A New Conceptual Framework and Approach to Decision Making in Public Policy

Abstract

In this conceptual paper, we bring forward a new theoretical proposition in the form of an Integrated Policy-Making Process Framework. This tool arose from the discussion and hypothesis that the main models used to analyse public policies have practical and theoretical limitations and/or are due for an update, especially when it comes to the study of phenomena with many actors involved, vast legislation, large timeframes, and high degrees of complexity. Our original model encompasses the three fundamental stages of the public policy cycle (agenda; decision making; implementation). Our approach can have a wide spectrum of applications and contribute to the field of knowledge of political sciences. Our proposal of using three frameworks in an integrated way enables researchers and users to gain a holistic vision concerning policy processes, and it offers the possibility to compare and rank categories. The Integrated Policy-Making Process Framework is thus proposed as a new tool to tackle research and studies on decision making in public policies and the policy-making cycle.

1. Introduction

This paper arose from the observation that the main models used to analyse public policies (i) have limitations and/or (ii) are due for an update, in order to encompass contemporary problems and interdisciplinary approaches. Therefore, based on the presuppositions that we will explain, and the merits of each independent framework we selected, we devised a framework of analysis to understand the phenomena related to public policies. This is our aim. As we will explain, this framework can be carried over to research in the fields of political studies, policy studies, public administration, and management.

According to Cejudo and Michel [1], solving complex problems is one of the crucial challenges for public policy and administration today. That is why a policy frame is required to better understand the problems that policies will address. Indeed, this framework can theoretically be applied (or at least adapted) to a myriad of contexts, and especially those with many actors involved, vast legislation, large timeframes, and high degrees of complexity. Furthermore, as we will see, the framework we will present at the end is particularly fitting to address situations that go through Lasswell’s Public Policy Cycle [2].

A framework is a synthetic abstract scheme that organises the elements of reality—and every framework is a compromise between the attempt to simplify and the need to reflect reality’s complexity. Thus, it provides a formal representation of ideas and knowledge concerning phenomena [3]. Frameworks are also empirically oriented. However, abstract conceptualizations are paramount to developing complete framings of analytical processes [4]. Moreover, models are mainly descriptive and are not predictive tools [5]. Hence, the pertinence of our contribution.

The new framework we propose is an articulation in three sequential stages of three seminal frameworks (Kingdon’s Multiple Streams Framework, Lindblom’s Muddling Through Framework, and Sabatier and Mazmanian’s Implementation Analysis Framework). After a thorough enquiry into different and well-established models, with the intent of analysing decision making with a high degree of complexity, and the weighting of their pros and cons, it became clear that a new synthetic model could be developed. Our proposal follows the three fundamental stages of the public policy cycle (agenda; decision making; implementation). During our research, we realised how these three previously existing frameworks had their strengths allocated to one of these stages, and we proceeded to adapt them into a new synthetic model. What follows is the explanation of the rationale that led to this conclusion, and the blueprint for the new model.

In order to integrate the models of reference, we had to carry out a structural and conceptual architecture exercise that required a full understanding of these models, as well as the needs that public policies with the characteristics described above require. We would like to pose this Integrated Policy-Making Process Framework as a new method/tool/hypothesis to tackle research and studies on decision making in public policies and the policy-making cycle. This model will also allow researchers (especially in social sciences) and users to approach the three theoretical models of reference in a fresh way, and to use them in various contexts. Naturally, the integrated model proposed may not be applicable to every possible situation in the areas of administration and public policies. However, from our explanation, its purpose, scope, and potential may become clearer.

In this paper, we will expose the blueprint for the creation of the integrated framework, presented at the end. Comprising three stages, our integrated framework allocates each framework of reference for the stage at which its strengths are more evident—this is the guiding principle and fundamental logical premise of our proposition. The use of more than one framework in an integrated way enables researchers to gain not only a more holistic vision concerning public policy processes, but also to quantify and homogenise the analysis variables. In the next pages, we lay out the theoretical basis of our framework, and we present its development and final form.

Background

Ever since Harold Lasswell’s [2] seminal contributions to the field of political science, three premises about public policy studies have been taken for granted: (1) interdisciplinarity; (2) the solution-oriented approach; (3) normativity. These presuppositions concerning the framing of public policies still guide the research in this area, although with a much wider scope than originally envisioned [4]. Indeed, public policy is a field in which many approaches come together. Understandably, the results, analyses, and methodologies from public policy studies have been used by academics and politicians alike in a vast array of sectors of public administration, decision making, policy making, and everyday life [6].

In order to visually represent and create meaning out of the different sequential and interdependent stages of public policies, Lasswell proposed a public policy cycle in 1957. For Howlett et al. [4], the public policy model is founded upon the premise that there is a dynamic interaction between the solutions to problems (even if not explicitly) and the corresponding stages of the cycle. The public policy cycle is thus illustrated in Figure 1.

