Loading field map
Public policy is the study and practice of how governments decide what to do and what not to do, and of the effects those decisions produce. It examines the entire lifecycle of collective action through the state: how problems come to be seen as public rather than private, how options are formulated and chosen, how decisions are implemented through laws, regulations, programs, and spending, and how those actions are evaluated and revised. As a field of study, it sits at the intersection of political science, economics, law, and administration, but it is distinguished from each by its practical orientation: the central question is not merely how politics works, but how governing can be made to work better.
The field is organized around a cluster of enduring questions. First, agenda-setting: why do some conditions become problems requiring government action while others remain private troubles or are ignored entirely? Second, decision: among possible responses, how do governments choose, and what role do evidence, interests, institutions, and ideas play in that choice? Third, implementation: why do policies so often fail to produce what their designers intended, and what determines whether a law on the books becomes a change on the ground? Fourth, evaluation: how can we know whether a policy achieved its goals, at what cost, and with what unintended consequences? Fifth, legitimacy and feasibility: what makes a policy acceptable to those who must comply with it and to the public that funds it, and how do political constraints shape what is even considered?
The stakes are correspondingly high. Public policy determines who gets what from the state—who is taxed, who is subsidized, who is protected, who is punished. It allocates life chances in areas as varied as health, education, income security, environmental protection, and public safety. Because these decisions are made under conditions of scarcity, uncertainty, and disagreement, the field is inherently normative as well as empirical. Analysts cannot avoid questions of value: what counts as a good outcome, whose welfare matters, and how trade-offs between competing goods should be resolved. The field's practical ambition—to improve the quality of collective decisions—requires both rigorous description of how policy actually works and reasoned argument about how it should work.
The modern study of public policy emerged in the mid-twentieth century, but it drew on much older traditions. Eighteenth- and nineteenth-century political economy, particularly the work of Adam Smith and later utilitarian thinkers, asked what functions government should perform in a market society. German and American public administration scholars in the late nineteenth and early twentieth centuries studied how bureaucracies could implement laws efficiently. The Progressive Era in the United States promoted the idea that expertise and scientific management could make government more effective. These precursors shared a concern with the state's capacity to act, but they did not yet constitute a distinct field of policy analysis.
The field proper took shape after the Second World War, especially in the United States. Several developments converged. The rise of the welfare state created a large apparatus of programs that demanded systematic evaluation. The behavioral revolution in political science shifted attention from formal institutions to actual decision-making processes. The development of economics, particularly welfare economics and cost-benefit analysis, offered tools for comparing policy options. The term "policy science" was coined in the early 1950s by the political scientist Harold Lasswell, who argued for a discipline that would integrate knowledge about the policy process with knowledge for use in that process. Lasswell's vision was explicitly democratic: policy scientists should serve the public interest, not merely the state.
A second major strand developed in the 1960s and 1970s, driven by the expansion of federal social programs in the United States and comparable developments in other wealthy democracies. Governments began to demand systematic evidence about whether programs worked. This gave rise to policy analysis as a professional practice—the application of economic reasoning, statistical methods, and program evaluation to concrete decisions. Graduate schools of public policy were founded to train analysts, and a distinct professional identity emerged. This strand was more technocratic than Lasswell's vision, emphasizing analytical technique over political deliberation.
A third strand, beginning in the 1970s and accelerating since, focused on explaining why policies take the forms they do. Political scientists studying agenda-setting, policy change, and comparative public policy developed theories of the policy process itself. This strand is more academic and explanatory, asking not "what should government do?" but "why does government do what it does?" It drew on institutional economics, organizational theory, and comparative politics, and it produced the frameworks that now organize much of the field's theoretical literature.
These three strands—normative analysis of what government should do, practical analysis of what works, and positive explanation of how policy is made—have never fully merged. They coexist in departments of political science, schools of public policy, and government agencies, with different weights in different settings. The field's identity is defined less by a single method or theory than by a shared subject matter and a shared commitment to connecting knowledge to action.
The oldest and most durable organizing framework for studying public policy is the stages or cycle model. It breaks the policy process into a sequence: problem identification, agenda-setting, policy formulation, adoption, implementation, and evaluation. This model originated in the 1950s and became a standard pedagogical device. Its appeal is intuitive: it mirrors how textbooks and government documents describe the policy process, and it provides a checklist for practitioners.
