I. Goals of Public Policy: Normative, Explanatory or Something Else?

While a number of scholarly reflections on the relationship between theory and practice in the field of public policy have appeared recently (Allison, Reference Allison, Goodin, Moran and Rein2008; Cairney, Reference Cairney2015; Gormley, Reference Gormley2007), Simeon's was one of the first. As Atkinson's article in this issue argues, Simeon also provides an unequivocal answer: the goal of public policy studies should be to explain and understand the scope, instrument choice and distributive impact of public policies, but definitely not to be prescriptive. Atkinson's article, “Richard Simeon and the Policy Sciences Project,” juxtaposes this view with that of Harold Lasswell who, on the one hand, believed in a science of policy problem solving with the use of integrated tools from economics, decision theory, systems theory and public administration and, on the other, believed that public policy analysis should be policy and socially relevant. Atkinson argues, “Simeon accepted the distinction between seeking an improved understanding of the policy process and actually improving it but wanted no part of the latter enterprise,” instead directing our attention to explaining and understanding.

Lasswell, in contrast, was a proponent of both knowledge of and knowledge in the decision process. He would have looked with favour at the proliferation of policy schools in the United States (Ellwood, Reference Ellwood2008) and Canada, 22 as of this writing, with McGill's slated to come online in September 2018 (Cappe, Reference Cappe2015), and the systematic methods and approaches taught to students such as rational decision making that are purported to “tame politics.” Atkinson's article provides a candid assessment of the challenges facing contemporary policy research and policy making, which he notes are subject to increased complexity, and the need for mechanisms to overcome these so-called “wicked” or insoluble problems. A “science” of policy making—and policy analysts willing to enter the fray of decision making—seems to be exactly what is needed even while we uncover empirical evidence of our limited “cognitive capacity to manage the political challenges policy problems.”

Simeon, alas, is not here to defend his arguments in favour of “knowledge of” alone. It should be noted that he was an eager enthusiast of the training of students at the University of Toronto's School of Public Policy and Governance where he ended his teaching career. His approach to teaching, though, was always to remind students of the importance of understanding the essentially political underpinning of decision making and the policy process. His research and teaching continually returned to the key dimensions of policy making highlighted in the 1976 article: the importance of the policy environment, the distribution of power, the importance of prevailing ideas and institutional frameworks, and the process of decision making (Simeon, Reference Simeon1976: 566), all of which help us to develop “theoretically relevant categories, typologies, or classifications of the dimensions of public policy” (553). In such a policy-making environment, we can see an even greater value of Simeon's call for a theoretically informed understanding of politics grounded in both means-end reasoning and value rationality, “that is, the collective deliberation on, and choice of, rational goals” (in Atkinson's words).

II. A Framework to Explain Policy Outcomes

As noted above, Simeon (following Lasswell) defined the goal of policy scholars as providing an explanation of “who gets what, when and how.” For Simeon, explaining “who gets what” is a matter of accounting for three dimensions of policy making: the scope of government social and economic action, the means or instruments used to make and enforce policy choices, and the distributional effects of policy choices on interests and values. To explain these three features of public policy, Simeon advanced an analytical “funnel of causality” framework that situated policy makers’ decision about what constituted important problems, their acceptable solutions, and appropriate procedures for considering policy alternatives, in their broader macro-level context of social and economic structures, dominant ideas and values, formal institutional structures and the distribution of power and influence (Reference Simeon1976: 555–56). These contextual features shape the policy process of state and non-state actors’ interaction and bargaining, with the process itself making its own contribution to the micro-level behaviour of decision makers. Simeon emphasized that it was decision makers’ perceptions of their context and how they interpreted the costs and benefits of the actions that ultimately determine their behaviour.

Simeon's case for a comprehensive model of policy making that integrated causal factors at macro- and micro-levels was intended to overcome what he saw as a regrettable tendency in the existing literature “to focus on one end of the funnel without taking account of the other,” for example, to focus on the environment of policy making while ignoring the political process, or on the policy process while neglecting the policy setting (556). Although he stressed it was the independent and interactive effects of a set of causal factors that explained policy outcomes, Simeon nonetheless hypothesized that some causal factors were especially important to explain the scope, means and distributional outcomes of policies. The distribution of power resources was “obviously related to all three dimensions of policy” (570) and cultural/ideological factors in the form of assumptions, norms and values concerning the substance and procedure of policies were “essential” in explaining the substance and means of policy (573).

