Governing the Contemporary Administrative State: Studies on the Organizational Dimension of Politics 3031280075, 9783031280078

Citation preview

EUROPEAN ADMINISTRATIVE GOVERNANCE

Governing the Contemporary Administrative State Studies on the Organizational Dimension of Politics Jarle Trondal

European Administrative Governance

Series Editors

Thomas Christiansen Luiss University Rome, Italy Sophie Vanhoonacker Maastricht University Maastricht, The Netherlands

The series maps the range of disciplines addressing the study of European public administration. In particular, contributions to the series will engage with the role and nature of the evolving bureaucratic processes of the European Union, including the study of the EU’s civil service, of organization aspects of individual institutions such as the European Commission, the Council of Ministers, the External Action Service, the European Parliament, the European Court and the European Central Bank and of inter-institutional relations among these and other actors. The series also welcomes contributions on the growing role of EU agencies, networks of technical experts and national officials, and of the administrative dimension of multilevel governance including international organizations. Of particular interest in this respect will be the emergence of a European diplomatic service and the management of the EU’s expanding commercial, foreign, development, security and defence policies, as well as the role of institutions in a range of other policy areas of the Union. Beyond this strong focus of EU administrative governance, the series will also include texts on the development and practice of administrative governance within European states. This may include contributions to the administrative history of Europe, which is not just about rules and regulations governing bureaucracies, or about formal criteria for measuring the growth of bureaucracies, but rather about the concrete workings of public administration, both in its executive functions as in its involvement in policy-­ making. Furthermore the series will include studies on the interaction between the national and European level, with particular attention for the impact of the EU on domestic administrative systems. The series editors welcome approaches from prospective contributors and are available to contact at [email protected] and s.vanhoonacker@maastrichtuniversity. nl for proposals and feedback. All books in the series are subject to Palgrave’s rigorous peer review process: https://www.palgrave.com/gb/ demystifying-peer-review/792492

Jarle Trondal

Governing the Contemporary Administrative State Studies on the Organizational Dimension of Politics

Jarle Trondal University of Agder Kristiansand, Norway University of Oslo Oslo, Norway University of California Berkeley, USA

ISSN 2524-7263     ISSN 2524-7271 (electronic) European Administrative Governance ISBN 978-3-031-28007-8    ISBN 978-3-031-28008-5 (eBook) https://doi.org/10.1007/978-3-031-28008-5 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Acknowledgement

This volume shares two aims: the first is to feature and theorize the administrative basis of contemporary democracy; the second is to offer an organizational approach to account for the endogenous and exogenous dynamics of the administrative state. The volume suggests that the contemporary administrative state has transformed some of the basic roles in democracy and that a theory of politics therefore needs to embed the role of public administration. The book identifies two major shifts in the role of administration state and how these feed into politics. In order to theorize these shifts, the book outlines an organizational theory approach that accounts for dynamics in the administrative state. These shifts also highlight how an analysis of the ‘state’ transcends methodological nationalism by including how state structures are deeply embedded in multilevel governance structures and processes, and moreover how a study of public administration is extended to the study of international bureaucracies of international organizations. The book suggests that organization theory may have two distinct contributions to such analyses: First, it offers a bridge between political science and public administration by outlining how dynamics of the administrative state is conditioned by the formal organization of its composite parts and by the social interaction patterns that emerge from these parts. The book argues that organization theory is a powerful tool for approaching the administrative state. Political processes and political systems can be neither adequately understood nor explained without including the organizational dimension(s) of it; secondly, it offers a bridge between the academic communities and the world of practice by offering v

vi 

ACKNOWLEDGEMENT

a theory of organizational design. Taken together, the volume aims to contribute with a coherent and comprehensive theory of the administrative state. The theory is, however, partial by including only a limited set of variables for analysis, thus indirectly challenging eclectic theorizing. The volume summarizes two decades of research in public policy and administration from my side. Building on prior work in organization theory and political science, the volume should be considered a companion volume of some main papers that brings them into a coherent whole. The volume thus partly collects papers that are published and partly frame them into one coherent theoretical framework. In so doing, the volume examines novel elements of the administrative state and offer a theory of its dynamics. Despite reworked for the purpose of this volume, parts of the following chapters have been previously published: Chap. 3 (Public Administration), Chap. 5 (Public Administration Review), Chap. 7 (Policy and Politics), Chap. 8 (Politics and Governance), Chap. 9 (Journal of European Public Policy), Chap. 11 (West European Politics), Chap. 12 (Global Policy). I would like to thank the editors of these journals for their kind permission to reprint parts of these articles. I should also mention that these parts are thoroughly rewritten for the purpose of this book. These chapters have also greatly profited from research collaboration with Nadja Kühn on Chaps. 5, 9, and 10, Gjermund Haslerud on Chaps. 5 and 6, Frode Veggeland, Martin Marcussen, Torbjørn Larsson on Chap. 11, and Stefan Gänzle and Thomas Tieku on Chap. 12. Valuable research assistance has been provided by Silva Malin Hoffmann, Silje Marie Thorstensen, and Siri Katrine Helland Stai at ARENA Centre for European Studies, University of Oslo. The volume has moreover benefited from editorial assistance by the publisher and feedback from numerous colleagues at seminars over the years. I would particularly mention insightful comments on parts of the manuscript from Chris Ansell, Michael W. Bauer, Morten Egeberg, Christoph Knill, Todd Laporte, Johan P.  Olsen, Jeremy Richardson, Jacob Torfing, and Eva Sørensen. I also acknowledge the Charles and Louise Travers Department of Political Science, University of California, Berkeley, for hosting me while writing most of the manuscript. The volume has greatly benefited from research grants and generous research time offered by my host universities. Financial support has been provided by the Norwegian Research Council on three occasions: ‘DISC: Dynamics of International Secretariats’ (2003–2009—‘FRIPRO’), ‘EURODIV: Integration and division: Towards a segmented Europe’ (2013–2018), and ‘Collaborative Strategies for Robust Governance in

 ACKNOWLEDGEMENT 

vii

Turbulent Times’ (2021–2024), by four consecutive awards from the Peder Sather Grant Program at the University of California, Berkeley (2013–2014, 2014–2015, 2015–2016, 2020–2022), and the Horizon programme of the European Union: ‘EU Differentiation, Dominance and Democracy (EU3D),’ H2020 SC6 (2019–2023). These grants have made it possible to collect huge amounts of data over the years, to organize workshops and invite colleagues, and to provide time for writing. Kristiansand, Oslo, Berkeley

Jarle Trondal

Contents

Part I Introducing and Theorizing the Administrative State   1 1 An Introduction to the Administrative State  3 2 Theorizing the Administrative State 33 Part II Outlooks on the Domestic Administrative State  53 3 Evolving Tensions in Public Governance 55 4 Public  Administration Sustainability and Its Organizational Basis 85 5 Agency Governance in Integrated Administrative Systems105 6 Restructuring  Public Governance in Integrated Administrative Systems141 7 Designing the Administrative State169

ix

x 

Contents

Part III Outlooks on the Multilevel and International Administrative State 195 8 Administrative  Sciences and the Multilevel Administrative State197 9 European Integration and the Administrative State221 10 Incomplete Contracting and Policy Influence245 11 The Autonomy of International Public Administration273 12 The Organizational Dimension of Global Governance295 13 Conclusion327 Index341

About the Author

Jarle  Trondal  is Professor of Political Science in the Department of Political Science and Management at the University of Agder, Norway, Professor of Political Science at ARENA Centre for European Studies at the University of Oslo, Norway, and a senior fellow at Institute of European Studies, University of California, Berkeley. His research interests focus on public governance, organizational theory, European Union integration and governance, and international executive governance. Some main books include An Emergent European Executive Order (2010), Governance in Turbulent Times (2016, with Chris Ansell and Morten Ogard), and An Organizational Approach to Public Governance (2018, with Morten Egeberg). https://orcid.org/0000-­0002-­6723-­7761

xi

List of Figures

Fig. 3.1 Fig. 5.1 Fig. 5.2 Fig. 5.3 Fig. 5.4 Fig. 7.1 Fig. 10.1 Fig. 10.2

The theoretical model (a, b) Relationship between perceived importance of the Commission and the perceived importance of own parent ministry (dependent variable), with interaction effects (a, b) Relationship between perceived importance of EU agencies and the perceived importance of own parent ministry (dependent variable), with interaction effects (a, b) Relationship between perceived importance of the Commission and the perceived importance of the government (dependent variable), with interaction effects (a, b) Relationship between perceived importance of EU agencies and perceived importance of the government (dependent variable), with interaction effects Framework for analysis Number of agreements between Norway and the EU, by year, 1973–2018. Source: Norwegian Government (2019) Accumulated number of agreements between Norway and the EU, 1973–2018. Source: Norwegian Government (2019)

65 131 132 133 134 175 256 257

xiii

List of Tables

Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 3.5 Table 3.6 Table 3.7

Table 4.1 Table 4.2

Sample size and response rates in the ministry and agency surveys, 1976, 1986, 1996, 2006, and 2016 Numbers of officials in ministries and agencies, over time (absolute numbers) Proportion of ministry officials reporting contacts toward the following domestic institutions and bodies during the last year (percent) Proportion of agency officials reporting contacts toward the following domestic institutions and bodies during the last year (percent) Proportion of ministry and agency officials reporting contacts toward the following international and EU-level institutions during the last year (percent) Proportion of ministry and agency staff participating in the following during the last year (percent) Summary of factors affecting ministry and agency officials’ patterns of contact and participation (standardized beta coefficients; linear regressions on ministry and agency officials; 2016 data) Civil servants in ministerial departments who report that the following institutions are fairly or very important when important decision are made within own dossier (percent) Civil servants in agencies who report that the following institutions are fairly or very important when important decision are made within own dossier (percent)

67 69 70 71 72 74

76 92 93

xv

xvi 

List of Tables

Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 4.7 Table 4.8

Table 5.1 Table 5.2 Table 5.3 Table 5.4

Table 5.5

Table 5.6

Table 5.7

Table 5.8 Table 6.1

Civil servants in ministerial departments who report that own unit has succeeded fairly much or very much in influencing the following institutions (percent) 93 Civil servants in agencies who report that own unit has succeeded fairly much or very much in influencing the following institutions (percent) 94 Civil servants within ministerial departments characterizing coordination as fairly good or very good within their own policy area along the following dimensions (percent) 94 Civil servants within agencies characterizing coordination as fairly good or very good within their own policy area along the following dimensions (percent) 95 Civil servants’ characterization of mutual trust relationships between ministries and agencies within their own policy area (percent) 95 Summary of factors affecting officials’ reported influence vis-a-vis own institution (own ministry/own agency) and the government (standardized beta coefficients; linear regressions on ministry and agency officials; 2016 data) 99 Descriptive statistics: dependent, independent, and moderator variables 114 Patterns of pairwise correlations 116 Inter-correlations matrix on the importance of institutions (Pearson’s r) 117 Relationship between perceived importance of the Commission and the perceived importance of own parent ministry (dependent variable), with interaction effects: Multivariate regression models 120 Relationship between perceived importance of EU agencies and the perceived importance of own parent ministry (dependent variable), with interaction effects: Multivariate regression models 122 Relationship between perceived importance of the Commission and the perceived importance of the government (dependent variable), with interaction effects: Multivariate regression models 124 Relationship between perceived importance of EU agencies and perceived importance of the government (dependent variable), with interaction effects: Multivariate regression models126 Summary of findings 128 Descriptive statistics: Dependent, independent, and moderator variables 154

  List of Tables 

Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table 6.6 Table 6.7 Table 6.8 Table 8.1 Table 9.1 Table 9.2 Table 9.3 Table 9.4

Table 9.5 Table 9.6

Table 9.7 Table 9.8 Table 9.9 Table 9.10 Table 10.1 Table 10.2 Table 10.3

xvii

Intercorrelation matrix (Pearson’s r) 155 Multivariate regression models (unstandardized coefficients) 156 Multiple interaction models 157 Conditional effects of the importance ascribed to own ministry/agency158 Conditional effects on the importance ascribed to own government159 Conditional effect on the importance ascribed to the Commission159 Conditional effects on the importance ascribed to EU agencies159 Conceptual elements 208 Summary of propositions 228 Sample size and response rates in the ministry surveys, by year 229 Percentage of officials who report being affected by the EU/EEA/Schengen, by year 232 Percentage of officials who report contact with and/or participation in EU-level institutions, by year and ministerial affiliation (sector ministries (SM)/Ministry of Foreign Affairs (MFA)) (percentages) 233 Percentage of officials who have met in ministerial coordination committees, by year 234 Percentage of officials who report contact with or participation in the following EU-level institutions, by year and rank (lower-level officials (L)/medium and higher-level officials (MH) (percentage) 235 Percentage of officials who report the following coordination behavior, by year and rank (lower-level officials (L)/medium and higher-level officials (MH)) 237 Inter-correlation matrix on coordination behavior (Pearson’s r, 2016 data) 238 Percentage of officials who report that the following institutions are important when making decisions on their own policy area, by year 239 Percentage of officials who report the following priorities if conflicts occur between the wishes of their national political leadership and the requirements of EU law 240 Operationalizations 253 Frequency distribution, influence vis-à-vis the Commission and EU agencies. Percent (N) 259 Frequency distribution, contact to the Commission and EU agencies. Percent (N) 261

xviii 

List of Tables

Table 10.4 Table 10.5 Table 10.6 Table 10.7 Table 10.8

Table 10.9

Table 11.1 Table 12.1 Table 12.2 Table 12.3 Table 12.4 Table 12.5 Table 12.6 Table 12.7 Table 12.8 Table 12.9 Table 12.10 Table 12.11 Table 12.12 Table 12.13

Frequency distribution, participation in the Commission and EU agencies. Percent (N) 261 Frequency distribution, policy development. Percent (N) 261 Frequency distribution, administrative capacity. Percent (N) 263 Inter-correlation matrix on influence, secondary structures, and administrative capacity 264 Controlled effects of participation in secondary structures and domestic administrative capacity on agency officials’ reported influence vis-à-vis the Commission (OLS regression analysis with standardized beta coefficients) 265 Controlled effects of participation in secondary structures and domestic administrative capacity on agency officials’ reported influence vis-à-vis EU agencies (OLS regression analysis with standardized beta coefficients) 265 List of interviewees among permanent officials, by formal rank281 Descriptive statistics, respondents, percent (N)304 Time spent on the following tasks (mean (N) and standard deviation, min = 1 max = 5) 309 Inter-correlation matrix on task profile (Pearson’s r) 311 Loading matrix of factor solution after orthogonal varimax rotation (salient loading values are printed in bold) 311 Index variables on policymaking and organizational manager (N, mean and standard deviation, min = 1, max = 5) 312 Importance ascribed to the following considerations and concerns (mean (N) with standard deviation, min = 1, max = 5) 312 Inter-correlation matrix on considerations and concerns (Pearson’s r) 313 Loading matrix of factor solution after orthogonal varimax rotation (salient loading values are printed in bold) 313 Importance assigned to arguments from the following (mean (N) and standard deviation, min = 1, max = 5) 314 Inter-correlation matrix on weight assigned to arguments (Pearson’s r) 315 Loading matrix of factor solution after orthogonal varimax rotation (salient loading values are printed in bold) 316 Role perceptions (mean (N) with standard deviation, min = 1, max = 5) 317 Multivariate regression analysis on task profile (using index variables—standardized beta coefficients) 318

PART I

Introducing and Theorizing the Administrative State

CHAPTER 1

An Introduction to the Administrative State

State formation has historically been driven by needs to administrate war and taxation and as responses to domestic rivalries, in which bureaucratization and the rise of public governments have been by-products (Ansell & Lindvall, 2021; Grzymala-Busse, 2019; Kelemen & McNamara, 2022; Tilly, 1992). The Administrative State, originally coined by Dwight Waldo (1947) and as revisited within the current volume, emphasizes the central role of public administration in democratic systems of governance (March & Olsen, 1984: 741; Olsen, 2018). Seventy years ago, Dwight Waldo (1952) wrote: ‘If administration is indeed “the core of modern government,” then a theory of democracy in the twentieth century must embrace administration.’ Departing from the invitation and lessons of Waldo, this volume suggests that the contemporary administrative state has transformed its basic role in democracy and that a theory of politics needs to embrace the role of public administration. Responding to recurrent debates about the claimed divorce between political science and public administration (e.g., Kettle, 2022) as well as between theory and practice (e.g., Pollitt, 2000), this book aims to identify two major shifts in the role of the administrative state and ultimately how these feed into politics, as well as outlines a theoretical approach that accounts for dynamics of these shifts in the administrative state. The book argues and demonstrates that organization theory has two distinct contributions: First, it offers a bridge between political science and public administration by arguing that © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_1

3

4 

J. TRONDAL

administrative structures fundamentally shape politics and ultimately policy outcomes; second, it offers a bridge between the academic communities and the world of practice by offering a design tool. Empirically, the volume establishes how the administrative state is facing endogenous reforms through administrative devolution (‘agencification’) and exogenous shifts by the rise of multilevel administrative systems and international bureaucracy. Facing dual shifts, the volume offers a comprehensive analysis of how the administrative state handles three interconnected challenges: first, the need for innovation and reform as well as requisite stability and robustness; second, administrative autonomy among regulatory bodies as well as political leadership and democratic accountability; and finally, nation-state autonomy as well as international collaboration and coordination. The traditional role of the administrative state has thus been profoundly modified since Waldo. Still, this volume also highlights the robust character of the administrative state by demonstrating profound stability in public governance even during times of profound turbulence. The volume establishes how governance robustness is sustained by organizational structures representing systemic capacities for stabile governance patterns as well as agile adaptations. Illuminating the organizational basis of robustness, the book demonstrates how the robustness of the administrative state is anchored in stable organizational architectures. Empirically, the volume establishes the organizational basis of the administrative state by offering novel large-N studies of public administrators embedded in multilevel administrative systems and international bureaucracies. The volume shares two ambitions: one is to feature and theorize the administrative basis of contemporary democracy; the second is to offer an organizational approach to account for the endogenous and exogenous dynamics of the administrative state. By theorizing the organizational nuts and bolts of the administrative state, the volume introduces an often-­ neglected theory in the study of public policy and administration. In doing so, organization theory represents a design approach that also establishes a bridge between theory and practice within this field of scholarship. Bringing organization theory to account for contemporary dynamics of the administrative state, the volume aims both to understand contemporary dynamics of the administrative state and suggesting how key components may be (re)designed. Tackling future policy challenges arguably calls for knowledge about the possibilities for organizational design. The book makes two distinct contributions:

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

5

–– Theoretically, it introduces an organizational theory approach to the administrative state. This theoretical approach emphasizes how organizational features bias governance dynamics of the administrative state by nudging institutions’ attention toward certain problems and solutions, making certain policy outcomes more likely than others. Being a theory of organization, this approach arguably represents a design tool that may bridge the gap between a theory of the administrative state and its practical craft. –– Empirically, the book offers multiple empirical studies that demonstrate how varieties of organizational architectures mobilize varieties of governance processes within the administrative state.

Twin Shifts in the Administrative State The book complements existing literature that addresses other core elements of the state, such as the policy state that provides vital welfare (Orren & Skowronek, 2017), the protective state that offers fundamental securities (Ansell, 2019; Kelemen & McNamara, 2022), the market state that manages ordered markets (Miller, 1990), and the state as a sovereign judicial order providing the rule of law (Strayer, 1970). This volume adds to this broad literature by focusing on the administrative components of the state and its role in structuring and coordinating the policymaking process—both within the state by central administrative services, across levels of authority by multilevel administrative structures, as well as at the international level by international bureaucracies. The empirical focus of the volume thus goes beyond methodological nationalism by studying the administrative state beyond its inherent national origins. It identifies two key shifts with profound implications for the administrative state: the first is endogenous administrative reforms and agencification of the administrative state; the second is the rise of international administrative systems, notably international public administration (IPA) and multilevel administrative systems (MLA). Each shift is briefly introduced and comprehensively examined throughout the volume. Endogenous Shifts The rise and growth of the administrative state has required that discretionary powers have been decentralized to government agencies. This

6 

J. TRONDAL

subsection asks three sets of related questions: What is agencification? What explains agencification? What implications does agencification yield for the administrative state? One persistent puzzle within public administration history has been if government portfolios should be mainly organized within integrated ministries or as differentiated structures composed of ministerial departments and semi-detached agencies (Verhoest et  al., 2012). ‘Agencification’ has, partly due to the New Public Management (NPM) wave, been high on the agenda of administrative policymakers and attracted considerable scholarly attention. After at least two decades of agencification research, it is time to take stock. Essentially, this book offers a state-of-the-art account on agencification. One novel ambition of this introductory chapter is to show how endogenous and exogenous shifts in the administrative state co-evolve and are mutually reinforcing. This section highlights a broader set of implications of agencification, particularly with respect to (1) political steering and autonomy within the administrative state and (2) the rise of multilevel administration beyond the ‘national’ administrative state. Little research has addressed the agency phenomenon in a comprehensive manner (Busuioc et al., 2012; Christensen & Lægreid, 2006; Pollitt et al., 2004; Verhoest et al., 2010). One characteristic of the agencification literature is the study of formal goals, resources, and structures of agencies. The literature has examined the constitutive rules on which agencies are based on and the legal framework in which they are operating. The role of agencies and their potential impact has often been derived from their legal competences and their de jure features (e.g., Chiti, 2000, 2009; Szapiro, 2005; Vos, 2000, 2005). While this constitutes a necessary starting point, such approaches need to be supplemented by the study of actual practices to verify to what extent the expected de jure cues manifest themselves in administrative practice. Students of agencification have also focused on both the causes of agencification and its consequences (Christensen & Lægreid, 2006; Lægreid & Verhoest, 2010; Pollitt & Bouckaert, 2004; Pollitt et  al., 2004). However, one noticeable bias in this literature is that a vast majority of the ‘agencification’ scholarship is geared toward administrative history, reform, and change, and less on the consequences of agencification (e.g., Pollitt et  al., 2004). Moreover, to the extent that this literature has explored effects of agencification, organizational structures, procedures, and legal capacities have served as key independent variables (see below). A comprehensive understanding of agencification thus needs to bring several literatures together.

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

7

 hat Is Agencification? W Historically, ministerial portfolios have been arranged either as ‘integrated ministries,’ meaning that a ministerial portfolio constitutes a unitary organization, or as a vertically specialized structure, meaning that a portfolio is split into a ministerial, or cabinet-level, department on the one hand and one or more separate agencies on the other (Verhoest et  al., 2012: 3). Over time, agencies have been moved out of and into ministerial departments, often in cyclical patterns (Aucoin, 1990; Hood & Jackson, 1991; Pollitt, 2008; Verhoest et al., 2007). An agency is an administrative body which is formally separated from a ministerial, or cabinet-level, department, and which carries out public tasks at a national level on a permanent basis, is staffed by public servants, is financed mainly by the state budget, and is subject to public legal procedures. Agencies are supposed to enjoy some autonomy from their respective ministerial departments as regards decision-making (Verhoest et al., 2012). However, the respective ministers normally keep the political responsibility for agencies’ activities (Pollitt & Talbot, 2004). Agencification thus signifies a transfer of government activities to bodies vertically specialized outside ministerial departments. Governments across continents have established agencies at arm’s length from ministerial departments to take care of different regulatory and administrative problems (Pollitt et al., 2004; Verhoest et al., 2004). Agencification is also often accompanied by geographical relocation of agencies away from national capitals and government headquarters. Although an old topic of administrative science (e.g., Gulick, 1937), in the study of state building, party formation, and voting behavior (Rokkan & Urwin, 1982), as well as the symbolic meaning of architecture (Goodsell, 1977), the significance of place has been largely neglected in scholarship on agencification (see however Egeberg & Trondal, 2011a). During the last couple of decades, agencification has also been observed in the European Union (EU) (Ongaro et al., 2012). More than 35 decentralized EU agencies have been established (and new agencies are pending). Apart from being geographically spread throughout Europe, EU agencies cover multiple policy areas, have various legal standings and formal powers, staffing and funding provisions, and engage in a web of relations with external stakeholders. EU agencies have been considered weak in most terms. However, the quantitative leap of EU-level agencification is increasingly gaining a qualitative shift in terms of the establishment of EU agencies with ever more regulatory power and within policy domains of core-state powers (e.g., economic affairs, foreign and security affairs).

8 

J. TRONDAL

Studies show that agencification at the EU level is characterized as a transfer of action capacity from the constituent states to the EU level (Egeberg & Trondal, 2017). For example, EU agencies seem to establish relatively stronger relationships toward the European Commission (Commission) than toward member-state governments. This seems to particularly be the case in relationship with the implementation of EU legislation and in areas where the Commission has established administrative capacities that partly overlap or duplicate the capacities of EU agencies (Egeberg & Trondal, 2011b). Agencification of the EU governing system therefore contributes to a consolidation of the executive powers of the Commission, although at arm’s length distance from political oversight by Commissioners and the European Parliament.  hat Explains Agencification? W A vast literature suggests competing mechanisms of agencification. Agencification may be accounted for by (1) organizational, (2) functional, (3) contingency, and (4) institutional mechanisms. 1. According to an organizational/institutional approach, agencies come about through power struggles and compromises conditioned by pre-existing organizational structures. Organizational change is framed by the heritage of structures and new agencies are thus likely to be embedded within existing organizational architectures (Radin, 2012: 17). 2. According to a functionalist account, agencification is a response to collective action problems. The principal-agent model is often the analytical expression of this functional logic, together with the notion of transaction costs (Tallberg, 2003: 25). The benefits of agencies ‘lie in the reduction of political transaction costs, by providing solutions to collective-action problems that prevent efficient political exchange’ (Tallberg, 2003: 26). 3. Contingent events may also help explaining institutional change and the timing of organizational birth (March & Olsen, 1989; Pierson, 2004). Decisions to create agencies have been motivated by needs to respond to particular circumstances of the moment, and in some cases to crisis. The COVID-19 pandemic was a global crisis that illustrates crisis-driven agencification of the EU since the pandemic led both to the strengthening of existing agencies and to the development of new ones.

