Editorial introduction

Contemporary policy-making considerably relies on policy advice contributed by a variety of actors and sources. “Policy advisory systems” as introduced by Seymour-Ure (1987) and later expanded on by Halligan (1995) have been central to moving beyond considerations regarding individual actors to assessments of the interactive effects of multiple interlocking sets of suppliers in specific jurisdictions, which provide policy advice to policy-makers. The concept of policy advisory systems focuses on the country-specific organization and institutionalization of policy advice. It refers to an interlocking set of actors with a unique configuration in each sector and jurisdiction, who provide information, knowledge, and recommendations for actions to policy-makers (Halligan 1995). This advice can be located in the public service, internal, and external to government (ibid.; Craft and Howlett 2013). The notion of policy advisory systems thus “transcends the boundaries of internal government expertise and knowledge transmission activities” (Howlett and Migone 2013a, p. 241). Each country has its own advisory system that consists of different actors, such as ministerial policy units, councils, (ad hoc) commissions, state-financed research institutes, think tanks, and (private) consulting organizations. Advisory actors can give solicited and unsolicited advice to the government, accord with the government or choose a highly critical approach and almost act as voice of external interests or even countervailing power. These varieties of the institutionalization of policy advice have been tackled by research examining their aggregate and sectoral policy dynamics beyond those typically associated to country-based assessments or discreet sets of actors.

Investigating policy advisory systems allows reconstructing and explaining both country and sector-specific characteristics of policy advice. The concept of policy advisory systems enables research to integrate the analysis of dimensions of power in policy-making and the knowledge dimension of policy content—instead of separating these dimensions in different analytical perspectives: In a conventional perspective on policy advice, the function of knowledge and expertise transferred to policy-making is to mitigate the implications of complex political environments and institutional and organizational conditions of policy-making by rationalizing policy processes (e.g., Nutley et al. 2007). Policy advice, however, not only serves this rationalization, but—at least as important—also legitimizes policies and policy processes (Vedung 1997; Weiss 1979). The concept of policy advisory systems addresses this link. Moreover, the focus on policy advisory systems allows us to understand the relation of advisors and political decision-makers as a multilateral relationship where different advisory actors and decision-makers interact with each other. Hence, research on policy advisory systems completes our understanding of the processes and mechanisms of policy-making. Scholarship has provided valuable insights with regard to the “mapping” of policy advisory systems in different countries (see for instance different contributions in Blum and Schubert 2013; Head and Crowley 2015; Veselý et al. 2016). Moreover, recent research has focused on changes in policy advisory systems in Western democracies, most prominently the externalization (e.g., Howlett and Migone 2013b; Veselý 2013) and politicization of advice (e.g., Craft and Howlett 2013), and discussed the implications of these changes for the policy process (Craft and Wilder 2015; Craft 2016).

Despite considerable progress in the theorizing of policy advisory systems and a growing body of empirical findings, some important gaps in the literature remain. First, comparative studies on the causes and consequences of changes in national policy advisory systems are missing. Second and related to this, there is a lack of theoretical development regarding how policy advisory systems evolve and why they change over time. Third, so far, little work has been invested to link empirical findings with policy process theory (Halligan and Craft this issue) or concepts of network or collaborative governance (Torfing and Ansell 2017). Fourth, existing empirical research is mainly focussing on Anglo-Saxon countries (Van den Berg this issue), whereas there are far fewer studies on jurisdictions grounded in other political-administrative traditions. Also, the more “visible” policy advisors, like professional policy analysts in government or think tanks, receive much more academic attention than the less visible actors who are “hired” by governments as managerial or policy consultants on a temporary basis (Howlett and Migone 2013b: p. 112; Van den Berg this issue). Finally, the politics of policy advisory systems—i.e., the interests of different (political) actors in actively shaping or changing different countries’ policy advisory systems—is so far exclusively discussed with regard to specific supplies, particularly internal policy analysts/bureaucrats, but not for the whole system.

Summing this up, efforts to systemize, theorize, or explain variation of policy advisory systems across countries—even under similarly changing conditions—are rare. This special issue seeks to contribute to filling the abovementioned research gaps and to deepen our understanding of the conditions under which policy advisory systems vary. Thus, the issue focuses on the sources of variation of policy advisory systems and their implications for the policy advisory process, i.e., for how knowledge is transferred into the policy cycle and for how policy advisory systems evolve over time. Studying policy advisory systems across countries and policy domains in an explanatory perspective enhances our understanding of factors underpinning policy processes in contemporary democracies and their complex political, organizational, and social points of reference. Extrapolating factors crucial for variation in policy advisory systems contribute to a systematic understanding of political decision-making processes in democratic systems of government and, hence, to core questions of policy research.