This model is essential to our work because it has to tackle each stage of the framework. As Harguindéguy [7] has argued, the setting up of an agenda determines which problems are recognised and will be addressed by way of public policy. This stage is crucial and will determine how the process will subsequently unfold. Not all societal problems can be addressed at the same time. Yet, according to Sjöblom [8], the acknowledgement of a public problem involves: (i) the perception of the problem (i.e., a situation becomes publicly unsatisfactory from the moment that it affects the perception of many social players); (ii) the definition or delimitation of the problem.

The second stage of the process, as per Howlett et al. [4], is the formulation of policy. It is here that we can see the process of creating options concerning a public problem. At this point, the options are outlined and laid out for refinement and discussion. This moment happens in state institutions and involves specific groups and political agents with more or less interest in the formulation of the policies for any determined problem and its outcome. These groups may participate in the formulation through the collection of information and data analyses, or by intervening directly with decisionmakers. This stage usually receives more coverage, and especially by the media, because it generates debate and puts analysts, political agents, and decisionmakers to the test. According to Heidemann and Salm [9], policy formulation has several general and broad characteristics that Howlett et al. [4] systematised according to the work of Jones [10], such as: (1) the formulation does not need to include only a small group of actors, as there may be two or more groups producing competing (or complementary) proposals; (2) the formulation can advance without a clear definition of the problem; (3) there is not necessarily an association between policy formulation and certain institutions, although policy formulation is often carried out by bureaucratic offices; (4) formulation and reformulation can happen over a long stretch of time, and there is not the certainty that one day there will be a prevailing majority in favour of a certain proposal; (5) it is frequent to find many instances of appeal available to those who lose in the formulation process; (6) the outcome of the process is never impartial, which means that there are winners and losers. A series of recent studies have indicated that the way that an issue is framed changes as policies spread within a diffusion network and their practical implications are better understood [11,12].

The third stage (decision making) is the political process whereby a course of action is officially approved [9] through the selection of a solution among the pool of proposed alternatives. Zahariadis [13] perceived the transition from agenda setting to decision making to be smooth, but Herweg et al. [14] assumed the need for a “decision window” and policy entrepreneurship to move the process from one stage to the next. Decisions made at this stage will have a direct impact on the type of policy that will be made [15]. This is why it is a moment of dialogue between intentions and actions. Unlike the previous stages, the number of people involved in decision making narrows to include only those with power (soft and/or hard) and formal authority to make binding public decisions [4].

The implementation represents the execution of political measures, and so it is the stage at which rules, routines, and social processes are put into action [16]. While initially, implementation was seen as an integral part of policy making, in the 1970s, it was considered to have its own dynamic within the policy cycle [17]. In the second generation of studies about implementation, the focus changed. This generation divided the implementation process into two main strands: top-down and bottom-up. The third generation of studies on implementation evolved to multilayered analyses and perspectives that consider not only political actors, but also their underlying interests. According to this point of view, implementation is regarded as an ongoing negotiation process between central authorities and implementers. Thus, political networks become key to understanding the implementation process, given that the type of policy to be implemented becomes critical [17]. The inherent ambiguity when adopting a policy demands that the people in charge of implementation make sense out of the policies when applying them to real-world conditions. This opens the door for administrative discretion in determining the appropriate implementation behaviour [18,19].

Finally, policy evaluation, which has been widely discussed, is the stage of the policy cycle at which the implementation process and policy design are examined to ascertain how the policy is producing an effect. Provided that the policy has effectively tackled the underlying cause, the evaluation process can also be a strategic feedback tool. By the same token, it can highlight problems and difficulties. Now that we have a clearer view of the public policy cycle, we will dissect how frameworks play a role in understanding how these different stages happen and advance.

Over the years, political science has developed a series of models to systematise and understand the political processes that generate public policies [6]. According to Dye [20], such models intend to: (i) simplify and clarify ideas about politics and public policy; (ii) identify relevant aspects of sociopolitical questions; (iii) help communication by giving emphasis to the essential characteristics of public life; (iv) direct efforts towards a better understanding of public policies, suggesting what does matter and what does not; (v) propose explanations for public policies and predict their consequences. In the political science literature, one can find many tried-and-tested models that reflect the way their authors reasoned about the ways they saw politics being formulated, decided, and implemented [21].

After a thorough reading of the different theoretical contributions from the main authors of public policies’ models of analyses, we identified a consensus about the most well-established frameworks. These models have served to help frame countless works in many fields of knowledge, and they are still appraised for their characteristics and versatility [22,23]. They are: (1) the already-mentioned policy cycle model; (2) Kingdon’s Multiple Streams Framework; (3) the Punctuated Equilibrium Theory; (4) the Advocacy Coalition Framework; (5) the Muddling Through Framework; (6) the Implementation Analysis Framework. These models are all logically coherent, empirically verifiable, clear, broad, and easily replicated in a plethora of contexts. They present syntheses in two significant ways: they are points of view on very complex nets of problems and questions, and they promote the integration of concepts and the analytical proposals of different authors, thereby exploring the complementarity of different approaches.