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

9

4. Finally, the creation of agencies can also be seen as a trend in public policy and as a fashionable idea within the realms of public management (Christensen & Lægreid, 2006). Meyer and Rowan (1977: 73) emphasize the importance of cultural rules within wider institutional environments which take the form of ‘rationalized myths.’ They are myths because they are widely held beliefs whose effects ‘inhere, not in the fact that individuals believe them, but in the fact that they “know” everyone else does, and thus that for all practical purposes the myths are true’ (Meyer & Rowan, 1977: 75). Delegating tasks to ‘independent’ agencies was increasingly fashionable across the OECD area from the late 1980s, and therefore appealed to many national governments. Implications of Agencification The rise of the administrative state may have implications for democracy since discretionary behavior delegated to agencies may affect established political accountability relationships. Agencification may have profound implications for political steering and bureaucratic autonomy. Before proceeding, however, we should emphasize that the sheer number of agencies across countries should not in itself matter with regard to the effects of agencification. These effects are arguably conditioned by particular organizational forms, not by the statistical distribution of such forms. Egeberg and Trondal (2009) revealed that agency officials pay significantly less attention to signals from executive politicians than their counterparts in ministerial departments. The relationship was also a robust one since it holds when controlling for the type of tasks, the amount of public debate, and contestation and officials’ rank. Last, but not least, the findings were highly consistent across time. At the agency level, the more modest attention to political signals from above seems partly ‘compensated for’ by more emphasis on user and client interests. Thus, the autonomous institution is seldom found; more autonomy gained in one relationship may be followed by more dependence in another relationship (Olsen, 2009; Thatcher, 2002). Officials routinely cope with competing expectations. However, since it is often assumed that the relationship between formal structure and actual behavior is relatively weak (Christensen & Lægreid, 2006), it might be expected that changing administrative doctrines (e.g., NPM, ‘whole of government’) made a difference in agency decision-making. This is, however, not the case: Egeberg and Trondal (2009) show that the proportion of agency personnel emphasizing

10 

J. TRONDAL

political signals was not smaller in the NPM era than in the pre-NPM period. However, the more organizational capacity available in the respective ministerial departments, the more agency personnel tend to assign weight to signals from their respective ministers (Verhoest et al., 2010). As mentioned earlier, one neglected variable within this field of study is agency location. Even though geographical relocation fairly often (but far from always) seems to have accompanied agencification, the effect of agency location has thus far escaped scholarly attention. Given that many decision processes are often hectic and intertwined, to be on the spot means that many actors and arenas can be reached in a relatively short time. Thus, under such circumstances, geographical proximity might be important. Hence, the impact of location might be conditioned by policy stage and the temporal dimensions of decision-making (Goetz, 2012). A large-scale survey (Egeberg & Trondal, 2011a) shows that agency site does not make any significant differences for agency autonomy, agency influence, and interinstitutional coordination. This study focused on already semi-detached, often highly specialized, agencies whose ‘need’ for being steered, influenced, or coordinated with others is relatively modest. It may thus still be possible for organizational location to make a difference if research focus is directed toward bodies that are relatively more involved in the policymaking process. However, notwithstanding missing effects of agency location on administrative behavior, agency locus might indeed also have symbolic effects of importance (Goodsell, 1977). Other studies of agencies look into the perceived functional necessity of agency creation or the political dominance of its principals on the agency through design. They emphasize that their creation reflects from the need for executive capacity or that agencies operate under a set of political constraints. Agencies are considered functional solutions addressing a perceived need by these actors or political instruments through which governments act (Dehousse, 1997; Kelemen, 2002; Kreher, 1997; Majone, 1997; Yataganas, 2001). Yet others interpret the agencification process as the emergence of a multilevel administration—primarily in an increasingly complex European executive order (Curtin & Egeberg, 2008; Egeberg, 2006; Trondal, 2010). One underlying assumption in contemporary agency literature is that agencies do what their inventors want them to do; agencies are expected to develop in ways intended by their makers. The above observation is also largely applicable to studies of the effects of agency creation and design. As delegation of tasks to independent agencies started gaining momentum

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

11

with an ever-growing number of agencies being delegated decision-­ making and quasi-regulatory tasks, anxiety arose at the possibility of them escaping accountability and control. Scholars have pointed to the possible consequences of placing too much power in the hands of such agencies operating at an arm’s length from traditional control and appointing agency heads who cannot easily be held accountable for their actions (Curtin, 2005; Dehousse, 2008; Everson, 1995; Flinders, 2004; Shapiro, 1997; Vos, 2000, 2005; Williams, 2005). A vast literature on agency accountability and control has focused on de jure arrangements as provided for in the agencies’ basic acts without delving into the de facto operation of such arrangements (but see Busuioc, 2010a, 2010b). Verhoest et al. (2012) contributed to rebalance this overly legal bias in much agency literature. In that same spirit, this volume aims to bring scholarship of public administration and European studies together by highlighting a broader set of implications of agencification, particularly with respect to (1) political steering and autonomy within the administrative state and (2) the rise of the multilevel administrative state (see below). The volume establishes that the process of agencification inside sovereign states and within the EU system have profound implications for the politico-­ administrative order in Europe, and for how we should understand it. Exogenous Shifts An often-neglected aspect in public administration literature is how endogenous organizational shifts within the administrative state, for example, agencification, may have profound and often unintended exogenous consequences. Public administration scholarship has paid scant attention to the broader implications of agencification. At a general level, the literature has failed to examine how agencification at one level may have profound consequences for public governance at another level, and thus how shifting features of the state—such as agencification and subsequent networking of agencies—may coexist with democratic governance, multilevel governance, and the rule of law (Danielsen & Yesilkagit, 2014; Egeberg, 2006; Trondal, 2014). This volume examines how ways of organizing the administrative state at one level of government (e.g., the EU level) may bias ways of making public policy across levels of authority and bias the policymaking and accountability process at another level of government (e.g., the national level).

12 

J. TRONDAL

Organizational decentralization and horizontal dispersion of portfolios within the state may provide requisite autonomy for domestic government institutions to become involved in European and transnational governance networks (Legrand & Stone, 2021). Decentralization of the administrative state may supply opportunities for its component parts—for example, agencies—to become partners in administrative networks above and beyond the state level (Moloney, 2021). This volume argues that national agencies organized at arm’s length from their parent ministerial departments and that are partly de-coupled from direct steering from these departments constitute an administrative infrastructure for agency capture beyond the state. National agencies may thus become building blocks of a multilevel EU administrative system. The main EU executive body, the Commission, lacks its own agencies at the national level for the implementation of EU policies. To generate uniform implementation across the EU, the Commission in cooperation with EU agencies have established partnerships with national agencies, partly circumventing ministerial departments. National agencies are thus becoming ‘double-hatted,’ serving both national ministries and EU-level bodies (Egeberg, 2006). Agency de-­ coupling (from ministerial departments) at the national level makes agency re-coupling across levels of governance possible. Integrated ministries would not have been conducive to such a development. Thus, organizational re-coupling (‘de-agencification’) at the national level would seriously challenge administrative integration across levels of governance (Egeberg & Trondal, 2011b). This volume addresses a research gap on the role of public agencies in democratic governance, by providing novel data on the shifting role of public agencies in the governing of a deeply integrated (yet differentiated) EU multilevel administrative system. A recent branch of empirical studies has documented an emergent European multilevel administrative system (MLA) consisting of strongly interconnected administrative bodies across levels of governance (Bach & Ruffing, 2018; Bauer & Trondal, 2015; Heims, 2019; Trondal, 2007). Administrative capacity building at the EU level is seen as challenging administrative autonomy among the member-­ state governments (Genschel & Jachtenfuchs, 2014). These studies also suggest how an organizationally heterogeneous EU sends a plethora of differentiated steering signals to the member states, for example, how the Council of the EU fuels strong member-state coordination and perceptions of national preferences, whereas the Commission fuels a circumvention of domestic political control and privileges non-majoritarian bodies

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

13

(Egeberg et  al., 2003). The more the EU becomes involved in policy coordination and implementation, the more important become issues of administrative interaction between the involved political levels—if only because the EU lacks an administrative basis to conduct ‘supranational’ policies independently from member states’ administrative systems (Bauer & Becker, 2014; Heidbreder, 2011). Contemporary studies have been preoccupied with both understanding the European administrative capacity building (Rittberger & Wonka, 2011) and the interconnected nature of the European public administration (Curtin & Egeberg, 2008; Egeberg & Trondal, 2009). Illustrative of the latter, the European administrative system has been conceived of as a multilevel and nested network administration, albeit sometimes loosely coupled (Benz, 2015), where institutions at different levels of government ‘are linked together in the performance of tasks’ (Hofmann & Turk, 2006: 583). These studies also suggest that whereas the executive branch of government is deeply integrated across levels of government (Trondal & Peters, 2015), accountability institutions and processes are much less interconnected and thus more easily on the losing side compared to executive institutions (Trondal & Wille, 2017). These are structurally embedded biases in the administrative state that are also examined in this book. This volume hereby contributes to a ‘public administration turn’ in EU studies (e.g., Bauer & Trondal, 2015; Trondal, 2007) as well as to an ‘EU turn’ in the study of the administrative state. Essential to this turn has been trying to understand the role of administrative institutions in the EU policy process (Olsen, 2018). The inherent state prerogative of preparing policymaking and getting them implemented has been challenged by the rise of independent and integrated administrative capacities. The supply of new supranational organizational capacities with the rise of EU agencies has strengthened the executive system of the EU to act independently of domestic government institutions (Egeberg & Trondal, 2017). Moreover, organizational capacities of EU-level administrative actors supply capacity to influence non-majoritarian institutions (agencies) at domestic level (Egeberg, 2006). Representing a vertical fragmentation of the administrative state, agencies are an organizational compromise between the need for political steering and the need for administrative autonomy and technical regulation. Moreover, parallel agencification at both member state and EU level not only causes polity fragmentation, but in turn contributes to a broader transformation of the EU’s political-administrative order. Order transformation in turn may cause accountability challenges given that

14 

J. TRONDAL

agencies are structurally situated in the periphery of political scrutiny and parliamentary oversight and embedded in multilevel governing processes. Consequently, parallel agencification across levels of authority may have significant consequences for the distribution of power and accountability across levels of government, pooling of administrative resources across countries, and adding genuinely European perspectives in the policy process (Egeberg & Trondal, 2017). In the policy process, parallel agencification may cause turbulence of scale in which policy solutions at one level cause policy problems at another (Ansell et al., 2017). Studies suggest that the accountability complexity is amplified by both nationaland EU-level agencies’ ambiguous institutional location between member states and the EU level. Since agencies are often ‘double hatted’ (Egeberg & Trondal, 2017), a simple principal-agent model of accountability is inadequate. Instead, one should take a more pragmatic approach, acknowledging that agencies might actually be held to account by several forums representing checks and balances (Busuioc, 2013; Curtin, 2007). However, close ties that have been observed between the Commission and EU agencies point in a more ordered direction, thus reducing the ‘accountability overload’ stemming from a multitude of accountability forums (cf. Busuioc, 2013). This volume shows how deep administrative integration across levels of governance also contributes to external differentiation of the EU (see Rittberger & Blauberger, 2018). This is illustrated by focusing on the administrative state of Norway. Compared to contemporary cases of EU external governance, Norway is by far the most strongly integrated EU nonmember through a dense web of treaties (Egeberg & Trondal, 1999; Fossum & Graver, 2018; Lavenex, 2009). This affiliated status grants the Norwegian central administration privileged access to most parts of the EU administration, which in turn paves the way for deep administrative integration. Moreover, administrative integration might go even further in affiliated nonmember states than in EU member states due to their exclusion from political representation at EU level. A lack of political representation with voting rights in the parliamentary institutions of the EU (the Council and the European Parliament) arguably mobilizes a bias toward the administrative state in strongly affiliated nonmember states.

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

15

Theorizing the Administrative State This section outlines two theoretical concepts that guide the volume— turbulence and organization theory. It is argued that turbulence as a ‘new normal’ in the administrative state—as illustrated by the twin shifts outlined above—may be accounted for by introducing an organizational theory perspective. Turbulence and the Administrative State We live in a near chronic state of crisis-induced turbulence that calls for a new type of governance that may be based on improvisation, agility, and fast learning, which is supported by institutions permitting negotiated knowledge production and adaptation to new and unforeseen problems. The section introduces an analytical vocabulary that captures this predicament and sets a new agenda for public governance and administration research. The term turbulence originated in public administration, where ‘turbulence’ depicted the perception of a need to learn how to manage existing levels of uncertainty, interdependence, and change (Ansell & Trondal, 2018). Turbulence literature is relatively new and builds on key contributions to political science which suggest that problem-solving depends on a complicated intermeshing of temporally interacting items (Cohen et al., 1972; Kingdon, 1984). The theoretical perspectives and more substantive contributions have come later (e.g., Rosenau, 1990), with the most nuanced conceptual take by Ansell and colleagues (Ansell et  al., 2017; Ansell & Trondal, 2018). They distinguish between three types of turbulence (environments, organizations, and scale) which feature different sources and dynamics of turbulence (shifting parameters, intercurrence, and temporal complexity) (Ansell & Trondal, 2018). Crises, as most recently demonstrated by the COVID-19 pandemic, confront the administrative state with situational and transitional challenges to react effectively, timely, and legitimately. This section argues that turbulence—which refers to situations where events, demands, and support interact and change in changing, inconsistent, uncertain, or unpredictable ways—challenges the very conditions for stable, long-term governance. Societal, political, and economic ruptures call upon the administrative state to adapt, anticipate, reform, and innovate at greater speeds in order to produce and deliver public value. Public organizations must not only ‘manage

16 

J. TRONDAL

the unexpected’ (Weick & Sutcliffe, 2011) but also set in motion processes and mechanisms that foster systemic adaptability to ongoing, accelerating, and amplified problems and crisis situations. Turbulence may thus push government institutions to make difficult tradeoffs, pulling them in contradictory, even paradoxical, directions. Societal turbulence may increasingly condition and challenge public governance at multiple levels and across policy domains. New research identifies the challenge that turbulence poses to contemporary governance, and particularly the ability of governance institutions to maintain basic goals, functions, and values in the face of crisis and turmoil (Ansell & Trondal, 2018). This volume theorizes how turbulence can be managed through the provision of robust governance structures that combine flexible adaptation in order to change what needs to be changed and maintain what should be maintained (‘dynamic conservatism’). Recent scholarship has suggested that turbulence is addressed by building a resilient society capable of ‘bouncing back’ and restoring the status quo ante (Ansell & Trondal, 2018). However, robustness suggests a more dynamic approach to turbulence that requires adaptive strategies aiming to ‘bounce forward’ and ‘build back better.’ In short, the book argues that robust policies, governance strategies, and institutions are essential for meeting the heightened turbulence that characterizes our present epoch. Times of societal ruptures and political unrest call upon the administrative state to adapt, anticipate, reform, and innovate—and at greater speeds. Turbulence thus faces the administrative state with a challenge of providing requisite continuity.  urbulence and the Challenge of Continuity T The overall structure of the administrative state may be stable and contain institutions that have lasted for centuries, but within most systems change is ubiquitous. Organizations are created and added to the governance of societal sectors and at different governance levels (Meyer et  al., 1997). New organizational forms arise that for instance reshape public-private boundaries or create new links between existing organizations (Ahrne & Brunsson, 2019; Brunsson & Sahlin-Andersson, 2000). Organizational births can be followed by their ‘deaths’ as established organizations are split into new ones, merged, or, less frequently in the case of public organizations, simply terminated (Boin & Lodge, 2016). Formal structures are frequently changed along the vertical and horizontal axes as tasks and responsibilities are (re)distributed between hierarchical levels, different

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

17

organizations, and organizational subunits (Rolland & Roness, 2011: 400). How can such changes be accounted for in theoretical terms? As March (1981) in a seminal article argues, theories of organizational change are not very different from theories of action in organizations. Consequently, the scholarship on the processes that produce such changes in organizations reflects the theoretical diversity of organization studies in general. This volume shows how public governance is characterized by tensions (Ansell, 2011; Ansell & Trondal, 2018; Emery & Giauque, 2014). One avenue of study has examined how national public administration balances competing steering signals (Olsen, 2010). Another strand of research has focused on the temporal dimension, that is, on how public administration balances continuity and change across time (Howlett & Goetz, 2014; Pierson, 2004). Societal transformations evoke concerns about the sustainability and resilience of public administration and public governance (Christensen & Lægreid, 2009; Pollitt, 2008). Contemporary public governance faces increased calls for reform (e.g., during the COVID-19 pandemic), triggering widespread institutional soul-seeking and questioning of the changing role of the state in society and the economy (Pollitt, 2011). Whereas one body of literature suggests that public sector organizations are responsive to reform demands (Ansell & Trondal, 2018), another strand of literature focuses on how government institutions and public governance processes are profoundly stable across time with an embedded status quo bias (Pierson, 2004: 42). While architects of administrative reforms claim to transform the nuts and bolts of public governance through design measures (Osborne & Gaebler, 1992), studies suggest that results come neither automatically nor efficiently (March & Olsen, 1989). These insights are important since times of administrative reform and turbulence increasingly test the stability of the administrative state and the reliability of public service provision (Alvesson & Spicer, 2019; Ansell & Trondal, 2018; Olsen, 2017).

An Organization Theory Approach to the Administrative State This section introduces an organization theory approach to the administrative state. Organizational factors are vital determinants of decision-­ making behavior in the administrative state, and essential variables in

18 

J. TRONDAL

public governance processes. Despite endogenous and exogenous shifts as outlined above, this book argues and empirically establishes that the administrative state is foremost characterized by a profound stability in which key parameters of public governance processes remain overwhelmingly firm. Stability in public governance implies that certain patterns of interaction are preserved within and across public organizations. The causal role of organizations in contemporary political science is contested. Two competing theories can be used to illustrate that point. The constructivist literature contends that organizational forms are symbolic and de-coupled from actual governing processes. They are underpinned by hegemonic scripts, rituals, and myths that, once adopted, infuse public organizations with positive legitimacy claims, with regard to being ‘modern,’ ‘responsive’ to environmental and/or stakeholders’ demands, ‘entrepreneurial,’ and so on. The historically oriented literature contends that organizational forms reflect incremental evolutionary processes laden with cultural features and local attributes. In this context, organizational continuity and change is a result of the extent to which reform scripts adapt to the layered set of institutionalized norms, values, and identities. In circumstances where there is a clash between reform logics/objectives and institutional imperatives, decoupling is likely to occur (Oliver, 1991), with local actors shielding their organizations from being co-opted by external influences and strategic interests (Selznick, 1966; Thompson, 2008). This book focuses on the structural element of the administrative state and how varieties of organizational forms shape and bias public governance processes. An organizational approach focuses on the organizational architecture of the administrative state and explains governance processes by its organizational architecture (Egeberg & Trondal, 2018). The independent variables include organization structure, organization demography, organization locus, and organization culture. ‘Organization structure’ is a codified system of positions and their respective role expectations concerning who is entitled to do what and how (Scott & Davis, 2016). The structure of the administrative state suggests what should be done, by whom, and how. The theory thus has a strong normative element to it (Stivers, 2020). A theory of organization is thus also a theory of politics (Waldo, 1952). ‘Organization demography’ encompasses the background characteristics that personnel bring with them into the organization, such as gender, geographical or social background, education, and previous work experience as well as length of service in the organization (Pfeffer,

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

19

1982). ‘Organization locus’ denotes the physical structure and geographical location of the organization (Pfeffer, 1982). Finally, ‘organization culture’ consists of the informal norms and values of the organization that enable trust building (Ruscio, 1997; Selznick, 1957). An organizational approach is based on the assumption that organizational characteristics can explain both how organizations act and how they change. It emphasizes how decision processes and human behavior respond to a set of stable organizational routines (Cyert & March, 1963). Essentially, stable premises for behavioral choices are past experiences encoded in rules and expressed in the organizational structure of a government apparatus (Olsen, 2017; Waldo, 1952). Organizational characteristics of the administrative state systematically enable and constrain public governance processes, making some policy choices more likely than others. Organizational factors focus and mobilize attention and action capacity around certain problems and solutions while ignoring others, focus attention along particular lines of conflict and cooperation, and so on (Simon, 1983: 21). An organizational approach posits that organizational factors are not merely an expression of symbol politics (Feldman & March, 1981; Meyer & Rowan, 1977), but create systematic biases in human behavior and collective decision processes by nudging individual and collective choices toward certain problems and solutions, thereby making certain outcomes more plausible than others (Egeberg & Trondal, 2020; Fligstein, 2001; Gulick, 1937; Hammond, 1986; Schattschneider, 1975; Thaler & Sunstein, 2009). The organizational structures of the administrative state are thus fundamentally architectures of choice.

Outline of the Volume The volume is presented in three parts: Each chapter specifies the analytical model used and the datasets applied. Part I: Theorizing the Administrative State This chapter (An Introduction to the Administrative State) sets the stage of the volume and sets out two ambitions: one is to feature and theorize the administrative basis of contemporary democracy; the second is to offer an organizational approach to account for the endogenous and exogenous dynamics of the administrative state. By theorizing the organizational nuts and bolts of the administrative state, the chapter introduces organization

20 

J. TRONDAL

theory as a design approach that establishes a bridge between theory and practice. Bringing in organization theory to understand contemporary dynamics of the administrative state, the volume aims both to understand contemporary dynamics of the administrative state and to suggest how key components may be (re)designed. Tackling future policy challenges arguably calls for knowledge about the possibilities for organizational design. Chapter 2 (Theorizing the Administrative State) outlines the key components of an organizational theory of the administrative state. Being a non-substantive theory, an organizational approach has a wide application to both public and private sector organizations, while also covering a wide range of organizational forms, such as networks and nongovernmental organizations. This chapter theoretically discusses how organizational structures may affect governance processes and considers how such structures make a difference in governance processes. The final section outlines implications of organizational forms on the decision-making cycle and in situations featured by turbulence. Part II: Outlooks on the Domestic Administrative State Chapter 3 (Evolving Tensions in Public Governance) brings two distinct contributions to the book. Theoretically, it demonstrates how the organizational architecture of government represents a robust and systemic capacity for public governance. The chapter establishes how turbulence serves as an enduring feature of public governance and how turbulence is ultimately anchored in the organizational structure of the civil service. Moreover, structured flexibility is illustrated by how the civil service has adapted to both international organizations and societal stakeholders. Empirically, these arguments are probed by a novel large-N longitudinal data set that spans 40 years and several generations of government officials, consisting of nine surveys of civil servants at both ministry and agency levels. These data encompass five observation-points, enabling a long-term perspective on government civil servants over nearly half a century, allowing a comprehensive study of the organizational basis for robust public governance. Chapter 4 (Public Administration Sustainability and Its Organizational Basis) probes the sustainability of the administrative state. Theoretically, the chapter examines how sustainable public governance rests on its organizational fabric. The chapter illuminates how organizational factors systematically influence decision-making behavior and thus public governance

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

21

sustainability. Moreover, the chapter argues that since organization structure is amendable to deliberate manipulative change, it may thus be an available design instrument of the context of choice in public governance. Accordingly, the chapter offers an avenue to build bridges between the academic and practitioner worlds of public administration. Benefiting from the survey outlined in Chap. 3, the chapter demonstrates both administrative sustainability and how public governance sustainability rests on its organizational architecture. Chapter 5 (Agency Governance in Integrated Administrative Systems) observes that government agencies are engaged in two parallel domains of executive governance that fuel contradictory opportunities and constraints. The chapter makes two contributions to the book: First, two conceptual images of agency governance are outlined that derive distinct predictions on how agencies are likely to maneuver when embedded in integrated multilevel administrative orders such as the EU. Second, benefiting from a novel large-N dataset on agency officials, the chapter observes that domestic agencies feature compound behavior characterized by compromises and abilities to navigate conflicting concerns—such as those of domestic governments and EU-level institutions. Moreover, the chapter suggests that this coordination dilemma is mediated through, and conditioned by, existing institutions, practices, and traditions, thus contributing to a differentiated (European) administrative order. However, the compound image of agency governance is shown to be robust because no moderator variables fundamentally change relationships from one governance type to another. Chapter 6 (Restructuring Public Governance in Integrated Administrative Systems) examines government ministries and agencies facing choice-architectures that are multiple, overlapping, ambiguous, and incompatible. The chapter has two ambitions: First, by using an organizational approach of public governance, it derives predictions on how government officials maneuver when taking part in integrated multilevel administrative orders. Second, benefitting from a large-N dataset as outlined in Chap. 3, the chapter demonstrates how organizational factors systematically shape behavioral patterns among government officials. Moreover, to probe the robustness of causal effects, the chapter outlines multiple interaction models. The analysis finds that few moderators have dramatic effects by not profoundly weakening any causal relationships. Moderating variables either strengthen or attenuate already apparent effects. These findings support one of this volume’s core messages, namely

22 

J. TRONDAL

the robustness of the administrative state, even when government institutions are embedded in integrated administrative systems. Chapter 7 (Designing the Administrative State) applies the organization theory approach to meta-governance by illustrating how public organizations may organize policy change and reform by (re-)designing organizational choice-architectures. Examining conditions for meta-­ governance is important since governments experience frequent criticism of existing inefficient organizational arrangements and calls for major reforms of the state. The chapter moreover offers an illustrative case by examining how public innovation processes are shaped by organizational designs. Two arguments are proposed: First, public meta-governance is an accessible tool for facilitating policy change; second, meta-governance may be systematically biased by organizational structuring. The chapter suggests how an organizational approach to meta-governance might both explain meta-governance and make it practically relevant for solving societal challenges in the future. Part III: Outlooks on the Multilevel and International Administrative State Chapter 8 (Administrative Sciences and the Multilevel Administrative State) makes two additional contributions to the book: First, the chapter outlines a conceptual framework that theorizes the administrative system of the EU, and second, offers empirical illustrations on administrative integration in the EU. Taken together, the chapter makes a plea for public administration scholarship to study political order in general and multilevel governance in particular. The political order consists of a relatively stable arrangement of institutions that are formalized and institutionalized. The chapter moreover argues that a common political order entails that relevant institutions are fairly independent of pre-existing institutions, relatively integrated and internally cohesive, and reasonably able to influence governance processes within other governing institutions. The chapter empirically suggests that EU-level administrative institutions may act integrated and independently vis-à-vis member-state governments and exert systematic influence on public governance processes within member-­ state institutions. Chapter 9 (European Integration and the Administrative State) demonstrates how the EU contributes to a self-reinforcing administrative bias in public governance processes of the nation state. The chapter argues that

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

23

European integration without membership reinforces a politico-­ administrative gap, and that this gap may expand over time. The chapter uses an extreme case of high integration without formal EU membership represented by Norway. The findings suggest that the EU reinforces the administrative state through strong unintended assimilation effects. These findings are probed by the large-N data set outlined in Chap. 3, offering observations on government officials at three points in time in the Norwegian central administration: 1996, 2006, and 2016. Theoretically, the chapter examines the role of organizational factors in administrative integration and how the impact of the EU is mediated by organizational variables at the national level. Chapter 10 (Incomplete Contracting and Policy Influence) examines how secondary structures in the administrative state affect policy influence. The chapter studies varieties of institutional access to the EU through differentiated secondary access structures. A central dimension of association agreements and ‘quasi-memberships’ relate to varieties of administrative cooperation across levels of governance. Similar to EU member states, government officials from associated states interact closely with EU-level administrative institutions, notably the Commission and EU agencies. This has spurred an extensive literature on the effect of EU institutions on national policy enforcement (policy downloading). Conversely, this chapter examines how secondary structures may create avenues for third countries to influence EU policies (policy uploading). The chapter offers two distinct contributions to the book: First, it presents novel data on how domestic agency officials perceive their influence vis-à-vis the European Commission and EU agencies, and second, it probes the effects of two distinct drivers of policy influence—secondary structures and available administrative capacity. Chapter 11 (The Autonomy of International Public Administration) moves beyond the EU by extending the study of the administrative state toward international bureaucracies. The influence of international organizations (IOs) is to a large extent supplied by the autonomy of its bureaucratic arm. The chapter has two ambitions: The most important ambition is to theorize conditions for the autonomy of international bureaucracies. One secondary ambition is to offer empirical illustrations of autonomy among office holders in international bureaucracies. Benefiting from interviews with civil servants from three international bureaucracies (the European Commission, the secretariat of the OECD, and the secretariat of the WTO), two observations are made: First, we observe profound

24 

J. TRONDAL

actor-level autonomy among civil servants within three international bureaucracies embedded in three seemingly different IOs. One second theoretical lesson is that international bureaucracies may possess considerable capacity to shape essential behavioral perceptions among its staff in particular and foster behavioral autonomization more generally, through two causal mechanisms: (1) behavioral and role adaptation through organizational rule-following and (2) behavioral and role internalization through ‘in-house’ socialization processes. Chapter 12 (The Organizational Dimension of Global Governance) broadens the study of international bureaucracies by incorporating cases of ‘Non-Western’ international bureaucracies. The rise of executive authority at international level has fueled scholarly interest in the policymaking role of international bureaucracies. Whereas researchers have addressed this phenomenon extensively within the United Nations and EU, we are short of studies of international bureaucracies outside the ‘Western world.’ This chapter makes two distinct contributions: First, it advances an organizational approach to the study of international bureaucracies, and second, it empirically probes the organizational basis for decision behavior among international civil servants of the Commissions of the African Union (AUC) and the Economic Community of West African States (ECOWAS). Benefiting from two original surveys, three distinct findings are put forth: First, even though embedded in intergovernmental IOs, these international bureaucracies are more than neutral secretariats for the member states and may contribute to policymaking in their own right; second, a multiple proxy analysis establishes that international civil servants adopt complex sets of sectoral, epistemic, and supranational behavioral patterns that are biased toward the internal affairs of these bureaucracies. Finally, since organizational variables may be ‘nudged’ to achieve desired outcomes, the chapter draws lessons that may add practical value for change. The volume closes with a conclusion in chapter 13.

