Policy Cycle

Michael Howlett and Sarah Giest, Simon Fraser University, Burnaby, BC, Canada
Ó 2015 Elsevier Ltd. All rights reserved.

Abstract

Analyzing policy development in terms of policy cycles and policy styles is useful for several reasons. Not only does it help to
make sense out of the different models put forward earlier in the history of policy studies, such as the rational, incremental,
and garbage can models of public policy decision making, it also helps to advance studies of these models by specifying the
conditions under which different styles of policy making occur. This article sets out the logic of the policy cycle model and
justifies its five-stage vision of that process.

The Policy Cycle Model of the Policy Process they can be applied (Kingdon, 1984; Howlett et al., 2009). It
also does not answer several key questions such as the actual
The idea that the policy process can be thought of as a series of substance of policy, the number and type of relevant actors
steps in a cyclical model of decision making was first broached involved in the process, the exact manner and sequence in
systematically in the work of Harold Lasswell, a seminal figure which actual policy development processes occur, and
in the development of the policy sciences at the University of whether there exist basic patterns of development in different
Chicago and Yale University in the early 1950s (Lasswell, issue areas, sectors, or jurisdictions (Sabatier, 1991).
1956, 1971). At present, a five-stage model of the policy Overall, when using the policy cycle idea, diligent attention
process is the most commonly used, although many other has to be paid to the fact that the model is a guideline for the
versions exist with more fewer substages. In this model, analysis of a far more complex reality.
‘agenda setting’ refers to the first stage in the process when
a problem is initially sensed by policy actors and a variety of
solutions put forward (see Agenda Setting, Public Policy in). Styles of Policy Behavior in the Policy Cycle
‘Policy formulation’ refers to the development of specific
policy options within government when the range of possible Empirical studies of policy-making processes undertaken to
choices is narrowed by excluding infeasible ones and efforts are answer these questions and generate more precise models of
made by various actors to have their favored solution ranked the policy-making process have developed more specific
highly among the remaining few. ‘Decision making’ refers to knowledge about the interactions of policy actors at each stage
the third stage in which formal actors in government adopt of the cycle, summarized below. These draw attention to the
a particular course of action. In the fourth stage of ‘policy limited number of common modes, or styles, which charac-
implementation,’ governments put their decisions into effect terize policy processes throughout the world. A policy style is
using some combination of the tools of public administration also often part of a larger ‘policy regime’ that emerges over time
in order to alter the distribution of goods and services in and includes not only how policy deliberations take place, but
society in a way that is broadly compatible with the sentiments also the kinds of actors and ideas present. As such, it is useful in
and values of affected parties. Finally, ‘policy evaluation’ refers identifying long-term patterns in content and process of public
to the fifth stage in the processes in which the results of policies policy making (Howlett et al., 2009). Common policy styles at
are monitored by both state and societal actors, often leading each stage of the policy cycle characterize the principal
to the reconceptualization of policy problems and solutions in components of policy dynamics. A policy style describes how
the light of experiences encountered with the policy in ques- governments go forward during, for example, the agenda-
tion and the start of a new iteration of the cycle (Howlett et al., setting stage and which effects result from that stage for the
2009). ongoing policy-making process. They are tools for
This idea of policy making existing as a set of interrelated understanding the problem-solving directions of actors in the
stages provides a general ‘framework’ for understanding the cycle and for revealing the institutional routines and
policy development process and points to several of the crucial procedures underlying policy processes.
