Three foundational works including: Emerson (1976), Thibaut & Kelley (1959), and Cropanzano & Mitchell (2005), argued that behavior is sustained by contingent returns from others. Rewards and costs. Interdependence. Comparison levels. Power and dependence. Norms and rules. Relationships that develop over time.
Together, these sources provide the conceptual content around longitudinal relations, network structures, and the imperfect social structures where exchange actually happens.
Thibaut & Kelley (1959) supply the analytic engine. The matrix of outcomes maps every possible interaction and its consequences—rewards and costs—for each party. They introduce Comparison Level (CL) as the standard for satisfaction and CLalt as the standard for staying or leaving. The dyad is the starting point; power, norms, roles, and tasks follow. The basic assumption: every individual voluntarily enters and stays in a relationship only as long as it is adequately satisfactory in terms of rewards and costs.
Emerson (1976) defines the frame of reference. Social exchange is limited to actions contingent on rewarding reactions from others—a two-sided, mutually contingent, mutually rewarding process. He confronts the charges of tautology, rationality, and reductionism head-on, and resolves them by adopting the social relation as the unit of analysis. From there, he extends exchange from micro to macro through productive exchange and exchange networks.
Cropanzano & Mitchell (2005) bring it into organizational science. They believe that relationship building is like climbing a ladder. They trace SET’s roots, then clarify what the field keeps muddling: exchange rules (reciprocity, negotiated, and alternatives like altruism, group gain, status consistency, competition), resources (six types, collapsing into economic vs. socioemotional outcomes), and relationships (POS, LMX, commitment, trust). Their key move: separating relationships-as-transactions from relationships-as-interpersonal-attachments. The causal ambiguity—does exchange cause the relationship or the relationship cause the exchange?—is answered through a developmental model.
Social exchange theory is not a single theory but a frame of reference within which many theories—some micro, some macro—can speak to one another. Emerson (1976) defines its scope: “Social exchange as here conceived is limited to actions that are contingent on rewarding reactions from others.” Implied is a two-sided, mutually contingent, and mutually rewarding process involving transactions or simply exchange. The approach can be described as the economic analysis of noneconomic social situations.
Emerson examines the psychological underpinnings of exchange theory, particularly Homans’ use of Skinnerian operant psychology. The operant research format typically involves (a) a single subject, (b) studied over time, (c) in a bounded environment, (d) where the experimenter enjoys real social power. For a social exchange theorist, operant research studies an organism-environment exchange system. Homans summarizes this in three basic propositions: the Success Proposition, the Stimulus Proposition, and the Deprivation-Satiation Proposition, supplemented by the Value Proposition and the Rationality Proposition.
Skinnerian. Relies on value as the result of prior conditioning in longitudinal relationships. Allows, but does not require, rationality. Behavior is sustained by reinforcing activity from others.
Assumes intelligent, well-informed actors formulate probabilities and estimate expected utilities prior to deciding. Rationality is either an assumption or a normative model.
Economic theory is organized around the perfectly competitive market—which assumes homogeneous goods, no friction, no personal ties. Social exchange theory, by contrast, deals with exactly those structures that violate these assumptions: longitudinal relations between interdependent actors, in which power is neither diffuse nor equally distributed. The problem of bilateral monopoly (isolated exchange) is where economic theory flounders and social exchange theory picks up.
The transition from micro to macro exchange theory requires recognizing three features distinguishing the levels: elementary behavior is subinstitutional, the actor is usually an autonomous individual, and the analysis is dyadic. To extend beyond the dyad, Emerson proposes productive exchange (group problem solving, division of labor) and exchange networks (the Kula ring, matrilateral cross-cousin marriage). Networks tie together both groups and individuals as actors.
Thibaut and Kelley present a theory of interpersonal relations and group functioning. The major analytic technique is a matrix formed by taking account of all the behaviors two individuals might enact together. Each cell represents one of the possible parts of the interaction and summarizes the consequences—rewards (pleasures, satisfactions, gratifications) and costs (factors that inhibit or deter performance)—for each person. The matrix defines all the different things they might possibly do in interaction.
Exogenous determinants are external to the relationship—abilities, propinquity, similarity of attitude, complementary needs, power and status. Endogenous determinants arise from the interaction itself—response interference, facilitation, and sequential effects (satiation and fatigue). Interference raises costs and usually lowers rewards; synchronization of sets and elimination of incompatible behaviors are means of avoiding interference.
Early interaction is a sampling process. Formation depends on whether experienced outcomes meet standards of acceptability (CLalt). Strangeness, autistic hostility, and autistic friendliness bias the sample of outcomes.
CL is a modal or average value of all known outcomes, weighted by salience. Outcomes above CL are satisfying; below CL are unsatisfying. Individual differences in confidence and perceived power affect CL.