References Ahrne, G., & Brunsson, N. (Eds.). (2019). Organization outside organizations. Cambridge University Press. Alvesson, M., & Spicer, A. (2019). Neo-institutional theory and organization studies: A mid-life crisis? Organization Studies, 40(2), 199–218. Ansell, B. W., & Lindvall, J. (2021). Inward conquest. Cambridge University Press.

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

25

Ansell, C. (2011). Pragmatist democracy. Oxford University Press. Ansell, C. (2019). The protective state. Cambridge Elements. Ansell, C., & Trondal J. (2017). Coping with turbulence. In C. Ansell, J. Trondal, & M. Ogard (Eds.), Governance in Turbulent Times. Oxford University Press. Ansell, C., & Trondal, J. (2018). Governing turbulence. An organizational-­ institutional agenda. Perspectives on Public Management and Governance, 1(1), 43–57. Ansell, C., Trondal, J., & Ogard, M. (Eds.). (2017). Governance in turbulent times. Oxford University Press. Aucoin, P. (1990). Administrative reform in public administration: Paradigms, principles, paradoxes and pendulums. Governance, 3(2), 115–137. Bach, T., & Ruffing, E. (2018). The transformative effects of transnational administrative coordination in the European multi-level system. In E.  Ongaro & S. Van Thiel (Eds.), The Palgrave handbook of public administration and management in Europe. Palgrave Macmillan. Bauer, M.  W., & Becker, S. (2014). The unexpected winner of the crisis: The European Commission’s strengthened role in economic governance. Journal of European Integration, 36(3), 213–229. Bauer, M. W., & Trondal, J. (Eds.). (2015). The Palgrave handbook of the European administrative system. Palgrave Macmillan. Benz, A. (2015). European Public Administration as a multilevel system administration: A conceptual framework. In M.  W. Bauer & J.  Trondal (Eds.), The Palgrave handbook on the European administrative system. Palgrave Macmillan. Boin, A., & Lodge, M. (2016). Designing resilient institutions for transboundary crisis management: A time for public administration. Public Administration, 94(2), 289–298. Brunsson, N., & Sahlin-Andersson, K. (2000). Constructing organizations: The example of public sector reform. Organization Studies, 21(4), 721–746. Busuioc, M. (2010a). The accountability of European agencies: Legal provisions and ongoing practices. Eburon. Busuioc, M. (2010b). European agencies: Pockets of accountability. In M. Bovens, D. Curtin, & P. ’t Hart (Eds.), The real world of EU accountability: What deficit? Oxford University Press. Busuioc, M. (2013). European agencies. Oxford University Press. Busuioc, M., Groenleer, M., & Trondal, J. (Eds.). (2012). The agency phenomenon in the European Union. Manchester University Press. Chiti, E. (2000). The emergence of a community administration: The case of European agencies. Common Market Law Review, 37(2), 309–343. Chiti, E. (2009). An important part of EU’s institutional machinery: Features, problems and perspectives of European agencies. Common Market Law Review, 46(5), 1395–1442.

26 

J. TRONDAL

Christensen, T., & Lægreid, P. (2006). Agencification and regulatory reform. In T. Christensen & P. Lægreid (Eds.), Autonomy and regulation. Edward Elgar. Christensen, T., & Lægreid, P. (2009). Living in the past? Change and continuity in the Norwegian Central Civil Service. Public Administration Review, 69(5), 951–961. Cohen, M. D., March, J. G., & Olsen, J. P. (1972). A garbage can approach to organizational choice. Administrative Science Quarterly, 17(1), 1–25. Curtin, D. (2005). Delegation to EU non-majoritarian agencies and emerging practices of public accountability. In D. Gerardin & N. Petit (Eds.), Regulation through agencies in the EU.  A new paradigm of European governance? Edward Elgar. Curtin, D. (2007). Holding (quasi-) autonomous EU administrative actors to public account. European Law Journal, 13(4), 523–541. Curtin, D., & Egeberg, M. (2008). Tradition and innovation: Europe’s accumulated executive order. West European Politics, 31(4), 639–661. Cyert, R. M., & March, J. G. (1963). A behavioral theory of the firm. Prentice-Hall. Danielsen, O.  A., & Yesilkagit, K. (2014). The effects of European regulatory networks on the bureaucratic autonomy of national regulatory authorities. Public Organization Review, 14(3), 353–371. Dehousse, R. (1997). Regulation by networks in the European community: The role of European agencies. Journal of European Public Policy, 4(2), 246–261. Dehousse, R. (2008). Delegation of powers in the European Union: The need for a multi-principals model. West European Politics, 31(4), 789–805. Ebinger, F., Veit S., & Fromm, N. (2019). The partisan–professional dichotomy revisited: Politicization and decisionmaking of senior civil servants. Public Administration, 97(4), 861–876. Egeberg, M. (Ed.). (2006). Multilevel Union administration. Palgrave Macmillan. Egeberg, M., & Trondal, J. (1999). Differentiated integration in Europe: The case of the EEA country Norway. Journal of Common Market Studies, 37(1), 133–142. Egeberg, M., & Trondal, J. (2009). Political leadership and bureaucratic autonomy. Effects of agencification. Governance, 22(4), 673–688. Egeberg, M., & Trondal, J. (2011a). Agencification and location: Does agency site matter? Public Organization Review, 11(2), 97–108. Egeberg, M., & Trondal, J. (2011b). EU-level agencies: New executive centre formation or vehicles for national control? Journal of European Public Policy, 8(6), 868–887. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies. What have we learnt (and where do we go from here). Journal of Common Market studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press.

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

27

Egeberg, M., & Trondal, J. (2020). The organizational basis for public governance. The Oxford Research Encyclopedia of Politics. https://doi.org/10.1097/ acrefore/9780190228637.013.1491 Egeberg, M., Schaefer, G.  F., & Trondal, J. (2003). The many faces of EU Committee Governance. West European Politics, 26(3), 19–40. Emery, Y., & Giauque, D. (2014). The hybrid universe of public administration in the 21st century. International Review of Administrative Sciences, 80(1), 23–32. Everson, M. (1995). Independent agencies: Hierarchy beaters? European Law Journal, 1(2), 180–204. Feldman, M. S., & March, J. G. (1981). Information in organizations as signal and symbol. Administrative Science Quarterly, 26(2), 171–186. Fligstein, N. (2001). Organizations: Theoretical debates and the scope of organizational theory. Manuscript, UC Berkeley. Flinders, M. (2004). Distributed public governance in the European Union. Journal of European Public Policy, 11(3), 520–544. Fossum, J.  E., & Graver, H.  P. (2018). Squaring the circle on Brexit. Bristol University Press. Genschel, P., & Jachtenfuchs, M. (Eds.). (2014). Beyond the regulatory polity? Oxford University Press. Goetz, K. (2012). Political time in the EU. In E. Jones, A. Menon, & S. Weatherill (Eds.), The Oxford handbook of the European Union. Oxford University Press. Goodsell, C. T. (1977). Bureaucratic manipulation of physical symbols: An empirical study. American Journal of Political Science, 21, 79–91. Grzymala-Busse, A. (2019). How the Catholic Church shaped early state formation in Europe. Unpublished paper, Stanford University. Gulick, L. (1937). Notes on the theory of organization: With special reference to government in the United States. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration, Columbia University. Hammond, T. H. (1986). Agenda control, organizational structure, and bureaucratic politics. American Journal of Political Science, 30(1), 379–420. Heidbreder, E. (2011). Structuring the European administrative space: Policy instruments of multi-level administration. Journal of European Public Policy, 18(5), 709–727. Heims, E. (2019). Building EU regulatory capacity. Palgrave Macmillan. Hofmann, H.  C. H., & Turk, A.  H. (Eds.). (2006). EU administrative governance. Edward Elgar. Hood, C., & Jackson, J. (1991). Administrative argument. Dartmouth. Howlett, M., & Goetz, K. H. (2014). Introduction: Time, temporality and timescapes in administration and policy. International Review of Administrative Sciences, 80(3), 477–492.

28 

J. TRONDAL

Kelemen, R.  D. (2002). The politics of ‘Eurocratic’ structure and the new European agencies. West European Politics, 25(4), 93–118. Kelemen, R. D., & McNamara, K. R. (2022). State-building and the European Union: Markets, war, and Europe’s uneven political development. Comparative Political Studies, 55(6), 963–991. Kettle, D. F. (2022). Public administration and political science: Can this marriage be saved? Governance. https://doi.org/10.1111/gove.12724 Kingdon, J. W. (1984). Agendas, alternatives, and public policies. Little, Brown. Kreher, A. (1997). Agencies in the European community—A step towards administrative integration in Europe. Journal of European Public Policy, 4(2), 225–245. Lægreid, P., & Verhoest, K. (Eds.). (2010). Governance of public sector organizations. Palgrave Macmillan. Lavenex, S. (2009). EU rules beyond EU borders: Theorizing external governance in European politics. Journal of European Public Policy, 16(6), 791–812. Legrand, T., & Stone, D. (2021). Governing global policy: What IPE can learn from public policy? Policy and Society, 40(4), 484–501. Majone, G. (1997). The new European agencies: Regulation by information. Journal of European Public Policy, 4(2), 262–275. March, J. G. (1981). Footnotes to Organizational Change. Administrative Science Quarterly, 26(4), 563–577. March, J. G., & Olsen, J. P. (1984). The new institutionalism: Organizational factors in political life. American Political Science Review, 78, 734–749. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions. Free Press. Meyer, J. W., Boli, J., Thomas, G. M., & Ramirez, F. O. (1997). World society and the nation-state. American Journal of Sociology, 103(1), 144–181. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Miller, D. (1990). Market, state, and community. Oxford University Press. Moloney, K. (2021). Deconcentrated global governance, transnational administration, and the public administration discipline. Global Public Policy and Governance, 1, 175–201. Oliver, C. (1991). Strategic responses to institutional processes. Administrative Science Quarterly, 16(1), 145–179. Olsen, J.  P. (2009). Institutional autonomy and democratic government. In P. G. Roness & H. Sætren (Eds.), Change and continuity in public sector organizations. Fagbokforlaget. Olsen, J. P. (2010). Governing through institution building. Oxford University Press. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford University Press. Olsen, J. P. (2018). Democratic accountability and the changing European political order. European Law Journal, 24(1), 77–98.

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

29

Ongaro, E., Barbieri, D., Bellè, N., & Fedele, P. (2012). European Union (EU) agencies. In K.  Verhoest, S.  Van Thiel, G.  Bouckaert, & P.  Lægreid (Eds.), Government agencies. Palgrave Macmillan. Orren, K., & Skowronek, S. (2017). The policy state. Harvard University Press. Osborne, D., & Gaebler, T. (1992). Reinventing government. Penguin Press. Pfeffer, J. (1982). Organizations and organization theory. Pitman. Pierson, P. (2004). Politics in time. Princeton University Press. Pollitt, C. (2000). Shadowland: The poorly-mapped, underdiscussed yet vital interface between public administration research and practice. In B. Geert & J.  Werner (Eds.), European perspectives for public administration. Leuven University Press. Pollitt, C. (2008). Time, policy, management. Oxford University Press. Pollitt, C. (2011). Not odious but onerous: Comparative public administration. Public Administration, 89(1), 114–127. Pollitt, C., & Bouckaert, G. (2004). Public management reform. Oxford University Press. Pollitt, C., & Talbot, C. (Eds.). (2004). Unbundled government. Routledge. Pollitt, C., Talbot, C., Caulfield, J., & Smullen, A. (2004). Agencies. How governments do things through semi-autonomous organizations. Palgrave Macmillan. Radin, B. A. (2012). Federal management reform. Georgetown University Press. Rittberger, B., & Blauberger, M. (2018). Introducing the debate section: The EU in crisis: EU studies in crisis? Journal of European Public Policy, 25(3), 436–439. Rittberger, B., & Wonka, A. (2011). Introduction: Agency governance in the European Union. Journal of European Public Policy, 18(6), 780–789. Rokkan, S., & Urwin, D. (1982). The politics of territorial identity. Sage. Rolland, V. W., & Roness, P. G. (2011). Mapping organizational change in the state: Challenges and classifications. International Journal of Public Administration, 34(6), 399–409. Rosenau, J. N. (1990). Turbulence in world politics. Princeton University Press. Ruscio, T.  P. (1997). Trust in the administrative state. Public Administration Review, 57(5), 454–458. Schattschneider, E. E. (1975). The Semisovereign people: A realist’s view of democracy in America. Harcourt Brace Jovanovich College Publishers. Scott, W. R., & Davis, G. F. (2016). Organizations and organizing. Routledge. Selznick, P. (1957). Leadership in administration. University of California Press. Selznick, P. (1966). TVA and the grass roots. Harper & Row. Shapiro, M. (1997). The problems of independent agencies in the United States and the European Union. Journal of European Public Policy, 4(2), 276–277. Simon, H. (1983). Reason in Human Affairs. Stanford: Stanford University Press. Stivers, C. (2020). The significance of the administrative state. Public Administration Review, 68(1), 53–56.

30 

J. TRONDAL

Strayer, J. (1970). On the medieval origins of the modern state. Princeton University Press. Szapiro, M. (2005, September 8–11). The framework for European regulatory agencies: A balance between accountability and autonomy. Paper Presented at the Third ECPR Conference, Budapest. Tallberg, J. (2003). Delegation to supranational institutions. In M. Thatcher & A. Stone Sweet (Eds.), The politics of delegation. Frank Cass. Thaler, R. H., & Sunstein, C. R. (2009). Nudge. Improving decisions about health, wealth and happiness. Penguin Books. Thatcher, M. (2002). Analysing regulatory reforms in Europe. Journal of European Public Policy, 9(6), 859–872. Thompson, D. F. (2008). Deliberative democratic theory and empirical political science. Annual Review of Political Science, 11, 497–520. Tilly, C. (1992). Coercion, capital, and European states: AD 990-1992. Blackwell Publishers. Trondal, J. (2007). The public administration turn in integration research. Journal of European Public Policy, 14(6), 960–972. Trondal, J. (2010). An emergent European Executive Order. Oxford University Press. Trondal, J. (2014). Agencification. Public Administration Review, 74(4), 545–549. Trondal, J., & Peters, B. G. (2015). A conceptual account of the European administrative system. In M. W. Bauer & J. Trondal (Eds.), The Palgrave handbook of the European administrative system. Palgrave Macmillan. Trondal, J., & Wille, A. (2017). The European ombudsman: A resilient institution in a turbulent, evolving administrative order. In H. Hofmann & J. Ziller (Eds.), Accountability in the EU. Edward Elgar. Verhoest, K., Bouckaert, G., & Peters, B. G. (2007). Janus-faced reorganization: Specialization and coordination in four OECD countries in the period 1980-2005. International Review of Administrative Sciences, 73(3), 325–348. Verhoest, K., Peters, B. G., Bouckaert, G., & Verschuere, B. (2004). The study of organisational autonomy: A conceptual review. Public Administration and Development, 24, 101–118. Verhoest, K., Roness, P.  G., Verschuere, B., Rubecksen, K., & MacCarthaigh, M. (2010). Autonomy and control of state agencies. Palgrave Macmillan. Verhoest, K., Thiel, S. V., Bouckaert, G., & Lægreid, P. (Eds.). (2012). Government agencies. Palgrave Macmillan. Vos, E. (2000). Reforming the European Commission: What role to play for EU agencies? Common Market Law Review, 37(5), 1113–1134. Vos, E. (2005). Independence, accountability and transparency of European regulatory agencies. In D.  Geradin, R.  Munoz, & N.  Petit (Eds.), Regulation through agencies in the EU.  A new paradigm of European governance. Edward Elgar. Waldo, D. (1947). The administrative state. Routledge.

1  AN INTRODUCTION TO THE ADMINISTRATIVE STATE 

31

Waldo, D. (1952). Development of a theory of the administrative state. American Political Science Review, 46(1), 81–103. Weick, K. E., & Sutcliffe, K. M. (2011). Managing the unexpected. Wiley. Williams, G. (2005). Monomaniacs or schizophrenics? Responsible governance and the EU’s independent agencies. Political Studies, 53(1), 82–99. Yataganas, X. A. (2001). Delegation of regulatory authority in the European Union: The relevance of the American model of independent agencies (Jean Monnet Working Paper 3).

CHAPTER 2

Theorizing the Administrative State

Modern governments formulate and execute policies with consequences for society (Hupe & Edwards, 2012). Yet, governance takes place within and through bureaucratic structures. These structures provide the resource basis for governments to govern, broadly speaking, and also the foundation of modern political order and civilized political life (Fukuyama, 2014; Kristof, 2016). This chapter applies an organizational approach to account for how bureaucratic structures shape governance of the administrative state. It is argued that bureaucratic structures intervene in governance processes, thereby creating a systematic bias that makes some process characteristics and outputs more likely than others. An emphasis on bureaucratic structure presupposes a theory of organizations that assumes that ways of organizing may affect how organizations and their members think and act (Hammond, 1986; March & Olsen, 1983a, 1983b). Structural factors include bureaucratic structure, demography, culture, and location. In this chapter, however, the role of organizational structure is central to the argument. There are at least two important reasons for this choice: First, some of the most promising research findings are related to this variable. Second, contemporary scholarship lacks a comprehensive analysis of how organizational structures affect governance. ‘Organizational structure’ consists of a formalized set of role expectations that specifies who is expected to do what, how, and when. In this sense, the organizational structure is a normative structure that is © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_2

33

34 

J. TRONDAL

analytically disconnected from the decision behavior or process. A particular bureaucratic structure can be illustrated by an organization chart. While there are a variety of definitions, this chapter defines ‘governance’ as encompassing the multidimensional set of decision-making processes taking place within and between government institutions. Few scholars in political science apply an organizational approach to their research. This comes even though few would deny that organizations matter to politics (Döhler, 2020). An organization theory approach is useful in at least two respects. First, it may add new knowledge on how different bureaucratic structures shape governance processes in the administrative state. Second, it may also add practical value for administrative policy. These dual usages of organization theory reflect the old controversy in public administration between seeing it as a ‘science’ for understanding (e.g., Simon) versus as a ‘craft’ for organizational design (e.g., Waldo). If we observe significant causal relations between elements of the bureaucratic structure and characteristics of governance, this provides a potentially important design instrument (Egeberg & Trondal, 2018; Krause & Jin, 2020). This is due to the bureaucratic structure being relatively open to deliberate design (Egeberg, 2012; Gulick, 1937; Meier, 2010). Yet, processes of organizational change may also be difficult to understand and control. Arguably, though, the more organization designers are informed about the complexity of structural change, the more realistic their design ambitions may be. Being a general theory, an organizational approach has a wide application in both public sector organizations and private sector organizations, covering a wide range of organizational forms. The next section discusses theoretically how bureaucratic structure may affect governance. The second section presents some empirical illustrations on how bureaucratic structures make a difference in governance processes. The final section outlines implications of organizational forms on the decision-making cycle and turbulence.

An Organizational Approach Contemporary public administration is conventionally portrayed as being based on a series of dichotomies: politics versus administration, coordination versus fragmentation, integration versus disintegration, trust versus distrust, and so on (Egeberg & Trondal, 2018; Ebinger et  al., 2018; Olsen, 2017; Orton & Weick, 1990; Trein et al., 2020). As an alternative, this volume conceptualizes and empirically demonstrates how government

2  THEORIZING THE ADMINISTRATIVE STATE 

35

institutions at different levels of governance are driven by pragmatism characterized by the co-existence of multiple decision-making premises (Ansell & Trondal, 2018). Public governance is thus seen as a positive-­ sum process in which officials evoke multiple decision-making premises. To account for the composite aspect of government, this volume illustrates how organizational factors shape key elements in the policymaking process. Moreover, the volume also suggests that hybrid structures, such as networks and collaborative arrangements, are established to master unruly public problems. Therefore, this volume also responds to the appeal from Gary King (2014: 165) that ‘the social sciences are undergoing a dramatic transformation from studying problems to solving them.’ Tackling future policy challenges, including improving implementation and law enforcement, calls for knowledge about the possibilities for organizational design. Taking the Weberian model of bureaucracy as a point of departure, it is typically assumed that public bureaucracies possess capacities to shape government staff through mechanisms such as socialization (behavioral internalization through established bureaucratic cultures), discipline (behavioral adaptation through incentive systems), and control (behavioral adaptation through hierarchical control and supervision) (Page, 1992; Weber, 1983). These mechanisms ensure that public bureaucracies perform their tasks relatively independently from outside influences but within the boundaries set by the legal authority and (political) leadership they serve (Weber, 1924/2007). Causal emphasis is thus placed on the organizational structures of the bureaucracy and how they contribute to mobilizing bias. The Weberian bureaucracy model provides a picture of organizations as creators of the ‘organizational man’ (Simon, 1997) and as a stabilizing element in politics more broadly (Olsen, 2010). According to this model, bureaucracies develop their own nuts and bolts quite independently of the societies to which they belong. The model implies that civil servants may act on roles that are shaped by the organization in which they are employed. Key to the nuts and bolts of bureaucracy is thus how the bureaucracy itself is organized and institutionalized, as well as how it is embedded in a wider political order. Organizational dynamics and decision-making behavior are thus primarily assumed to be defined by the ‘in-house’ organizational structures of the government in question (Radin, 2012: 17). Public organizations and the study of contemporary politics create elements of robustness to bureaucratic processes, and concepts such as ‘historical inefficiency’ and ‘path dependence’ suggest that the match between environments,

36 

J. TRONDAL

organizational structures, and decision-making behavior is not automatic and precise (Olsen, 2010). An organizational approach suggests that the supply of organizational capacity has certain implications for how organizations and incumbents act. An organizational approach starts from the assumption that organizational structures mobilize biases in public policy processes because organizations supply cognitive and normative shortcuts and categories that simplify and guide decision-makers’ search for problems, solutions, and consequences (Ellis, 2011; Schattschneider, 1975; Simon, 1997). There may be several reasons why civil servants enact certain behavioral logics. The literature suggests two main mechanisms: adaptation through organizational rule-following and internalization through ‘in-house’ socialization processes. Therefore, literature suggests a distinction between actor-level behavioral internalization of roles and behavioral patterns on the one hand and actor-level behavioral and role adaptation through control and discipline on the other (Checkel, 2007; Trondal et al., 2008). Yet, Lipsky (1980: 19) famously claimed that the nuts and bolts of public bureaucracies are ultimately determined by actors’ conspicuous desires to maximize their own autonomy. By contrast, an organizational approach argues that public governance is organizationally contingent. An organizational approach posits that the rules and routines established in a bureaucracy regulate, constitute, and bias the decision-making behavior and role perceptions that civil servants are likely to evoke (Barnett & Finnemore, 2004: 3). A theory of organization thus also provides a theory of politics (Waldo, 1952). Several mechanisms may serve to connect structure and governance. First, ‘bounded rationality’ means that decision-makers are unable to consider all possible alternatives and their consequences (March & Simon, 1958; Simon, 1965). They therefore need to simplify the world. Organizational structure may provide a systematic and predictable selection of problems, solutions, and choice opportunities (March & Olsen, 1976). Thus, officials in a particular organizational unit are expected to consider only certain alternatives and their consequences, while leaving it to other units to consider other problems and effects. Second, organizational structures are incentive systems in which participants may find it rational to act in accordance with their respective role expectations in order to achieve rewards or avoid punishments (Scott, 1981). Third, institutions may be seen as collectivities in which role compliance is deemed appropriate (March & Olsen, 1989).