temporal activities and relationships that should be examined
in furthering the study of the issue. Although a useful
Agenda Setting
conceptual or analytical tool, the policy cycle model has some
limitations as an empirical description of policy-making reality In the scholarly literature on agenda setting, a useful distinc-
due to its simplification of highly contingent and complex tion is drawn between the systemic or unofficial public agenda
policy processes. The reality of policy making is not as and the institutional or formal, official, agenda. The systemic
systematic and linear as the model might suggest. Also, the agenda “consists of all issues that are commonly perceived by
stages are often compressed, skipped, or change their order members of the political community as meriting public
entirely. For example, policy formulation can sometimes attention and as involving matters within the legitimate juris-
precede agenda setting, as ‘solutions seek problems’ to which diction of existing governmental authority” (Cobb and Elder,

288 International Encyclopedia of the Social & Behavioral Sciences, 2nd edition, Volume 18 http://dx.doi.org/10.1016/B978-0-08-097086-8.75031-8
 Policy Cycle 289

1972, p. 85). This is essentially a society’s agenda for the The power of ideas and their stability in policy subsystems,
discussion of public problems, such as crime or health care, in particular, has been a subject of much attention in studies of
water quality or wilderness preservation. The formal or policy formulation in recent years. Carstensen (2011), for
institutional agenda, on the other hand, consists of only example, has suggested that ideas change incrementally,
a limited number of issues or problems to which attention is when new elements of meaning are added to them, resulting
devoted by policy elites (Baumgartner and Jones, 1991; in a characteristic process of policy formulation in which
Kingdon, 1984). similar ideas inform similar policy options developed over
Research conducted by John Kingdon into the dynamics of long periods of time. Focusing on the information or
agenda setting in the US Congress in the early 1980s focused on knowledge available to decision makers has led policy
the timing of agenda entrance. In his work, he focused on the research in this vein to focus on the role of epistemic or
role played by policy entrepreneurs both inside and outside policy communities in policy formulation (see Haas, 2012).
governments in taking advantage of agenda-setting opportu- These are loose groupings of experts who serve as knowledge
nities – policy windows – to move items onto formal govern- ‘providers’ for the decision-making process – opening up
ment agendas (Kingdon, 1984). Kingdon argued that the opportunities for them to influence the content and timing of
characteristics of issues (the problem stream) combined with policies (Dunlop, 2009, p. 290; Haas, 1992).
the characteristics of political institutions and circumstances Scholars over the years have developed a variety of taxon-
(the politics stream) and the development of policy solutions omies to help identify who the key actors in these policy
(the policy stream) lead to the opening and closing of subsystems are, what brings them together, how they interact,
opportunities for agenda entrance. Such opportunities can and what effect their interaction has on policy development.
be seized upon or not, as the case may be, by policy Most of these distinguish between a larger set of actors with
entrepreneurs who are able to recognize and act upon them. some knowledge of the policy issue in question, and a smaller
The policy windows resulting from this include routine, set in which actors not only have requisite knowledge, but also
spillover, discretionary, and random types (Kingdon, 1984). have established patterns of more or less routine interactions
Empirical evidence suggests that the frequency of occurrence with each other (Knoke, 1993).
of the window types varies by level of institutionalization – Membership in knowledge-based policy communities
with the most institutionalized types occurring much more extends to actors such as state policy makers (administrative,
frequently than the least institutionalized. This lends the political, and judicial), members of nongovernmental organi-
model a degree of predictability in terms of the occurrence zations concerned with the subject, members of the media who
and type of policy window. report on the subject, academics who follow or research the
Almost 40 years ago, the American political scientists Cobb, area, and members of the general public who, for whatever
Ross, and Ross developed a model of typical agenda-setting reason, have taken an interest in the subject (see Thatcher,
styles. In their analysis, they argued that three basic patterns of 2012, Issue Networks: Iron Triangles, Subgovernments, Policy
agenda setting could be discerned, distinguished by the origins Communities, Policy Networks, Sabatier, 1987). In many issue
of a policy issue as well as the resources utilized to facilitate its areas, the policy community also involves members of other
inclusion on the policy agenda. In the outside initiation organizations such as businesses, labor unions, or various
pattern, “issues arise in nongovernmental groups and are then formalized interest groups and even international actors such
expanded sufficiently to reach, first, the public [systemic] as multinational corporations (Haas, 1992). A subset of these
agenda and, finally, the formal [institutional] agenda” (Cobb actors who interact within more formalized institutions and
et al., 1976). The mobilization case is quite different and procedures of government are defined as members of policy
describes ‘decision-makers trying to expand an issue from networks (see Coleman, 2012, Policy Networks). These policy
a formal to a public agenda.’ In this model, issues are simply networks include representatives from the community, but
placed on the formal agenda by the government with no are ‘inner circles’ of actors, who effectively hold the power to
necessary preliminary expansion from a publicly recognized veto or disparage many policy options as untenable or
grievance. In the third type of agenda setting, inside infeasible.