When considering the conceptual context of the public policy cycle, the introduction of a topic in the policy-making process all the way through to its implementation, as we will explain in detail, three models proved more advantageous: the Multiple Streams Framework, the Muddling Through Framework, and the Implementation Analysis Framework [6]. Indeed, none of these models of reference, just by themselves, were fitting enough to assess in detail all the stages of the public policy cycle concerning the abovementioned topic and thus provide us with a clear insight and better understanding of the situation. This realisation led us to create a new up-to-date synthetic framework that could address the complex structures of decision making.

As mentioned, these three models are characterised by their versatility, which means that their application is broad. This, however, does not mean that they overlap. Following the example of Gestel et al. [24], rather than treating these frameworks as competing, we will bring to the fore their complementary and interdependent contributions to explaining the policy processes that underpin reforms in the public sector.

Various authors have been following this line of thought and the practice of integrating frameworks, successfully using a mix of theoretical approaches in their works. For instance, Vaughn and Cortner [25] used a combination of the Stage Framework with the premisses of Kingdon’s Multiple Streams Framework “primeval soup” to analyse changes in the forest policy during the George W. Bush administration. Birkland [26] developed a model in which he fused the Multiple Streams Framework with the Advocacy Coalition Framework (ACF), developed by Paul Sabatier and Jenkins-Smith in the 1980s, to study oil spills and accidents in nuclear power plants. In our research, the integrated model we devised proved to be a good methodological strategy to attain a holistic perspective of the phenomena.

After carefully examining the strengths and weaknesses of each model of reference, it became clear that they were a very good fit to a sequential model comprising three stages. Effectively, both the content and form of the framework intertwine seamlessly, without the need to change the chosen parts of the previously existing frameworks to fit the model in which they were included. Before we present our synthetic model, we will now delve into some specificities of each model of reference.

Based on Coen, March, and Olsen’s Garbage Can model [27], John Kingdon formulated his Multiple Streams Framework in 1984 [28] in an attempt to explain how social problems become political, and conversely, how (and why) certain topics leave the political agenda. This framework analyses the fluxes through which a decision goes through until it becomes enshrined in legislation. According to Capella [29], Kingdon sees public policies as a body formed by four processes: the establishment of a public policy agenda; the consideration of alternatives capable of being transformed into public policy; the prevailing choice among the pool of alternatives; the implementation of the decision. This means that Kingdon primarily focuses on the two first stages of the public policy cycle (i.e., the predecision stages of policy formulation).

For Kingdon, changes in the political agendas of governments (characterised as “organised anarchies”) depend on the result of the convergence of three streams of decision: the policy stream, solutions stream, and problems stream [28]. In this way, Kingdon’s model analyses how problems are identified, and how they are then included in the governmental agenda. Kingdon concludes that not all problems can be addressed, and that there may be situations in which a problem becomes pressing and quickly escalates to become a top priority. A good example of this is the COVID-19 pandemic and the need to not only find long-lasting solutions, but also to contain the contagion among the population through legally muscled measures.

The second stream is a set of possible solutions generated inside political communities [28]. The primeval soup metaphor, put forward by the author, proposes that a great number of ideas flow inside political communities. Some will remain intact, while some others will be discarded, but the majority will change through debate [29]. This leads to a process of the narrowing of possibilities, which culminates in a restricted list of alternatives that includes the most acclaimed, technically validated, and economically viable [29]. The third stream consists of the sociopolitical dimension of policy making. It oscillates according to the specific rules within a society, and it is alien to both the acknowledged problems and solutions [28].

Under certain circumstances, these three streams—problems, solutions, and political dynamics—converge, and a politically favourable moment happens (i.e., a window of opportunity to introduce solutions and advance a certain agenda opens). However, when one of the three variables destabilises, the opportunity to change the agenda ceases. The author highlights certain moments when an opening is predictable (such as periodical budget cycles, changes in governance, and unexpected events) [28].

Along with this view of the policy cycle and the idea of a political window of opportunity, it is also relevant for us to retain the way Kingdon conceptualises the policy entrepreneur in his seminal work Agendas, Alternatives and Public Policies [28], in which the author posits the Multiples Streams Framework. Resorting to a conceptual iteration of the classical definition of entrepreneurship, in his book, Kingdon divides these entrepreneurs into two main groups: the first consists of people in visible institutions and high-profile positions; the second includes people that belong to less visible groups, such as lobbying groups, bureaucrats, researchers, academics, political parties, and media, among other groups or communities in which ideas are generated and widespread, usually ending up as alternatives proposed by the people belonging to the first group. Kingdon’s Multiple Streams Framework points out that what matters most to us can be summed up in Figure 2, which stems from Zahariadis’ vision of the model developed by Kingdon:

While Kingdon’s framework is extremely useful to understand in depth the first two stages of the policy cycle (i.e., setting up the agenda and policy formulation), we found that a model with a similar impact and strength to cover the remaining stages would be advantageous. That is why we turned to Lindblom’s Muddling Through Framework (for the decision-making stage) and Sabatier and Mazmanian’s Implementation Analysis Framework (for the implementation stage).