2  THEORIZING THE ADMINISTRATIVE STATE 

37

Civil servants live with a constant overload of potential and inconsistent information that may be drawn upon during decision situations. Organizational routines guide the decision-making behavior of civil servants due to computational limitations and the need for selective search. Organizations create collective order out of cognitive disorder by establishing local rationalities among organizational members (March & Shapira, 1992). Organizations are systematic devices for simplifying, classifying, routinizing, directing, and sequencing information toward particular problems, solutions, and decision situations (Cohen et al., 1972; Schattschneider, 1975: 58). Organizations ‘are collections of structures, rules and standard operating procedures that have a partly autonomous role in political life,’ guiding incumbents to systematically emphasize certain aspects of organizational realities (March & Olsen, 2006: 4). This volume theoretically and empirically illustrates that the administrative state and public problem-solving require independent administrative resources and capacities. One necessary, albeit insufficient factor in building an administrative state for the collective pursuit of common goods is the establishment of a permanent and independent government apparatus that serves a common interest (Trondal & Peters, 2013). The rise of political order through institutional capacity building and bureaucratic ‘autonomization’ is seen as a key ingredient of state formation (Bartolini, 2005). With the gradually increased role of bureaucracies, the literature has been occupied in studying the extent to which and under what conditions such institutions can formulate their own policies and transcend a mere neutral and passive role. The craft of political order, according to this volume, is to a large extent brought about by the autonomy of its bureaucratic arm, that is, by the ability of bureaucracies and their staff to act relatively independently of mandates and decision premises from exogenous actors. Part II of the book therefore focuses on government ministries and agencies as vital components of the executive branch of government playing fundamental roles in the democratic governing of modern societies (Orren & Skowronek, 2017; Vibert, 2007). Theorizing the structure-governance relationship entails identifying generic dimensions of structure that might affect decision behavior in systematic and predictable ways. In short, we need to ‘unpack’ organizations to identify such dimensions. If one is interested in how structures matter, one must unpack the different dimensions of structure. These structural dimensions may include variables like capacity, horizontal and vertical specialization, primary or secondary affiliation, and tight or loose coupling,

38 

J. TRONDAL

among others. Such structural dimensions can significantly affect interaction, loyalty, coordination, and information processing. First, capacity should be taken into account (Mathews, 2012). Organizational capacity refers to the resources available in organizations to solve problems, coordinate, learn, and so on (Joaquin & Greitens, 2022). In this volume it refers to the existence of departments, units, or positions devoted to a particular policy area. The idea is that in an information-­rich world systematic interest articulation, problem attention, and problem solving are highly dependent on the degree to which such activities are underpinned by organizational capacity (Egeberg & Trondal, 2018). Second, we should examine how bureaucratic structures are specialized. Horizontal specialization entails how work is divided horizontally within or between organizations. Organizational boundaries help to coordinate activities within units but tend to hamper such activities across units. According to Gulick (1937), horizontal specialization may take place by geography (territory), purpose (sector), function (process), or clientele. For example, territorial specialization tends to focus decision-­ makers’ attention along territorial lines of cooperation and conflict and to create policy consistency within geographical units while fostering variation across such units. Sectoral specialization, on the other hand, evokes sectoral cleavages among decision-makers and fuels policy standardization within a particular policy field across territorial units. Moreover, a functionally specialized structure emphasizes the importance of how things are handled rather than for what purpose. Thus, a structure signals what kind of expertise is deemed relevant. Vertical specialization entails how tasks are allocated vertically within or between organizations. For example, by hiving off regulatory tasks from a ministry to a semi-detached agency one may gain less political interference into agency decisions and/or more involvement of particularly affected interest or user groups. Vertical specialization also characterizes a system of government which spans two or more geographical levels (Hooghe et al., 2016). Third, bureaucratic structures involve primary and secondary structures. A primary structure is a structure within which a decision-maker is expected to use most of his or her time and energy. While a ministry department constitutes a ministry official’s primary affiliation, committees or organized networks represent secondary structures since participation is expected to be part-time. Although secondary structures might affect

2  THEORIZING THE ADMINISTRATIVE STATE 

39

actual decision behavior, the impact is assumed to be significantly less profound than in primary structures (Egeberg, 2012). Finally, bureaucratic structures may be more or less loosely coupled. Loose coupling may lead to more anarchical decision-making processes and thus more surprising results (March & Olsen, 1976). By design, ‘organized anarchy’ may represent a route to discovery or innovation, a way of breaking the bonds of groupthink and building bridges across policy domains. It may also be a counterweight to opportunism and corruption. Key characteristics of such governance processes come close to a general understanding of improvisation: the absence of consistent and shared goals, trial-and-error learning, shifting attention, and fluid participation. Each chapter throughout the volume makes a distinctive operationalization of the different organizational variables that are then empirically tested.

How Organizational Structures Shape Governance This section presents empirical illustrations of how characteristics of bureaucratic structures intervene in governance processes. Assuming that governance is sequenced in stages, the discussion is presented in two stages—the policymaking stage and the implementation stage. The first subsection illuminates how organizational structures may affect the policy cycle. Moreover, assuming that governance is turbulent (see Chap. 1), the discussion will also address how organizational structures may be coping mechanisms in times of turbulence. Implications for the Policymaking Cycle First, organizational structures might affect processes at the policymaking stage. Studies show how horizontal specialization affects the distribution of information in national government bureaucracies (Egeberg & Trondal, 2018). Organizational borders, for example, between different ministries, establish semipermeable walls that shape the informational basis for ‘silo-­ thinking.’ Similar effects are found within international bureaucracies (Trondal, 2014). Inter-ministerial information exchange is positively related to officials’ rank and to the existence of inter-ministerial committees (secondary structures). A study by Kassim et  al. (2013: 188–89)

40 

J. TRONDAL

illustrates behavioral consequences of horizontal specialization within a supranational bureaucracy, showing that policy coordination is simpler within than across bureaucratic units of the Commission. Through horizontal specialization, certain interests and concerns become routinely supported by organizational capacity while others are ignored: A study of the transfer of the Commission’s pharmaceutical unit from DG Enterprise and Industry to DG Health and Consumers showed that the unit’s policy focus changed from being biased toward business interests to becoming more attentive to patient and public health concerns. The unit’s external environment also changed from being dominated by industry organizations toward being more populated by patient and consumer groups (Vestlund, 2015). The relationship between a public bureaucracy and its stakeholder groups may thus be transformed through organizational reforms. To illustrate how horizontal specialization affects governance, EU institutions provide a valuable laboratory. One distinction is whether a bureaucratic structure is structured according to territory or a nonterritorial principle of specialization, such as sector or function. In the case of the EU, an inherent territorially specialized system is reflected in the Council of Ministers. The Council, mainly a legislative body, reflects the territorial composition of the nation-state system where each member state is represented by an executive politician. Studies show, accordingly, that decision-­ makers mainly upload national preferences in the Council and that patterns of cooperation and conflict tend to follow territorial (national) lines (Naurin & Wallace, 2008). By contrast, the Commission is mainly structured according to sector and function. Executive politicians at the top (commissioners) oversee sectoral and functional departments, and they have the Commission as their primary organizational affiliation. Studies suggest that commissioners behave significantly different from ministers in the Council: Sectoral and supranational concerns are considerably more emphasized (Egeberg, 2006; Trondal et al., 2010). At the administrative level, a departmental structure based on sector and function evokes sectoral or functional identities among Commission officials. Commissioners’ supranational, or European, orientation reflects their primary organizational affiliation at the EU level. A recent study shows that the primary organizational affiliation of administrative staff in the Commission toward the EU level explains their supranational role orientation (Trondal et al., 2017). This observation holds not only for the Commission services but

2  THEORIZING THE ADMINISTRATIVE STATE 

41

also for other international bureaucracies (Trondal et al., 2010), and public administration in general (Egeberg, 2012). Second, organizational structures might affect processes at the implementation stage. Empirical illustrations also show how vertical specialization and capacity affect governance processes at the implementation stage. Government bureaucracies can be specialized vertically into separate institutions at the national level, for example a ministerial (cabinet level) department on the one hand and a central (subordinated) agency on the other. Studies show that policy choices at the implementation stage are shaped by the organizational context in which they are made: In contrast to cabinet-level departments, agencies exercise their discretion in comparative insulation from ongoing political processes at the cabinet level (Bach, 2010; Egeberg & Trondal, 2009a; Verhoest et  al., 2010). They have relatively little contact with the political leadership of the ministry, with other ministerial departments than their ‘own’ or with the parliament. When exercising discretion, they attach most importance to expert considerations. Agency personnel are also attentive to the concerns of affected interest and user groups. Less weight is assigned to signals from the political leadership of the ministry. In ministerial departments, by contrast, priority is given to signals from the minister (Egeberg & Trondal, 2009a). Concomitantly, vertical specialization reduces the potential for political steering and control. Studies indicate, however, that this loss of political direction can be partly compensated for by creating organizational capacity in the form of a unit in the ministerial department that duplicates work being done in the agency (Egeberg & Trondal, 2009a; Verhoest et al., 2010). Even though political salience and ministerial steering over national agencies are positively associated, this does not seem to reduce the relationship between agencification (vertical organizational specialization) and agency autonomy (Egeberg & Trondal, 2009a). One second example is how horizontal specialization affects policy implementation across levels of governance. A first illustration involves governance across levels in a unitary state. Implementation of central government policies may take place through sectoral bodies at the regional or local level that are owned by the central government. This organizational architecture leads to standardized public services and regulations across territories, but, simultaneously, intersectoral coordination and the need for local adaptation of policies may suffer. Studies indicate that the latter effects may be compensated by establishing territorially integrated government offices at the regional level. By setting up Government Offices

42 

J. TRONDAL

for the Regions (GORs) in the UK, the reformers aimed at improving coordination between the regional offices of Whitehall departments and to meet the demand for a single point of contact. Research shows that GORs led to greater coordination in the regions and became important mechanisms for developing ‘holistic governance’ within regions (Mawson & Spencer, 1997; Rhodes, 2000). Regional or local self-government strengthens the territorial component further, thus increasing the potential for policy coherence within subnational territories, but, at the same time, increasing policy variation across territories. Federalization is a step going even further in that direction (Hooghe et al., 2016). Rules adopted by IOs are implemented by member states. Thus, implementation is at the outset territorially arranged. The EU is based on this model: EU legislation is implemented by the member states ministries and agencies. This creates considerable variation in implementation practices across member states (e.g., Treib, 2008). EU executive bodies, mainly the Commission and EU agencies, do not possess their own bodies at the national level; however, EU bodies have established direct networked partnerships with national regulatory authorities in charge of implementation, most commonly through networks of agencies working within the same policy area (e.g., Dehousse, 1997; Eberlein & Grande, 2005). In this way, national agencies become ‘double-hatted’ and part of two administrations: a national one and a European one (Egeberg, 2006; Egeberg & Trondal, 2009b). Although national ministries may still be the strongest institutions influencing agencies’ application of EU law, the Commission and EU agencies also exercise considerable impact. The latter entails more uniform practicing of EU legislation across member states. A shift from a territorially arranged structure to sectorally structured implementation rests on two institutional requirements (Egeberg, 2006): first, agencification and vertical specialization at the national level. This has made parts of the bureaucratic structure of national administration available for re-coupling to EU-level bodies. Second, the existence of the Commission and EU agencies embodies organizational capacity for controlling implementation practices at the national level and for constituting hubs of transnational agency networks. Moreover, studies show that such networks tend to increase national agencies’ autonomy and power vis-à-vis their parent ministries (Bach & Ruffing, 2013; Danielsen & Yesilkagit, 2014; Maggetti, 2014).

2  THEORIZING THE ADMINISTRATIVE STATE 

43

Implications for Turbulence Organizations are often created to cope with or manage dilemmas caused by turbulence. These are typically imperfect coping mechanisms rather than perfect solutions to such dilemmas. An organizational approach highlights the role of organizations as both structures and agents of governance. To understand governance in turbulent times, we argue that it is important to pay more attention to how organizations channel, create, and manage turbulence. Contemporary scholarship lacks a comprehensive analysis of the organizational dimension of governance—particularly in understanding public governance. Despite vast scholarship on both public governance and organization theory, respectively, these strands of research have been in mutual disregard (e.g., Kettl, 2002; Olsen, 2010). This volume suggests that governance, and particularly its turbulent features, cannot be adequately explained without making organizations more central to the discussion. Our goal is not only to demonstrate that organizations matter, but also how they matter. In order to elaborate on this perspective, we now illustrate how organizations may matter in situations of turbulence.  rganizations Channel Turbulence O One argument is that organizational structures may channel how turbulence is experienced and managed by organizations. It is well recognized in organization theory that structural dimensions can significantly affect interaction, loyalty, coordination, and information processing. For example, horizontal specialization refers to how work is divided horizontally within or between organizations. The idea is that organizational boundaries help to coordinate activities within entities, but they also tend to hamper such activities across units. For example, territorial specialization may focus the attention of decision-makers along territorial lines of cooperation and conflict, creating policy consistency within geographical units, but inducing variation across such units. Sectoral specialization, on the other hand, is thought to evoke sectoral cleavages among decision-makers and to foster policy standardization within a particular policy field across territorial units. Thus, a structure may consist of sections signaling what kind of expertise is deemed relevant, such as legal, economic, or technical. In her analysis of the American intelligence community prior to 9/11, Zegart provides a good example of how such structural dimensions of organizations may channel turbulence. Overall, US intelligence efforts were highly fragmented, spread across thirteen different agencies with

44 

J. TRONDAL

different missions, objectives, and loyalties, as well as limited overall coordination. Two of the most important intelligence agencies for dealing with the emerging terrorist threat—the CIA and the FBI—encountered their own structural challenges that made it difficult for them, individually and collectively, to ‘connect the dots.’ The CIA’s organizational structure delegated power to its field offices, which was a structure tailored to the relatively stable threats of the Cold War. This territorial structure did a poor job of tracking terrorists moving across field office boundaries. The FBI suffered analogous problems. It also had a decentralized, field office-­based structure focused on prosecuting crime and, as a result, it was unable to connect critical information held by different field offices. The way that organizational structure shapes interaction, loyalty, cooperation, and information processing is well recognized in the organization theory literature. But they are far less well attended to by the public governance literature. Many governance challenges, however, are analogous to the American intelligence community failing to ‘connect the dots.’ Organizational structures channel how turbulence is governed.  rganizations Manage, Absorb, and Amplify Turbulence O Organizations do not passively channel turbulence. They actively manage, absorb, and amplify it. Turbulence creates problems and challenges for organizations which they must find ways to resolve or cope with. While turbulence may be productive or even restorative for some organizations, and create new opportunities for action for others, it may also be constraining, destructive, or even perverse. Thus, organizations may be agents of transforming turbulence, either by dampening it or amplifying it through their actions. We begin with an example from water management in the US, where expanding demands for water have interacted with periodic droughts to produce turbulent conditions for water governance. Lach and colleagues conducted a study of water utilities as a way to understand their adaptation to this turbulence. The study is useful because it describes three different modes of response, illustrating a number of ways in which organizations and governance can interact. The authors describe how utilities respond to the mounting stresses on water supply and demand. In the first mode of response ‘(...) water is viewed [by the utilities] as a benign resource that can be managed through the application of expertise, authority, and money’. Basically, utilities adopt an engineering approach to the challenges they face, and they seek to reduce complexity as a strategy for dealing with uncertainty.

2  THEORIZING THE ADMINISTRATIVE STATE 

45

As the stress on the water system increases, incremental innovations can prove insufficient. Agencies are pushed to envision more significant strategies for ensuring supply and as they do conflict arises with other organizations, including courts, interest groups, and other public agencies. Turbulence is amplified by these interorganizational conflicts, and the water utilities learn that they will not be able to cope with these conflicts through incremental and unilateral engineering strategies. As a result, utilities move toward a second mode of action that the authors call ‘domestication,’ which they describe as a strategy of ‘risk spreading’ by developing collaborative relations with other partners who control some aspect of the risk related to water management. The utilities still maintain an ‘optimization’ view of the problem, but they acknowledge that this optimization can only be achieved by working with other stakeholders to ‘tame’ the problems they face. New institutions that pool information are created to manage this collaborative decision-making and new expert groups are brought into the process to address specific issues. This new governance strategy, however, creates its own turbulence as the utilities struggle to coordinate their different perspectives and agendas. Lach note that water utilities often find it difficult to really tame water management problems using a domestication strategy. A few utilities have therefore begun to experiment with a third mode of governance, pushed in part by the very multistakeholder process created to domesticate the problem. This ‘third mode’ strategy adopts a much more flexible, context-specific, and knowledge-­intensive process of managing water problems. New ideas emerge from the collaborative process that would never have been considered under the engineering paradigm, and new institutions—such as watershed councils—are created to institutionalize the management of multiple worldviews and the ‘adaptive management’ of water. These different modes of governance suggest some general points about how organizations, governance, and turbulence interact. In the first instance, public agencies may prefer unilateral strategies that protect their own prerogatives and expertise. They will try to adapt to challenges through incremental modification of their own governing paradigms. If conflict with external organizations makes this strategy unworkable, however, these agencies may reach out to address the sources of turbulence more directly. In an organized society, these sources of turbulence often arise from, or are influenced by, other organizations. Therefore, directly addressing these sources of turbulence requires public agencies to work

46 

J. TRONDAL

cooperatively with these organizations to share governance risks. Doing this typically requires ‘domesticating’ turbulence by working within new collaborative institutions. These collaborative institutions, in turn, may generate new ideas and institutions. More flexible, adaptive, and knowledge-­intensive problem-solving strategies may arise as a result.  rganizations Make Sense of and Respond to Governance Surprises O and Uncertainty Organizations are the institutional frameworks through which decisions get made that both create and manage turbulence. A good example is provided by the BP oil spill in the Gulf of Mexico, where turbulence was amplified by BP’s decisions. To save money and time, BP decided on a single tube design for the Deepwater Horizon oil rig and this design made it difficult to achieve a proper cement seal of the tube. As a result, the cement contractor Halliburton requested 21 ‘centralizers’ to guide the cement into place. However, BP could not locate the desired number of centralizers from another contractor and, to avoid delay, decided to seal the tube with a reduced number. They also took several additional critical decisions to save time and expenses that further compromised the cement seal. Ultimately, it was the failure of the seal that directly caused the rig explosion and the disastrous oil spill that followed. In the face of turbulence, decision-makers ‘tend to perceive increased urgency to act, narrowing options and the need for dramatic response’. A particular challenge for organizations in managing turbulence is to deal with surprises. Pina and colleagues identify three distinct types of surprises: creeping developments, where ‘emergent, complex, and interactive processes lead to unexpected situations’; sudden events, where ‘new themes emerge from existing processes’; and losses of meaning, where ‘novel, incomprehensible situations’ occur. Since surprises are difficult to predict, and hence, to prepare for, they lead organizations to an emphasis on resilience—that is, on creating the capacity to adapt to surprises. They also accentuate the importance of ‘sensemaking.’ A substantial body of literature describes how organizations engage in sensemaking in the face of rapid change, uncertain circumstances, and surprise. By generating plausible interpretations of situations that create the basis for joint action, ‘[s]ensemaking organizes flux’. Often this occurs in situations characterized by scattered and temporary organization—that is, among a dispersed group of actors thrown together to address specific, time-delimited needs.

2  THEORIZING THE ADMINISTRATIVE STATE 

47

Although the crisis-­management literature is attentive to the importance of sensemaking (Boin et al., 2005), the governance literature has not been as receptive. However, sensemaking theory would seem to be particularly useful for appreciating how multiple stakeholders operating in turbulent environments ‘(...) seek to understand ambiguous, equivocal, or confusing issues or events’. One example of where sensemaking literature has been used to understand interorganizational processes in turbulent conditions is in the context of humanitarian relief. Such situations are characterized by ‘[m]ultiple organizations with multiple missions and different frames of accountability [operating] on multiple scales’. These loosely coupled multiorganizational settings must build on trust and shared understanding to produce effective joint action. Such conditions characterize many turbulent governance situations. In polycentric governance settings, sensemaking may be a process of collective examination and trial-and-error that is likely to require compromise and may lead to hybrid outcomes. In a study of sensemaking at the US Federal Reserve in the face of the financial crisis of 2007–2008, Abolafia observed that the organization drew on a repertoire of narrative plots to make sense of relatively routine events. However, surprise led to ‘narrative innovation’ designed to maintain the organizations’ identity and self-image. He describes sensemaking as ‘(...) a ritual through which members build consensus and create legitimate policy’.  rganizations Manage Turbulence by Creating New Organizations O Governance is often understood to be an emergent process. Organizations are agents because they mold their environments by creating new organizations and institutions (Trist, 1983), and here we may list some examples of how governance is shaped by this: Transnational private regulatory bodies are created as ‘settlements of conflict’. Private transnational regulatory organizations have proliferated, which engage directly with transnational governance by adopting standards, promoting, monitoring, and enforcing such standards, and conducting administrative activities. In peace-building efforts, organizations are sometimes not in a position to engage in long-term planning and must adapt to turbulent local contexts; they compensate by building coalitions. The US National Oceanic and Atmospheric Administration, in cooperation with other international weather and climate research organizations, created the International Research Institute for Climate Prediction to help forecast and provide information about uncertain El Niño weather events. Preceding the

48 

J. TRONDAL

COVID-19 pandemic, national disease response agencies, the World Health Organization, and leading public health NGOs created the Global Outbreak Alert and Response Network to facilitate faster and more effective global response to novel disease outbreaks. In each of these cases, organizations are the agents of creating new organizations and institutions to respond to turbulent conditions.

Conclusion Contemporary scholarship lacks a comprehensive analysis of the organizational dimension of governance. Despite vast scholarship available both on governance and organization theory, respectively, these strands of research have been in mutual disregard. This chapter has argued that applying organization theory to governance may be useful in at least two respects. First, it may add new knowledge on how different organizational structures shape governance processes. Second, it may also add practical value for change. If organizational structures affect governance processes in particular ways as illustrated above, these structures may subsequently be ‘manipulated’ to achieve desired goals. In this way, theoretically informed empirical research may serve as an instrumental device for policy change.

References Ansell, C., & Trondal, J. (2018). Governing turbulence. An organizationalinstitutional agenda. Perspectives on Public Management and Governance, 1(1), 43–57. Bach, T. (2010). Policy and management autonomy of federal agencies in Germany. In P. Lægreid & K. Verhoest (Eds.), Governance of public sector organizations. Palgrave Macmillan. Bach, T., & Ruffing, E. (2013). Networking for autonomy? National agencies in European networks. Public Administration, 91(3), 712–726. Barnett, M., & Finnemore, M. (2004). Rules for the world. Cornell University Press. Bartolini, S. (2005). Re-structuring Europe. Oxford University Press. Boin, A., ’t Hart, P., Stern, E., & Sundelius, B. (2005). The politics of crisis management. Cambridge University Press. Checkel, J. T. (2007). International institutions and socialization in Europe: Introduction and framework. In J. T. Checkel (Ed.), International institutions and socialization in Europe. Cambridge University Press. Cohen, M. D., March, J. G., & Olsen, J. P. (1972). A garbage can approach to organizational choice. Administrative Science Quarterly, 17(1), 1–25.

2  THEORIZING THE ADMINISTRATIVE STATE 

49

Danielsen, O.  A., & Yesilkagit, K. (2014). The effects of European regulatory networks on the bureaucratic autonomy of national regulatory authorities. Public Organization Review, 14, 353–371. Dehousse, R. (1997). Regulation by networks in the European Community: The role of European agencies. Journal of European Public Policy, 4(2), 246–261. Döhler, M. (2020). The architecture of organizations as missed opportunity in political research. Public Administration, 98(4), 891–904. Eberlein, B., & Grande, E. (2005). Beyond delegation: Transnational regulatory regimes and the EU regulatory state. Journal of European Public Policy, 12(1), 89–112. Ebinger, F., Sylvia, V. & Fromm, N. (2018). The partisan-professional dichotomy revisited: Politicization and decision-making of senor civil servants. Public Administration, 97(4), 861–876. Egeberg, M. (Ed.). (2006). Multilevel Union administration. Palgrave Macmillan. Egeberg, M. (2012). How bureaucratic structure matters: An organizational perspective. In B.  G. Peters & J.  Pierre (Eds.), The SAGE handbook of public administration. SAGE. Egeberg, M., & Trondal, J. (2009a). Political leadership and bureaucratic autonomy. Effects of agencification. Governance, 22(4), 673–688. Egeberg, M., & Trondal, J. (2009b). National agencies in the European administrative space: Government driven, commission driven, or networked? Public Administration, 87(4), 779–790. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Ellis, D. C. (2011). The organizational turn in international organization theory. Theorizing International Organizations (http://www.journal-iostudies.org/ sites/journal-iostudies.org/files/JIOS1012.pdf). Fukuyama, F. (2014). Political order and political decay. Farrar, Straus and Giroux. Gulick, L. (1937). Notes on the theory of organizations. With special references to government in the United States. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration, Columbia University. Hammond, T. H. (1986). Agenda control, organizational structure, and bureaucratic politics’. American Journal of Political Science, 30(1), 379–420. Hooghe, L., Marks, G., Schakel, A. H., Niedzwiecki, S., Osterkatz, S. C., & Shair-­ Rosenfield, S. (2016). Measuring regional authority. Oxford University Press. Hupe, P., & Edwards, A. (2012). The accountability of power: Democracy and governance in modern times. European Political Science Review, 4(2), 177–194. Joaquin, M. E., & Greitens, T. J. (2022). American administrative capacity. Springer. Kassim, H., Peterson, J., Bauer, M. W., Connolly, S., Dehousse, R., Hooghe, L., & Thompson, A. (2013). The European Commission of the twenty-first century. Oxford University Press. Kettl, D.  F. (2002). The transformation of governance. The John Hopkins University Press.