initiation, influential groups with special access to decision In this view, the likely results of policy formulation are
makers initiate a policy and do not necessarily want it to be contingent upon the nature and configuration of the policy
expanded and contested in public. community and network in the specific sector concerned. A key
variable that many observers have argued affects the structure
and behavior of policy networks is their number of members,
Policy Formulation
which affects aspects of networks such as their level of inte-
Studies of policy formulation have emphasized the importance gration and the types of interactions members undertake with
of actors interacting to develop and refine policy options for each other (van Waarden, 1992). What is important for policy
government (Freeman, 1955; Linder and Peters, 1990). But communities, on the other hand, is not the number of
unlike agenda setting, where many members of the public participants in the community so much as the number of
may be actively involved, in policy formulation the relevant relatively distinct ‘idea sets’ that exist within it. This affects
policy actors are typically restricted to those who not only the nature of conflict and consensus that exist in the
have an opinion on a subject, but also have some minimal community and, as a result, impacts the behavior of
level of knowledge that allows them to comment, at least community actors.
hypothetically, on the nature and feasibility of options put In open subsystems where networks have many members
forward to resolve a policy problem. and communities share many idea sets, it can be expected
290 Policy Cycle

that a propensity for new, radical alternatives to the status The well-known incremental model of policy making
quo may be generated in the policy formulation process. In developed by Yale University political scientist Charles Lind-
closed subsystems, where networks have few members and blom incorporated these insights into the best known alterna-
communities are dominated by a single idea set, on the tive to the classic rational model (Dahl and Lindblom, 1953;
other hand, a status quo orientation is likely to emerge in Lindblom, 1959). Lindblom summarized the elements of his
the policy options developed and put before decision model as consisting of ‘mutually supporting set of simplifying
makers. and focusing stratagems.’ These resulted, as he put it in his
Although the networking framework has been criticized as oft-cited 1959 article on “The Science of ‘Muddling Through,’
being primarily a descriptive metaphor, due to the fact that “the in decision makers working through a process of ‘continually
independent variables, (of policy-making) are not network building out from the current situation, step-by-step and by
characteristics per se but rather characteristics of components small degrees’ or ‘increments’” in which policies were
within the networks” (Dowding, 1995, p. 137), it provides invariably developed through a process of ‘successive limited
a general mechanism to organize the complex reality of comparisons’ with earlier ones with which decision makers
multiple actors, institutions, and ideas found in policy were already familiar (Lindblom, 1959).
formulation and generates a relatively parsimonious model Taking these insights into account, Braybrooke and
of formulation styles (Howlett et al., 2009). Lindblom (1963) argued that different styles of decision
making could be discerned depending upon the amount of
knowledge at the disposal of decision makers, and the amount
Decision Making
of change the decision involved from earlier decisions. Other
Many early studies of policy making in companies, govern- authors, like Graham Allison, also developed similar models
ments, and organizations, conducted largely by the students of of distinct decision-making styles (Allison, 1971), but did not
public and business administrations, argued decision makers specify in any detail the variables that led to the adoption of
attempt to follow a systematic method for arriving at logical, a particular style.
efficient decisions. They argued that policy makers achieved Attempting to improve upon these models, John Forester
superior results when they first established a goal, explored (1984) suggested that decision making was affected by the
alternative strategies for achieving it, attempted to predict its number of agents involved in a decision, their organizational
consequences and the likelihood of each occurring, and then setting, how well a problem is defined, the information
chose the option that maximized potential benefits at least cost available on the problem, its causes and consequences, and
or risk (Cahill and Overman, 1990). the amount of time available to decision makers to consider
This model was ‘rational’ in the sense that it prescribed possible contingencies and their present and anticipated
a standard set of procedures for policy making, which were consequences. In this model, decision makers situated in
expected to lead in all circumstances to the choice of the most complex subsystems are expected to undertake adjustment
efficient means of achieving policy goals. Such pure ‘rational’ strategies while those dealing with simple configurations of
models of decision making often thought of policy makers as actors and ideas will be more prone to undertake search-type
neutral ‘technicians’ or ‘managers,’ who after identifying ‘rational’ strategies. The nature of the decision criteria, on the
a problem would find the most effective or efficient way of other hand, varies with the severity of the informational, time
solving it. Many of the latest efforts to enhance the efficiency and other resource constraints under which decision makers
and effectiveness of public policy decision making, such as the operate. Hence, decision makers faced with high constraints
‘evidence-based policy movement’ (see Policy Analysis: will tend to favor satisficing over optimization, itself an
Evidence-Based Policy Making), focus on the application of outcome more likely to occur in situations of low constraint.