To what extent has the funnel of causality framework advanced by Simeon for studying public policy been taken up? And to what extent have other scholars affirmed the causal significance of the factors that he identified as especially important for explaining distributional outcomes, that is, the distribution of power and dominant ideas? Two articles in this symposium shed light on these questions.

Matt Wilder, in his article on “Whither the Funnel of Causality?” provides an equivocal answer to the first question. On the one hand, he observes that Simeon's comprehensive explanatory model of integrating “a range of inputs from numerous levels of abstraction” (as Wilder puts it) is not the current norm. On the other, Wilder argues that a number of assumptions in the funnel of causality model are tenets of neo-institutionalist approaches: namely, that input variables interact to produce outputs, relationships between structures and agents are multidirectional, and outputs at one point become inputs at another. The interpretivist assumption in the funnel of causality model, he argues, has also been taken up by neo-institutionalist approaches, especially those that recognize the power of discourse. More specifically, Wilder argues that neo-institutionalist approaches that conceptualize policy subsystems, policy communities and networks and policy regimes as the locus of policy making incorporate funnel of causality and Simeon's assumptions of factors from various levels of analyses having causal effects in interaction and via feedback mechanisms. Tracking the conceptual evolution of the literature on policy networks and communities, policy subsystems and policy regimes, Wilder describes how this literature has built-in funnel of causality themes, including the interpretivist assumption that reflexive actors cognitively filter their structural context. He concludes that the funnel of causality “remains useful for parsing out the relative influence of environment, power, ideas, institutions, and process.”

In their article, “Ideas, Institutions, and the Politics of Federalism and Territorial Redistribution,” Daniel Béland and André Lecours demonstrate the importance of examining the interactive effects of input variables in order to explain policies that transfer fiscal resources across constituent units of a federation. Documenting the presence of ideas countering such territorial redistribution in both Canada and Belgium, they also demonstrate that the impact of these ideas on the two countries’ respective equalization and social security programmes has been different. While ideas supporting a significant reduction in territorial redistribution have had a profound and enduring impact on policy debates in Belgium, permeating electoral politics and resulting in reductions in territorial redistribution in Belgium, they have not had similar effects in Canada.

To explain this cross-national difference, Béland and Lecours point to differences in the territorial organization of the two federations’ political party systems and how these institutional differences filter redistribution ideas. Belgium's territorially fragmented system of regional parties gives some regional (Flemish) parties incentives to spread ideas against territorial redistribution and place them on the federal policy agenda. By contrast, Canada's system of pan-national parties that rely on gaining votes across the entire country does not. As the authors conclude, their analysis is consistent with Simeon's view that policy scholars should pay systematic attention to both ideas and institutions and, further, that comparative case studies are an appropriate methodology to highlight their interactive effects.

As the literature reviews in the articles by Wilder and Béland and Lecours make clear, input factors, like institutions and ideas, that Simeon correctly understood affected “who gets what, when and how” have since been given considerable theoretical flesh. How these and other macro- and micro-level factors have causal impacts on policy outcomes is now better theorized and, arguably as well, more clearly demonstrated. As scholars focus on how institutions or power or cultural values and norms affect the behaviour of decision makers, they appear to be increasingly likely to demonstrate that such factors have causal effects only in interaction with features of the context in which they operate (Falleti and Lynch, Reference Falleti and Lynch2009: 1147). To the degree they do so, the field of policy studies has progressed in a way that is consistent with Simeon's contextually situated model of the policy process.

