Every day, governments make decisions that shape the lives of ordinary people. A city council votes to change zoning rules, and a neighborhood changes character. A state legislature cuts funding for community colleges, and tuition rises. A federal agency issues new safety standards, and manufacturers must redesign their products. A president signs an executive order, and thousands of workers face new workplace requirements. An international trade agreement takes effect, and the prices of goods in grocery stores shift. None of these outcomes happened automatically. Each one was the result of a long, contested, complicated process — a process called public policy.

Public policy is how governments decide what to do and what not to do. It is the set of choices, rules, priorities, and actions through which public authority addresses collective problems. Policy is not simply a law or a regulation. It is a pattern of decisions stretching from the identification of a public problem through the design of a response, the adoption of rules, the implementation of programs, the evaluation of results, and the ongoing revision of what government does. Understanding policy is understanding government in action — not government as an abstraction of institutions and constitutions, but government as the engine that distributes benefits, sets rules, protects rights, builds infrastructure, manages risk, and shapes daily life.

This chapter introduces the public policy process. It is written for students who want to understand not just what governments decide, but how and why those decisions happen. The policy process is neither a clean administrative procedure nor a purely rational exercise in problem-solving. It is a political process, shaped by competing interests, contested values, limited information, resource constraints, public pressure, institutional rules, and the unpredictable rhythms of political attention.

Public policy is the set of decisions, actions, priorities, rules, and choices through which governments address collective problems and pursue public goals. The word "public" in public policy does two things at once. It signals that policy is about matters that affect people collectively, not merely private individuals, and it signals that policy is carried out by public institutions — governments — using public authority and public resources.

A useful starting definition: public policy is what governments choose to do and choose not to do. This definition has the virtue of reminding students that inaction is itself a form of policy. When a government chooses not to regulate a particular industry, not to fund a particular program, or not to intervene in a particular conflict, that choice has consequences just as surely as active intervention does. Policy exists in both action and omission.

Policy is not simply a law, although laws are one important form that policy takes. Public policy includes statutes passed by legislatures, regulations issued by administrative agencies, executive orders signed by presidents or prime ministers, court decisions that establish legal rules, budget appropriations that determine how money is spent, treaties with foreign governments, administrative guidelines that govern how agencies behave, and informal practices that have solidified over time into consistent patterns of government behavior.

A second important feature of public policy is that it involves choices. Every policy decision reflects a judgment about what problems deserve attention, what solutions are acceptable, what values should take priority, and who should bear the costs. A decision to fund public education through local property taxes rather than state or federal taxes reflects a set of judgments about equity, local control, and the role of government. These judgments are not purely technical. They involve values, politics, interests, and power.

Third, public policy is a process, not a moment. When a law is signed, that is not the end of the policy story. The law must be implemented by agencies. Agencies must write regulations and guidelines. Programs must be administered, staffed, and funded. Results must be monitored. Problems with implementation must be identified and addressed. Affected groups react to the policy and may organize to change it. Courts review the policy for constitutional and legal compliance. A policy that looks clear on paper may look very different by the time it reaches the citizen it was designed to serve.

Students sometimes think of policy as the technical management of public problems: identify the problem, apply the expert solution, monitor results, adjust as needed. This technocratic view has some merit — expertise and evidence genuinely matter in policy-making — but it misses the fundamentally political character of public policy.

Policies always involve choices among competing values. Should public funds be used to support parents who choose religious schools, or only secular public schools? Should environmental regulations prioritize economic efficiency or ecological protection? Should immigration policy prioritize family reunification, economic need, or national security? These are not questions that can be answered by technical analysis alone. They require political judgments about values, priorities, and what kind of society a community wants to create.

Policies also involve choices about who pays and who benefits. A tax cut benefits some taxpayers more than others. A regulation places costs on some businesses while protecting some workers or consumers. Because policy distributes costs and benefits unequally, it is always contested. Those who pay more than they receive have strong incentives to oppose policies, while those who receive more than they pay have strong incentives to support them. This is not selfishness or irrationality — it is the normal political dynamic of democratic governance.

Finally, policies involve choices under uncertainty. Governments rarely know with confidence what the effects of a policy will be before they adopt it. Economic models make assumptions that may not hold. Social programs interact with complex human behavior. Good policy-making requires not only technical expertise, but intellectual humility, adaptive management, and willingness to revise course when evidence indicates that a policy is not achieving its goals.