a systemic evaluative rationality to policy problems in classic
rational style.
Policy Implementation Styles
Empirical research into decision-making processes, however,
has discovered that political processes of bargaining and nego- Generally speaking, comparative studies have shown that
tiation often outweigh ‘rational’ deliberations and calculations governments tend to develop specific implementation styles
of costs and benefits in public policy decision making. Policy as they go about the course of their business (Knill, 1998;
makers were often found to be neither necessarily neutral nor Hawkins and Thomas, 1989; Kagan, 1991). These styles
competent and other models of the public policy decision- combine various kinds of policy instruments – such as public
making processes have argued this is not an accidental situation enterprises, regulation, or public consultation – into a more or
but rather an inherent and unavoidable characteristic of the less coherent whole, which is consistently applied in specific
policy-making exercise. Therefore, critics of the rational model, sectors of government activity – such as health policy or
such as Herbert Simon, noted that there are definite cognitive industrial policy. These policy tools provide the substance or
limits to the decision makers’ ability – ‘bounded rationality’ – to content needed to implement what was planned in the
consider all possible options, which forces them to selectively formulation stage and decided upon afterward (Salamon, 1989).
consider alternatives. ‘Satisficing’ behavior, in which decision These tools fall into two types. Substantive instruments such
makers did not consider all options in an effort to maximize as regulation or public enterprises are those directly providing
policy impact but rather simply agreed on a path that they goods and services to members of the public or governments.
found ‘satisfactory,’ was argued to better reflect the reality of They include a variety of tools or instruments relying on
decision making in a public policy context (Simon, 1955; different types of governing resources – such as money,
Smith and May, 1980). information, authority, or organizational resources (Hood,
 Policy Cycle 291

1986) – for their effectiveness. Procedural instruments, on the ‘trajectory’ following some ‘critical juncture’ can hinder policy
other hand, such as public hearings or commissions of change and learning (Pierson, 2000). Organizational–
inquiry, are different from substantive ones in that their institutional designs which limit ‘absorptive capacity’ are also
impact on policy outcomes is less direct. Rather than affect often seen as barriers to learning from policy evaluations.
the delivery of goods and services, their principal intent is to A significant variable in this regard is the capacity of
modify or alter the nature of policy processes themselves. an organization to absorb new information (Cohen and
Studies such as the work of Gunningham et al.’s (1998) ‘smart Levinthal, 1990). Only when state absorptive capacity is high
regulation’ led to the development of efforts to identify would one expect any kind of learning to occur. If a relatively
complementarities and conflicts within instrument mixes or closed network dominates a policy subsystem, then learning is
tool ‘portfolios’ involved in more complex and sophisticated likely to be restricted to some form of ‘lesson drawing,’ in
policy designs. Each tool mix decision combines advantages which policy makers draw lessons from past uses of specific
and disadvantages of each tool in its relationship to other tools policy instruments (Rose, 1991). Open subsystems allow for
as well as the effect it has on costs and benefits for government. ‘social learning’ when administrative capacity is high and
Why a particular combination of procedural and substantive more informal evaluation takes place among all those affected
instruments is utilized in particular policy issue areas is a key by a policy (Howlett et al., 2009; Bennett and Howlett, 1992).
question affecting implementation and implementation studies.
