Firm. North-Holland, Amsterdam. Berglo¨f, E. and Von Thadden, E.-L. (1999) The changing corporate governance paradigm: Implications for transition and developing countries. Paper presented to Annual World Bank Conference on Development Economics, Washington D.C. Available at SSRN: http:// ssrn.com/abstract=183708. Charreaux, G. and Desbrie`res, P. (2001) Corporate governance: Stakeholder value versus shareholder value. Journal of Management and Governance 5(2), 107–128. Davis, J., Schoorman, F. and Donaldson, L. (1997) Toward a stewardship theory of management. Academy of Management Review 22(1), 20–47. Dewatripont, M. and Legros, P. (2005) Public-private partnerships: Contract design and risk transfer. European Investment Bank Papers 10(1), 120–145. Donaldson, L. (1990) The ethereal hand: Organizational economics and management theory. Academy of Management Review 15(3), 369–381. Donaldson, T. (1999) Making stakeholder theory whole. Academy of Management Review 24(2), 237–241. Donaldson, T. and Preston, L. (1995) The stakeholder theory of the corporation: Concepts, evidence, and implications. Academy of Management Review 20(1), 65–91. Esty, B. (2003) The Economic motivations for using project finance. Unpublished working paper, Harvard Business School. Esty, B. (2004) Why study large projects? An introduction to research on project finance. European Financial Management 10(2), 26–42. Fama, E. and Jensen, M. (1983) Separation of ownership and control. Journal of Law and Economics 26, 301–325. Flyvberg, B., Holm, M.C. and Buhl, S. (2002) Underestimating costs in public works projects. Journal of the American Planning Association 68(3), 279–295. Flyvberg, B., Bruzelius, N. and Rothengatter, W. (2003) Megaprojects and risk: an anatomy of ambition. Cambridge University Press. Freeman, R.E. (1984) Strategic management: A stakeholder approach. Pitman, Boston. Freeman, R.E. and Evan, W.M. (1990) Corporate governance: a stakeholder interpretation. Journal of Behavioral Economics 19(4), 337–359. Frooman, J. (1999) Stakeholder influence strategies. Academy of Management Review 24(2), 191–205. Genus, A. (1997) Managing large-scale technology and inter-organizational relations: The case of the Channel Tunnel. Research Policy 26, 169–189. Grant, M. (1997) Financing Eurotunnel. Japan Railway & Transport Review, 46–52. Hart, O. and Moore, J. (1990) Property rights and the nature of the firm. Journal of Political Economy 98, 1119–1158. 161

NEITHER SHAREHOLDER NOR STAKEHOLDER MANAGEMENT

Hill, C.W. and Jones, T.M. (1992) Stakeholder-agency theory. Journal of Management Studies 29(2), 131–154. Jensen, M. (1986) Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76(2), 323–329. Jensen, M. (1993) The modern industrial revolution, exit, and the failure of internal control systems. Journal of Finance 48(3), 831–880. Jensen, M. (2001) Value maximization, stakeholder theory, and the corporate objective function. Journal of Applied Corporate Finance 14(3), 8–21. Jensen, M. and Meckling, W. (1976) Theory of the firm, managerial behaviour, agency costs and ownership structure. Journal of Financial Economics 3, 305–360. Jensen, M. and Meckling, W. (1994) The Nature of Man. Journal of Applied Corporate Finance Summer, 4–19. Jones, T. and Wicks, A. (1999) Convergent stakeholder theory. Academy of Management Review 24(2), 206– 221. Kirkland, C. (1995) The Channel Tunnel – Lessons learned. Tunnelling and Underground Space Technology 10(1), 5–6. Kleimeier, S. and Megginson, W. (2000) Are project finance loans different from other syndicated credits? Journal of Applied Corporate Finance 13(1), 75–87. Kochan, T. and Rubinstein, S. (2000) Toward a stakeholder theory of the firm: The Saturn Partnership. Organization Science 11(4), 367–386.

Mitchell, R., Agle, B. and Wood, D. (1997) Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts. Academy of Management Review 22, 853–886. Pajunen, K. (2006) Stakeholder influences in organizational survival. Forthcoming in Journal of Management Studies. Pfeffer, J. and Salancik, G.R. (1978) The external control of organizations. Harper & Row. Preston, L. (1998) Agents, stewards and stakeholders. Academy of Management Review 23(1), 9. Rappaport, A. (1990) The staying power of the public corporation. Harvard Business Review 68(1), 96–104. Shleifer, A. and Vishny, R. (1997) A survey of corporate governance. Journal of Finance 52(2), 737–783. Stonham, P. (1995) The Eurotunnel Rights Issue. Part One: Strategy. European Management Journal 13(2), 201–211. Tirole, J. (2001) Corporate governance. Econometrica 69(1), 1–35. Tirole, J. (2006) The theory of corporate finance. Princeton University Press. Williamson, O. (1991) Strategizing, economizing, and economic organization. Strategic Management Journal 12, 75–94. Winch, G. (1998) The Channel fixed link: Le projet du sie`cle. Unpublished case study, Manchester University.

LAURENT VILANOVA, University of Lyon 2, IEF – MFB, 93 Chemin des Mouilles, 69131 Ecully, France. E-mail: Laurent. Vilanova@univ-lyon2. fr Laurent Vilanova is Professor of Finance at University of Lyon 2. His research interests include financial intermediation, law and finance, project finance and corporate restructurings.

162

European Management Journal Vol. 25, No. 2, pp. 146–162, April 2007