III. Simeon and the “Comparative Turn” in Public Policy

A second major theme in “Studying Public Policy” was its prescription for a comparative study of policy “across both space and time” (1976: 550). In place of what he saw as the existing practice of unique case studies asking different questions and using different methods and approaches, Simeon argued that building up cumulative knowledge and theory required either co-ordinated case studies that asked similar questions and applied similar frameworks and methods, or, alternatively, the application of several different models to a single case (551.). He expressed a preference for longitudinal studies that looked at “the evolution of patterns of policy over long periods within countries, provinces, and other units” when the goal was explaining differences in how they dealt with similar problems (550–51). Although he certainly didn't make light of the challenge, Simeon also believed it was both important and possible to find observable measures of the explanadum, that is, the scope, means, and distributive effects of a policy. The use of empirical methods—of measurement, quantification, and categorization of these dimensions—was necessary in order to compare across units and build theoretical knowledge (557).

Has the study of public policy adopted the comparative methodology that Simeon recommended as a way to build knowledge and theory? And, further, was Simeon's faith in comparative analyses as the method for building policy theory warranted? These questions are taken up by two articles in this symposium issue.

In their article, “Has Simeon's Vision Prevailed Among Canadian Policy Scholars?” Éric Montpetit, Christine Rothmayr Allison and Isabelle Engeli provide empirical evidence that the study of public policy has moved consistently in the direction Simeon advocated, that is, toward comparative policy analysis and toward theory building and explanation, rather than description. They do so by undertaking a content analysis of the articles published in five major generalist public policy journals from 1980 to 2015: Governance, Journal of European Public Policy, Journal of Public Policy, Policy Sciences, and Policy Studies Journal. The authors’ analyses show that the comparative trend (across countries, policy sectors and time) and the shift away from single case studies hold for policy scholars as a whole. But it is especially true of policy scholars based at Canadian institutions who are also part of the broader trend away from descriptive studies toward explanations that rely on methods of systematic observation. Despite Canadians and Australians presumably both sharing the same incentive to shift toward comparative and explanatory analyses in order to garner a larger readership and advance knowledge, Canadian policy scholars have been decidedly more comparative than their Australian counterparts. For the authors, this Canadian-Australian difference indicates Simeon's foresight on the course of Canadian policy studies.

Despite evidence of an increasingly comparative turn in policy studies, Montpetit, Rothmayr Allison and Engeli emphasize the diversity of research designs that characterizes policy studies today. As they observe, the field is rife with debates over the merits of single cases versus comparative studies, over theory building versus thick description and over positivist versus interpretive methods of building knowledge.

Some of the flavour of this debate—and the merits of comparative case studies—is revealed in the article, “Studying Public Policy: Historical Institutionalism and the Comparative Method” by Gerry Boychuk. The literature on historical institutionalism, he argues, has uncovered causal mechanisms of policy development over time that undermine the assumption (which he attributes to Simeon) that independent variables have the same effects across cases. Mechanisms, such as policy feedback, mean that the causal effects of a policy event can be quite different, depending upon when and in what order in the policy sequence it occurs. Although Boychuk acknowledges that Simeon called for longitudinal studies over long periods (as noted above), he argues that Simeon's language of dependent and independent variables suggests a causal logic at odds with the reality of causal relationships that are complex and reciprocal over time. This ontological understanding of causality, argues Boychuk, renders the “comparative imperative” in policy studies less powerful than it was at the time Simeon was writing. He cites developments in “comparative historical analysis” consistent with a shift toward historical case-based studies whose methods include process tracing and archival research. Aligning with the latter approach, Boychuk argues for within case studies that identify the causal mechanisms that explain shifts in policy paths over time either within the same case or in different cases.

IV. A Refocused Research Agenda

We have argued that Simeon accurately articulated much of the agenda that has unfolded in policy research over the last forty years. In this section, we reflect on what he may have overlooked and what we can anticipate as items on the research agenda for the next decade.

In thinking about a future policy studies agenda, we do so as political scientists who share Simeon's belief that our discipline's contribution to policy studies lies in linking policy studies to the perennial political question: who gets what, when and how? Explaining why some alternatives (in terms of scope, means, and distributive effects) are chosen and others are not remains the foremost goal. In quest of this objective, policy scholars have directed their attention to understanding the dynamics of the policy process: the crucial “impact point” as Simeon (Reference Simeon1976: 578) describes it, where structural (material, ideational, institutional) and agential inputs into decision making meet. It is this knowledge of the policy process, says Atkinson, that is a distinct contribution political scientists can make to the world of practical policy making.