The Muddling Through Framework, developed by Lindblom in 1959 [31] and deepened in 1979 [32], is a model that focuses on the study of decision making applied to public policies, and it is therefore especially relevant for the study of the decision-making stage. It is based on the premise that public administrators, in general, are faced with complex problems that hinder them from identifying all the possible options and their consequences. Based on this premise, this model tries to find ways to simplify the reality that is inherent to decision making by establishing that political actors decide through a restricted series of comparisons between marginally different alternatives [31,32]. Lindblom also concludes that, from a certain point onwards, decisions represent more what is politically more viable than what might be technically more desirable [4]. We summarise the process proposed by the Muddling Through Framework in Figure 3.

According to Lindblom, decision making is usually formalised as a means-to-an-end relation: the means are to be evaluated and chosen in light of the ends, selected independently, and before the choice of the means [31]. Thus, this incremental method, according to Weick [33], boils down to an idea of continuous and incremental change through small wins. In short, the minor costs of both the failure and benefits of small steps lower the resistance and political competition for success [34]. This kind of insight into decision making is what gives the Muddling Through Framework its strength, and how it has helped us.

Moving on to the final stages of the public policy cycle, the Implementation Analysis Framework has presented itself as the best model to tackle the implementation-of-policies stage. Based on several studies carried out in the United States during the 1960s and 1970s on the unsatisfactory implementation of public policies and juridical decisions, Sabatier and Mazmanian, in 1980, developed a synthetic model based on previously existing theories and proposals entitled the Implementation Analysis Framework [35]. According to the authors, implementation is the execution of a basic political decision, usually enshrined in the body of laws. The implementation process is normally a process with various stages, such as the approval of legislation, directives of the institutions that are to implement the measure, impacts of the policy, and finally, revision of the initial bill.

Sabatier and Mazmanian’s proposal had several premises attached to three main factors: the treatability of a problem, the capacity of the law to structure implementation, and nonstatutory variables that determine implementation, which would then break down into a few categories and variables. Acknowledging that the use of this framework for analytical/interpretation purposes by academics and politicians could be too complex, Sabatier and Mazmanian [35] decided to organise the statutory variables into a checklist of six conditions that a policy or decision that would interfere with the status quo had to meet to be effectively implemented [35]. Figure 4 illustrates that checklist.

From the analysis of the different stages of the public policy cycle, and based on a set of well-established models, we came to see how important it was to critically adopt these theoretical hooks to understand the mechanisms that govern the setting up of the political agenda, decision making, and implementation in complex and multidimensional processes that operate in different layers (i.e., supranationally, nationally, regionally, and locally).

2. Methods

For the purpose of this paper, we assume an abstract case that follows every stage of the public policy cycle. Virtually every policy devised, or reform undertaken, by governments, institutions, or corporations adheres to this premise [2,4]. There is thus the advantage —academic and practical—of analysing it through the lenses of the models of reference we have presented. The first step is to deconstruct the public policy process in the following stages: (i) “Agenda”—which is a predecision stage; (ii) “Decision”—revolving around the net of problems in the decision-making stage; (iii) “Implementation”—which is a postdecision stage. This way, we had to resort to a specific theoretical reference to study the underlying issues of these stages. Thus, our synthetic conceptual framework aimed to consider the theoretical aspects linked to the construction of explanatory models for a certain reality, as well as the practical aspects that follow from the compliance of any case study with the premises exposed (i.e., many actors involved, vast legislation, a large timeframe, and a high degree of complexity). What follows next is an explanation of our course of action to achieve and create the conceptual framework.

Given that these three stages occur in the order Agenda–Decision–Implementation, the adoption of a sequential model in which these stages remain interconnected is not only fitting, but also a logical step to take. Despite the criticisms around a sequential model [5], as we will see, this model is not an oversimplification of reality, but rather an interpretation of it. The goal of this systematization is to acknowledge and represent the reality that is inherent to public policies, and to find clusters of common elements and identify patterns. Indeed, it is necessary to understand policies as a continuous stream of decisions and proceedings, of which we must create meaning [36].

A sequential model offers various advantages. Among them is the clarity of analysis it provides, as well as the fact that it allows for a more concrete and detailed analysis of each step of the decision process, which, in other models that are not (strictly) sequential, can become too diffuse and complicated to scrutinise.

Having established the legitimacy and adequacy of adopting a sequential model to understand policies in all their complexity and interdependence, the question of how many stages it should comprise arose. In our framework, the evaluation stage was removed (because it may trigger another process of its own, coming from inside the policymaking cycle itself, that does not allow us to have a clear vision of the agenda stage), and we have sectioned our framework into three (instead of five) clearly distinct parts:

• Agenda—comprising the two first steps of the public policy cycle, as described in Figure 1 (“Setting the Agenda” and “Policy Formulation”), corresponding to a predecisional stage;
• Decision—which, as the name suggests, corresponds to the third step of the public policy cycle, “Decision-Making”;
• Implementation—the fourth step of the public policy cycle, “Policy Implementation”.