Several contributions in this special issue further our empirical knowledge on national policy advisory systems and the drivers of change within these. By doing so, the authors surpass a pure “mapping exercise” by explaining the observed changes and discussing their implications. In their study of policy advisory change dynamics, three contributions move beyond identifying externalization and politicization as core trends. The articles of Craft and Halligan and Van den Berg point to more specific categories to the study of change. Both contributions underline temporality and composition dynamics according to which change dynamics vary. Veit, Hustedt, and Bach highlight changing advisory logics represented in the policy advisory system to assess change. The included country cases in this special issue comprise a group of continental European countries (Denmark, Germany, the Netherlands) and a group of Anglo-Saxon countries (Australia, Canada, New Zealand, UK) allowing for comparative conclusions and findings. Craft and Halligan compare Westminster systems, and the contribution of Fraussen and Halpin focuses on the Australian case. Kelstrup compares think tanks in Germany and Denmark with the UK. Veit, Hustedt, and Bach investigate change dynamics in the German internal policy advisory system in a longitudinal perspective. Van den Berg employs a similar focus on the Dutch policy advisory system. What makes his study particularly interesting is that he also includes “less visible actors” of policy advice—ad hoc commissions and consultants—in his analysis. In terms of theory, the contributions of this special issue center on the concept of the policy advisory systems, but study that from various angles. While some articles focus on the origins and implications of this concept in the preceding literature, others use it as the analytical tool to guide their empirical analysis, while again others seek to refine it. In particular, the contributions by Van den Berg and Veit, Hustedt, and Bach address the rationalization versus legitimization function of policy advice. As Van den Berg argues, this question is of particular importance for systems with neo-corporatist structures where both functions are institutionally intertwined. Both articles show that rationalization and legitimization represent distinct demands that policy advice is confronted with and that actors can play with strategically.

The special issue comprises six contributions, starting with a literature review that takes stock of scholarly work on policy advisory systems in the last thirty years, followed by four empirical studies that all include different innovative elements in conceptual terms and concludes with a conceptual contribution that proposes a new analytical framework for the study of policy advice.

The article by John Halligan and Jonathan Craft reflects on the concept of policy advisory system thirty years after Halligan’s seminal chapter. This article offers retrospective analysis in tracing the contributions of the initial advisory system concept, highlights key subsequent developments, and offers prospective critical analysis of a future research agenda. Emphasizing advisory change dynamics, Halligan and Craft study the (de)institutionalization of select advisory units in the “classic” Westminster systems finding that general characterizations of externalization and politicization can be better specified through attention to the tempo, intensity, and sequencing of (de)institutionalization dynamics in the cases. These dimensions, Halligan and Craft argue, add much precision to the comparative study of advisory system change as they reveal differences that are often masked by the broader trends of politicization and externalization.

The contribution by Caspar van den Berg expands the empirical domain beyond the Westminster family by studying change dynamics in the Dutch policy advisory system since the 1960s based on three surveys (conducted with ministerial officials in 2007 and 2013, and with external consultants in 2015) as well as personal interviews with ministerial officials. Representing a consensus-driven, neo-corporatist polity, the author finds fragmentation, externalization, and politicization as core trends, which, however, developed rather differently from the Anglo-Saxon systems with regard to timing, rhetoric, and substance. In particular, the Dutch policy advisory system is portrayed as rather dynamic developing toward a pluralist scenery of advisory actors that represent coexisting, but competing ideas about good policy-making.

Whereas public sector reform is an important driver of change in the Dutch policy advisory system, this is not the case in Germany. In their study of the German internal policy advisory system, Sylvia Veit, Thurid Hustedt, and Tobias Bach identify mediatization, scientization, and the increasing relevance of wicked problems as key drivers of change for the German system. The authors assess advisory change dynamics since the 1990s according to three logics of policy advice—political salience, credibility, and representativeness. These serve as analytical dimensions to assess change dynamics. The authors show that the components of the internal policy advisory system have changed differently over time, but all share a hybridization of advisory logics changing the nature of policy advice.

