Structure, Action, and Outcomes: The Dynamics of Power in Social Exchange

Abstract

Two levels of power - the structure of power in exchange networks, and the strategic use of power resources by actors - affect the frequency and distribution of exchange outcomes. Theories of power and exchange suggest alternative hypotheses about how structural power and strategic action are related to each other and to outcomes. These are tested with data from a series of experiments in which power relations are based on control over rewards and punishments. Both structural power and its strategic use have substantial effects on exchange outcomes, but structure and action are related only weakly to each other. Their relative effects on exchange outcomes vary with the base of power. Structural power has stronger effects on reward exchange, and strategic action on punishment exchange. The asymmetry of reward exchange in the relation, the traditional measure of power use in powerldependence theory, is affected by both. Social scientists have conceptualized power in various ways: as a structural potential (Bierstedt 1950; Wrong 1968; Emerson 1962), as a process of behavioral or tactical influence (Rubin and Brown 1975; Michener and Suchner 1972; Tedeschi and Bonoma 1972), and as the successful outcome of influence (Dahl 1957; Mayhew, Gray, and Richardson 1969; Simon 1957). I argue that a complete analysis of power must include all three facets. Understanding the relations among them requires integrating concepts at micro and macro levels of analysis (Alexander 1987; Munch and Smelser 1987). The macro level of power is the structure of control that provides the opportunities and constraints within which the micro level of power, the strategic behavior of actors, operates. Together, structure and action produce outcomes of consequence for actors and relationships. I investigate how structure and action affect the amount and distribution of exchange in power/dependence relations. Previous research on power and exchange has studied only the relations between power structure (the struc* This work was supported by a grant from the National Science Foundation (SES-84 19872). I am