Having established a framework with three stages, we decided on the theoretical substance to include in each stage.

2.1. Selection Criteria

With the main goal being the achievement of a holistic knowledge of the cycle, our choice regarding which model of reference should be attached to each one of the three abovementioned stages was governed by two main criteria: on the one hand, the theoretical suitability of the model, and on the other hand, the possibility of restructuring the models in order to break them down and allow for a more refined analysis. Based on these criteria, the theoretical references chosen to analyse each stage of the public policy cycle are as follows:

• The Agenda Stage (A)—Multiple Streams Framework (based on Kingdon, 1984);
• The Decision Stage (B)—Muddling Through Framework (based on Lindblom, 1959 and 1979);
• The Implementation Stage (C)—Implementation Analysis Framework (based on Sabatier and Mazmanian, 1980).

2.2. Theoretical References Adopted for Each Stage

Kingdon’s Multiple Streams Framework was chosen for Stage A because it is widely regarded in the academic literature as the most fitting to understand the predecision stages. According to Zohlnhofer et al. [37], academics have rekindled their interest in Kingdon’s model in the last few years because it has proven relevant and well suited to analysing the political agendas and policy formulations in mature democracies in the 21st century. As we have seen, when conditions are met, and when policy entrepreneurs step into the process in moments when the window of opportunity opens, changes in the agenda may happen [28,30].

Regarding the decision-making stage, we could have extended the use of Kingdon’s framework to this stage as well, as Zahariadis [38] suggested. Zahariadis was perhaps Kingdon’s most prominent follower, the one who further developed the Multiple Streams Framework, and the person responsible for giving this framework a great part of the dimension that it has [30]. However, other authors argue that the Multiple Streams Framework is not suitable or useful beyond the agenda stage and advise against it. According to Herweg et al. [14], if we follow Zahariadis’s suggestion, then we will not be able to know why political change fails. Given that the pairing process is the decisive concept that elevates the MSF from the level of description to the level of explanation of the agenda and political changes, insufficiencies of the framework in the following stages still need to be addressed. Bearing this limitation of the MSF in mind, our choice to theoretically harness the decision stage fell upon Lindblom’s Muddling Through Framework.

As we mentioned previously, policy making and decision making advance in a way that is neither rationally controlled nor controllable, but rather by several attempts, agreements, and concessions that are built upon an ever-changing platform of previously made decisions [6]. Moreover, Lindblom’s incremental model focuses on the human and environmental characteristics of the people that have to decide in political contexts (i.e., risk minimisation, negotiation with potentially divergent parts, political compromise and commitment, and efficiency). This puts the Muddling Through Framework in an ideal position to be used to understand the decision stage of the policy cycle.

Given its pragmaticism and realist view of the political processes involved in public policy making [4], we found the Muddling Through Framework to be a better-suited model to equate to the decision-making stage. It should also be noted that this is one of the most used models in the public administration and organisational theory fields [39]. Because the Muddling Through Framework focuses on decision making, it does not lend itself to be used at the agenda stage or implementation stage.

The Implementation Analysis Framework proposed by Sabatier and Mazmanian is a theoretical hook fitting to grasp the implementation stage. The clear structure presented by this model, the proposed checklist with six items to ascertain the efficiency or failure of public policy, and the set of tasks that need to be carried out to understand how goals were achieved (or not), are substantial advantages of this model over others, even though the six items present some limitations, such as not clearly assuming the relevance of technical and technological conditions, for example (even if these and other cases can theoretically be included inside the economic conditions). This model has also been widely used by researchers in studies focused on the implementation of public policies [40,41]. Sabatier and Mazmanian’s framework also has the strength of relating various factors that affect the efficiency of policies and public programmes by proposing a separation between the formulation and implementation of public policies. The authors argue that the decisionmakers involved in the formulation stage may provide an important contribution to the effective implementation of policies by helping with their design, the legal framing, and the choosing of competent people.

The orientation provided by this model, in the ambit of implementation processes, is relevant to understanding and framing the many levels and layers of governance (national, regional, local, etc.) involved in the implementation of public policy, with each level having its own specific interests and ambitions [4]. Due to the abovementioned characteristics, the Implementation Analysis Framework is suitable to address all issues around implementation, but it was not designed to provide insights into the agenda or decision-making stages.

3. Results

The option for different theoretical references is particularly adequate and important when the subject of study is complex, which is often the case when it comes to public policy. Sabatier [30] argues that scientists should be capable of applying various types of theoretical hooks and not just one. In fact, being aware of various perspectives encourages the analyst to clarify the differences in the assumptions that permeate frameworks, instead of implicitly accepting the set of presuppositions that are presented. Moreover, multiple perspectives create the conditions for the development of competing hypotheses that should ideally lead to a robust inference. The knowledge and application of multiple perspectives should gradually clarify the conditions in which one can ascertain that one perspective is more useful than another. Finally, multiple perspectives encourage a comparative approach: instead of questioning if theory X produces statistically significant results, we should question if theory X answers more questions than theory Y.