Public policy covers nearly every domain of collective life. The traditional core areas include national security and defense, public order and criminal justice, economic management and regulation, taxation and public finance, and social welfare. Over time, the scope has expanded enormously. Modern governments make policy about education from preschool through higher education, food safety, drug approval, environmental pollution, financial markets, workplace conditions, transportation safety, telecommunications, immigration, natural disasters, public health emergencies, and infrastructure. The reach of modern government is enormous, and public policy is the mechanism through which that reach operates.

This breadth means that policy is not the exclusive province of elected officials. Thousands of administrators, regulators, judges, technical experts, and service providers make decisions every day that shape the policy experience of ordinary citizens. The teacher who interprets attendance rules, the caseworker who decides whether a family qualifies for assistance, the building inspector who grants or denies a permit — all of these actors are making policy in practice, even if they are not making policy in theory.

Not every problem that exists in society becomes a public problem that government tries to solve. The transformation of a condition into a recognized public problem requiring government action is itself a political process. It involves identifying the condition, framing it in ways that make government action seem appropriate, mobilizing political support, overcoming opposition, and gaining the attention of decision-makers. This process is called problem identification and problem definition, and it is the first stage of the policy cycle.

The most important insight here is that problems do not define themselves. A condition — say, a rising rate of accidental overdose deaths — exists in the world. But whether that condition is defined as a law enforcement problem, a public health problem, a regulatory problem, or a social problem depends on how different actors frame and interpret the situation. Each framing leads to different policy responses and creates different winners and losers. Problem definition is not neutral. It is itself a form of political power.

Several factors influence whether a condition becomes a recognized public problem. The first is visibility. Problems that are easy to see, measure, and document are more likely to attract attention than problems that are hidden or diffuse. A dramatic industrial accident is more visible than chronic exposure to low-level pollution. Media coverage plays a major role in making some conditions visible while others remain invisible.

The second factor is organized interests. Groups that are well-organized, well-funded, and politically connected are more effective at bringing their problems to government attention. Farmers, manufacturers, veterans, teachers' unions, and medical associations all have organized representation that can translate their members' problems into policy demands. Diffuse groups — consumers, low-income individuals, future generations — are typically less well-organized and less effective, even when their problems are serious and widespread.

The third factor is political opportunity. Even when a condition is visible and organized groups care about it, government may not act unless the political moment is right. A party with strong views on the issue may have just won an election. A crisis may have made inaction politically untenable. A policy entrepreneur — someone who champions a solution and works to connect it to receptive political decision-makers — may have found the right moment to push.

The fourth factor is the existing definition of government responsibility. Some conditions are widely accepted as appropriate subjects for government action; others are not. The definition of what government should do shifts over time. What was once considered a purely private matter — domestic violence, food safety, workplace safety, child labor — has in many countries been redefined as a legitimate public problem requiring government response.

When a condition is brought to political attention, it must be framed. Framing refers to the way a problem is defined — what kind of problem it is, who is responsible, what causes it, and what kinds of solutions are appropriate. Different framings activate different political coalitions, different policy repertoires, and different distributional consequences.

Consider homelessness. If homelessness is framed as a housing affordability problem, the appropriate response involves subsidizing housing construction and providing rent assistance. If it is framed as a mental health problem, the appropriate response involves expanding mental health services. If it is framed as a problem of urban disorder, the appropriate response may involve policing and enforcement. In reality, homelessness is caused by multiple interacting factors. But the political process rarely accommodates complex multi-causal frames. Different coalitions rally around different frames, and the dominant framing at any given moment shapes the kinds of policies that are considered.

The policy agenda is the list of issues that government institutions are actively considering and treating as priorities. Agenda setting is the process by which issues get onto that list. It is one of the most important stages of the policy process because issues that never reach the agenda are never acted upon, regardless of how serious they are.

Political scientists distinguish between different levels of agenda. The systemic agenda includes all issues that are publicly recognized as legitimate subjects of political concern. The institutional agenda is the narrower set of issues that specific government bodies are actively considering. The decision agenda is the even narrower set of issues on which action is actually imminent. Most social conditions never reach the institutional agenda. Of those that do, most never reach the decision agenda. Of those that reach the decision agenda, many do not result in policy action.