The stages model has been heavily criticized. It implies a neat linearity that real processes rarely exhibit; policies are often reformulated during implementation, evaluation feeds back into agenda-setting, and multiple stages proceed simultaneously. It also says little about causation—it describes where things happen but not why. Few scholars today treat it as a theory. Yet it remains influential as a heuristic, and much of the field's research can be located within one of its stages. The most important theoretical work has been devoted to explaining particular stages, especially agenda-setting and implementation.
Why do some problems capture government attention while others languish? The most influential answer is the multiple streams framework, developed by John Kingdon in the 1980s, building on the "garbage can" model of organizational choice. Kingdon argued that three independent streams flow through the policy system: problems (conditions that come to be seen as needing attention), policies (proposed solutions developed by experts and advocates), and politics (electoral, partisan, and interest-group dynamics). These streams normally operate separately, but occasionally they converge during a "policy window"—a moment when a problem is salient, a solution is available, and political conditions are favorable. Policy entrepreneurs exploit these windows to couple the streams and push their proposals onto the agenda.
The multiple streams framework explains why agenda-setting is often chaotic and contingent rather than rational and orderly. Its limits are that it is better at describing how issues rise than at predicting which issues will rise, and its metaphors can be applied so loosely that they explain everything and nothing. Nevertheless, it remains a standard tool for understanding why some problems become policy problems at particular moments.
A second influential account of policy change, developed by Frank Baumgartner and Bryan Jones in the 1990s, draws on bounded rationality and institutional theory. Their punctuated equilibrium theory holds that policy typically proceeds through long periods of stability, during which a particular "policy image" (how an issue is understood) and a particular institutional arrangement reinforce each other. Occasionally, however, a shift in the policy image—often triggered by new information, a focusing event, or a change in venues—can produce rapid, dramatic change. The theory explains why policy change is usually incremental but sometimes abrupt.
The key mechanism is friction: institutions are designed to resist change, filtering out most pressures for alteration. When pressures accumulate beyond a threshold, the system lurches. This theory has been tested extensively with data on budgets, legislation, and media attention across many countries. Its contribution is to explain both stability and change within a single framework, and to connect cognitive limits (bounded rationality) to institutional design. Its limits include difficulty in specifying exactly when friction will break and in accounting for the content of change, as opposed to its timing and magnitude.
The advocacy coalition framework, developed by Paul Sabatier and Hank Jenkins-Smith in the 1980s, explains policy change over periods of a decade or more. It begins from the observation that policy subsystems—the set of actors who are active on a particular issue—are organized into coalitions bound together by shared belief systems. These beliefs are arranged in a hierarchy: deep core beliefs (fundamental values), policy core beliefs (basic positions on the issue), and secondary aspects (instrumental details). Coalitions translate their beliefs into policy through strategies that use resources such as money, expertise, and political access.
Policy change occurs when coalitions learn, when external events (economic shocks, elections, changes in public opinion) shift the balance of resources, or when internal shocks destabilize a dominant coalition. The framework emphasizes that policy conflict is often durable and belief-driven, not simply interest-driven, and that learning across coalitions is rare because actors filter information through their beliefs. Its strength is its attention to the long time horizons and complex actor networks that characterize real policy subsystems. Its limits include the difficulty of defining subsystem boundaries and the framework's complexity, which makes it hard to test cleanly.
A fourth major approach applies rational choice theory and institutional economics to policy. The most prominent version is the institutional analysis and development framework, developed by Elinor Ostrom and colleagues. It focuses on how the "rules of the game"—formal laws, informal norms, and enforcement mechanisms—shape the incentives facing actors in a policy arena. The framework is used to analyze common-pool resource problems, such as fisheries, forests, and irrigation systems, and to ask when communities can govern resources sustainably without state coercion or privatization.
Ostrom's work showed that neither state control nor market privatization is the only solution to collective action problems; communities often devise their own institutional arrangements that work well. The framework specifies the components of an action situation—participants, positions, actions, information, control, and costs—and analyzes how different rule configurations produce different outcomes. Its contribution is to make institutional design a central object of study and to provide a systematic language for comparing governance arrangements. Its limits include the demanding data requirements for full application and the difficulty of predicting which institutional arrangements will emerge from a given context.
A more recent strand focuses on policy design: the deliberate choice of policy instruments and their configuration. This approach, associated with scholars such as Anne Schneider, Helen Ingram, and Michael Howlett, examines the "tools of government"—regulations, taxes, subsidies, information campaigns, service provision, and so on—and asks how the choice of tools affects policy outcomes and political support. A key insight is that policies are not neutral instruments; they construct target populations in ways that shape their political standing. Schneider and Ingram argued that policies send messages about which groups are deserving and which are not, and that these messages feed back into politics, creating a "degenerative policy process" when punitive policies are directed at weak groups.