Power derives from dependence. A has fate control over B if he can affect B’s outcomes regardless of what B does. A has behavior control if he can make it desirable for B to change his behavior. Fate control can be converted to behavior control through matching rules. In any dyad both members are interdependent; each has some power over the other. Power strategies include developing better alternatives, reducing the other’s alternatives, improving one’s ability to deliver rewards, and building up the value of one’s product.
A norm is a behavioral rule accepted to some degree by both members. Norms serve as substitutes for informal influence—they control behavior without the costs, uncertainties, resistances, and power losses involved in the unrestrained use of interpersonal power. A role is the class of norms that applies to a person’s behavior with regard to a specific problem or class of others. Tasks have states; conjunctive tasks require coordination; disjunctive tasks permit independent work. Task structure affects the pattern of interdependence and the need for norms.
Social exchange theory is among the most influential conceptual paradigms for understanding workplace behavior. Its roots trace back to at least the 1920s, bridging anthropology, social psychology, and sociology. However, theoretical ambiguities within SET remain. As a consequence, tests of the model, as well as its applications, tend to rely on an incompletely specified set of ideas. The core ideas that comprise SET have yet to be adequately articulated and integrated. The review addresses four issues: (a) the roots of the conceptual ambiguities, (b) norms and rules of exchange, (c) the nature of the resources being exchanged, and (d) social exchange relationships.
Three types: (a) reciprocity as interdependent exchanges—contingent interpersonal transactions; (b) reciprocity as a folk belief—the cultural expectation that people get what they deserve; (c) reciprocity as a moral norm—a standard that describes how one should behave.
More explicit and quid pro quo than reciprocal exchanges. Duties and obligations are detailed and understood. May continue beyond short-term agreements and may be bound by legal or contractual sanctions.
Foa and Foa’s resource theory organizes six resources along two dimensions: particularism (worth varies by source) and concreteness (tangibility). Within organizational science, these collapse into economic outcomes (financial, tangible) and socioemotional outcomes (social and esteem needs, symbolic, particularistic). Tsui et al.’s typology of employee-employer relationships—quasi-spot, mutual investment, underinvestment, overinvestment—is based on the types of resources exchanged.
The review proposes a critical distinction: relationships-as-transactions versus relationship-as-interpersonal-attachment. Separating the exchange relationship from the form of exchange yields four cells: Cell 1 (social transaction in a social relationship), Cell 2 (economic transaction in a social relationship), Cell 3 (social transaction in an economic relationship), and Cell 4 (economic transaction in an economic relationship). Cells 2 and 3 are mismatches that provide fertile ground for future research.
Does the exchange cause the relationship or does the relationship cause the exchange? The resolution: relationship development is not a single stimulus-response but more analogous to climbing a ladder. The outcome from one stage becomes the input for the next. Cause and effect are determined by where in the sequence the researcher elects to examine the process. Relational benefits can be both a result and a resource for exchange.
All three works converge on a central viewpoint: social exchange as a frame of reference. They share a common vocabulary—reward, reinforcement, cost, value, utility, resource, transaction, outcome—and a common premise: behavior is sustained by contingent returns from others.
Defines exchange as a frame of reference, not a theory. Resolves the tautology and reductionism controversies by adopting the social relation as the unit of analysis. Extends exchange from micro to macro through productive exchange and exchange networks.
Provides the matrix of outcomes as the major analytic technique. Introduces CL and CLalt as standards for evaluating relationships. Analyzes the determinants of rewards and costs, the formation and evaluation of dyads, and the dynamics of power, norms, roles, and tasks.
Traces SET’s roots, clarifies exchange rules (reciprocity, negotiated, and alternatives), examines resources (six types; economic vs. socioemotional), and analyzes social exchange relationships in work settings (POS, LMX, commitment, trust). Proposes a typology separating transactions from relationships and resolves the causal ambiguity.
Social exchange theory has become one of the most influential conceptual paradigms for understanding workplace behavior. It has been applied to social power, networks, board independence, organizational justice, psychological contracts, and leadership, among other areas. Yet conceptual ambiguities remain, and tests of the model often rely on an incompletely specified set of ideas.
Nearly a half-century ago, Homans presented a concept of social behavior based on exchange. He introduced the notion that exchanges are not limited to material goods but also include symbolic value—approval and prestige. His writings bridged a variety of disciplines and sparked differing theories of social exchange. The three works examined here—Emerson’s frame of reference, Thibaut and Kelley’s matrix of outcomes, and Cropanzano and Mitchell’s interdisciplinary review—together provide the conceptual tools needed to deal with exactly those topics that economic theory has trouble with: market imperfections, longitudinal relations, and imperfect social structures.