Accordingly, and despite what some might regard as an existing over-abundance of policy process theories, a first item on the policy studies agenda is a continuing, but also renewed focus on the policy process. Our point of departure for understanding the nature of the policy process is Atkinson's observation that, from a political science point of view, policy is “about tracing the battle to define problems and the consequences of various solutions.” Referencing Stone (Reference Stone2002), Atkinson argues this battle of “defining policy problems, framing them, and constructing policy agendas is fundamentally about the exercise of political power.” Certainly no political scientist would disagree. But if politics is also “an open-ended struggle for authority and voice” (as Atkinson also observes), the policy process is not just as an exercise of political power, it is also as an exercise in constituting legitimate political authority.

What leverage on the policy process can be gained by distinguishing between the exercise of political power (the ability to get someone to do something he or she would not otherwise have done) and political authority (the ability to induce deference in others)? This question is important because leading frameworks of the domestic policy process (as with Simeon's) rarely make explicit reference to the importance of political authority in the policy process. Instead, they focus on power, with the identification of the actors who exercise political power and the sources (institutional, structural, ideational, and contingent) of their political power occupying a central place in all accounts of the policy process. Insofar as authority is usually defined as legitimate power, this implicit equation of “political power” and “political authority” is understandable, at least when it comes to state actors vested with legal authority.

Yet, to the degree that political authority is a socially constructed concept, it is useful to distinguish it from (materially based) political power since ideas about sources and sites of political authority are more mutable across contexts and time. Although much of normal policy making is characterized by considerable consensus about the actors and processes of constituting legitimate political authority, authority contests—that is, competing claims about who gets to speak in the processes of defining policy problems and their solutions—are also characteristic of policy processes, especially those that surround technically complex issues and/or matters of significant societal dissensus (Skogstad and Whyte, Reference Skogstad and Whyte2015). Authority contests are also evident with respect to transboundary policy issues, where the scope of the problem transcends the regulatory capacity (and legal authority) of states, creating incentives for governments to turn to the regulatory authority of international institutions or, alternatively, private market actors for an effective solution (Genschel and Zangl, Reference Genschel and Zangl2014: 248). Although the need to construct a legitimate basis for binding decision making is especially evident in situations such as these, paying close attention to how and where the process of constituting a legitimate base for the exercise of binding authority unfolds provides us with a window into understanding who gets what.

The adequacy of existing policy process theories in capturing the dynamics of constituting legitimate political authority is uneven. On the one hand, policy process theories that situate policy networks (or policy regimes) at the heart of the policy process (see Wilder's article) are well positioned to capture the dynamics by which state actors seek to augment their own rule- and institutionally based authority by incorporating into the policy process individuals or organizations with credentials of superior knowledge, skill or other resources (Scharpf, Reference Scharpf1999; Skogstad, Reference Skogstad, White, Simeon, Vipond and Wallner2008). On the other hand, and with a few exceptions (Bernstein and Cashore, Reference Bernstein and Cashore2007; Auld et al., Reference Auld, Cashore, Renckens, Kanil, Andresen and Haas2014), public policy scholars have been less attentive to the dynamics of transnational private authority construction, that is, the process by which private (non-state) actors construct authority to develop and monitor rules and standards for their own conduct and that of others in what is usually a global supply chain (Cutler et al., Reference Cutler, Haufler, Porter, Cutler, Haufler and Porter1999). As examples with respect to labour rights, human rights and the environment illustrate, these private standard-setting initiatives have increased in number and complexity over time (Abbott and Snidal, Reference Abbott, Snidal, Mattli and Woods2009; Büthe, Reference Bűthe2010; Porter and Ronit, Reference Porter, Ronit, Porter and Ronit2011). To the (considerable) extent that it impacts the scope for authoritative state action, the exercise of private authority rule making warrants greater attention from policy scholars. While state actors remain pivotal actors and usually retain ultimate control over domestic policy networks, this same assumption cannot be made of transnational private actors’ setting, monitoring and/or enforcing rules for their and other actors’ economic conduct.