This approach connects policy analysis to democratic theory and social construction. It asks not only whether a policy works but what kind of political community it creates. Its limits are that it is more diagnostic than predictive, and its normative commitments can make it difficult to apply in settings where the analyst does not share its democratic premises.
Alongside these explanatory theories, the field includes a strong practical tradition: policy analysis as a craft. This tradition is rooted in welfare economics and decision theory. Its core method is cost-benefit analysis: enumerate the consequences of alternative policies, value them in monetary terms, discount future effects, and choose the option with the greatest net benefit. In practice, analysts also use cost-effectiveness analysis (when benefits cannot be monetized), risk analysis, and distributional analysis to examine who gains and who loses.
This tradition has been enormously influential in government, particularly in the United States, where regulatory impact analysis and budget review processes institutionalize it. Its strengths are clarity, comparability, and a commitment to evidence. Its limits are equally well documented. Cost-benefit analysis requires valuing things that are difficult to value—human life, environmental quality, distributional fairness—and the methods used to do so are contested. It tends to privilege efficiency over equity, and it can obscure the political judgments embedded in technical choices. Critics argue that it falsely presents value-laden decisions as neutral calculations.
A related practical tradition is program evaluation, which uses social science methods—randomized controlled trials, quasi-experimental designs, qualitative case studies—to estimate the effects of existing programs. The rise of "evidence-based policy" in the 1990s and 2000s pushed governments to demand rigorous evaluations before scaling up programs. This movement has produced genuine advances in knowledge about what works in areas like education, employment, and social welfare. Its limits include the difficulty of generalizing from one context to another, the long time lags between evaluation and use, and the risk that evidence is used selectively to justify predetermined positions.
Public policy is not confined to the nation-state. Comparative public policy studies how different countries address similar problems, asking why policy choices diverge and what explains convergence. This literature has identified the importance of political institutions (federal versus unitary systems, electoral rules, veto points), welfare state regimes, and policy legacies in shaping national approaches. It has also documented the spread of policy ideas across borders through learning, emulation, and coercion—processes captured by the concept of policy diffusion.
A related development is the growth of multilevel governance. In the European Union, in federal systems, and in international regimes, policy is made at multiple levels simultaneously, and the boundaries between levels are increasingly blurred. This complicates the traditional picture of a single government choosing and implementing policy. It also raises new questions about accountability, legitimacy, and coordination.
The field has also become more attentive to the Global South. Early policy studies were heavily based on the experience of wealthy democracies, and their models often assumed institutional conditions—strong states, capable bureaucracies, functioning legal systems—that do not hold everywhere. Scholars of development policy have shown that policy processes in low-income countries are shaped by different constraints: weaker state capacity, greater reliance on international donors, more pervasive informality, and different patterns of state-society relations. This work has enriched the field by challenging assumptions that were previously taken for granted.
Contemporary public policy is characterized by several durable features. First, it is methodologically plural. Quantitative and qualitative approaches coexist, and the field increasingly values mixed methods that combine causal inference with attention to process and meaning. Second, it is substantively broad. Policy scholars work on every issue governments touch, from climate change to artificial intelligence to pandemic preparedness, and the field's theories are applied across these domains. Third, it is institutionally diverse. The field is practiced in universities, think tanks, government agencies, international organizations, and advocacy groups, and the norms of these settings differ.
The relationship between analysis and politics remains the field's central tension. The technocratic impulse—to find the best policy and persuade decision-makers to adopt it—coexists with the political science insight that policy is the product of power, interests, and ideas, not just evidence. The most sophisticated work in the field treats this tension as a subject of study rather than a problem to be solved. It asks how evidence is used and abused in political contexts, how expertise is legitimated, and how democratic accountability can be maintained in a world of complex, technical decisions.
The field's future directions are shaped by current challenges: the rise of algorithmic governance and the question of how to regulate artificial intelligence; the urgency of climate policy and the difficulty of making decisions under deep uncertainty; the erosion of trust in expertise and institutions; and the persistence of inequality, which raises distributional questions that efficiency-oriented analysis has often neglected. These challenges are not new in kind, but they press on the field's unresolved questions: how to combine analytical rigor with democratic legitimacy, how to make policy under uncertainty, and how to ensure that the study of policy contributes to better collective choices rather than merely to more sophisticated descriptions of how choices are made.