The preferences of state decision makers and the nature of the
constraints within which they operate are among the factors Conclusion: Policy Development as Policy Style
affecting policy implementation and instrument choices that lead
to the development of specific preferences or styles (Freeman, As Lasswell noted in the 1950s, envisioning policy develop-
1985). A state must have a high level of administrative capacity, ment as a staged, sequential, and iterative process is a useful
for example, in order to utilize authority, treasure, and analytical and methodological device. Methodologically, such
organization-based instruments in situations in which they wish an approach reduces the complexity of public policy making by
to affect significant numbers of policy targets. When it has few of breaking down that complexity into a small number of stages
these resources, it will tend to utilize instruments like incentives and substages, each of which can be investigated alone or in
or propaganda or rely on existing voluntary, community, or terms of its relationship to any or all the other stages of the
family-based instruments (Howlett et al., 2009). Similarly, a key cycle. The policy cycle idea also helps to answer many key
feature of procedural instrument choice is a government’s questions about public policy making regarding the effective-
capacity to manipulate policy subsystems (Mayntz, 1975). ness of different tools and the identification of bottlenecks in
policy processes. The stages model allows for the identification
of specific actors and actions in different phases of tackling
Policy Evaluation Styles
a problem, which makes it easier to measure the impact of key
The last stage of the policy cycle is policy evaluation. For many variables on policy outcomes and to identify existing policy
early observers, policy evaluation was expected to consist of process styles.
assessing if a public policy was achieving its stated objectives
and, if not, what could be done to eliminate impediments
to their attainment. Thus David Nachmias (1979) defined Bibliography
policy evaluation as “the objective, systematic, empirical
examination of the effects ongoing policies and public Allison, G., 1971. Essence of Decision: Explaining the Cuban Missile Crisis. Little
Brown, Boston.
programs have on their targets in terms of the goals they are Baumgartner, Frank R., Jones, Bryan D., 1991. Agenda dynamics and policy
meant to achieve.” However, while analysts often resorted to subsystems. Journal of Politics 53 (4), 1044–1074.
concepts such as ‘success’ or ‘failure’ to conclude their Braybrooke, David, Lindblom, Charles E., 1963. A Strategy of Decision: Policy Eval-
evaluation, Ingram and Mann (1980) caution that such uation as a Social Process. Free Press of Glencoe, New York.
Bennett, Colin J., Howlett, Michael, 1992. The lessons of learning: reconciling theories
concepts are slippery and often highly subjective. That is,
of policy learning and policy change. Policy Sciences 25 (3), 275–294.
public policy goals are often not stated clearly enough to find Cahill, Anthony G., Overman, E. Sam, 1990. The evolution of rationality in policy
out if and to what extent they are being achieved, nor are they analysis. In: Nagel, S.S. (Ed.), Policy Theory and Policy Evaluation: Concepts,
shared by all key policy actors. Knowledge, Causes, and Norms. Greenwood Press, New York.
What is significant in the evaluative process is thus not so Carstensen, Martin B., 2011. Ideas are not as stable as political scientists want
them to be: a theory of incremental ideational change. Political Studies 59 (3),
much ultimate success and failure, but that policy actors and 596–615.
the organizations and institutions they represent can learn from Cobb, Roger W., Elder, Charles D., 1972. Participation in American Politics: The
the formal and informal evaluation of policies in which they Dynamics of Agenda-Building. Allyn and Bacon, Boston.
are engaged. This can lead them to modify their positions in the Cobb, Roger W., Ross, J.K., Ross, M.H., 1976. Agenda building as a comparative
political process. American Political Science Review 70 (1), 126–138.
direction of greater substantive or procedural policy change, or
Cohen, Wesley M., Levinthal, Daniel A., 1990. Absorptive capacity: a new perspective
it can lead them to resist any alteration to the status quo. on learning and innovation. Administrative Science Quarterly 35, 128–152.
Policy evaluations do not necessarily result in major policy Dahl, Robert A., Lindblom, Charles E., 1953. Politics, Economics and Welfare: Plan-
change. That is, while the concept of evaluation suggests an ning and Politico-Economic Systems Resolved into Basic Social Processes. Harper
implicit ‘feedback loop’ is an inherent part of the policy cycle, and Row, New York.
Dowding, K., 1995. Model or metaphor? A critical review of the policy network
in many cases this loop may not be operationalized or may approach. Political Studies 43, 136–158.
result in only incremental changes to the status quo (Pierson, Dunlop, Claire A., 2009. Policy transfer as learning: capturing variation in what decision-
1993). Path dependence, in which policies are set on makers learn from epistemic communities. Policy Studies 30 (3), 289–311.
