Loading field map
Public policy scholars have long wrestled with a deceptively simple question: why do governments do what they do? The answers have shifted dramatically over the past century, as researchers moved from describing formal institutions to modeling the behavior of actors, then to explaining the messy, multi-actor processes that produce policy change. The history of public policy analysis is not a smooth accumulation of knowledge but a series of competing frameworks, each offering a different diagnosis of what matters most—institutions, interests, ideas, or networks.
The earliest approach to studying public policy focused on the formal structures of government. The Institutional/Traditional Approach treated policy as the output of constitutional arrangements, legislative procedures, and administrative hierarchies. Scholars described how bills became laws, how agencies implemented statutes, and how courts reviewed decisions. This work assumed a clear separation between politics (which set goals) and administration (which carried them out). By the 1950s, however, many researchers grew frustrated with purely descriptive accounts that could not explain why similar institutions produced different policies or why policies changed over time.
The Behavioralism movement swept through political science in the 1950s, demanding that scholars study observable behavior rather than formal rules. In public policy, this meant tracking how legislators voted, how interest groups lobbied, and how bureaucrats exercised discretion. Behavioralism did not replace the institutional approach so much as narrow its focus: institutions still mattered, but only as arenas where actors pursued their interests.
Three frameworks emerged from this behavioral ferment, each offering a different account of who really drives policy. Elite Theory argued that a small, cohesive group of economic and political elites makes the important decisions, while the public is largely passive. Pluralist Theory countered that power is dispersed among many competing groups, and policy outcomes reflect the shifting balance of group pressures. Systems Theory took a more abstract view, treating the policy process as a system that converts demands and supports from the environment into authoritative outputs. These three frameworks coexisted in lively disagreement throughout the 1960s and 1970s, with pluralism dominating American political science while elite theory found support among critics of corporate power.
In 1959, Charles Lindblom published a landmark article challenging the rational-comprehensive model of decision-making. Incrementalism argued that policymakers rarely start from scratch; instead, they make small, marginal adjustments to existing policies. Lindblom portrayed this as both realistic and defensible, since it reduces conflict and allows learning from experience. Incrementalism narrowed the scope of earlier frameworks by focusing on the cognitive and political constraints on decision-making rather than on broad power structures.
By the 1970s, scholars wanted a more systematic way to organize the policy process. The Policy Stages Heuristic divided the process into a sequence of stages: agenda-setting, formulation, adoption, implementation, and evaluation. This framework provided a useful teaching tool and a common vocabulary, but it soon drew criticism for implying that policy moves in neat, linear steps. Researchers found that stages often overlap, that feedback loops are common, and that the heuristic says little about why policies change. The stages approach gradually gave way to more dynamic theories.
Rational Choice Theory entered public policy from economics, applying the assumption that actors are self-interested utility maximizers. Scholars used formal models to analyze legislative behavior, bureaucratic discretion, and interest group competition. Rational choice offered parsimony and deductive power, but its assumptions often seemed unrealistic to scholars who observed the complexity of real policymaking. The framework coexisted uneasily with behavioral and pluralist approaches, each side accusing the other of missing important features of political life.
New Institutionalism emerged in the 1980s as a direct response to the limitations of both behavioralism and rational choice. It revived the old institutionalist interest in formal structures but transformed it by incorporating insights from organization theory and cognitive psychology. New institutionalists argued that institutions—understood as formal rules, informal norms, and taken-for-granted scripts—shape actors' preferences and strategies, not just constrain them. This framework absorbed the behavioral focus on actors while insisting that institutions are more than arenas; they are constitutive of political life. New Institutionalism remains one of the most influential frameworks today, especially in comparative policy studies.
By the mid-1980s, a new generation of scholars sought theories that could explain specific patterns of policy change without claiming universal applicability. The Multiple Streams Framework (MSF), developed by John Kingdon, adapted the garbage can model of organizational choice to the policy process. MSF argues that policy change occurs when three streams—problems, policies, and politics—converge during a window of opportunity, often pushed by a policy entrepreneur. The framework differs from earlier approaches by emphasizing ambiguity and timing rather than rational calculation or institutional constraint.
Punctuated Equilibrium Theory (PET), introduced by Frank Baumgartner and Bryan Jones in 1993, explains why policy is usually stable but occasionally undergoes dramatic shifts. Drawing on agenda-setting research, PET argues that institutional friction and issue redefinition produce long periods of incremental change interrupted by bursts of major reform. This framework complements MSF by providing a more systematic account of stability and change, and it challenges the incrementalist assumption that small adjustments are the norm.
The Advocacy Coalition Framework (ACF), developed by Paul Sabatier and Hank Jenkins-Smith, focuses on the role of belief systems and coalitions over periods of a decade or more. The ACF assumes that policy subsystems are composed of advocacy coalitions that hold deep core beliefs and policy core beliefs, and that policy change occurs through learning, external shocks, or negotiated agreements. The ACF differs from MSF by emphasizing long-term coalition dynamics rather than short-term windows, and it differs from PET by focusing on belief change rather than agenda access. These three mid-range theories—MSF, ACF, and PET—now dominate empirical policy process research, each with its own strengths: MSF for explaining sudden change, ACF for understanding coalition behavior, and PET for linking agenda-setting to institutional friction.
While mid-range theories largely accept positivist assumptions about explanation and measurement, a different set of approaches emerged in the 1990s that questioned those assumptions altogether. Constructivist/Critical/Poststructuralist Approaches draw on discourse analysis, feminist theory, and poststructuralism to examine how policy problems are socially constructed, how power operates through language, and how certain groups are marginalized in policy debates. These approaches do not aim to predict policy change but to reveal the hidden assumptions and power relations embedded in policy frameworks. They coexist in productive tension with the mid-range theories, often critiquing them for ignoring the constitutive role of ideas and the exclusionary effects of seemingly neutral policy tools.
At the same time, a separate line of research shifted attention from government to governance. Governance Frameworks argue that policymaking increasingly involves networks of public, private, and nonprofit actors rather than hierarchical state institutions. Scholars study how these networks coordinate, how accountability works in the absence of clear authority, and how the state's role has changed from rowing to steering. Governance frameworks differ from earlier institutional approaches by emphasizing horizontal relationships and the blurring of boundaries between state and society. They have been especially influential in European policy studies and in research on regulation, public management, and multi-level governance.
Today, public policy analysis is a pluralistic field. The leading frameworks—New Institutionalism, Multiple Streams, Advocacy Coalition, Punctuated Equilibrium, and Governance—agree that policy processes are complex, multi-actor, and shaped by both institutions and ideas. They disagree on what drives change: MSF emphasizes timing and entrepreneurship, ACF emphasizes belief systems and learning, PET emphasizes agenda access and institutional friction, and governance scholars emphasize network dynamics. Critical and constructivist approaches add a further layer by questioning the very categories used in mainstream research. This diversity is a strength: each framework illuminates a different facet of the policy process, and researchers increasingly combine them to address specific questions. The field has moved far from the days of formal institutional description, but the core question remains: why do governments do what they do?