50 

J. TRONDAL

King, G. (2014). Restructuring the Social Sciences: Reflections from Harvard’s Institute for Quantitative Social Science. PS: Political Science and Politics, 47(1), 165–172. Krause, G. A., & Jin, R. Q. (2020). Organizational design and its consequences for administrative reform: Historical lessons from the U.S.  Budget and Accounting Act of 1921. Governance, 33(2), 365–384. Kristof, N. (2016). ‘Big government’ looks great from here. International New York Times, Friday, March 18. Lipsky, M. (1980). Street-Level Bureaucracy. Russel Sage Foundation. Maggetti, M. (2014). The rewards of cooperation: The effects of membership in European regulatory networks. European Journal of Political Research, 53(3), 480–499. March, J.  G., & Olsen, J.  P. (1976). Ambiguity and choice in organizations. Scandinavian University Press. March, J. G., & Olsen, J. P. (1983a). Organizing political life: What administrative reorganization tells us about government. American Political Science Review, 77(2), 281–296. March, J. G., & Olsen, J. P. (1983b). The new institutionalism: Organizational factors in political life. American Political Science Review, 78(3), 734–749. March, J. G., & Shapira, Z. (1992). Behavioral decision making theory and organizational decision theory. In M. Zey (Ed.), Decision Making. SAGE. March, J.  G., & Olsen, J.  P. (2006). Elaborating the ‘new institutionalism’. In R. A. W. Rhodes, S. A. Binder, & B. A. Rockman (Eds.), The Oxford handbook of political institutions. Oxford University Press. March, J. G., & Simon, H. A. (1958). Organizations. Wiley. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions. The Free Press. Matthews, F. (2012). Governance and state capacity. In D. L. Faur (Ed.), The Oxford handbook of Governance. Oxford University Press. Mawson, J., & Spencer, K. (1997). The origin and operation of the government offices for the English regions. In J.  Bradbury & J.  Mawson (Eds.), British regionalism and devolution. Jessica Kingsley. Meier, K.  J. (2010). Governance, structure, and democracy: Luther Gulick and the future of public administration. Public Administration Review, 70(1), 284–291. Naurin, D., & Wallace, H. (Eds.). (2008). Unveiling the Council of the European Union. Palgrave Macmillan. Olsen, J. P. (2010). Governing through institution building. Oxford University Press. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford: Oxford University Press. Orren, K., & Skowronek, S. (2017). The policy state. Harvard University Press. Orton, J. D., & Weick, K. E. (1990). Loosely coupled systems: A reconceptualization. Academy of Management Review, 15, 203–223. Page, E. C. (1992). Political authority and bureaucratic power. Harvester Wheatsheaf.

2  THEORIZING THE ADMINISTRATIVE STATE 

51

Radin, B. A. (2012). Federal management reform. Georgetown University Press. Rhodes, R. A. W. (2000). The governance narrative: Key findings and lessons from the ESRC’s Whitehall programme. Public Administration, 78, 345–363. Schattschneider, E. E. (1975). The Semisovereign People. Harcourt Brace Jovanovich College Publishers. Scott, W.  R. (1981). Organizations. Rational, natural, and open systems. Prentice Hall. Simon, H. A. (1965). Administrative behavior. Free Press. Simon, S. A. (1997). Administrative behavior. New York: The Free Press. Treib, O. (2008). Implementing and complying with EU governance outputs. Living Reviews in European Governance, 3(5) http://www.livingreviews.org/ lreg-­2008-­5 Trein, P., Biesbroek, R., Bolognesi, T., Cejudo, G. M., Duffy, R., Hustedt, T., & Meyer, I. (2020). Policy coordination and integration. A research agenda. Public Administration Review, https://doi.org/10.1111/puar.13180 Trist, E. (1983). Referent Organizations and the Development of InterOrganizational Domains. Human Relations, 36(3), https://doi.org/10.1177/ 001872678303600 Trondal, J. (2014). Agencification. Public Administration Review, 74(4), 545–549. Trondal, J., Marcussen, M., Larsson, T., & Veggeland, F. (2010). Unpacking international organizations. Manchester University Press. Trondal, J., Murdoch, Z., & Geys, B. (2017). How pre- and post-recruitment factors shape role perceptions of European Commission officials. Governance, 31(1), 85–101. Trondal, J., Suvarierol, S., & van den Berg, C. (2008). The compound machinery of government. The case of seconded officials in the European Commission. Governance, 21(2), 253–274. Trondal, J., & Peters, B. G. (2013). The rise of European administrative space. Lessons learned. Journal of European Public Policy, 20(2), 295–307. Verhoest, K., Roness, P.  G., Verschuere, B., Rubecksen, K., & MacCarthaigh, M. (2010). Autonomy and control of state agencies. Palgrave Macmillan. Vestlund, N. M. (2015). Changing policy focus through organizational reform? The case of the pharmaceutical unit in the European Commission. Public Policy and Administration, 30(1), 92–112. Vibert, F. (2007). The rise of the unelected. Cambridge University Press. Waldo, D. (1952). Development of a theory of the administrative state. American Political Science Review, 46(1), 81–103. Weber, M. (1924/2007). Legitimate Authority and Bureaucracy. In D. S. Pugh (Ed.). Organization theory. Selected readings. 3rd edition. Penguin Books. Weber, M. (1983). On capitalism, bureaucracy and religion. Harper Collins Publishers. Weick, K. E., Sutcliffe, M. K., & Obstfeld, D. (2005). Organizing and the process of sensemaking. Organization Science, 16(4), 409–421.

PART II

Outlooks on the Domestic Administrative State

CHAPTER 3

Evolving Tensions in Public Governance

How do government officials cope with times of turbulence in their everyday decision-making processes and how may robust public governance be explained? This question is pertinent since societal transformations and environmental turbulence evoke concerns about the robustness of public policy and administration (Adam et  al., 2021; Christensen & Lægreid, 2009; Pollitt, 2008). Times of turbulence increasingly force governments to reform their organizational systems and routines (OECD, 2021; Weick & Sutcliffe, 2007). Contemporary public administration faces increased calls for change—as illustrated during the COVID-19 pandemic—triggering widespread debates on the legitimate and efficient role of public authority (Alvesson & Spicer, 2019; Ansell et al., 2017; OECD, 2021: 19; Olsen, 2017, 2018; Osborne & Gaebler, 1992; Pollitt, 2011; Rainey & Steinbauer, 1999; Riddervold et  al., 2021; Sørensen & Ansell, 2021). This chapter shows that civil servants’ decision-making behavior are profoundly robust—as measured by its long-term stability. The chapter illustrates how stability serves as an enduring feature of public governance and how this is anchored in the organizational architecture of government systems. The argument is empirically probed by a longitudinal research design across half a century. As outlined in Chap. 1, public governance is increasingly characterized by endogenous hybridity and governing paradoxes (Ansell, 2011; Ansell & Trondal, 2017; Emery & Giauque, 2014). One avenue of research © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_3

55

56 

J. TRONDAL

examines how public administration strives to balance competing signals and preferences, for example, those originating from elected office holders, stakeholder groups, and transnational regulatory networks (Moloney & Stone, 2019; Olsen, 2018). Robust public service is seen as the capacity of public institutions to balance and reconcile highly valued but competing concerns, such as the importance of political/democratic steering (majority rule), knowledge concerns (expert rule), the interests of particularly affected parties (stakeholder rule), and the impartiality in law application (the rule of law) (Rothstein, 2012). This chapter contributes to a second avenue of studies interested in the temporal dimension of public administration (Ansell & Trondal, 2017; Douglas et al., 2021: 5). Extant literature has focused on dilemmas of continuity and discontinuity in public governance across time and how temporal robustness in public governance may be understood (Ansell & Trondal, 2017; Howlett & Goetz, 2014; Olsen, 2009; Pierson, 2004). One challenge plaguing these studies has been a dearth of long-term data that enables a lasting perspective on government civil servants (Sørensen & Ansell, 2021). Already in the 1970s, Meyer (1979) observed that studies of public governance were often drawn from one or a few cases, often based on cross-sectional data. To fill this void, this chapter examines temporal robustness in public governance over a period of 40 years. Temporal robustness refers to civil servants’ ability to combine behavioral stability and change across time, however, accentuating their ability to preserve stability by resisting abrupt change. In this chapter, robustness-as-stability in public governance implies that significant behavioral patterns among civil servants are established within and across administrative systems, thus maintaining the ability to allocate sustained and stable value over time (Trondal, 2021). The time period examined (see below) is characterized by internal and external shifts to the government apparatus. These include shifting administrative reform programs questing remodeling of public administrative structures (Aucoin, 1990; Christensen & Lægreid, 2007; Dibben et al., 2004; Emery & Giauque, 2014; Hood & Jackson, 1991; Kettl, 2002; Pollitt & Bouckaert, 2017; Trondal, 2021), competing concepts of legitimate and just modes of public governance (Olsen, 2017; Self, 1972), as well as increasing integration in the EU that ultimately challenges domestic public administration (Benz et al., 2021; Egeberg & Trondal, 2018). Public organizations have experienced sequences of restructuring due to reform doctrines, sectoral reform programs, but also self-standing changes prompted by the environment, political entrepreneurship, organizational

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

57

initiatives, or a combination thereof (Greve et  al., 2018; Kuipers et  al., 2018; Wynen et  al., 2019). Similarly, the Norwegian government has faced increased numbers of public organization reforms during the period of study (Christensen & Lægreid, 2006; Greve et  al., 2018; Verhoest, 2017). Facing a range of stressors, the chapter may be considered as a least likely case of governance stability—and thus temporal robustness. The chapter demonstrates how the civil service is characterized by profound stability in which essential elements of public governance processes are fairly change resistant. Theoretically, it is argued that organizational structures of government institutions foster elements of temporal stability over time, showing that contact patterns and participatory behavior among government officials are unaffected by the above shifts. This suggests that government officials remain unaffected because of isolated structural clusters of the government apparatus. These arguments are probed by a sizable dataset with observations across 40 years, consisting of nine surveys from ministerial and agency officials (N = 13,173). Three distinct contributions are made to extant literature: –– First, the chapter applies an organizational theory perspective on public governance. The chapter thus complements existing literature that emphasizes the role of de-institutionalized informal practices in public policy and administration (White, 2022). Three rationales motivate our theoretical focus: First, the chapter demonstrates how organizational structures of government institutions foster elements of temporal stability in the behavioral patterns among civil servants (e.g., Ansell & Trondal, 2017; Olsen, 2006; Simon, 1983: 22); second, organizational theory is not frequently used in contemporary studies of public policy and administration (e.g., Bevir, 2010; LeviFaur, 2012). Third, an organizational approach is instrumental: Organizational structures are flexibly available to deliberate design and may thus be a potential design tool in public governance (Egeberg & Trondal, 2018; Goodin, 1996; Peters, 2018; Self, 1972). By bringing in the organizational dimension of governance, the chapter introduces a design tool that may be applied for deliberately designing robust governance. The chapter responds to Ostrom’s challenge of building ties between the academic study and the practical conduct of public administration (Egeberg & Trondal, 2018; O’Leary et al., 2010: 292; Trondal, 2021), as well as to Johan Olsen’s call for

58 

J. TRONDAL

making political science an ‘architectonic discipline’ based on organizational sciences (Olsen, 2010). –– Second, these arguments are probed by a sizable dataset with observations across 40 years among top civil servants within both government ministries and agencies in Norway. Offering a long-­ term perspective on government civil servants over almost half a century, the dataset contains multiple observations at several observation points in time. It consists of nine surveys among top civil servants conducted every tenth year between 1976 and 2016 (see below). In sum, this data establishes profound stability in governance processes over study period, and thus overall governance robustness in the central administration overall. The Norwegian public administration is structurally stable over the period of this study, contributing to long-term stability in governing patterns. –– Finally, the chapter establishes how varieties of organizational structures in the government apparatus fuel some consistent tensions within the central administration. These tensions moreover illuminate structured flexibility in the government civil service. A first tension is observed between the open and inclusive versus the sealed and disclosed central administration; the second is between primary versus secondary organizational structures within the administrative apparatus; while the third is between the central administration as predominantly the national system of governance versus its role as part of a EU multilevel administrative system, meaning that national government institutions and EU institutions connect across levels of governance in their everyday activities. As argued below, the central administration has become ‘double-hatted,’ in which its role as a ‘national’ central administration has been supplemented with its role as an administrative infrastructure for the EU. In sum, these tensions provide structured flexibility by allowing the civil service to act flexibly but systematically over time by adapting to internal and external institutions, which includes societal stakeholders and IOs (cf. OECD, 2021). More generally, these observations highlight how diversity inherent in overlapping administrative structures fuels tensions within the central administration in which some ministers are oriented toward politics and the EU, while subordinated agencies are more oriented internally, professionally, and toward s­ takeholder

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

59

groups. In short, organizational specialization represents stable checks and balances inherent in the administrative system. As a result, the chapter establishes how such structural stability serves as a robust character in the central administration. The chapter proceeds in the following steps: The next section theorizes the organizational dimension of governance robustness. Succeeding a presentation of data and methods, the empirical section displays observations of both behavioral continuity and change in the behavioral patterns of top civil servants across time. The chapter concludes by discussing implications for future studies of governance robustness.

Theorizing Governance Robustness The theoretical argument is built on the conjecture that organizational features may account for how organizations perform and reform. Organization theory suggests that governance processes and human behavior reflect some fairly stable organizational routines (Cyert & March, 1963). Premises for choices within organizations are arguably based on previous experiences encrypted in organizational rules and articulated in the organizational architecture (Frederickson et al., 2012; Olsen, 2017). The organizational features are thus likely to systematically enable and constrain certain governing processes, thus making some particular organizational choices more likely than others. Moreover, an organizational approach also features similar ideas on organizational change, emphasizing ideas of structured flexibility in which sustained or stable governing patterns over time may be structurally conditioned (Olsen, 2009). However, we may envisage two kinds of flexibility. The first is associated with the ability of public administration to change over time. The second is related to the informal ‘secondary’ structures that may develop both within and across sectors and levels within organizations and political life (White, 2022). To illustrate, structured flexibility was observed in the German financial administration facing the 2008 financial crisis. Whereas the hierarchically structure of the Federal Ministry of Finance exhibited/ retained stability throughout the crisis, it also developed collegial structures (networks) to respond to the financial crisis, which helped it adapt flexibly to episodic demands (McCowan, 2016). Different conceptions on resilience and robustness can also be found in extant literature. As outlined by the OECD (2021: 26), resilience is not a

60 

J. TRONDAL

binary concept. Studies distinguish between static and dynamic resilience (Ansell & Trondal, 2017). The OECD (2021: 25) also distinguishes how different logics of resilience may unfold at different stages of a disruption. Organizations may be likely to select a logic of static resilience to maintain and restore equilibrium conditions, particularly during early stages of a disruption. This strategy encourages path-dependent solutions that do not transform the essence or character of institutions (Selznick, 1958). This solution is likely to strive to get (or bounce) back to basics—reducing uncertainty and complexity to achieve order and stability. Resilience is enhanced by improving the fitness of the organization vis-à-vis new conditions. To do this, organizations are likely to establish and strengthen buffering capacity such as organizational units and resources whose core task is protecting the organization from changing conditions (Egeberg & Trondal, 2018). By contrast, a logic of dynamic resilience does not draw a sharp distinction between stability and change (Easton, 1965; Farjoun, 2010). Rather, organizations may flexibly and pragmatically use stability to help them change and use this change to help them stabilize (Ansell, 2011). Moreover, no clear equilibrium between the organization and its environment is discerned and the organization appears to be continually changing as a reforming organization (Brunsson & Olsen, 1993). Dynamic resilience emphasizes the importance of building flexibility into organizational arrangements by absorbing complexity and incorporating requisite variety (Ansell & Trondal, 2017). Hence, it emphasizes the importance of maintaining multiple repertoires that can be flexibly used to meet changing situations. Rather than the sharp distinction between minor path-­ dependent incremental change and major exogenously produced punctuated change, this logic anticipates endogenous change that continuously reforms and updates the organization. The two logics of resilience moreover resonate with the distinction between exploitation and exploration (March, 1991), that is learning how to do better what you already do (exploitation) and learning about new opportunities or how to do new things (exploration). Studies show how public organizations tend to balance exploitation and exploration, for example, in order to build capacities for continuous reform at the same time as organizing for everyday service delivery (March, 1991). Exploitation and exploration are thus often combined. A recent study of the COVID-19 pandemic shows how efficient and resilient crisis response includes creative and pragmatic combinations of ‘core government functions’ and at the same time temporary and agile crisis response units and

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

61

skills (Mazzucato et al., 2021). This suggests a complex interplay between strategies of resilience and institutional change. In this chapter, temporal robustness refers to the ability of civil servants to combine behavioral stability and change across time, however, accentuating their ability to preserve stability by resisting abrupt centrifugal forces stemming from political intrusion, democratic backsliding, and disruptive events (Bauer et  al., 2021; Sørensen & Ansell, 2021). Essentially, however, this chapter does not test the effects of abrupt disruptions, such as the COVID-19 pandemic (OECD, 2021). Robustness-as-stability in public governance implies that significant behavioral patterns among civil servants are established within and across administrative systems, thus maintaining the ability to allocate sustained value over time (Trondal, 2021). The next paragraph outlines the general argument and then suggests four propositions for empirical study: (#1) on stability and change; (#2) on interorganizational vertical specialization; (#3) on intraorganizational vertical specialization; (#4) and on organizational (re-)socialization. The Argument in Brief By suggesting that organizational features are likely to mobilize actors’ attention to some problems, solutions, and consequences while discounting others, Dwight Waldo inferred that a theory of organization is equally a theory of politics (Waldo, 1952). The pursuit of organizing and reorganizing are thus not neutral activities but are consequential (Gulick, 1937; Hammond, 1986; Schattschneider, 1975). Organizational variables are not merely manifestations of symbols and signals (Feldman & March, 1981; Meyer & Rowan, 1977), but are likely to generate systematic bias in human behavior and collective decision-making processes (Egeberg, 2012; Fligstein, 2001; Olsen, 1997; Simon, 1983; Thaler & Sunstein, 2009). In this vein, organization theory is built on theories of decision-­ making (Simon, 1965). Contemporary research in organization theory has been particularly interested in organizational structure (Egeberg & Trondal, 2018). Organizational structures represent normative rules identifying a distribution of tasks among a set of roles (Scott & Davis, 2016). The structure submits how power and accountability should be allocated and is thus likely to shape actors’ behavior by offering them with ‘a systematic and predictable selection of problems, solutions and choice opportunities’ (March & Olsen, 1976: 13). Organizational structures are not

62 

J. TRONDAL

likely to determine the behavior of actors and organizations; they are likely to make some choices more likely than others (Egeberg & Trondal, 2018). Moreover, organizational variables are not assumed to have direct societal impacts but to generate indirect effects by shaping policy processes within governing organizations. Organization theory posits multiple mechanisms as to why actors are likely to comply with organizational rules and routines. Bounded rationality (Simon, 1965) is one such mechanism that connects roles to behavior. In short, organizational structures help to simplify actors’ cognitive search for problems and solutions by nudging their attention toward a minor set of problems and solutions, and ways of linking them. Decision-makers work under three constraints disregarded by the ‘consequentialist theology’ of economic man (March, 2020: 2): incomplete information; imperfect cognitive capacity to assess and process information; and shortage of time to make decisions. Accordingly, actors are likely to select satisfactory alternatives rather than optimal ones, and are to search for solutions that are proximate in time and space (Simon, 1965). The second mechanism—the logic of appropriateness—views human action as driven by internalized perceptions of appropriateness (March & Olsen, 1989). Finally, actors may comply with rules based on their self-interests. Organizations are seen as providers of individual returns and penalties for organizational members (e.g., Ostrom et  al., 2015; Simon, 1983). In sum, an organizational approach conjectures that structural features of administrative systems are likely to influence the behavior of administrative staff over time. This general theoretical argument offers a first general proposition regarding stability and change: #1 Stable organizational structures in the government apparatus are likely to fuel stable contact patterns and participatory behavior among top civil servants over time. Next, three organizational variables are delineated that specify how organizational structure may affect civil servants’ behavior: interorganizational vertical specialization, intraorganizational vertical specialization, and organizational (re)socialization. Based on extant literature (see below), this variable is treated as a core organizational variable in this chapter, and thus serves as the basis for descriptive measurements (see Tables 3.3, 3.4, 3.5, and 3.6). All other independent variables (see below) are included in OLS regression models (Table 3.7).

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

63

I nterorganizational Vertical Specialization The division of tasks and labor between levels of authority in a government apparatus is expressed by a structural division between ministries and subordinate agencies (Pollitt & Talbot, 2004). Vertical specialization anchors relatively autonomous expertise within confined administrative units such as agencies subordinated to ministerial departments, which contributes to balance political loyalty against bureaucratic impartiality among office holders (e.g., Bach et  al., 2015; Egeberg & Trondal, 2018; Lægreid & Verhoest, 2010). Government agencies established at arm’s-length distance from political control are created both to serve as advisory expert institutions and as loyal agents of the government. Such institutions may supply stability in governing processes and function as a buffer preventing shifting political signals from rotating governments. Vertical specialization allows decentered agencies to act shielded from direct political intervention in everyday affairs. This also increases the likelihood that agencies establish strong ties toward institutions outside the government apparatus, such as affected stakeholder groups and EU-level institutions (Egeberg & Trondal, 2009, 2017; Gornitzka & Sverdrup, 2011; Veit et al., 2017; Verhoest, 2017). By contrast, ministerial departments and ministry officials are structurally exposed to the political leadership and likely to become active in interorganizational processes of the government apparatus at large, such as participation in inter-ministerial coordination committees. Structural proximity to executive heads in the ministerial leadership also diminishes potential role of ambiguities and interpretive leeway in task execution (Egeberg & Trondal, 2018). In sum, vertical specialization is conducive to robustness since different vertical levels of the administrative apparatus are receptive to different constituencies and concerns. This leads to the second proposition: #2 Whereas ministry officials are likely to prioritize contact patterns and participatory behavior vertically toward the politico-administrative leadership and horizontally toward a broader spectrum of the administrative apparatus, agency personnel are likely to prioritize similar activities toward external institutions, such as stakeholder groups, EU-level and international institutions. I ntraorganizational Vertical Specialization The division of tasks and labor within levels of authority and within institutions is measured in this chapter by officials’ rank within their own

64 

J. TRONDAL

organization. Previous studies show that top-ranked personnel in government ministries and agencies tend to assign more weight to steering signals from the top compared to low-ranked staff (e.g., Christensen & Lægreid, 2009; Egeberg & Sætren, 1999). Studies also suggest that government personnel in higher echelons of the hierarchy enjoy a horizontally broader attention span than officials in lower positions, and they tend to identify more regularly with the government apparatus as a whole compared to staff located in lower level positions. On top of this, higher ranked government officials tend to be exposed to larger sets of relevant decision information compared to lower level staff. In sum, government officials in higher ranks are more likely to be attentive to broader organizational perceptions and outlooks than lower ranked personnel (Egeberg & Trondal, 2018). By contrast, government staff in lower ranked positions are more likely to be loosely coupled to the government and enjoy local perspectives on responsibilities and tasks. This leads to the third proposition: #3 Whereas officials in higher ranked positions are likely to prioritize contact patterns and participatory behavior vertically toward the politico-­ administrative leadership and horizontally toward a broader spectrum of the administrative apparatus, officials in lower ranked positions are likely to prioritize similar activities toward external institutions, such as stakeholder groups and EU-level and international institutions. Organizational (Re)socialization Time spent in organizations is likely to influence incumbents’ behavior. This argument follows from a broad constructivist and institutionalist research literature in political science arguing that the effect of pre-­ socialization outside institutions is likely to be shaped and filtered by processes of resocialization inside institutions (Checkel, 2005; Trondal, 2001; Trondal et al., 2018). It is argued that socialization are dynamic processes in which organizations acquire some sense of shared values and meaning (or institutional ‘character’) beyond the technical requirements at hand (Selznick, 1958). In short, organizations become institutions, and members of the organizations are induced to some shared rules, norms, and beliefs, which filter out the causal effect of background (demographic) factors such as gender and age (e.g., Peters et al., 2015; Selden, 1997). At the actor level, processes of socialization within organizations relate to the processes in which organizational members are induced into a core set of institutional norms and beliefs that are deeply felt by senses of loyalty and

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

65

long-term commitment (Selznick, 1992). Hence, processes of organizational (‘in-house’) socialization are likely to take time. Moreover, this argument also suggests that behavioral perceptions may occur over time due to enduring exposure to organizational rules and norms, leading to the development of new perceptions of appropriate behavior among the actors (March & Olsen, 1989). It is likewise argued that processes of resocialization are positively associated with the length of exposure to organizations—or the length of service within organizations (Beyers, 2010). This chapter measures resocialization by the number of years that government staff are hired in the central administration. Organizational members with longer tenure are assumed to make different behavioral priorities compared to organizational members with shorter terms in office. This leads to the fourth proposition (Fig. 3.1): #4 Officials with longer terms in office are likely to be more integrated into the administrative services, measured by more frequent contact patterns and participatory behavior vis-à-vis the politico-administrative leadership, as well as horizontally between organizations (ministries and agencies) compared to officials with shorter terms of office.

