Strategic Decisions in Complex Stakeholder Environments: A Theory of Generalized Exchange

Stakeholder theory offers a strong starting point for this analysis. Work following Freeman’s first conceptualization (Freeman, 1984) has focused on the definition and prioritization of actors related to the firm in a variety of contexts. Mitchell (1997) offered a chronology of stakeholder definitions, ranging from actors necessary for the survival of the firm, to actors in the environment which could be affected by the firms actions. Our approach is decidedly instrumental, in that we seek to connect stakeholder outcomes to actions of the firm. Previously in Cameron (2008), we outlined how the construction of a network can operationalize stakeholder definitions. By requiring that stakeholders provide or enable important inputs to the firm, we established completeness criteria for the set of relevant stakeholders. As a corollary to the theory developed in this article, we will prioritize the stakeholders, using emergent properties of the network. Mitchell (Mitchell, 1992; Agle, Mitchell, and Sonnenfeld, 1999) advanced an influential framework in the instrumental tradition (Donaldson and Preston, 1995) for quantitatively valuing stakeholders along three dimensions. We build Mitchell’s stakeholder quantification out in an important dimension: rather than parameterizing the stakeholder, we parameterize the relationship to the stakeholder, which can be broken out into separate transactions – some important, some not. While our characterization does not capture the same richness in the time/ urgency or legitimacy dimensions as Mitchell, we ground the understanding of power in economic exchange conditions. The economic foundations for this work lie in the theory of General Equilibrium in microeconomics, originally developed by Arrow, Debreu, and McKenzie (1954). In this framework exchange prices depend on quantity demanded, location, and timing. Relaxations include information asymmetries (Akerlof, 1970) and uncertainty in the future state of the system (Magill and Quinzii, 2009). The branch of economics that has dealt with the relaxation for indirect transactions is principal – agent theory. In an agent problem, the principal transfers risk across a contract to an agent. The literature explores the structuring of contracts to reduce costs (Prendergast, 1999), choose the right agents (Nalebuff and Stiglitz, 1983), and prevent moral hazard (lack of effort or withholding of information by the agent) (Arrow, 1963). The majority of the literature focuses on principal-agent pairs; whereas, we will use the concept recursively, such that actor 1 can be both the principal for actor 2 and the agent of actor 3. Said otherwise, we will explicitly create cycles in the graph of actors. As such, the contracts represented in this model will be significantly simpler than most studied under principal-agent theory. Where these contracts represent non-monetary exchanges, utility theory has been used to record the preferences of actors. For simplicity, we also consider these actors internally consistent – we ignore the