Thus, the choice of resorting to three tried-and-tested theoretical models seems justifiable and useful in that it allows us to build a clear structure to understand the complex phenomena related to the public policy decision and implementation processes. Figure 5 presents our synthetic model.

The model we propose is the result of an integrated and coherent scheme that allows us to better explore, understand, and explain the forces and operating factors present in each stage of the public policy cycle [6]. In short, it is an Integrated Policy-Making Framework (IPMF). Thus, the approach we have described provides a structure that aims to identify: (i) the processes that must happen for a topic to be included in the political agenda; (ii) how the different institutions initiate and carry out decision making; (iii) how institutions outline the implementation of the chosen public policy. The combination of the three models we have chosen in a single integrated framework (the IPMF) proved its analytical potential in a context in which we had to deal with ambiguous processes entailing a high number of agents and complex interactions.

The result we were able to achieve in terms of both the theoretical understanding of the phenomena and subsequent data analysis provided a significant and valuable interconnectivity between the different stages of the public policy cycle. This leads us to believe that the framework we propose can help frame various parts of empirical research methods. The next section will explain the process of the conception and construction of the “dimensions” and “categories” that allow the operationalisation of this framework.

Constitutive Elements of the Adopted Conceptual Model

To theoretically reinforce the presented model, we created two sublevels of analysis inside each of the adopted theoretical references, which we called “dimensions” and “categories”. These sublevels constitute the organisational axes of the indicators and indexes, which, in their turn, allow us to identify the most relevant factors for the successful implementation of public policies. In line with Bardin’s thought [42], the dimensions are built based on the constitutive elements of each theoretical model of reference. These elements originate subsets of criteria/questions/mechanisms, which we use to understand the elements according to the key ideas provided by their respective authors. A thorough literature review enabled us to identify and organise the main components into what we here call categories. These are classes that gather a group of elements under an identifiable title indicative of the message the author wants to convey. In harmony with Capella [29], we considered the existence of five fundamental components in Kingdon’s model [28]. These five components were then broken down according to the basic mechanisms that are associated with each of the key components of the model. For instance, the component “problems” creates the need for action through five basic mechanisms: indicators; specific studies; events; crises and symbols; and feedback on political action. In line with the works of Schweizer [43], Kopecka et al. [44], Bürstrom [45], and Bendor [39], five categories were created that allow us to grasp the essential of Lindblom’s model [31,32]. Finally, the categories associated with Sabatier and Mazmanian’s model [35] flow from the six conditions that these authors have identified in their work.

The following tables (

Table 1. Dimensions and categories of the agenda stage.

Stage	Dimension	Category
Agenda	Streams of Problems	Indicators
		Specific studies
		Feedback on political action
		Events, crises, and symbols
	Policy stream (alternatives and solutions)	Community of specialists (think tank)
		Stream of ideas inside the community of specialists (primeval soup)
		Evolutive process of maturing of solutions
		Technical feasibility
		Compatibility with values
		Financial viability
		Political receptivity
		Acceptance by the community
		Restrictions in the system
	Policy stream	Supranational/national/regional feeling
		Governmental changes
		Stance of the involved institutions
		Organised political forces
	Political entrepreneurs	Visible/invisible actors
		Access to decisionmakers
		Resources spent
		Strategies used
	Window of opportunity	Opening of the window of opportunity

Source: authors’ own study.

,

Table 2. Dimensions and categories of the decision-making stage.

Stage	Dimension	Category
Decision	Goals	Connection to other problems
		Prevalence of ambiguous goals instead of clear ones
		Transversal solutions to other problems
	Alternatives	Few proposals, which are more or less similar
		Interconnection between the evaluation of the action to accomplish and goals to achieve
	Resources	Previous experiences
		Decision support information
		Little relevance of theory
		Financial situation
	Interaction between decisionmakers	Decisionmakers with different visions
		Previous inexistence of clear goals by decisionmakers
		Cultural idiosyncrasies
		Leadership
	Choice of solution	Trial-and-error process
		Increased preoccupation with problem mitigation and desirable results
		Consensus as a sign of good decision

Source: authors’ own study.

and

Table 3. Dimensions and categories of the implementation stage.

Stage	Dimension	Category
Implementation	Goals	Clear
		Consistent
	Adequate causal theory and influencing power	Involvement of public institutions/the state in the concretisation of the program
		Suitable delegation of powers (decision-making powers)
	Legally framed structure	Allows access to institutions that support the program
		Suitable hierarchical integration
		Possibility of assigning responsibility to specialised entities
		Enough financial resources
		Adequate decision-making rules
	Dedicated and competent people	Commitment to the achievement of goals
		Training in management and political science
	Support from lobbying groups and people in high positions	Formalisation of support
		Decision power
	Changes in social and economic conditions	Economic
		Social

Source: authors’ own study.