292 Policy Cycle

Forester, John, 1984. Bounded rationality and the politics of muddling through. Public Linder, Stephen H., Peters, B. Guy, 1990. Research perspectives on the design of
Administration Review 44 (1), 23–31. public policy: implementation, formulation, and design. In: Palumbo, Dennis J.,
Freeman, J. Leiper, 1955/1965. The Political Process: Executive – Bureaucratic – Calista, D.J. (Eds.), Implementation and the Policy Process: Opening up the Black
Legislative Relations, rev. ed. Random House, New York. Box. Greenwood Press, New York.
Freeman, Gary P., 1985. National styles and policy sectors: explaining structured Mayntz, Renate, 1975. Legitimacy and the directive capacity of the political
variation. Journal of Public Policy 5 (4), 467–496. system. In: Lindberg, Leon N., Alford, Robert, Crouch, Colin, Offe, Claus (Eds.),
Gunningham, N., Grabosky, P., Sinclair, D., 1998. Smart Regulation: Designing Stress and Contradiction in Modern Capitalism. Lexington Books, Lexington,
Environmental Policy. Clarendon Press, Oxford. MA, pp. 261–274.
Haas, Peter M., 1992. Introduction: epistemic communities and international policy Nachmias, David, 1979. Public Policy Evaluation: Approaches and Methods.
coordination. International Organization 46 (1), 1–36. St. Martin’s Press, New York.
Hawkins, Keith, Thomas, John M. (Eds.), 1989. Making Regulatory Policy. University of Pierson, Paul, 1993. When effect becomes cause: policy feedback and political
Pittsburgh Press, Pittsburgh. change. World Politics 45, 595–628.
Hood, Christopher, 1986. The Tools of Government. Chatham House, Chatham, NJ. Pierson, Paul, 2000. Increasing returns, path dependence, and the study of politics.
Howlett, Michael, Ramesh, M., Perl, Anthony, 2009. Studying Public Policy: Policy American Political Science Review 94 (2), 251–267.
Cycles and Policy Subsystems. Oxford University Press, Toronto. Rose, Richard, 1991. What is lesson-drawing? Journal of Public Policy 11 (1),
Ingram, Helen M., Mann, Dean E., 1980. Why Policies Succeed or Fail. Sage, Beverly 3–30.
Hills, California. Sabatier, Paul A., 1987. Knowledge, policy-oriented learning, and policy change.
Kagan, Robert A., 1991. Adversarial legalism and American government. Journal of Knowledge: Creation, Diffusion, Utilization 8 (4), 649–692.
Policy Analysis and Management 10 (3), 369–406. Sabatier, P.A., 1991. Toward better theories of the policy process. PS: Political
Kingdon, John, 1984. Agendas, Alternatives and Public Policies. Little, Brown, Boston. Science and Politics 24 (2), 144–156.
Knill, Christoph, 1998. European policies: the impact of national administrative tradi- Salamon, Lester M. (Ed.), 1989. Beyond Privatization: The Tools of Government Action.
tions. Journal of Public Policy 18 (1), 1–28. Urban Institute, Washington.
Knoke, David, 1993. Networks as political glue: explaining public policy-making. In: Simon, Herbert A., 1955. A behavioral model of rational choice. Quarterly Journal of
Wilson, W.J. (Ed.), Sociology and Public Agenda. Sage, London, pp. 164–184. Economics 69 (1), 99–118.
Lasswell, H.D., 1956. The Decision Process: Seven Categories of Functional Analysis. Smith, Gilbert, May, David, 1980. The artificial debate between rationalist and
University of Maryland, College Park. incrementalist models of decision-making. Policy and Politics 8 (2), 147–161.
Lasswell, H.D., 1971. A Pre-view of Policy Sciences. American Elsevier, New York. van Waarden, F., 1992. Dimensions and types of policy networks. European Journal of
Lindblom, C.E., 1959. The science of muddling through. Public Administration Review Political Research 21 (1/2), 29–52.
19 (2), 79–88.