Garbage Can Model

framework · management · philosophical-tradition

Cohen, March, and Olsen's framework characterizing decision-making in organized anarchies as the chance confluence of problems, solutions, participants, and choice opportunities.

The Garbage Can Model is the organizational decision-making framework articulated by Michael D. Cohen, James G. March, and Johan P. Olsen in their 1972 Administrative Science Quarterly article 'A Garbage Can Model of Organizational Choice.' The model characterizes decisions in 'organized anarchies' — organizations with three defining features: problematic preferences (participants disagree about goals and values), unclear technology (means-ends relationships are poorly understood), and fluid participation (participants enter and leave decisions inconsistently). In such organizations, decisions emerge from the contingent intersection of four temporally-decoupled streams: problems looking for solutions, solutions looking for problems, choice opportunities (decision-making occasions), and participants who happen to be present. The 'garbage can' metaphor captures how decisions result from temporal-fortuitous mixing rather than rational means-ends reasoning. The model substantially challenged rational-choice and bounded-rationality models of organizational decision-making and remains foundational in organization theory.

Originators

Michael D. Cohen (University of Michigan, foundational co-author); James G. March (Stanford University, foundational co-author and substantial subsequent body of work); Johan P. Olsen (University of Bergen, foundational co-author); Foundational 1972 Administrative Science Quarterly article; intellectual antecedents in Herbert Simon's bounded-rationality work (March and Simon's Organizations 1958, Cyert and March's A Behavioral Theory of the Firm 1963), broader Carnegie School organization-decision tradition; subsequent development through March's substantial body of work including A Behavioral Theory of the Firm (with Cyert), Decisions and Organizations (1988), A Primer on Decision Making (1994) high

Year / Decade

1972 (foundational publication); ongoing development through Cohen, March, Olsen subsequent work and broader organization-theory tradition high

Primary sources

Cohen, M.D., March, J.G. & Olsen, J.P. (1972). 'A Garbage Can Model of Organizational Choice', Administrative Science Quarterly, Cohen, M.D. & March, J.G. (1974). Leadership and Ambiguity: The American College President, Kingdon, J.W. (1984, multiple editions). Agendas, Alternatives, and Public Policies (Multiple Streams Framework), Bendor, J., Moe, T.M. & Shotts, K.W. (2001). 'Recycling the Garbage Can: An Assessment of the Research Program', American Political Science Review high

Core components

Organized anarchy: organization with problematic preferences (no clear shared goals), unclear technology (means-ends relationships poorly understood), fluid participation (participants engage inconsistently)
Four streams in organizational decision-making: (1) Problems — issues seeking attention
(2) Solutions — answers actively seeking questions, developed independently of specific problems
(3) Choice opportunities — formal occasions where decisions are expected (meetings, review cycles, budget approvals)
(4) Participants — individuals who happen to be engaged in particular choice opportunities
Decision dynamics: decisions emerge from contingent temporal intersection of streams in choice opportunities, rather than from rational problem-identification then solution-development
Three resolution modes: (1) Resolution — decision actually solves problem
(2) Oversight — choice made without attaching to active problems
(3) Flight — problems leave the choice to attach to other choices
Computer-simulation foundation: 1972 article includes simulation results from formal model of stream intersection
Application contexts: universities (Cohen-March 1974 Leadership and Ambiguity studied US college presidents)
public bureaucracies
political bodies
scientific research institutions
Subsequent extensions: Kingdon's 1984 Agendas, Alternatives, and Public Policies adapted the model for policy analysis (Multiple Streams Framework)
March's subsequent 'organizational learning' work
new-institutional-economics organizational-decision research
Distinction from rational-choice and bounded-rationality models: garbage-can model treats irrationality not as deviation from rationality but as normal organizational dynamic

Primary use case

Foundational framework in organization theory and organizational decision-making research; applied principally in: research on universities (Cohen-March's foundational US-college-presidents study), public bureaucracies, government agencies, scientific research institutions, policy-making processes; Kingdon's Multiple Streams Framework adaptation for policy-process analysis; academic and professional reference in organization theory, public administration, political science, and management literature; intellectual foundation for organizational learning research (March's substantial subsequent work), organizational ambiguity research, post-rational-decision-making frameworks; pedagogical reference in organization-theory and decision-making graduate curricula globally; complementary to bounded rationality, behavioral decision research, and broader Carnegie School tradition; subject of extensive subsequent empirical and theoretical development with substantial citation and research engagement.

Common criticisms

The Garbage Can Model has been substantially debated as descriptive versus normative model — Cohen-March-Olsen presented it as descriptive of decision-making in organized anarchies, but the framework has been substantially used as critique of organizations that do not match its assumptions, with implications about decision rationality that some critics argue exceed the model's actual scope
the 'organized anarchy' designation has been critiqued as tautological — the model applies to organizations with the three characteristics it specifies, but identification of those characteristics often involves judgment that may import the analytical conclusion
empirical testing of the original computer-simulation results in actual organizations has been mixed — some empirical research supports the model's predictions while other research finds rational-choice and bounded-rationality frameworks fit specific decision contexts better
the 1972 article's narrow empirical base (university decision-making) limits generalization to other organizational contexts, with subsequent research extending applicability but producing substantial variation in empirical fit
Bendor, Moe, and Shotts's 2001 critique 'Recycling the Garbage Can: An Assessment of the Research Program' substantially challenged the model's empirical and theoretical foundations, generating ongoing scholarly debate
the four-streams framework is parsimonious but the boundaries between streams are sometimes contested in actual decision contexts — specifically, distinguishing 'solutions looking for problems' from 'problems looking for solutions' in observed decisions can be analytically slippery
Kingdon's Multiple Streams Framework adaptation for policy analysis has been productive but has produced its own intra-tradition debate about the relationship between original Cohen-March-Olsen model and Kingdon's policy-process variant
the model's implicit normative implications (rational decision-making is hard, organizational anarchy is unavoidable in many contexts) have been argued by rational-design advocates to underweight the available techniques (formal decision processes, structured decision-making, strategy infrastructure) for reducing garbage-can dynamics
cross-cultural applicability has been incompletely studied.

Lineage

Siblings: Bounded Rationality