) detail the constitutive elements of each one of the adopted references (i.e., dimensions and categories used in the agenda, decision, and implementation stages) [6].

Indeed, the IPMF covers each of the three stages of the cycle of a given public policy. The ability to be broken down into constituent parts, as described above, is an advantage that allows for a better understanding of which dimensions and categories have been important for the success (or lack thereof) of their respective stages. Figure 6 shows what can conceptually be understood as success at each stage, and the necessary conditions for it to occur.

Considering the above, resorting to the list of predefined variables, based on the existing theory (that is, based on the three seminal models adopted in an integrated manner), and bearing in mind the theoretical rationale proposed for the definition of success at each stage, the researcher, politician, or decisionmaker is able to have access to a more detailed look at the particular aspects that influenced the implementation (or nonimplementation) of a given public policy.

4. Discussion

The present proposal of an integrated model for the analysis of public policies, the IPMF, contemplates three main stages, and it is based on three pre-existing theoretical references that explain the process of decisions associated with these three main stages. The IPMF overcomes the limitations of the models on which it is based by blending them seamlessly and thus achieving a way to tackle the full public policy cycle, whereas the models on which the IPMF is based are only suitable to properly address one of the three stages, as we have discussed above, and the IPMF enables researchers to gain knowledge about the whole cycle.

Having established the premise that the main contribution of the IPMF is the ability to overcome the limits of each model,

Table 4. Summary of the Advantages and Disadvantages of each Framework or Reference.

Models	Advantages	Disadvantages
“Multiple Streams Framework” (Kingdon, J., 1984)	- The MSF deals with various variables (such as the time, different groups of people involved, and subjective perceptions of events, situations, and solutions) and scenarios. It also asserts that policy making is not an inherently rational process, and thus it takes into account real-world situations (such as ambiguity and uncertainty) that influence the political agenda [23]; - The MSF gives users unparalleled flexibility: there is no need for a detailed codebook to test hypotheses or advance general policy theory. Researchers can read one book (or a couple of chapters) and generate a theoretically informed and publishable empirical case study [46]; - The MSF is considered adequate to explain the initial stages of public policy making, especially in ambiguous processes, where information is scarce and the decision is time-sensitive [28,30]; - The “barrier to entry” of the framework is low compared with other policy process approaches, such as the Institutional Analysis Framework and Development Framework [46]; - This is a very popular, well-known, and widely used framework. The high number of citations confirms this [37], as does its resurgence in papers on COVID-19 measures [23]; - Kingdon identifies what we call “universal” policy-making issues that can arise at any time or in any place [46]; - The MSF seeks to provide tools for understanding the policy-making process, rather than focusing on predicting future events [29].	- Framework focused on the predecision stage (i.e., in the setting of the agenda and policy formulation) [37]; - Focus on case studies, and thus, qualitative analysis. This leads to very little quantitative research use [38]; - Limitative in the analysis of why the policy change fails, either due to the implementation of an agenda that fails, or the failure to achieve a majority to pass the initial proposal during the decision process [14]; - Most applications do not fully use the MSF’s problem stream, policy stream, politics stream, policy entrepreneur, or policy window concepts, nor all their subcomponents [47].
“Muddling Through Framework” (Lindblom, C., 1959, 1979)	- This framework allows for an approach that is aligned with the decisionmaker’s daily life. By verifying the limitations inherent to the decision process, this framework seeks a method capable of reducing the complexity of the reality that permeates it [48], in which political actors decide through a restricted series of successive comparisons between alternatives that are marginally different from each other [31,32]; - It enables the identification of common elements in the decision strategies effectively followed by public policy decisionmakers [10]; - Faced with many participants (stakeholders) in decision making, this framework organises reality in a simple way, and it focuses on stratagems that allow for the identification of the fragmentation of the analytical work arising from the attention that decisionmakers pay to their part in the domain in which the problem under consideration is inserted [4]; - The decisions taken represent what is politically feasible, rather than what is technically desirable and what is possible [4]; - The framework integrates analytical categories and dimensions. This makes it possible to observe the elements of integration versus fragmentation that affect the processes and actions of decisionmakers [4,49]; - Given the knowledge limitations to which policymakers are subject, simplification through a narrow focus on small variations around an existing policy allows for the maximum advantage to be taken of the available knowledge [31,32]; - It is one of the most recognised frameworks, which makes it one of the most widely cited articles in the field of public administration and organizational theory [39].	- This framework is unlikely to assume the novelty of a policy, the number of decisionmakers involved, or the degree of consensus among them on the policy-making goals [39]; - This framework is better suited to relatively stable environments than it is to situations that are uncommon, such as a crisis or new political problem [50,51]; - The possibility of a nonincremental process mediating this relationship, overriding it, or simply interacting with it in some way to modify the direct link was largely overlooked in the development of the model [52]; - The multiplicity of decision processes that have acquired the title “incremental” has deprived the concept of descriptive and explanatory value [52].
“Implementation Analysis Framework” (Sabatier, P. and Mazmanian, D., 1980)	- The framework allows: (i) the explanation of the political cycle as a causal theory; (ii) the integration of top-down and bottom-up approaches; (iii) the assumption of the importance of technical-scientific knowledge in the political process [53,54]; - The framework allows an understanding of how the process of formulation—as well as the mandate that comes with it—either attracts or detracts the chance that programmes and public policies are implemented [55]; - This framework provides a clear and objective structure and defined checklist, thus enabling the kind of analysis that is not just “descriptive”, but more pragmatic [6,35]; - The model has been used to identify critical variables in the implementation processes, allowing an analysis of which factors facilitated and/or limited the effectiveness of this implementation [41].	- Because the framework is focused on “how well” programs and policies are implemented, the substance and quality of these are underestimated [56]; - The framework tends to focus on the policymaker rather than on those affected by the policy, and it ignores the role of policy opponents, who make demands during the policy process [53]; - The original approach of the model does not provide an accurate empirical description and explanation of the interactions and problem-solving strategies of the actors involved in the policy delivery [57].