In democratic systems, many actors compete to place issues on the agenda. Political leaders — presidents, prime ministers, party leaders, legislative committee chairs — have significant agenda-setting power. A president who identifies climate change as a priority can focus executive agencies, direct regulatory activity, propose legislation, deliver speeches, and frame the issue for public discussion. Interest groups invest heavily in agenda setting, working to get their preferred issues onto the agenda and to keep threatening issues off it. Media organizations shape the agenda by deciding which issues to cover and how prominently to feature them. When major news outlets devote sustained attention to an issue, political leaders find it harder to ignore it.

Public opinion shapes the agenda as well, though its influence is complex. Strong, sustained, and broadly shared public concern about an issue can push it onto the institutional agenda. But public opinion is often fluid, poorly informed about specific policy questions, and susceptible to framing effects. Skilled political communicators can shift public attention away from some issues and toward others, at least temporarily.

Political scientist John Kingdon developed one of the most influential models of agenda setting. Kingdon argued that three largely independent streams flow through the political system: the problem stream (conditions that might be defined as public problems), the policy stream (possible solutions developed by experts, advocates, and policy communities), and the politics stream (the current political climate, including elections, public opinion, and partisan competition).

Most of the time, these streams flow separately. A policy window — a brief period during which agenda change becomes possible — opens when the three streams come together. A focusing event, such as a disaster or a crisis, suddenly elevates a problem and creates political pressure for action. A solution that has been waiting in the policy stream is quickly attached to the elevated problem. Political conditions make action feasible. Policy windows do not stay open long. This is why policy entrepreneurs — advocates who work to connect problems, solutions, and political moments — are so important.

Once a problem has been identified and has reached the agenda, the question becomes what to do about it. Policy design is the stage at which governments develop specific responses: what tools to use, how to structure programs, who will administer them, how they will be funded, who will benefit, and who will bear costs. Good policy design requires understanding both the problem to be solved and the institutional context in which the solution will operate. A policy that works well in one setting may fail in another.

Governments have a range of tools — often called policy instruments or policy tools — for addressing public problems. Different instruments work through different mechanisms, impose different costs and benefits, and face different political obstacles.

Policy designers must also consider targeting. Means-tested programs direct benefits to those below a certain income threshold, which concentrates help on the most disadvantaged but requires administrative machinery for determining eligibility and may stigmatize recipients. Universal programs provide benefits to everyone in a category — all children, all retirees — which reduces administrative complexity and stigma but is more expensive. The choice between these approaches reflects both technical considerations and deeply held values about the purpose of public programs.

Policy adoption is the stage at which a proposed policy is formally approved by an authoritative decision-making body. In most democratic systems, major policy adoption happens through legislation — the enactment of laws by a legislature — though it also happens through executive orders, administrative rulemaking, court decisions, and budget appropriations.

The legislative process varies significantly across political systems, but its common features include proposal, deliberation, revision, negotiation, and formal vote. A bill is introduced — either by a government minister (in parliamentary systems) or by an individual legislator (common in congressional systems). The bill may be referred to a committee that holds hearings, receives expert testimony, takes public comment, and marks up the bill with amendments. In bicameral systems, the bill must pass both chambers, and differences must be reconciled. The final bill is then sent to the executive for signature or veto.

This process involves constant negotiation. In parliamentary systems where the government typically controls the legislative majority, the government's bills are likely to pass, but opposition parties use the legislative process to scrutinize and criticize. In presidential systems like the United States, where the legislature and executive are independently elected and may be controlled by different parties, passing major legislation requires assembling coalitions across partisan lines and making compromises that may significantly alter the original proposal.

Much policy adoption happens outside the formal legislative process. Administrative agencies adopt regulations through rulemaking — a process in which the agency proposes a rule, publishes it for public comment, responds to comments, and issues a final rule. This process creates binding legal rules but happens within the executive branch and is governed by administrative law rather than by legislative procedure.

Executive orders allow presidents and similar executives to direct the activities of executive agencies without legislative approval. They operate within existing law but can significantly shape how laws are implemented. They can be issued quickly and can be reversed by successor executives.

Court decisions adopt policy when judges interpret statutes, apply constitutional principles, and review executive action. A court decision that strikes down a law on constitutional grounds, establishes a new standard for regulatory review, or expands the scope of a civil rights statute is effectively adopting policy — shaping the rules that govern public and private behavior.

One of the most well-documented patterns in policy adoption is incrementalism — the tendency of political systems to make small, marginal adjustments to existing policies rather than comprehensive reforms. Political scientist Charles Lindblom observed that decision-makers rarely evaluate all possible policy alternatives against all possible criteria. Instead, they work at the margin of the status quo, considering modest changes to existing programs. Incrementalism is a realistic response to bounded rationality — the cognitive and informational limits that prevent decision-makers from processing all relevant information — and to political fragmentation, which makes comprehensive reform politically difficult.