Interorganizational vertical specialization Intraorganizational vertical specialization

Organizational (re)socialization

Fig. 3.1  The theoretical model

Contact patterns and participatory behavior

66 

J. TRONDAL

Empirical Strategy Studies benefiting from large-N datasets allow for robust probes. This chapter benefits from nine novel large-scale surveys on the Norwegian central administration which includes officials within ministries (N  =  7662) and subordinated agencies (N  =  5511), with a total of 13,173 respondents. Questionnaires have been allocated to government officials from 1976 to 2016, including batteries of similar question over time regarding bureaucrats’ backgrounds, careers, patterns of contact and involvement, issue prioritizations, role perceptions, perceptions of power distribution, reform experiences, trust relationships, and internationalization and Europeanization. Taken together, these surveys have enabled studying sequences of variables across time (Pierson, 2004: 173). This chapter selected a subset of these variables that are often used to study decision-­ making behavior (e.g., Bouckaert et  al., 2010; Christensen & Lægreid, 2008; Egeberg & Trondal, 2018), that is, officials’ patterns of contact and participation in various institutions. This choice is made in order to offer a hard case of robustness-as-stability. Since patterns of contact and participation may arguably change more easily over time than, say, identities and role perceptions, we may consider contact and participation as least likely candidates for robustness. By measuring contacts and participation, we learn which institutions that are shaping discretionary behavior among civil servants, how frequent this happens, and ultimately how stabile it is in terms of structured flexibility. Contacts entail all forms of interaction—oral, written, formal, and informal. The survey asked which contacts civil servants emphasize during everyday decision-making, and what kind of participatory behavior they engage in at national and international level. These include primary structures such as ministries and agencies and secondary structures such as project groups, advisory boards, and committees (see below). By surveying government officials who take part in the daily processes of decision-making, this chapter examines how they perceive the role and power of different government institutions over time. In particular, the chapter examines continuity and change along these variables as well as how statistical variation is associated with the structure of the central administration. Whereas the surveys from 1986, 1996, 2006, and 2016 include individual data files for ministry and agency officials, the 2016 survey in addition includes a merged data file of both ministry and agency officials (N = 4285). The merged file enables careful analyses of the effect of interorganizational vertical specialization (#2), concomitantly, and was used in the OLS regression analyses (Table 3.7). Moreover, ever since the

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

67

first survey was fielded in 1976, the Norwegian central administration has become increasingly integrated into the EU administrative system, centered around the Commission and EU agencies. This has caused governing dilemmas that are hard to handle separately by governing institutions and require processes of multilevel governance, that is, coordination of political authority across scales of authority. Consequently, the surveys fielded since 1996 have included questions related to officials’ contacts toward and participation within a variety of EU administrative bodies. The surveys include respondents from all ministries and all agencies. At the ministerial level they have been circulated to all civil servants at a so-­ called A level, which includes staff having non-clerical portfolios and who have at least 1 year in office. The sample thus includes the full population of ‘A-level’ ministerial officials. By contrast, the agency surveys have been allocated to a random selection of every third official at the ‘A level’ who have a minimum of 1 year in office. Due to the huge number of potential respondents at the agency level, random selection was largely a pragmatic choice. Survey technologies have fundamentally shifted during these surveys— from the first survey in 1976 to the final survey in 2016. Whereas the studies in 1986 and 1996 approached respondents by physical post, the Norwegian Centre for Research Data (NSD) has been responsible for fielding the 2006 and 2016 surveys online. Yet, notwithstanding both changing survey technology as well as increase in survey exhaustion among respondents, response rates in these surveys have declined only slightly over the years (see Table 3.1). However, a minor reduction of responses from 1996 onwards may still partly reflect shifting survey technologies and rising survey exhaustion among our respondents. Since both surveys in 1986 and 1996 used similar postal survey technologies, the drop in the response rates may indeed reflect such fatigue. Table 3.1  Sample size and response rates in the ministry and agency surveys, 1976, 1986, 1996, 2006, and 2016 Ministry

1976 1986 1996 2006 2016 Total

Agency

Responses

%

Responses

%

784 1185 1497 1874 2322 7662

72 72 72 67 60

– 1072 1024 1452 1963 5511

– 68 64 59 60

68 

J. TRONDAL

One caveat is warranted: Empirical observations are based on the perceptions of the respondents. The chapter thereby risks that behavioral perceptions do not reflect actual behavior and/or are subject to perceptual errors. Admittedly, there is no guarantee that the perceptions of civil servants directly materialize in actors’ behavior and organizational decisions. However, perceptions serve as frames for behavior, rending it more likely than not that decision-making patterns are associated with perceptional patterns (Aberbach et al., 1981: 86). There are, however, ways of reducing such risks. One way is to use multiple observations and from different groups of respondents who respond to comparable questions over time. This is likely to render the conclusions more robust and less subject to random distributions and methodological errors. Considering the challenges of self-reporting and respondents overrating the importance of certain institutions, the chapter emphasizes variation in proportions across multiple variables.

Empirical Demonstration This section examines three behavioral proxies over time: first, civil servants’ contact patterns toward a variety of institutions involved in the policy process (Tables 3.3 and 3.4), contacts toward international and EU-level institutions (Table  3.5), and participations in secondary structures within the government apparatus (Table 3.6). Finally, based on the merged 2016 datafile, an OLS regression model probes the controlled effects of interorganizational vertical specialization (#2), intraorganizational vertical specialization (#3), and organizational (re-)socialization (#4) (Table 3.7). Despite our surveys being fielded before, during, and after the New Public Management (NPM) reform wave, the organizational structure of the government apparatus has remained profoundly stable and robust over time, notably as regards the key organizational principles of vertical and horizontal specialization. This finding supports recent observations on the stability and robustness of the organizational architectures of government institutions (Bertels & Schulze-Gabrechten, 2021). Essential to our argument, organizational stability is conducive to behavioral continuity (#1). However, Table 3.2 also shows a steady growth in organizational capacities measured by an increased number of staff, providing the government with increased administrative capacities for public governance. Robustness, however, is also a capacity to modify practices without fundamentally transforming organizational structures (see below).

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

69

Table 3.2  Numbers of officials in ministries and agencies, over time (absolute numbers)a

Ministries Agencies Total

1976

1986

1996

2006

2016

2812 – 2812

3491 – 3491

3945 9182 13,127

4350 11,040 15,390

4752 15,359 20,111

Comparable numbers are missing for agencies in 1976 and 1986 (Christensen et al., 2018: 27)

a

Long-Term Contact Patterns First, Tables 3.3 and 3.4 report long-term patterns of contact among ministerial and agency officials, respectively. These include contacts within the central administration, toward the political leadership, but also toward subnational administrations, interest groups, and business organizations. In sum, Tables 3.3 and 3.4 suggest that key contact patterns are associated with organizational boundaries within the central administration and thus that decision-making behavior, as measured, reflects the organizational architecture of the government apparatus. A second observation is profound stability in patterns of contact across 40 years (#1). Despite facing a range of stressors in this period, these data demonstrate how long-term contact patterns are characterized by overall stability, in which essential elements of public governance remain robust. By zooming in on variation in these observations, ministerial departments are consistently better anchored into the political leadership than agencies across time. Second, ministerial officials report consistently more horizontal anchoring within own institution than vis-à-vis other parts of the central administration, notably other ministries as well as agencies subordinated to the ministry. Contacts toward parliamentary institutions are also less frequent, reflecting that these relationships remain a responsibility for the ministerial political leadership. In essence, Tables 3.3 and 3.4 demonstrate profound stability over time (#1). Despite an increased call for structural and policy reforms of government systems, and an increased internationalization of the civil service (see below), government officials report profound continuity in their behavioral patterns. Yet, structured flexibility is also illustrated by behavioral change. The most profound change observed in our data is less frequent contact between ministerial officials and regional and local administration over time—notably toward regional and local administrative bodies in their own constituency. We also observe a decline in ministerial officials’

70 

J. TRONDAL

Table 3.3  Proportion of ministry officials reporting contacts toward the following domestic institutions and bodies during the last year (percent)a 1976 1986 1996 2006 2016 Intra-governmental contacts The minister 41 45 State secretary/political advisors 49 51 Administrative leadership within own ministry (general 90 87 secretary, heads of unit) Other units within own ministry 74 87 Other ministries 78 78 The Parliament and its internal bodiesb 12 10 Own subordinate agencies and public enterprises 38 73 Agencies and governmental bodies beneath other 21 20 ministries Interorganizational contacts Regional and local governmental bodies beneath own 38 55 ministry Regional and local administration 35 28 Labor and business organizations 42 28 Other organizations – –

33 44 77

34 49 82

33 51 76

81 68 6 56 17

85 68 4 61 18

86 69 10 61 17

33

23

17

22 19 23

9 16 19

9 26 23

Civil servants reporting contacts ‘almost every month,’ or more often. The original value scale on the variable was as follows: almost every week (value 1), almost every month (value 2), a few times (value 3), never (value 4), not relevant (value 5) a

This includes contacts toward the Parliament and its internal units: Parliamentary committees within own policy area, individual Parliamentary representatives, and Parliamentary party groups b

contacts toward their own political and administrative leadership. This does, however, not suggest a weakening of political steering or a depoliticization of the government apparatus since we are seeing a rising number of political advisors around the ministers during the same time period, relieving the politico-administrative leadership from direct interaction with ministerial officials (Krick et al., 2019; Veit et al., 2017). Moreover, agency officials have less frequent interaction with agencies other than their own and more frequent contacts toward business organizations. Supporting #2, high frequency of business contacts among agency officials reflects the agencies’ decentered role and distance from the political leadership, making agency officials biased toward stakeholder groups. Reflecting the organizational specialization of the central administration, patterns of contacts vis-à-vis stakeholder groups are also biased by policy portfolios: Ministerial officials tend to prioritize outward contacts that match their own task portfolios by enjoying more intensive contacts toward agencies, regional and/or local administration, and stakeholders

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

71

Table 3.4  Proportion of agency officials reporting contacts toward the following domestic institutions and bodies during the last year (percent)a 1976 1986 1996 2006 2016 Intra-governmental contacts Agency board – 44 The top management of own agency – 45 Other units within own agency – 78 The political leadership in parent ministry – 2 Civil servants in parent ministry – 25 Other ministries (political leadership and/or civil – 11 servants) Other agencies – 18 Own external governmental bodies (if such exists) – 84 Other external governmental bodies – 14 Inter-governmental contacts The parliament (and its internal bodies) – 11 Regional and/or local administration – 13 Labor organizations – 19 Business organizations – 14 Other organizations – –

52 42 69 3 32 11

6 42 67 4 31 9

– 43 67 3 28 8

19 69 16

24 40 14

33 33 13

0 17 14 13 44

4 16 9 11 35

16 16 14 15 35

Civil servants reporting contacts ‘almost every month,’ or more often. The original value scale on the variable was as follows: Almost every week (value 1), almost every month (value 2), a few times (value 3), never (value 4), not relevant (value 5) a

that are compatible to their portfolios. Similar observations are seen at agency level. Moreover, Tables 3.3 and 3.4 establish that ministerial and agency officials have much less frequent contact toward regional and local administration—that is, toward regional and local bodies subordinate to own ministry or agency, including the country governor’s office. Reflecting a quest for order and ‘better coordination’ of the government apparatus, our finding shows a stronger integration of the central government than an integration of national and subnational administrative institutions (Egeberg & Trondal, 2016). Long-Term International Contacts Norway is characterized as an associated EU member state without politically being a member. There are 101 living treaties in force between Norway and the EU. As a consequence, more than 14,000 legislative acts originating from the EU have been incorporated into Norwegian law since 1994. Figures 10.1 and 10.2 (in Chap. 10) illustrate the structured flexibility of these agreements, which leads to a continuous transposition

72 

J. TRONDAL

of EU legislative acts into Norwegian law. It also reflects incomplete contracting in which the original EEA agreement gradually has become incomplete vis-à-vis the EU legal development and subsequently been accompanied by new agreements with the EU, leading to dynamic updating of the set of contracts—or treaties—over time. In effect, legal transposition increasingly occupies the working hours of Norwegian civil servants. Table 3.5 examines patterns of contacts among ministerial and agency officials vis-à-vis EU and international institutions over time. Despite the dynamic character of Norway’s relationship with the EU, Table 3.5 establishes profound continuity in civil servants’ general international contact patterns across time. However, reflecting the dynamic character of Norway’s relationship with the EU, observations made in Table 3.5 mask an increase in the absolute numbers of officials who are in regular contact with EU institutions. Two supplementary observations are noticeable: First, both ministerial and agency officials report fairly similar contact Table 3.5  Proportion of ministry and agency officials reporting contacts toward the following international and EU-level institutions during the last year (percent)a 1976 1986 1996 2006 2016 Ministerial officials Norway’s delegation to the EU – The European Commission – The (Union) Council – The European Parliament – EU agencies – Nordic institutions (e.g., Nordic Council of Ministers, 7 Nordic Council) Other international organizations 11 Governments in other countries 12 Agency officials Norway’s delegation to the EU – The European Commission – The (Union) Council – The European Parliament – EU agencies – Nordic institutions (e.g., Nordic Council of Ministers, – Nordic Council) Other international organizations – Governments in other countries –

– – – – – –

16 7 – – – 9

18 8 – – 2 7

16 7 1 1 2 7

22 22

12 7

14 14

18 22

– – – – – –

1 2 – – – 7

2 4 – – 4 8

2 4 0 2 6 5

11 8

7 4

9 4

7 12

Civil servants reporting contacts ‘almost every month,’ or more often. The original value scale on the variable was as follows: almost every week (value 1), almost every month (value 2), a few times (value 3), never (value 4), not relevant (value 5) a

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

73

patterns, most importantly toward governmental institutions in other countries and international institutions. The European Commission and Nordic institutions are perceived as important, yet less than other international institutions. This is most notable at ministerial level (#2). One significant difference is that national agency officials prioritize contact toward policy-­relevant EU-level agencies compared to ministry officials (Egeberg & Trondal, 2009). Ministry staff enjoy a slightly more politicized international contact profile, inter alia focusing on contacts toward the Norwegian EU delegation. Continuity and Change of Secondary Structures Government officials participate in multiple organizational affiliations that mobilize varieties of agendas, commitments, and loyalties. These affiliations in sum provide structured flexibility for government officials and thus avenues for robust governance. A ‘primary affiliation’ is characterized as a structure that represents a primary guide to organizational members—such as ministries and agencies. In addition to the primary structure of government, central administrations also harbor a vast set of secondary structures such as collegial bodies, committee systems, projects and working groups that involve civil servants’ part-time as well as external stakeholder groups (Self, 1972: 105). A ‘secondary affiliation’ is characterized as a structure that represents a secondary guide to organizational members (Egeberg, 2012). These structures are often established to reconcile administrative conflicts as well as increase coordination and integration between often loosely coupled primary structures. Such network structures tend to be established in the ‘shadow of hierarchy’ (Hèritier & Lehmkuhl, 2008), aimed to support activities in the primary structure. They might facilitate mutual learning and socialization among primary structures (March & Olsen, 1989). Secondary structures can be temporary or semi-permanent. Moreover, they can be internal to the central administration, that is, involving only ministry officials, or external and invite participation across ministries, agencies, levels of governance, as well as involving external stakeholder groups. Table 3.6 establishes that secondary structures such as working groups and project groups that are internal to the government apparatus are common both at ministerial level and at agency level, yet most frequent at the agency level. This is a robust organizational form with a fairly long history. At ministerial level, by contrast, secondary structures became particularly important during the 1980s. Still, 60 percent of ministerial staff attended collegial structures also during the 1970s.

74 

J. TRONDAL

Table 3.6  Proportion of ministry and agency staff participating in the following during the last year (percent) 1976a 1986a 1996a 2006a 2016b Ministerial officials Collegial bodies within own ministry Collegial bodies across ministries Collegial bodies with subordinated agencies Government advisory groups/bodies Nordic committee, working groups, etc. Committee, working group in international organizations Agency officials Collegial bodies within own ministry Collegial bodies across ministries Collegial bodies with subordinated agencies Government advisory groups/bodies Nordic committee, working groups, etc. Committee, working group in international organizations

57 40 – – 36b 37b

71 53 – 29 61b 61b

79 58 42 19 37b 41b

75 54 40 6 – –

79 63 50 12 15 26

– – – – – –

82 – 34 14 28b 20b

80 20 33 9 29b 29b

8 20 36 5 30b 35b

87 30 29 6 20 27

Civil servants reporting ‘Yes’ Civil servants reporting ‘Yes, a few times, or more often.’ The original value scale is: ‘Yes, several times’ (value 1), ‘Yes, one time’ (value 2), ‘never’ (value 3) a

b

Table 3.6 reveals two additional observations. First, we observe profound continuity in the role played by secondary structures across time (#1), particularly at the agency level (#2); second, we observe that participation in secondary structures reflects the primary structuring of the central administration in two ways. First, Table  3.6 reveals continuity over time as regards to participation in working groups and project groups in both international and Nordic institutions. Agency staff participate in international collegial bodies at a similar frequency as in national ones— notably with parent ministry, other agencies, as well as with subnational bodies. Moreover, ministerial officials report increased participation in working groups and project groups, both internally to the ministry, across ministries and toward subordinate agencies. Second, Table  3.6 suggests that participation in secondary structures is systematically shaped by primary structures, notably the organizational embedment of staff (#3). In effect, secondary structures tend to support primary structures since participation in collegial secondary structures is most evident within own institutions than across institutions, and more extensive within the central administration than internationally. Horizontally, this pattern is substantiated both by more extensive participation within own ministry and agency

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

75

than between ministries and subordinate agencies (for ministry staff), as well as toward parent ministry (for agency staff). Participation in collegial structures across ministries and agencies is thus much less developed than participation in intraorganizational groups, and such structures are more extensively used at ministerial than at agency level. While we see a growth in the use of secondary structures at ministerial level, no parallel growth is observed at agency level. Moreover, participation in working groups and project groups across levels of governance is less frequent than horizontally within each level of governance, particularly among ministerial officials. A final observation is that ministerial and agency officials tend to reduce their involvement in ad hoc advisory commission, typically established by the Cabinet (see Christensen and Hesstvedt (2018) on the expertization of government advisory commissions). Finally, to estimate controlled effects, four OLS regression models are presented using the merged 2016 dataset that includes both ministry and agency officials (N = 4285). Four dependent variables were selected: civil servants’ contact toward the leadership within ‘own’ ministry/agency (model 1), contact toward other units within ‘own’ ministry/agency (model 2), participation in collegial bodies within ‘own’ ministry/agency (model 3), and participation in collegial bodies between ­ministries/agencies (model 4). A general finding is that organizational structure significantly shapes the patterns of contact and participation among civil servants. Moreover, officials in ministerial departments (#2) and in higher ranks (#3) tend to have significantly more contact toward the administrative leadership in their ‘own’ ministry/agency and other units in their ‘own’ ministry/agency compared to officials at the agency level and in lower ranks. One implication of these findings is that senior ministerial officials are more likely to ‘have a common mind with the minister’ (Self, 1972: 165) and co-govern compared to lower ranked agency officials. In addition, higher ranked officials tend to participate more in secondary collegial structures—both inter and intraorganizationally—compared to lower ranked staff (#3). However, whereas agency officials are more likely to participate in collegial bodies within their own organization, ministerial officials are more likely to prioritize inter-ministerial participation. Finally, whereas organizational embedment and rank leave significant explained variance, organizational (re-)socialization (#4) as measured by the seniority of staff has significant, but weaker effects than the two other explanatory variables. The finding moreover supports extant literature suggesting that seniority is a profoundly weaker predictor of role obedience among civil servants than organizational structure (see Egeberg & Stigen, 2021) (Table 3.7).

.26**

.19** .06* 0.092

.35**

.31** .08* 0.122

Contacts toward other units in own ministry/agency

Contacts toward top leadership in own ministry/agency

0.082

.07*

.11**

–.10**

Participation in collegial bodies within own ministry/agency

Model 3

0.098

.05*

.17**

.35**

Participation in collegial bodies between ministries/agencies

Model 4

b

Coding of the independent variables: Vertical interorganizational specialization: ministry (value 1), agency (value 0); vertical intraorganizational specialization (director general or higher levels/adviser/director or equivalent (value 1), deputy director general (value 2), assistant director general/adviser (value 3), principal officer/adviser (value 4), executive officer, higher executive officer/adviser (value 5); organizational affiliation (seniority): original natural continuous variable

a

Coding of the dependent variables: The original value scale in the survey of variables 1 and 2: almost every week (value 1), almost every month (value 2), a few times (value 3), never (value 4), not relevant (value 5). The original value scale in the survey of variables 3 and 4: ‘Yes, several times’ (value 1), ‘Yes, one time’ (value 2), ‘never’ (value 3)

* p ≤ 0.05, **p ≤ 0.01

(#1) Interorganizational vertical specialization (ministry/agency) (#2) Intraorganizational vertical specialization (rank) (#3) Organizational (re-) socialization (seniority) R Square

Model 2

Model 1

Table 3.7  Summary of factors affecting ministry and agency officials’ patterns of contact and participation (standardized beta coefficients; linear regressions on ministry and agency officials; 2016 data)a, b

76  J. TRONDAL

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

77

Conclusions This chapter confirms Self’s (1972: 151) seminal argument that an important role of public administration is to add stability and routine to the fluidity and ambiguity that characterize democratic governance. Studying public governance across time is crucial in order to understand the conditions for robust political order (Olsen, 2018) and the long-term provision of public value (Dahlstrom & Lapuente, 2017; Douglas et al., 2021). This chapter carries one distinct contribution to the volume. It demonstrates how the organizational architecture of government represents a stable and systemic capacity for public governance across time. The chapter establishes how stability serves as an enduring feature of public governance and how this is anchored in the organizational architecture of government systems. Patterns of contact and participation among government civil servants are shaped and biased by the organizational affiliation of civil servants within the central administration. Moreover, observing that contact patterns and participatory behavior among government officials are unaffected by these shocks, suggests that government officials remain unaffected because of isolated structural clusters of the government apparatus. Behavioral processes are profoundly driven by fairly stable routines embedded in formal rules. Moreover, structured flexibility is illustrated by how the civil service has adapted to both international organizations and societal stakeholders. Second, the theoretical argument and empirical findings are probed by a novel large-scale and longitudinal dataset that spans four decades. These novel data encompass five observation-points, enabling a long-term perspective on government civil servants over nearly half a century, allowing a comprehensive study of the organizational basis for robust public governance. One theoretical implication of these findings is that organizational structure (#2, #3) represents a stronger predictor of robust governance than processes of organizational socialization (#4). One subsequent implication relates to design thinking in public policy and administration, in which the findings of this chapter establish that the robustness of public governance may be subject to deliberate intervention through organizational engineering (Egeberg & Trondal, 2018). One additional empirical implication is that robust coordination of governance processes within the central administration is primarily anchored among ministerial officials who are tied close to the politico-administrative leadership. These observed patterns differ according with the actors’ organizational belonging, which

78 

J. TRONDAL

as such remain profoundly stable across time. Stabile organizational architectures are thus conducive to robust behavioral patterns over fairly long periods of time. Taken together, these observations establish the organizational foundation for robust public governance in which administrative systems harbor some fundamental stable organizational capacities. Moreover, this chapter shows that collegial (secondary) structures largely support hierarchical (primary) structures over time, and thus supply two complementary organizational capacities for robustness in the civil service. Finally, the chapter establishes how varieties of organizational structures in the government apparatus fuel stable tensions within the central administration. The chapter features three tensions in the central administration that together illuminates structured flexibility: A first tension is observed between the open and inclusive versus a sealed and disclosed central administration, the second is between the primary and secondary structures of the administrative apparatus, while the third is between the central administration as genuinely national and its role as part of an EU multilevel administrative system. In sum, these tensions suggest structured flexibility by allowing the civil service to act flexibly but systematically over time by adapting to internal and external institutions, which includes societal stakeholders and international organizations. Organizational specialization of the civil service represents stable checks and balances inherent in the administrative system. As a result, the chapter establishes how such structural stability serves as a robust character in the central administration.

References Aberbach, J., Putnam, R. D., & Rockman, B. (1981). Bureaucrats and politicians in Western democracies. Harvard University Press. Adam, C., Hurka, S., Knill, C., & Steinebach, Y. (2021). Policy accumulation and the democratic responsiveness trap. Cambridge University Press. Alvesson, M., & Spicer, A. (2019). Neo-institutional theory and organization studies: A mid-life crisis? Organization Studies, 40(2), 199–218. Ansell, C. (2011). Pragmatist democracy. Oxford University Press. Ansell, C., & Trondal, J. (2017). Governing turbulence. An organizational-­ institutional agenda. Perspectives on Public Management and Governance, 1(1), 43–57. Ansell, C., Trondal, J., & Ogard, M. (Eds.). (2017). Governance in turbulent times. Oxford University Press.

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

79

Aucoin, P. (1990). Administrative reform in public administration: Paradigms, principles paradoxes and pendulums. Governance, 3(2), 115–137. Bach, T., Ruffing, E., & Yesilkagit, K. (2015). The differential empowering effects of Europeanization on the autonomy of national agencies. Governance, 28(3), 285–304. Bauer, M.  W., Peters, B.  G., Pierre, J., Yesilkagit, K., & Becker, S. (Eds.). (2021). Democratic backsliding and public administration. Cambridge University Press. Benz, A., Broschek, J., & Lederer, M. (Eds.). (2021). A research agenda for multilevel governance. Edward Elgar. Bertels, J., & Schulze-Gabrechten, L. (2021). Mapping the black box of intraministerial organization: An analytical approach to explore structural diversity below the portfolio level. Governance, 34(1), 171–189. Bevir, M. (Ed.). (2010). The SAGE handbook of governance. SAGE. Beyers, J. (2010). Conceptual and methodological challenges in the study of European Socialization. Journal of European Public Policy, 17(6), 909–920. Bouckaert, G., Peters, B. G., & Verhoest, K. (2010). The coordination of public sector organizations. Palgrave Macmillan. Brunsson, N., & Olsen, J. P. (1993). The reforming organization. Fagbokforlaget. Checkel, J.  T. (2005). International institutions and socialization in Europe. Introduction and framework. International Organization, 59, 801–826. Christensen, J., & Hesstvedt, S. (2018). Expertisation or greater representation? Evidence from Norwegian advisory commissions. European Politics and Society, 20(1), 3–100. Christensen, T., & Lægreid, P. (Eds.). (2006). Autonomy and regulation. Edward Elgar. Christensen, T., & Lægreid, P. (Eds.). (2007). Transcending new public management. Routledge. Christensen, T., & Lægreid, P. (2008). The challenge of coordination in central government organizations: The Norwegian case. Public Organization Review, 8(2), 97–116. Christensen, T., & Lægreid, P. (2009). Living in the past? Change and continuity in the Norwegian central civil service. Public Administration Review, 69(5), 951–961. Christensen, T., Egeberg, M., Lægreid, P., & Trondal, J. (2018). Sentralforvaltningen. Stabilitet og endring gjennom 40 år. Universitetsforlaget. Cyert, R., & March, J. G. (1963). A behavioral theory of the firm. Wiley-Blackwell. Dahlstrom, C., & Lapuente, V. (2017). Organizing leviathan. Oxford University Press. Dibben, P., Wood, G., & Rober, I. (Eds.). (2004). Contesting public sector reform. Palgrave Macmillan.