Source: authors’ own study.

summarises the discussion on the advantages and disadvantages of each model of reference. The internal analysis of each framework of reference is not the aim of our paper. Rather, we focus on proposing ways to adequately address the three stages that the researcher, politician, or decisionmaker should tackle to understand public policy. However, in order to help the reader, we present some of the key aspects of each framework of reference.

4.1. Implications for Theory

First, the IPMF, as a conceptual proposal, in our view, represents a potential theoretical and methodological contribution to the fields of management, administration, public policy, and political science, given that it provides an integrated analysis of the public policy cycle and a holistic view of the policy processes. Second, another aspect of this paper and our framework is that they coherently revisit and repurpose three well-established models, thereby encouraging readers to also look at these models in a critical and fruitful way. Third, none of the frameworks of reference incorporate dimensions or categories in the way we have presented, which is because they stem from a systematisation and digestion of the literature. Fourth, the strategy of harmonising the models allows for both collecting data and proceeding with the complex processing of that data. We not only propose new hierarchies of dimensions and categories, but also new interpretations for the models of reference. In fact, the adoption of three theoretical references, separated by decades, was not detrimental to our own theoretical exercise.

4.2. Limitations

We acknowledge that our proposition is a framework that shall benefit from empirical utilisation. As we have asserted previously, the IPMF rests upon the stages of the public policy cycle. Even though much of the conceptualisation and reasoning developed in this paper lean towards the decisions and implementation of public policies, we believe that the framework has the potential to be used in other contexts (i.e., to adhere to other realms of decision making). However, this needs to be further explored. Finally, we also concede that the categories and dimensions might require adjustments based on the application of the framework and the experience attained while using it.

4.3. Recommendations for Further Research

The complementarity we were able to forge between the frameworks of reference allowed us to establish a valuable interconnectivity between the different stages of decision making. Thus, the IPMF has potential usefulness in understanding the patterns that pertain to reforms in healthcare, education, transportation, and housing, among many other political realms that go through the public policy cycle. Moreover, the broad nature of the different dimensions and categories means that they can be used in various contexts. The IPMF thus opens avenues for research, application, and even evidence-based policy making because the IPMF lends itself to mixed methods research, and not just the processing of qualitative inputs. Hence, research can focus on the internal structure of the framework and/or on how it can be expanded, perfected, and used in tandem with other methods that researchers and users of the IPMF may deem necessary or supporting. This means that the IPMF may be used as a standalone tool, or as part of a more comprehensive research protocol, if the case or the user calls for that.

5. Conclusions

The IPMF allows for the integrated understanding and explanation of the forces and operative factors in each stage of the public policy cycle. In short, it couples the stages of the public policy cycle with three frameworks of reference, thereby allowing a holistic overview of the process. This coupling is innovative and may be useful to scholars and people in charge of institutions alike. An integrated framework, such as the one that we propose, due to the fact that it allows for the analysis of each of the three stages, may contribute to the success of each individual stage, and thus the effective implementation of the public policy being studied and analysed.

The IMPF offers the possibility of walking the same path over and over for successive interactions inside the cycle in a perfecting effort, based on the effects and impacts of the policy. This representation of policy making as a cycle of attempts to solve problems that result in policy learning from the reiterated analysis of the problems and experimental solutions is (we hope) a virtue of our proposal.

Our aim was to provide a theoretical proposition—a tool—that may inspire empirical applications. The density of the conceptualisation exposed enables us to predict that the integrated model is built in such a way that it may be applied, utilised, and expanded in future research. Considering all the advantages that we have explained, we would like to propose the IPMF as a new tool to tackle decision making in public policies, as well as a way to critically approach the three theoretical models of reference, and as a tool to better understand the stages of the policy-making cycle.