Major, non-incremental policy change does occur — at moments of crisis, political realignment, or extraordinary political opportunity. The New Deal of the 1930s, the Great Society programs of the 1960s, and the Affordable Care Act of 2010 in the United States; the creation of the National Health Service in the United Kingdom after World War II; the construction of welfare states across Western Europe in the postwar decades — these represent moments of significant departure from the incremental pattern. They typically occurred when a policy window opened, a political coalition was assembled, and the normal constraints on comprehensive change were temporarily relaxed.

Policy adoption is not the same as policy success. A bill that is signed into law, a regulation that is published in the official record, an executive order that is announced — these are the beginning of the policy story, not its conclusion. The real test of a policy is whether it achieves its goals in practice, and that depends on implementation.

Policy implementation is the process by which adopted policies are put into effect. It involves organizing administrative capacity, staffing programs, issuing guidance and regulations, contracting with service providers, communicating with affected populations, monitoring compliance, enforcing rules, and handling the countless problems that arise when general policy meets specific circumstances.

A central finding of implementation research is that there is frequently a significant gap between policy as designed and policy as implemented. Programs that look good on paper fail to reach intended beneficiaries. Regulations that look strict on paper are routinely evaded. Programs that are adequately funded at the national level are underfunded at the local level where they are actually delivered. Staff who are supposed to apply rules consistently exercise judgment that produces wide variation in outcomes.

Implementation gaps occur for many reasons. Policy designers may not fully understand how the programs they design will operate in practice. Legislation may be deliberately vague — the price of assembling a coalition — leaving implementing agencies to resolve disagreements that the legislature could not. Agencies may lack the resources, staff, expertise, or political support needed to implement a program effectively. Regulated entities may use their informational advantages to resist or evade compliance.

Implementation is also affected by federalism. In federal systems, national policies are often implemented by state or local governments, which may have different priorities, different administrative capacity, and different political orientations. A national program that allocates funds to states for health care or education may be implemented very differently in different states, producing significant variation in outcomes even under the same national law.

Political scientist Michael Lipsky introduced the concept of street-level bureaucracy to describe the implementing actors who interact directly with citizens: teachers, police officers, social workers, public health nurses, housing inspectors, unemployment counselors. These are the front-line workers who make policy real for ordinary people.

Street-level bureaucrats exercise enormous discretion in practice, even when formal rules constrain their choices. A police officer decides whether to make an arrest or issue a warning. A caseworker decides whether a family meets the eligibility criteria for a benefit. A teacher decides how strictly to enforce attendance policies. These decisions, made millions of times each day across the public sector, collectively constitute the actual implementation of policy. Discretion allows for judgment, flexibility, and humane application of rules — but it also creates risks: inconsistency, bias, arbitrariness, and deviation from legislative intent.

Policy evaluation is the systematic assessment of whether a policy is achieving its goals, at what cost, and with what unintended consequences. It is the stage at which governments and citizens ask: is this working? The answers to that question inform decisions about whether to continue, expand, reform, or terminate a policy.

Evaluation is both a technical and a political activity. Technically, evaluation involves defining what success looks like, collecting relevant data, measuring outcomes, and assessing costs and benefits. Politically, evaluation involves questions about who defines success, whose data is used, who conducts the evaluation, and who is empowered to act on its findings. Evaluations that find policies working well support political actors who championed them; evaluations that find policies failing threaten those actors. Evaluation is therefore not insulated from politics.

Process evaluation assesses whether a program is being implemented as designed — whether services are being delivered, whether targets are being reached, whether funds are being spent as authorized. Outcome evaluation assesses whether the policy is producing the intended changes — whether a job training program increases employment, whether a pollution regulation reduces emissions. Impact evaluation attempts to establish causal effects by comparing what happened to people or places affected by the policy with what would have happened in the absence of the policy. Cost-benefit analysis compares the total costs of a policy with its total benefits, attempting to determine whether the benefits justify the costs — a methodology that requires placing monetary values on things that may not naturally have prices, such as human lives or clean air, which is methodologically complex and normatively contested.