80 

J. TRONDAL

Douglas, S., Schillemans, T., ’t Hart, P., Ansell, C., Andersen, L. B., Flinders, M., Head, B., Moynihan, D., Nabatchi, T., O’Flynn, J., Peters, B. G., Raadschelders, J., Sancino, A., Sørensen, E., & Torfing, J. (2021). Rising to Ostrom’s challenge: an invitation to walk the bright side of public governance and public service. Policy Design and Practice, 4(4), 441–451. Easton, D. (1965). A systems analysis of political life. Wiley. Egeberg, M. (2012). How bureaucratic structure matters: An organizational perspective. In B.  G. Peters & J.  Pierre (Eds.), The SAGE handbook of public administration. SAGE. Egeberg, M., & Sætren, H. (1999). Identities in complex organizations: A study of ministerial bureaucrats. In M.  Egeberg & P.  Lægreid (Eds.), Organizing Political Institutions. Scandinavian University Press. Egeberg, M., & Stigen, I. M. (2021). Explaining government bureaucrats’ behavior: On the relative importance of organizational position, demographic background, and political attitudes. Public Policy and Administration, 36(1), 3–18. Egeberg, M., & Trondal, J. (2009). Political leadership and bureaucratic autonomy. Effects of agencification. Governance, 22(4), 673–688. Egeberg, M., & Trondal, J. (2016). Why strong coordination at one level of government is incompatible with strong coordination across levels (and how to live with it). The case of the European Union. Public Administration, 94(3), 579–592. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies. What have we learnt (and where do we go from here). Journal of Common Market studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Emery, Y., & Giauque, D. (2014). The hybrid universe of public administration in the 21st century. International Review of Administrative Sciences, 80(1), 23–32. Farjoun, M. (2010). Beyond dualism: Stability and change as a duality. Academy of Management Review, 35, 202–225. Feldman, M. S., & March, J. G. (1981). Information in organizations as signal and symbol. Administrative Science Quarterly, 26(2), 171–186. Fligstein, N. (2001). Organizations: Theoretical debates and the scope of organizational theory. Manuscript, UC Berkeley. Frederickson, H. G., Smith, K. B., Larimer, C. W., & Licari, M. J. (2012). The public administration theory primer. Westview Press. Goodin, R. E. (1996). The theory of institutional design. Cambridge University Press. Gornitzka, Å., & Sverdrup, U. (2011). Access of experts: Information and EU decision-making. West European Politics, 34(1), 48–70. Greve, C., Lægreid, P., & Rykkja, L. (2018). Nordic bureaucracy beyond new public management. In Bureaucracy and society in transition (Comparative social research, vol. 33) (pp. 205–224). Emerald.

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

81

Gulick, L. (1937). Notes on the theory of organization. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration. Hammond, T. H. (1986). Agenda control, organizational structure, and bureaucratic politics. American Journal of Political Science, 30(1), 379–420. Hèritier, A., & Lehmkuhl, D. (2008). Introduction: The shadow of hierarchy and new modes of governance. Journal of Public Policy, 28(1), 1–17. Hood, C., & Jackson, M. (1991). Administrative argument. Dartmouth. Howlett, M., & Goetz, K. H. (2014). Introduction: time, temporality and timescapes in administration and policy. International Review of Administrative Sciences, 80(3), 477–492. Kettl, D.  F. (2002). The transformation of governance. Johns Hopkins University Press. Krick, E., Christensen, J., & Holst, C. (2019). Between “scientisation” and a “participatory turn”. Tracing shifts in the governance of policy advice. Science and Public Policy, 46(6), 927–939. Kuipers, S., Yesilkagit, K., & Carroll, B. (2018). Coming to terms with termination of public organizations. Public Organization Review, 18(2), 263–278. Lægreid, P., & Verhoest, K. (Eds.). (2010). Governance of public sector organizations. Palgrave Macmillan. Levi-Faur, D. (Ed.). (2012). The Oxford handbook of governance. Oxford University Press. March, J.  G. (1991). Exploration and exploitation in organizational learning. Organization Science, 2, 71–87. March, J. G. (2020). A scholar’s quest. Journal of Management Inquiry, 2–3. March, J.  G., & Olsen, J.  P. (1976). Ambiguity and choice in organizations. Scandinavian University Press. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions. The Free Press. Mazzucato, M., Kattel, R., Quaggiotto, G., & Begovic, M. (2021). COVID-19 and the need for dynamic state capabilities: An international comparison. UCL Development Futures Series Working Papers. McCowan, M. (2016). Turbulent times for ministries of finance. In C.  Ansell, J.  Trondal, & M.  Orgard (Eds.), Governance in turbulent times. Oxford University Press. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations. Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Meyer, M. W. (1979). Change in public administration. Cambridge University Press. Moloney, K., & Stone, D. (2019). Beyond the state: Global policy and transnational administration. International Review of Public Policy, 1(1), 104–118. OECD. (2021). Government at a glance. OECD. O’Leary, R., Van Slyke, D.  M., & Kim, S. (Eds.). (2010). The future of public administration around the world. Georgetown University Press.

82 

J. TRONDAL

Olsen, J. P. (1997). Institutional design in democratic contexts. Journal of Political Philosophy, 5(3), 203–229. Olsen, J. P. (2006). Maybe it is time to rediscover bureaucracy. Journal of Public Administration Research and Theory, 16(1), 1–24. Olsen, J. P. (2009). Change and continuity: an institutional approach to institutions of democratic government. European Political Science Review, 1(1), 3–32. Olsen, J. P. (2010). Governing through institution building. Oxford University Press. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford University Press. Olsen, J. P. (2018). Democratic accountability and the changing European political order. European Law Journal, 24(1), 77–98. Osborne, D., & Gaebler, T. (1992). Reinventing government. Addison-Wesley. Ostrom, E., Ostrom, V., Aligica, P.  D., & Sabetti, P. (2015). Choice, rules and collective action. ECPR Press. Peters, B. G. (2018). Policy problems and policy design. Edvard Elgar. Peters, B. G., von Maravic, P., & Schroter, E. (Eds.). (2015). Politics of representative bureaucracy. Edward Elgar. Pierson, P. (2004). Politics in time. Princeton University Press. Pollitt, C. (2008). Time, policy, management. Oxford University Press. Pollitt, C. (2011). Not odious bot onerous: Comparative public administration. Public Administration, 89(1), 114–127. Pollitt, C., & Bouckaert, G. (2017). Public management reform. Oxford University Press. Pollitt, C., & Talbot, C. (Eds.). (2004). Unbundled government. Routledge. Rainey, H. G., & Steinbauer, P. (1999). Galloping elephants: Developing elements of a theory of effective government organizations. Journal of Public Administration Research and Theory, 9(1), 1–32. Riddervold, M., Trondal, J., & Newsome, A. (Eds.). (2021). The Palgrave handbook of EU crises. Palgrave Macmillan. Rothstein, B. (2012). Political legitimacy for public administration. In B. G. Peters & J. Pierre (Eds.), The SAGE handbook of public administration. SAGE. Schattschneider, E. E. (1975). The Semisovereign people. Harcourt Brace Jovanovich College Publishers. Scott, W. R., & Davis, G. F. (2016). Organizations and organizing. Routledge. Selden, S. C. (1997). The promise of representative bureaucracy. M. E. Sharpe. Self, P. (1972). Administrative theories and politics. George Allen & Unwin. Selznick, P. (1958). Leadership in administration. University of California Press. Selznick, P. (1992). The moral commonwealth. University of California Press. Simon, H. A. (1965). Administrative behavior. The Free Press. Simon, H. A. (1983). Reason in human affairs. Stanford University Press. Sørensen, E., & Ansell, C. (2021). Toward a concept of political robustness. Political Studies. https://doi.org/10.1177/00323221721999974

3  EVOLVING TENSIONS IN PUBLIC GOVERNANCE 

83

Thaler, R. H., & Sunstein, C. R. (2009). Nudge. Improving decisions about health, wealth and happiness. Penguin Books. Trondal, J. (2001). Is there any social constructivist—Institutionalist divide? Unpacking social mechanisms affecting Representational roles among EU decision-­makers. Journal of European Public Policy, 8(1), 1–23. Trondal, J. (2021). Public administration sustainability and its organizational basis. International Review of Administrative Sciences, 87(2), 399–415. Trondal, J., Murdoch, Z., & Geys, B. (2018). How pre- and postrecruitment factors shape role perceptions of European Commission officials. Governance, 31(1), 85–101. Veit, S., Hustedt, T., & Bach, T. (2017). Dynamics of change in internal policy advisory systems: the hybridization of advisory capacities in Germany. Policy Sciences, 50(1), 85–103. Verhoest, K. (2017). Agencification in Europe. In E.  Ongaro & S.  Van Thiel (Eds.), The Palgrave handbook on public administration and management in Europe. Palgrave Macmillan. Waldo, D. (1952). Development of theory of democratic administration. American Political Science Review, 46(1), 81–103. Weick, K. E., & Sutcliffe, K. M. (2007). Managing the unexpected. Wiley. White, L. (2022). The de-institutionalisation of power beyond the state. European Journal of International Relations, 28(1), 187–208. Wynen, J., Kleizen, B., & Verhoest, K. (2019). Are public organizations suffering from repetitive change injury? A panel study of the damaging effect of intense reform sequences. Governance, 32(4), 695–713.

CHAPTER 4

Public Administration Sustainability and Its Organizational Basis

Whereas a vast body of scholarly literature suggests that public sector organizations are profoundly unstable in the long run (e.g., Ansell & Torfing, 2014; Ansell & Trondal, 2017), few studies offer longitudinal data. Following Chap. 3, this chapter also shows how public governance processes are profoundly sustainable. Whereas most studies of public governance rely on cross-sectional datasets, this chapter benefits from a novel dataset that spans 40 years and is thus able to probe the sustainability of public governance by using a longer timeframe. Theoretically, it is argued that public governance profoundly rests on its organizational fabric. The chapter thus adds to an organizational approach to public governance studies. These twin contributions are important since times of administrative reform and turbulence increasingly question the sustainability of public sector organizations and reliable public service delivery (e.g., Alvesson & Spicer, 2019; Ansell et al., 2017; Olsen, 2017, 2018). Periods of political and economic turbulence call for studying conditions for sustainable public service provision. However, political science and public administration harbor competing ideas on the robustness of governments and public organizations (Bouilloud et al., 2020; Fligstein, 2001). A vast literature suggests that public organizations are particularly responsive to its task environment (e.g., Meyer et  al., 1997; Selznick, 2015; Tolbert & Zucker, 1983). This argument advocates that public organizations are profoundly embedded into their adjacent environments © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_4

85

86 

J. TRONDAL

(Fligstein, 2001; Meyer & Rowan, 1977; Pfeffer & Salancik, 1978) and that they have to continuously reform if they wish to survive (Selznick, 1958). A pragmatist literature also suggests a middle ground between stability and responsiveness in which organizations both defend core values at the same time as they permanently adapt to ongoing problems they confront (Ansell et al., 2015; Bouilloud et al., 2020). Despite critical literature arguing that environmental demands are often diffuse or at times inconsistent (Kraatz, 2015), a general empirical forecast from this literature has been that public administration tends to adapt effectively to its relevant task environment, and consequently that public governance processes—and public administration at large—are unsustainable over longer courses of time. This chapter, by contrast, suggests that public administration harbors profound sustainability and that sustainable public governance rests on its organizational fabric (e.g., Mahoney & Thelen, 2010; March & Olsen, 1989; Streek & Thelen, 2005). Two contributions are thus made: –– Theoretically, the chapter applies an organizational approach to the study of administrative sustainability. Three reasons motivate this choice: First, an organizational approach does not figure prominently in the sustainability literature (e.g., Bevir, 2010; Levi-Faur, 2012). Secondly, organizational factors arguably make a systematic and patterned impact on decision-making behavior, and thus are able to account for governance sustainability (see Christensen & Lægreid, 2007; Egeberg & Trondal, 2018; Olsen, 2006; Simon, 1983: 22). Finally, organizational structure, compared to other factors that might intervene in the policy process, may be a design instrument of the context of choice in public governance (Egeberg & Trondal, 2018; Thaler & Sunstein, 2009). As such, the chapter also offers an avenue to build bridges between the academic and the practitioner worlds of public administration (O’Leary et al., 2010: 292). An organizational approach should thus also serve as an action program for political science, making it practically relevant for solving societal problems (see Egeberg & Trondal, 2018). –– Empirically and methodologically, the chapter offers a novel large-­N (13,173) and longitudinal dataset that spans 40 years as presented in Chap. 3. Taken together, the chapter demonstrates both sustainability in public governance and probes its organizational basis.

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

87

The time period studied is signified by profound change in the task environment of the civil service, such as the ups and downs of competing administrative reforms (e.g., Klenk & Reiter, 2019; Pollitt & Bouckaert, 2017), rival calls for making the administrative state smarter (e.g., Aucoin, 1990; Christensen & Lægreid, 2007; Emery & Giauque, 2014; Kettl, 2002), as well as competing and shifting prescriptions for organizational design in the public sector (Bulmer & Joseph, 2016; Olsen, 2017). Nevertheless, this chapter argues and empirically suggests that public administration is foremost characterized by profound sustainability in which key parameters of public governance processes remain profoundly sustainable (see also Christensen & Lægreid, 2009; Pollitt, 2008). The chapter proceeds as follows: The next section outlines an organizational theory approach to sustainable public governance. Following a presentation of data and methods, the empirical section highlights two key observations: First, the data documents systematic effects of the organizational structuring of the central administration. Secondly, the data shows a surprisingly degree of behavioral continuity over time. The chapter concludes with a discussion and outline for future research.

Theorizing Organizational Sustainability The etymological meaning of sustainability is someone’s or something’s ability to last over time. Sustainability thus defined does not imply coordinated human interaction in organizations (e.g., Pfahl, 2005) nor any particular role for public administration in societal sustainability (e.g., Delmas & Young, 2009; Leuenberger & Bartel, 2009), but merely that certain patterns of interaction are preserved within or across public organizations. Sustainability is measured in this chapter by a stabile set of behavioral patterns among civil servants (see below). An organizational approach to sustainable governance emphasizes how decision processes and human behavior respond to a set of fairly stable organizational factors (Kettl, 2002; Olsen, 2017). Essential to the argument, stabile premises for behavioral choices are reflected in the organizational structure of the government apparatus (Frederickson et al., 2012; Olsen, 2017; Waldo, 1952). An organizational approach is grounded on the assumption that internal organizational characteristics may explain both how organizations act and how they change. Organizational characteristics of the government apparatus systematically enable, constrain, and shape public governance processes (Waldo, 1952). Moreover, an organizational approach

88 

J. TRONDAL

emphasizes the complexity of public administration (Frederickson et al., 2012: 252). Organizational factors focus and mobilize attention and action capacity around certain problems and solutions while ignoring others, that is, focus attention along particular lines of conflict, cooperation, and so on (Simon, 1983: 21). An organizational approach posits that organizations are not merely expressions of symbol politics or reflections of power politics (Brunsson, 1989; Feldman & March, 1981; Meyer & Rowan, 1977) but have an independent effect on human behavior and collective decision-­making processes by directing and nudging individual and collective choices (Fligstein, 2001; Gulick, 1937; Hammond, 1986; Schattschneider, 1975; Thaler & Sunstein, 2009). Contemporary studies in organization theory have focused particularly on the explanatory clout of organizational structure (Egeberg, 2012; Egeberg & Trondal, 2018). An organizational structure is a normative structure composed of rules and roles specifying expectations of who, what, when, and how (Scott & Davis, 2016). It suggests how roles, power, and responsibilities are distributed, controlled, and coordinated. It influences behavior and politics by providing individuals with a systematic and predictable selection of problems, solutions, and choice opportunities (March & Olsen, 1976). Whilst organizational structure does not necessarily predict nor determine actual decision-making behavior, it does make some choices become more likely than others (e.g., Egeberg & Trondal, 2018). As such, organization theory builds on decision theory with its focus on explaining decision-making behavior (Simon, 1965). One empirical prediction is that public governance processes—and the decision-making behavior of civil servants—are likely to be systematically biased by the organizational structures of the central administration rather than by demands from its immediate task environment. One caveat should however be mentioned: The above inference holds in strongly settled and institutionally established orders but not necessarily in orders that are in the making or in transformation (Olsen, 2017). That being said, an organizational approach implies that structural characteristics within central administrations bias behavior. Three such structural variables are discussed in the following section: interorganizational vertical specialization (ministry/agency), intraorganizational specialization (rank), and organizational affiliation (seniority). All are expected to pattern governance in particular ways across time. Interorganizational vertical specialization denotes division of responsibility and labor between levels of authority. Vertical specialization is an

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

89

organizational tool and an organizational signal to anchor relatively independent expertise in certain organizational units—such as government agencies—and thereby balance political loyalty and professional neutrality (e.g., Bach et al., 2015; Egeberg & Trondal, 2017; Lægreid & Verhoest, 2010; Pollitt & Talbot, 2004). Government agencies are typically established to pursue the dual role as independent advisory expert bodies and at the same time acting as loyal agents to the elected political leadership. Agencies established at an arm’s length distance may for example safeguard continuity in the governing process independently of shifting governments. Interorganizational vertical specialization enables agencies to operate relatively insulated from political steering on the one hand but being relatively influenced by affected interests and professional concerns, often organized into advisory systems (Egeberg & Trondal, 2009; Gornitzka & Sverdrup, 2011; Veit et al., 2017; Verhoest, 2017). Contrary to agencies, ministries are organizationally more exposed to direct political guidance and control. Proximity to political leadership significantly reduces ambiguities stemming from interpretive leeway in task execution (Egeberg & Trondal, 2009). One empirical prediction is that ministry officials, as opposed to agency officials, are likely to report that political institutions are relatively more important, that the administrative apparatus is relatively well coordinated, that they enjoy relatively more political influence and power, and that they have greater trust in the government apparatus.

Intraorganizational vertical specialization denotes division of responsibility and labor within levels of authority. This variable is measured here by official’s rank within their respective ministry and agency. Higher ranked staff in both ministries and subordinated agencies are shown to be more attentive to political signals than lower ranked personnel (e.g., Christensen & Lægreid, 2009; Egeberg & Sætren, 1999). Studies show that top-­ ranked staff have a wider range of attention and tend to identify more frequently with organizations as a whole than staff located at lower echelons. Additionally, these officials are exposed to a broader range of information than lower level staff and thus may be more attentive to broader organizational perspectives than lower ranked personnel (Egeberg & Trondal, 2018). It follows that government officials with lower ranks are more loosely coupled to the political leadership and have a more local perspective on task execution compared to higher ranked staff.

90 

J. TRONDAL

One empirical prediction is that officials in higher ranks, as opposed to officials in lower ranks, are likely to report that political institutions are relatively more important, that the administrative apparatus is relatively well coordinated, that they enjoy relatively more political influence and power, and that they have greater trust in the government apparatus.

Organizational affiliation consists generally of varying degrees of organizational attachment, time, energy, and loyalty. A ‘primary affiliation’ can thus be defined as the structure to which participants are expected to devote most of their loyalty, time, and energy. One empirical proxy would be high organizational seniority or length of service. A ‘secondary affiliation’ can be defined as the structure to which participants are expected to devote medium to low loyalty, time, and energy. One empirical proxy would be medium to low organizational seniority or length of service. It is assumed that intraorganizational socialization takes time and that it will induce actors to internalize certain patterns of organizational rules, norms, and beliefs, and dilute the impact of background factors such as gender and age (e.g., Peters et al., 2015; Selden, 1997). It follows that secondary affiliations cannot be expected to shape actors’ decision behavior to the same extent as primary affiliations do. Secondary affiliations often include participation in collegial bodies, committees, and networks (Egeberg, 2012). Although one can assume that decision-making behavior is a result of both primary and secondary affiliations, a logic of primacy would suggest that primary affiliation is likely to affect behavior more extensively than the secondary (March, 1994). One empirical prediction is that officials with high seniority, as opposed to officials with lower seniority, are likely to report that political institutions are relatively more important, that the administrative apparatus is relatively well coordinated, that they enjoy relatively more political influence and power, and that they have greater trust in the government apparatus.

Empirical Strategy Enabling a study on governance sustainability, the chapter relies on the same dataset as outlined in Chap. 3, consisting of 9 large-N surveys within the Norwegian central administration—both at the ministry level (N = 7662) and the agency level (N = 5511). This chapter uses a selection of variables that are commonly used in studies of decision behavior (e.g.,

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

91

Bouckaert et al., 2010; Christensen & Lægreid, 2008; Egeberg & Trondal, 2018). That is, civil servants’ perceptions on the importance of institutions, their coordination patterns, and trust relationships. By asking officials involved in policymaking processes, the chapter documents how civil servants over time assess the influence of various institutions, the extent to which the civil service is deemed fragmented or coordinated, the degree to which civil servants enjoy influence and power, and the degree of mutual trust in different parts of the governing system. Essentially, the chapter assesses continuity and change on these variables and how they vary with the organizational structuring of the central administration. As to the latter, whereas the 1986, 1996, 2006, and 2016 surveys consist of separate data files for ministry and agency officials, the 2016 survey in addition consists of a combined data file that encompasses ministry and agency officials (N  =  4285). This combined 2016 file renders separate analyses possible on the relationship between interorganizational vertical specialization (ministerial department versus agency) and their decision-making behavior. Consequently, the theoretical potential of the dataset is vast by encompassing this central structural characteristic of the central administration. Two caveats are needed: As in most social sciences based on interview and survey data, the observations reported rest on the perceptions of the respondents. Admittedly, there are no guarantees that actors’ perceptions of administrative behavior always reflect actual behavior. Studying actors’ perceptions renders the conclusions vulnerable to perceptual errors. However, by using large-N data from two different groups of respondents, across 40 years, that have been exposed to similar questions, the conclusions are less subject to random distributions and methodological errors. Secondly, are the empirical observations generalizable? Norway has been pictured as a reluctant reformer and a slow learner of administrative reforms (Olsen, 1997). As such, the data might over-emphasize governance sustainability. Although a reluctant reformer, the Norwegian government is nevertheless an integral part of the OECD area, sharing key characteristics of the constituent states (Christensen & Lægreid, 2006; Verhoest, 2017). Even more, administrative characteristics of the central administration—such as agencification—are also prevalent at EU level where more than 30 agency-like bodies have emerged at arm’s length from the Community institutions in Brussels. Thus, the results reported should be of relevance also for understanding sustainable governance beyond the case at hand (Egeberg & Trondal, 2017).

92 

J. TRONDAL

Public Governance Across Time This section examines civil servants’ perceptions of institutional importance in the policy process (Tables 4.1 and 4.2), their views on power and influence (Tables 4.3 and 4.4), and coordination and fragmentation of the administrative service (Tables 4.5, 4.6, and 4.7). Before entering survey findings, however, Table 3.2 in Chap. 3 showed a steady growth of staff numbers in the central administration supplying public governance with ever-greater administrative capacities. The same period is characterized by increased administrative reform: Whereas the number of ministries and agencies has remained constant, they have been subject to internal reforms, such as enlargements, reshuffling of divisions, and merging of units (Christensen et al., 2018). However, the overall organizational architecture of the central administration has remained stable over time as regards the core structural characteristics. This observation is conducive to sustainable governance (see Table 3.2 in Chap. 3) Table 4.1  Civil servants in ministerial departments who report that the following institutions are fairly or very important when important decision are made within own dossier (percent)a

Own ministry Own subordinated agencies, public enterprises Own regional and local agencies Other ministries Regional and local administration The Parliament The government Employee organizations Interest organizations Consultancies/think-tanks Private business Research and educational institutions Mass media The European Commission The EU Council The EU Parliament EU agencies Other international governmental organizations

1976

1986

1996

2006

2016

93 48 46 57 32 67 73 48 16 – 12 24 18 – – – – 14

96 45 48 58 24 71 82 20 25 – 14 28 22 – – – – 19

95 50 25 62 20 77 83 16 21 – 9 24 26 17 – – 11 15

95 61 22 60 15 71 84 16 20 – 9 30 33 20 – – 8 17

96 65 24 69 17 76 87 12 20 7 10 32 22 18 11 10 7 19

The original value scale in the survey: very important (value 1), fairly important (value 2), both/and (value 3), fairly unimportant (value 4), very unimportant (value 5), don’t know/not relevant (value 6) a

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

93

Table 4.2  Civil servants in agencies who report that the following institutions are fairly or very important when important decision are made within own dossier (percent)a

Own agency Own parent ministry Own regional and local agencies Other ministries and agencies Regional and local administration The Parliament The government Employee organizations Interest organizations Private business Research and educational institutions Mass media The European Commission The EU Council The EU Parliament EU agencies Other international governmental organizations

1976

1986

1996

2006

2016

– – – – – – – – – – – – – – – – –

87 64 41 26 13 53 50 37 22 20 25 19 – – – – 16

75 69 48 27 13 47 45 17 16 17 25 21 12 – – – 13

76 75 35 27 15 52 54 14 17 15 27 26 17 – – 11 14

81 78 43 34 17 51 56 15 16 15 25 17 20 8 8 13 11

The original value scale in the survey: very important (value 1), fairly important (value 2), both/and (value 3), fairly unimportant (value 4), very unimportant (value 5), don’t know/not relevant (value 6) a

Table 4.3  Civil servants in ministerial departments who report that own unit has succeeded fairly much or very much in influencing the following institutions (percent)a

The government The Parliament Own ministry Other ministries Own subordinate agencies Regional and local administration The European Commission

1976

1986

1996

2006

2016

66 65 – 54 65 – –

57 52 71 32 57 22 –

61 53 77 44 62 24 11

64 55 79 51 65 17 11

68 59 81 57 65 16 9

The original value scale in the survey: very much (value 1), fairly much (value 2), both/and (value 3), fairly little (value 4), very little (value 5), don’t know/not relevant (value 6) a

94 

J. TRONDAL

Table 4.4  Civil servants in agencies who report that own unit has succeeded fairly much or very much in influencing the following institutions (percent)a

The government The Parliament Own parent ministry Other ministries and agencies Own regional and local agencies Regional and local administration The European Commission EU agencies/EU agency networks

1976

1986

1996

2006

2016

– – – – – – – –

34 32 48 17 – 17 – –

41 37 60 20 – 16 5 –

45 41 64 27 44 25 9 –

40 39 62 24 41 23 7 8

The original value scale in the survey: very much (value 1), fairly much (value 2), both/and (value 3), fairly little (value 4), very little (value 5), don’t know/not relevant (value 6) a

Table 4.5  Civil servants within ministerial departments characterizing coordination as fairly good or very good within their own policy area along the following dimensions (percent)a

Between different government institutions within own policy area With government institutions in other policy areas With regional and local authorities With supranational/international organizations With private sector/civil society

2006

2016

59

64

35

38

16 33 24

16 34 27

The original value scale in the survey: very good (value 1), fairly good, value 2), both/an (value 3), fairly bad (value 4), very bad (value 5), not relevant (value 6) a

Institutional Importance Tables 4.1 and 4.2 provide a thorough review of how civil servants in ministries and agencies judge the importance of various institutions when important decisions within their own dossier are made. Both tables suggest how perceived importance reflects the organizational structure of the civil service—both intra and interorganizationally. First, own organizational affiliation is generally viewed as more important than other organizations. For example, ministerial officials view own ministry as far more important than other ministries. Similar patterns are discerned for agency officials. Secondly, assessments of institutional importance tend to drop profoundly when crossing organizational boundaries—both horizontally

95

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

Table 4.6  Civil servants within agencies characterizing coordination as fairly good or very good within their own policy area along the following dimensions (percent)a

Between different government institutions within own policy area With government institutions in other policy areas With regional and local authorities With supranational/international organizations

2006

2016

43

44

22

22

18 30

18 25

The original value scale in the survey: very good (value 1), fairly good, value 2), both/an (value 3), fairly bad (value 4), very bad (value 5), not relevant (value 6) a

Table 4.7  Civil servants’ characterization of mutual trust relationships between ministries and agencies within their own policy area (percent)a 2006 2016 Ministry officials: Trust between own ministry and subordinate agencies and 70 institutions Agency officials: Trust between own agency and own parent ministry 74

69 74

The original value scale in the survey: very good (value 1), fairly good (value 2), both/and (value 3), fairly bad (value 4), very bad (value 5), don’t know (value) a

and vertically. One interesting observation is that ministerial officials increasingly view ‘own agencies and public business’ as more important over time (48 percent in 1976 and 65 percent in 2016), whilst agency officials deem ‘own ministry’ as increasingly important (64 percent in 1986 and 78 percent in 2016). This observation signifies increasing integration between ministries and agencies, reflecting the ‘whole of government’ trend with a focus on administrative coordination and cohesion. A surprising observation is stability in the role of mass media. Whereas studies have shown that mass media increasingly influence government agendas (Figenschou et al., 2017; Strömbäck & Esser, 2015), our data suggests partly otherwise, even at the ministerial level. While we observe a steady increase in the reported role of mass media from 1976 to 2006, we see a sudden drop from 2006 to 2016. The latter observation might partly be explained by the inroad of information and communication specialists in

96 

J. TRONDAL

government during this period, and correspondingly fewer direct contacts between government officials and mass media. Reflecting organizational structure, respondents in ministerial departments view political institutions as more important than respondents at agency level: The government and the parliament are deemed far more important among ministerial staff than among agency staff. Table 4.1 also demonstrates that ministerial staff increasingly emphasize the role of political institutions. The latter finding suggests increased politicization of ministerial departments. Yet, a parallel process of politicization is not observed at agency level, thus suggesting a growing politics-­administration gap over time. Yet, at a more general level Tables 4.1 and 4.2 document profound continuity over time in how civil servants view institutional importance. Similar continuity is observed with respect to the importance ascribed to supranational institutions and international organizations (IOs). More generally, Tables 4.1 and 4.2 also report overall stability as regards to internationalization of public governance processes over time. Reflecting organizational structure, agency staff tend to ascribe more weight to technocratic EU-level institutions such as EU agencies, whereas ministerial officials tend to emphasize political institutions at EU level. It is primarily ministerial-level officials who report change over time, notably by ascribing less importance to subnational authorities. Even the Commission and other IOs score higher than subnational authorities in this regard. In addition, we observe a declining role of employee organizations. This might reflect an internal de-corporatization of the state in which employees are decreasingly organized and/or supported by employee organizations. Moreover, agency staff report increased importance of own ministries, which might reflect increased quest for order in the state and more emphasize on administrative coordination (e.g., Christensen & Lægreid, 2008). Power and Influence It is methodologically difficult to measure actual power and influence through survey questionnaires. One solution is to ask civil servants how they view the success of their own unit in getting viewpoints and preferences accepted by other institutions. Tables 4.3 and 4.4 offer a comprehensive picture of such assessments over time. The data suggests profound continuity over time and predicted variation between ministerial and agency personnel.