The two subsequent contributions focus on the role of think tanks—again in different jurisdictions and with varying analytical lenses. Bert Fraussen and Darren Halpin investigate the capacity of think tanks to contribute to strategic policy-making which they argue depends on their research capacity, organizational autonomy, and ability for long-term policy planning. By studying Australian think tanks through structured interviews with their CEO or Director, the authors find considerable variation in their ability to provide strategic policy advice. They argue that a deeper investigation of the actual organizational features of policy advisory actors is required, rather than assuming that particular policy capacities are inherent to certain organizational types, such as advocacy groups or think tanks.

In contrast, the quantitative study by Jesper Dahl Kelstrup asks if and why think tanks’ mobilization of expertise varies across different policy advisory systems. The author compares the dissemination activities of think tanks in Germany, UK, and Denmark. Based on an analysis of outputs (publications, events) and newspaper impact of think tanks, he finds considerable variation across these countries with the UK’s think tanks as the most active disseminators in the sample. While think tanks have so far been predominantly viewed as suppliers of evidence-based policy advice (Craft and Howlett 2012), Kelstrup’s study suggests that think tanks rather balance the need for supply activities with a need for dissemination to reach multiple audiences. This balance, however, varies across institutional systems depending on the overall degree of coordination within policy advisory systems.

Finally, Arnost Veselý proposes an innovative analytical framework for the study of policy advice by relating research on policy advisory systems to the literature on policy work. He understands policy advice as a continuous interaction of advisory actors embedded in a given institutional system through which advisory routines and norms are continuously reproduced. By conceptualizing policy advice as a multi-level phenomenon that is, moreover, deeply institutionalized, Veselý’s study seeks to overcome the implicit dichotomy of the locational model inherent to much thinking about policy advisory systems as well as the inherent dichotomy of knowledge and values at the core of much (often normative) reasoning on policy advice. By doing so, the author aims to achieve a more realistic conceptual framework to depict a phenomenon as complex as policy advice in the real world.

In sum, the contributions to this special issue show that the configuration of a policy advisory system is shaped by the overall political system it is embedded in. The articles in this issue point to distinct configurations of policy advisory systems in the Anglo-Saxon Westminster systems that traditionally relied on bureaucratic advice to a large extent combined with a multitude of highly active think tanks. These Westminster-type policy advisory systems can be distinguished from those embedded in the continental European parliamentary democracies such as Denmark, Germany, or the Netherlands that combine a traditionally strong bureaucracy with a complex and highly institutionalized landscape of advisory bodies across a range of organizational formats including academics and interest group representatives.

The articles collected here suggest that the openness of a policy advisory system to include external, i.e., non-bureaucratic knowledge affects the policy advisory process. This openness is pre-structured by the general degree of coordinated interaction between government and external actors in policy-making, i.e., the proximity to corporatist policy-making styles. If a corporatist policy-making style predominates, the policy advisory system is complex and well institutionalized including external actors whose knowledge is demanded in policy processes. In contrast, in pluralist systems with a less coordinated policy-making style, policy advisory systems are less institutionalized and, if needed, external knowledge is contracted by competition among advisory actors.

The contributions to this special issue in general indicate that changes in policy advisory systems can occur both as results of larger administrative or managerial reforms that also more or less directly conceive advisory systems as reform objects and as processes of a gradual adaptation to changing conditions of governing. Such an encompassing understanding of change as both reform and adaptation calls for a wide perspective in the study of change dynamics in policy advisory systems. Yet, having identified this pattern of variation and its sources directs our attention to the implications of this variation for future research.

The findings and research in this issue point to demands for further research in at least two respects. Firstly, distinguishing policy advisory systems alongside political system variables and policy process variables bears the question whether performance of policy advisory systems is systematically related to distinct political system or policy process variables? Do we find policy advisory systems in some political systems to be systematically more or less efficient than in others? Secondly, but related to the first area, is the “million-dollar-question” in policy advisory research, namely seeking to measure, assess or determine the influence of policy advice on policy output. Here, the works in this special issue again point to thinking alongside systematic types of advisory systems and their linkage to policy advisory processes in order to approach the concrete study of influence. However, more conceptual or typological work on the systematic classification of policy advisory systems alongside political system and policy process variables would be needed in a first step to investigate the above mentioned areas in future research. This special issue aspires to provide conceptual legwork paving the way for a new round of systematic comparative work on policy advisory systems, their working dynamics, and their impact on public policies.