Second, in the years since the writing of “Studying Public Policy,” an ever-growing group of scholars has taken a constructivist turn and emphasized the centrality of ideas, culture and identity in the public policy process, both in terms of key inputs into policy and as important outputs and outcomes. Sharing ontological space with bounded rationality models of policy making (Jones, Reference Jones1999; Simon, Reference Simon1985; Tversky and Kahneman, Reference Tversky and Kahneman1981) that highlight the importance of social preferences in decision making (Wilson, Reference Wilson2011), and non-positivist interpretivist methods (Yanow, Reference Yanow1999), the study of cultural factors, identities, and norms has moved far beyond the political culture tradition with which Simeon was familiar (Almond and Verba, Reference Almond and Verba1963; Elkins and Simeon, Reference Elkins and Simeon.1979; Lipset, Reference Lipset1990). The meaning of those identities for politics, and how policy making can shape and even transform those identities, comprises a major plank of research on advocacy coalitions (Weible et al., Reference Weible, Sabatier and McQueen2009), the social construction of target populations (Schneider and Ingram, Reference Schneider and Ingram1993) and narrative policy analysis (Shanahan et al., Reference Shanahan, Jones and McBeth2011), as well as the next generation of policy research on race (Thompson, Reference Thompson2013), indigeneity (Coulthard, Reference Coulthard2014), multiculturalism and citizenship (Banting and Kymlicka, Reference Banting and Kymlicka2006) and nationalism (Béland and Lecours, Reference Béland and Lecours2008).

Third, advances in technology and the advent of the era of “big data” allow for much more methodologically sophisticated scrutiny of the policy process through the use of large-n statistical analyses and survey experiments as well as medium-n studies using qualitative comparative analysis or network analysis (Lachapelle et al., Reference Lachapelle, Montpetit and Gauvin2014; Soroka and Wlezien, Reference Soroka and Wlezien2004). Combined with important theoretical work in comparative politics, including historical institutionalism, more sophisticated causal models, such as the comparative agendas project, have emerged that track and analyze processes of policy stasis and policy punctuations over time and across jurisdictions (Baumgartner and Jones, Reference Baumgartner and Jones1993; Baumgartner et al., Reference Baumgartner, Green-Pedersen and Jones2008; Jones and Baumgartner, Reference Jones and Baumgartner2005). These and other causal theories of path dependence, policy feedback effects, and feedback mechanisms (Campbell, 2003, Reference Campbell2012; Jacobs, Reference Jacobs2011; Pierson, Reference Pierson1993; Skogstad, Reference Skogstad2017; Weaver and Lerman, Reference Weaver and Lerman2010) shed new light on the complex relationship between politics and policy.

The technological capacity for big data-driven studies, however, which has also driven an increased focus on the technical aspects of policy making, such as cost-benefit analysis, has also generated a countermovement. As policy analysis becomes ever more complex and expert driven, resting on what some researchers describe as an overly positivist bent in policy research (De Leon, Reference De Leon1994), other literature has articulated the value of non-positivist and more democratically driven and socially innovative methods, such as human-centred policy design.Footnote ²

What would Simeon 2.0 have thought of these trends? Simeon's Reference Simeon1976 article argued in support of empiricism in policy research. Reflecting over the past forty years, Atkinson's article, in turn, reminds us of the imperative to do policy relevant work that also recognizes the complexity of human decision making. The appropriate balance between the call for empirically driven social research, explanatory theory building and theory testing (using ever more rigorous models and methods) and more normative reflections on what our findings mean for the human condition is likely to be one area where public policy scholars engage in a “stakes” discussion similar to that of constructivists, liberals and realists in international relations.Footnote ³ In thinking about the normative implications of public policy scholarship, like Simeon, we do not believe the task of public policy scholars is to prescribe the substance of good policy. However, and consistent with our attention above to political authority, we do believe that the normative remit of public policy scholars can, and perhaps should, extend to the constitutive features of legitimate policy processes (Scharpf, Reference Scharpf1997: 14–15).