One of the most important insights from policy studies is that policies create politics. Once a policy is adopted, it creates constituencies, institutions, interests, and expectations that make it difficult to change. Social Security in the United States has built a broad coalition of beneficiaries and their families who mobilize politically to defend it. The mortgage interest tax deduction has created a large constituency of homeowners who benefit from it. These feedback effects mean that policies tend to persist even when evidence suggests they are not working well, and major policy reform requires overcoming the political forces that existing policies have created.

Path dependence refers to the tendency of policy systems to stay on the tracks established by earlier choices. Early policy decisions constrain later options, not because the original choice was necessarily optimal, but because the institutions, interests, and expectations built around it make change costly. Health care systems, education systems, and pension systems all exhibit path dependence: they are shaped profoundly by the decisions made decades earlier.

The preceding sections have described the policy process as a sequence of stages from problem identification through evaluation. But this description may create a misleadingly orderly picture. In practice, the policy process is not linear, sequential, or orderly. It is messy, contested, iterative, and political.

Politics pervades every stage of the policy process. Problem definition is contested. Agenda setting is contested. Policy design is contested. Policy adoption is contested. Implementation is contested. Evaluation is contested. Reform or termination is contested. None of this is pathological. It is the normal operation of democratic politics. Democracy is a system for making collective decisions peacefully, under conditions of genuine disagreement about values, interests, and facts. The disagreements do not disappear; they are managed through procedures — votes, deliberation, negotiation, legal review — that give all affected parties some opportunity to participate.

Two models of political influence on the policy process deserve attention. The iron triangle model identifies three sets of actors who exert disproportionate influence over a given policy domain: a legislative committee that has jurisdiction over the issue, an executive agency that administers the relevant programs, and an interest group that represents the most directly affected economic interests. These three actors form a triangle of mutual dependence. The triangle is "iron" in the sense that it is stable and difficult for outside actors — including the general public — to penetrate.

Political scientist Hugh Heclo argued that the iron triangle model is too simple. In most policy domains, influence is exercised not by a tight triangle of three players but by a much larger issue network — a fluid, loosely organized set of participants who include legislators and their staff, agency officials, interest group representatives, academic researchers, think tank analysts, journalists, and advocates. Issue networks are larger, more diverse, and more open than iron triangles, but they are also more fragmented and less stable.

Both models recognize that policy-making is not primarily responsive to the undifferentiated preferences of "the public." It is responsive to organized, informed, persistent political actors who are actively engaged in specific policy domains. Citizens who want to influence policy need to organize, become informed, and engage persistently — not just vote once every few years and then disengage.

Political scientists Michael Cohen, James March, and Johan Olsen developed a provocative alternative model of policy-making that they called the garbage can model. In their view, organizational decision-making — including government decision-making — does not proceed in a rational, sequential fashion. Instead, problems, solutions, decision-makers, and choice opportunities all flow independently through organizations, like items thrown into a garbage can. Decisions happen when these streams happen to connect — when a decision opportunity arises, decision-makers who are paying attention at that moment connect it to a problem they care about and a solution that happens to be available.

The garbage can model captures something important: the policy process is often less rational than it appears. Solutions often precede problems — advocates develop policy ideas and then look for problems that justify them. Decision-makers often do not know what problem they are solving until after they have made a decision. Chance, timing, and the accidental juxtaposition of problems, solutions, and political moments shape outcomes as much as deliberate design. This does not mean that rationality and evidence are irrelevant — it means that rationality operates within political and institutional constraints that shape which problems are visible, which solutions are available, and which political moments make action possible.

Public policy is the primary mechanism through which governments honor or fail to honor their obligations to citizens. The rights that constitutions guarantee — to equal protection, to due process, to free expression, to economic security — are realized or denied not only through courts, but through the everyday operation of public programs, regulations, and administrative decisions. When a government cuts funding for legal aid, low-income individuals may be unable to vindicate their legal rights in court. When a government fails to enforce anti-discrimination law, protected groups lose the practical enjoyment of their rights. Policy failure is often rights failure.

Citizens have responsibilities in relation to public policy as well. Democratic governance requires an informed, engaged citizenry. Citizens who understand how the policy process works are better equipped to hold their governments accountable. Citizens who participate — not only by voting, but by engaging with public comment processes, attending local government meetings, contacting their representatives, joining civic organizations, and engaging in public debate — exercise their democratic rights and help to ensure that policy reflects something closer to the genuine public interest. Civic responsibility also includes a responsibility for accuracy and honesty in policy debate: engaging honestly, acknowledging uncertainty, and treating those with whom one disagrees with the respect owed to fellow citizens.