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

97

First, respondents report that ‘own’ unit are more successful in getting their interests and viewpoints across within own policy sector than across sectors, and within own institution than across institutions. To illustrate, ministerial officials report greater success in influencing own ministry than other ministries. Similarly, three times as many agency officials report that own unit have succeeded influencing own ministry than other ministries and agencies. While ‘own ministry’ score highest among ministerial staff, the Parliament, other ministries, and agencies share a second place. Reflecting interorganizational vertical specialization, the data report predicted variation between ministry and agency staff, particularly as regards to influencing political institutions such as the Parliament and the government. Ministerial staff report greater influence in this regard than agency personnel. Taken together, these observations suggest how ministries are tighter coupled with the policy process than agencies. Moreover, they show how the central administration is loosely coupled with the subnational level since civil servants in the central administration report fairly low influence toward local and regional administration. Similar observations are made vis-a-vis the Commission as well. Agency respondents received one additional question as to their perceived success in influencing EU agencies and European agency networks, and Table 4.4 suggests a generally low level of influence toward both EU agencies, networks of agencies, and the Commission. This finding most likely reflects Norway’s form of affiliation toward the EU which grants the Norwegian civil service weak formal access to the EU decision-making cycle. Coordination and Fragmentation Coordination is not a technical-neutral question; it affects the distribution and concentration of power. Questions about coordination can become controversial and lead to contestation: It is most often less popular to be subject to coordination than to be the one that coordinates (Bouckaert et al., 2010). The following tables demonstrate how government officials view coordination within their own policy area over time. This question first entered the survey in 2006, but despite a shorter time horizon Tables 4.5 and 4.6 reveal profound continuity over time in officials’ judgments on coordination. Despite increased quest for politico-administrative order, coordination, and cohesion in administrative doctrines and national government programs, the civil servants report small changes in how they characterize coordination within their own policy area. Moreover,

98 

J. TRONDAL

contrary to the hole-of-government image of increased coordination in government, Tables 4.5 and 4.6 report continuity in administrative coordination. Public governance thus harbors sustainability endogenously embedded in the government structure. Similarly, both tables illustrate how organizational boundaries affect decision-making behavior: Coordination is reduced by almost 50% when crossing organizational and sectoral boundaries in the central administration. Twice as many report good coordination between different government institutions within own policy sector than across policy sectors. Finally, both tables reveal high degree of both Europeanization and internationalization of the civil service: Twice as many civil servants report good coordination with supranational institutions and IOs than with subnational authorities. This observation suggests lower levels of national administrative integration between the central administration and subnational authorities than between the central administration and supranational institutions and international bureaucracies. Luther Gulick (1937) claimed that coordination between organizations can be achieved both through organizing—for example by establishing separate coordination units—and through the minds and wills of actors. Coordination through organizing often appears when attitudes and trust relationships alone do not suffice. Coordination through trust relationships might make the social glue within and across organizations stronger, making collaboration easier, organizations more robust, and contribute to legitimize the government decisions and actions. Mutual trust are signifiers of institutionalization ‘by which we move from something that is diffuse, unstable, and unfixed into something more settled, stable, and integrated’ (Selznick, 2015: 15). Organizations infused with value in the Selznick sense have distinct institutional character that help cement both external societal support as well as lead to internalized guidelines that shape behavior. Table 4.7 shows how civil servants in the Norwegian central administration characterize mutual trust relationships between ministries and agencies. This question first entered the survey in 2006, which offers only two observation points, but overall, the data suggests high degree of mutual trust inside the central administration over time. Moreover, trust relationships seem to be institutionalized since it varies marginally—both over time and between institutions (ministries and agencies). Finally, to gauge the relationships between interorganizational vertical specialization (ministerial department versus agency) and decision-making

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

99

behavior, while also incorporating rank and seniority, an OLS regression analysis was conducted on the combined 2016 data file encompassing ministry and agency officials (N = 4285—Table 4.8). Two dependent variables were selected for analysis: civil servants’ perceived influence vis-à-vis own ministry/agency and civil servants’ perceived influence vis-à-vis the government. The OLS regression models suggest that organizational structure significantly affects (p ≤ 0.01) power relationships. As predicted, officials in higher ranks and in ministerial departments are significantly more influential than those in lower ranks and in agencies. By contrast, seniority does not yield strong effect on power relationships, yet officials with longer terms of service tend to experience somewhat more influence than newcomers to office. Thus, organizational socialization is a much less salient predictor of influence than argued in much socialization literature and when compared to organizational factors. Moreover, this finding resonates with recent findings which show that organizational socialization does not seem to be a precondition for role obedience (Egeberg & Stigen, 2021).

Table 4.8  Summary of factors affecting officials’ reported influence vis-a-vis own institution (own ministry/own agency) and the government (standardized beta coefficients; linear regressions on ministry and agency officials; 2016 data)a, b Influence on own institution (own ministry/own agency) Vertical interorganizational specialization (ministry/agency) Vertical intraorganizational specialization (rank) Organizational affiliation (seniority)

Influence on the government

–.25**

–.23**

–.09**

–.12**

.05*

0.04

*p ≤ 0.05, **p ≤ 0.01 Coding of the dependent variable: The original value scale in the survey: very much (value 1), fairly much (value 2), both/and (value 3), fairly little (value 4), very little (value 5), don’t know/not relevant (value 6) a

Coding of the independent variables: Vertical interorganizational specialization: ministry (value 1), agency (value 0); vertical intraorganizational specialization (director general or higher levels/adviser/ director or equivalent (value 1), deputy director general (value 2), assistant director general/adviser (value 3), principal officer/adviser (value 4), executive officer, higher executive officer/adviser (value 5); organizational affiliation (natural continuous variable) b

100 

J. TRONDAL

Conclusions The chapter has made two main contributions to the volume: Theoretically, the chapter demonstrates that organizational factors may account for the sustainability of public governance. Civil servants’ perceptions of power, coordination, and trust over time systematically reflect their organizational embedment in the central administration. These items vary systematically with civil servants’ inter and intraorganizational embedment and remain profoundly stabile across time. Thus, key observations essentially reflect the organizational architecture of the civil service and not the time periods studied. The findings thus showcase the organizational basis for sustainable public governance and that public administration harbors some fairly stable resources expressed in the formal organization of the apparatus. Second, the chapter has offered a novel large-N and longitudinal empirical probe. By building on a large number of observations at five observation points over time, the dataset has enabled a study of public governance processes over nearly half a century. Consisting of nine surveys from both the ministry and agency level from 1976 to 2016, the chapter provides robust analyses of the organizational basis of public governance sustainability. The most puzzling empirical contribution is thus the notable continuity in administrative behavior. Both ministry and agency officials consistently report that power is located in the same government institutions in 2016 as in 1976. Some findings, however, suggest change and thus less sustainable patterns: Governmental and parliamentary influence is deemed weaker over time and particularly among agency officials. Whereas ministerial staff increasingly emphasizes the importance of political institutions, agency officials deem political institutions increasingly less important. This suggests a growing politics-administration gap over time. Yet, we also observe increased administrative integration within the central administration over time: Whilst ministerial officials ascribe increased importance to subordinate agencies, agency officials similarly deem own parent ministry as increasingly more important. This illustrates administrative integration, or executive center formation, within the polity. At the same time, we see less integration of the central administration and subnational authorities: Respondents view regional and local administrations as gradually less important. We also observe lower degrees of internal corporatism in the state by the fact that employee organizations are considered less important. During the same period, however, mass media is viewed as a stabile

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

101

influencer (Figenschou et al., 2017). This is a surprising finding since a vast field of research on mass media suggest that they play an increased role as policy agenda setters. Our observations challenge this observation in literature. Whereas a vast scholarly literature has emphasized that public sector organizations tend to be profoundly unsustainable in the long run, this chapter demonstrates profound sustainability in public governance. To advance our understanding of public governance sustainability, future studies should invest in (collaborative) longitudinal datasets both in national core executives as well as in the public administration of international organizations. Understanding the nuts and bolts of sustainable public governance at national and international levels is instrumental in safeguarding sustainable public service delivery and important in our understanding of stabile political order.

References Alvesson, M., & Spicer, A. (2019). Neo-institutional theory and organization studies: A mid-life crisis? Organization Studies, 40(2), 199–218. Ansell, C., Boin, A., & Farjoun, M. (2015). Dynamic conservativism: how institutions change to remain the same. In M. S. Kraatz (Ed.), Institutions and ideals: Philip Selznick’s legacy for organizational studies. Emerald. Ansell, C., & Torfing, J. (Eds.). (2014). Public innovation through collaboration and design. Routledge. Ansell, C., & Trondal, J. (2017). Governing turbulence. An organizational-­ institutional agenda. Perspectives on Public Management and Governance, 1(1), 43–57. Ansell, C., Trondal, J., & Ogard, M. (Eds.). (2017). Governance in turbulent times. Oxford University Press. Aucoin, P. (1990). Administrative reform in public administration: Paradigms, principles paradoxes and pendulums. Governance, 3(2), 115–137. Bach, T., Ruffing, E., & Yesilkagit, K. (2015). The differential empowering effects of Europeanization on the autonomy of national agencies. Governance, 28(3), 285–304. Bevir, M. (Ed.). (2010). The SAGE handbook of governance. SAGE. Bouckaert, G., Peters, B. G., & Verhoest, K. (2010). The coordination of public sector organizations. Palgrave Macmillan. Bouilloud, J.-P., Pèrezts, M., Viale, T., & Schaepelynck, V. (2020). Beyond the stable image of institutions: Using institutional analysis to tackle classic questions in institutional theory. Organization Studies, 41(2), 153–174.

102 

J. TRONDAL

Brunsson, N. (1989). The organization of hypocrisy. Wiley. Bulmer, S., & Joseph, J. (2016). European integration in crisis? Of supranational integration, hegemonic projects and domestic politics. European Journal of International Relations, 22(4), 725–748. Christensen, T., Egeberg, M., Lægreid, P., & Trondal, J. (2018). Sentralforvaltningen. Stabilitet og endring gjennom 40 år. Universitetsforlaget. Christensen, T., & Lægreid, P. (Eds.). (2006). Autonomy and regulation. Edward Elgar. Christensen, T., & Lægreid, P. (Eds.). (2007). Transcending new public management. Routledge. Christensen, T., & Lægreid, P. (2008). The challenge of coordination in central government organizations: The Norwegian case. Public Organization Review, 8(2), 97–116. Christensen, T., & Lægreid, P. (2009). Living in the past? Change and continuity in the Norwegian central civil service. Public Administration Review, 69(5), 951–961. Delmas, M. A., & Young, O. R. (Eds.). (2009). Governance for the environment. Cambridge University Press. Egeberg, M. (2012). How bureaucratic structure matters: An organizational perspective. In B.  G. Peters & J.  Pierre (Eds.), The SAGE handbook of public administration. SAGE. Egeberg, M., & Sætren, H. (1999). Identities in complex organizations: A study of ministerial bureaucrats. In M.  Egeberg & P.  Lægreid (Eds.), Organizing political institutions. Scandinavian University Press. Egeberg, M., & Stigen, I. M. (2021). Explaining government bureaucrats’ behavior: On the relative importance of organizational position, demographic background, and political attitudes. Public Policy and Administration, 36(1), 3–18. Egeberg, M., & Trondal, J. (2009). Political leadership and bureaucratic autonomy. Effects of agencification. Governance, 22(4), 673–688. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies: What have we learnt (and where do we go from here). Journal of Common Market Studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Emery, Y., & Giauque, D. (2014). The hybrid universe of public administration in the 21st century. International Review of Administrative Sciences, 80(1), 23–32. Feldman, M. S., & March, J. G. (1981). Information in organizations as signal and symbol. Administrative Science Quarterly, 26(2), 171–186. Figenschou, T. U., Karlsen, R., Kolltveit, K., & Thorbjørnsrud, K. (2017). Serving the media ministers: A mixed methods study on the personalization of ministerial communication. The International Journal of Press/Politics, 22(4), 411–430.

4  PUBLIC ADMINISTRATION SUSTAINABILITY AND ITS ORGANIZATIONAL… 

103

Fligstein, N. (2001). Organizations: Theoretical debates and the scope of organizational theory. Manuscript, UC Berkeley. Frederickson, H. G., Smith, K. B., Larimer, C. W., & Licari, M. J. (2012). The public administration theory primer. Westview Press. Gornitzka, Å., & Sverdrup, U. (2011). Access of experts: Information and EU decision-making. West European Politics, 34(1), 48–70. Gulick, L. (1937). Notes on the theory of organization. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration. Hammond, T. H. (1986). Agenda control, organizational structure, and bureaucratic Politics. American Journal of Political Science, 30(1), 379–420. Kettl, D.  F. (2002). The transformation of governance. Johns Hopkins University Press. Klenk, T., & Reiter, R. (2019). Post-New Public Management: reform ideas and their application in the field of social services. International Review of Administrative Sciences, 85(1), 3–10. Kraatz, M.  S. (Ed.). (2015). Institutions and ideals: Philip Selznick’s legacy for organizational studies. Emerald. Lægreid, P., & Verhoest, K. (Eds.). (2010). Governance of public sector organizations. Palgrave Macmillan. Leuenberger, D.  Z., & Bartel, J.  R. (2009). Sustainable development for public administration. M.E. Sharpe. Levi-Faur, D. (Ed.). (2012). The Oxford handbook of governance. Oxford University Press. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Cambridge University Press. March, J. G. (1994). A primer on decision making. Free Press. March, J.  G., & Olsen, J.  P. (1976). Ambiguity and choice in organizations. Scandinavian University Press. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions. Free Press. Meyer, J. W., Boli, J., Thomas, G. M., & Ramirez, F. O. (1997). World society and the nation-state. American Journal of Sociology, 103(1), 144–181. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations. Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. O’Leary, R., Van Slyke, D. M., & Kim, S. (2010). The future of public administration around the world. Georgetown University Press. Olsen, J. P. (1997). Institutional design in democratic contexts. Journal of Political Philosophy, 5(3), 203–229. Olsen, J. P. (2006). Maybe it is time to rediscover bureaucracy. Journal of Public Administration Research and Theory, 16(1), 1–24. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford University Press.

104 

J. TRONDAL

Olsen, J. P. (2018). Democratic accountability and the changing European political order. European Law Journal, 24(1), 77–98. Peters, B. G., von Maravic, P., & Schroter, E. (Eds.). (2015). Politics of representative bureaucracy. Edward Elgar. Pfahl, S. (2005). Institutional sustainability. Sustainable Development, 8(1–2), 80–96. Pfeffer, J., & Salancik, G. (1978). The external control of organizations. Harper and Row. Pollitt, C. (2008). Time, policy, management. Oxford University Press. Pollitt, C., & Bouckaert, G. (2017). Public management reform. Oxford University Press. Pollitt, C., & Talbot, C. (Eds.). (2004). Unbundled government. Routledge. Schattschneider, E. E. (1975). The Semisovereign people: A realist’s view of democracy in America. Harcourt Brace Jovanovich College Publishers. Scott, W. R., & Davis, G. F. (2016). Organizations and organizing. Routledge. Selden, S. C. (1997). The promise of representative bureaucracy. M. E. Sharpe. Selznick, P. (1958). Leadership in administration. University of California Press. Selznick, P. (2015). On sustaining research agendas: Their moral and scientific basis. In M. S. Kraatz (Ed.), Institutions and ideals: Philip Selznick’s legacy for organizational studies. Emerald. Simon, H. A. (1965). Administrative behavior. The Free Press. Simon, H. A. (1983). Reason in human affairs. Stanford University Press. Streek, W., & Thelen, K. (Eds.). (2005). Beyond Continuity. Oxford University Press. Strömbäck, J., & Esser, F. (Eds.). (2015). Making sense of mediatized politics. Routledge. Thaler, R. H., & Sunstein, C. R. (2009). Nudge. Improving decisions about health, wealth and happiness. Penguin Books. Tolbert, P. S., & Zucker, L. G. (1983). Institutional sources of change in the formal structure of organizations: The diffusion of civil service reform, 1880-1935. Administrative Science Quarterly, 28(1), 22–39. Veit, S., Hustedt, T., & Bach, T. (2017). Dynamics of change in internal policy advisory systems: the hybridization of advisory capacities in Germany. Policy Sciences, 50(1), 85–103. Verhoest, K. (2017). Agencification in Europe. In E.  Ongaro & S.  Van Thiel (Eds.), The Palgrave handbook on public administration and management in Europe. Palgrave Macmillan. Waldo, D. (1952). Development of theory of democratic administration. American Political Science Review, 46(1), 81–103.

CHAPTER 5

Agency Governance in Integrated Administrative Systems

What do public organizations do when subjected to conflicting influences on how to maneuver? What do government officials do when embedded in what Woodrow Wilson (1887: 221) described as ‘systems within systems’? This chapter conceptually outlines two possible responses and illustrate these with a large-N dataset on national government agencies embedded in two politico-administrative systems: The central administration of a unitary state and the administrative system of a quasi-federal order—the EU. When agency officials are thus subject to dual stimuli and influences, the question is how they are likely to respond. The relevance of the question is abundant since government agencies increasingly maneuver in situations where state sovereignty is under persistent stress (Ansell et  al., 2017: 1; Easton, 1965; Gunnell, 2011; Miller, 1971). Mutual dependence, international cooperation, and delegation to nonelected bodies challenge nation states’ capacity to make democratically accountable decisions. These developments have spurred vibrant scholarly debate on the quality and conditions of political sovereignty (e.g., Dahl, 2000; Olsen, 2018; Rosenau, 1990). Much less understood, however, is the shifting role of government agencies and what role public administration plays in making sovereignty resilient (Egeberg & Trondal, 2015). As the public administration of states serves as core capacities for state building, the role of public agencies is key to understanding sustained state sovereignty (Genschel & Jachtenfuchs, 2014: 10). Putting public © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_5

105

106 

J. TRONDAL

administration center stage in the study of democratic governance, Dwight Waldo (1952) emphasized the semi-autonomous role of public administration. Since Waldo, government agencies have become essential opportunity structures in advanced democracies (Orren & Skowronek, 2017; Vibert, 2007). Their tasks range from collecting and analyzing information, coordinating, and regulating. Organizationally, government agencies represent a vertical fragmentation of the polity and a supply of administrative capacities to solve regulatory challenges. They are thus organizational compromises between a need for political steering and requisite professional autonomy and technical regulation. Government agencies are organized at arm’s length from their parent ministries, ensuring agencies to operate relatively insulated from political steering but at the same time organizationally exposed to ‘capture’ from EU-level institutions and processes (Egeberg & Trondal, 2017). This chapter examines how government agencies maneuver when embedded in integrated multilevel administrative orders. Domestic central administration is engaged in two parallel domains of executive governance and may experience opportunities and constraints at both levels because different institutions may send different information, signals, and mandates to domestic civil servants (Dehousse, 2008; Egeberg & Trondal, 2018). The contribution of the chapter is twofold: –– First, it outlines two conceptual images of agency governance: a dyadic (Image I) and a compound (Image II) approach. These images derive distinct propositions as to how domestic government agencies act when embedded in integrated multilevel administrative orders such as the EU. The chapter also suggests how these images depend on moderator variables. –– Secondly, the analysis benefits from a novel large-N (1963) dataset on agency officials. The dataset is comprehensive by spanning the whole population of government agencies in a central administration (Norway). The chapter suggests that agency officials mainly feature compound behavior (Image II) characterized by compromises and abilities to navigate conflicting concerns—such as those of domestic governments and EU-level institutions. Secondly, the data suggests that this dilemma is mediated through, and conditioned by, existing domestic institutions, practices, and traditions, thus contributing to a differentiated European administrative order (e.g., Bauer & Trondal, 2015; Leruth et al., 2019). However, the

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

107

compound image of agency governance (Image II) is shown to be robust since no moderator variables do fundamentally change relationships from one type of governance to another. The chapter thus expands our understanding of the ‘coordination dilemma’ that are faced by domestic agencies embedded in EU governance processes (Egeberg & Trondal, 2016), and also advances our understanding of governance robustness. This chapter adds to a mounting literature on the changing role of public governance in an integrated multilevel European executive order (Bauer & Trondal, 2015; Goetz & Meyer-Sahling, 2008; Trondal, 2010). This literature has been preoccupied with studying EU-level agencification (e.g., Egeberg & Trondal, 2017; Levi-Faur, 2011; Rittberger & Wonka, 2011) and the interconnected nature of the EU-level and national-­ level agencies both with respects to substantive policymaking and implementation, as well as to administrative change and reform (Bach & Ruffing, 2018; Curtin & Egeberg, 2008; Egeberg & Trondal, 2018; Trein & Maggetti, 2018). This literature has shown how the European administrative system represents a multilevel network administration (e.g., Bach & Ruffing, 2018), where administrative bodies at different levels of government ‘are linked together in the performance of tasks…’ (Hofmann & Turk, 2006: 583). However, one challenge plagues contemporary scholarship which this chapter aims to address: How to understand the role of public agencies in the governing of the deeply integrated (yet differentiated) EU multilevel system. Existing literature has failed to understand how agencification at one level affects public governance at another, and thus how shifting features of the state—such as agencification and subsequent networking of agencies—influence democratic governance (Danielsen & Yesilkagit, 2014; Egeberg, 2006; Egeberg & Trondal, 2016; Vantaggiato, 2020; Verhoest et al., 2012). Whereas existing literature has mapped patterns of agencification, less is known about how and under what conditions public agencies contribute to the transformation of the EU’s politico-administrative order. Empirically, we lack studies of the actual role of agencies in the policymaking process in the EU. Without it, scholars and practitioners face the double danger of misunderstanding and mistreating it. The dataset is collected within by far the most integrated EU nonmember (Norway). Norway is not a formal EU member, yet it is closely affiliated through 101 agreements, notably a comprehensive and dynamic

108 

J. TRONDAL

European Economic Area (EEA) agreement. Norway has been considered a quasi-EU-member (Fossum, 2019b) as Norway’s affiliated status grants the Norwegian central administration privileged access to most parts of the EU administration, which in turn opens up for administrative integration quite similar to the central administrations of EU member states. Consequently, despite a lack of political representation in the Council (and the European Parliament), the Norwegian core executive apparatus is tightly integrated with and influenced by the EU’s core executive apparatus (Kühn & Trondal, 2018). The ‘Norway model’ is thus coined by administrative integration without political integration through a dense web of administrative agreements, with the EEA agreement being the most encompassing and which has the greatest implications for public policy and law (Egeberg & Trondal, 1999; Fossum, 2019b; Fossum & Graver, 2018; Lavenex, 2009). The chapter proceeds as follows: The next section outlines a framework for analysis and propositions for empirical inquiry. The next section presents the dataset and research strategy applied succeeded by a presentation and discussion of empirical findings. The chapter concludes with key findings and implications for this field of study.

A Two-Step Theoretical Strategy The theoretical strategy is brought about in two steps: The first step outlines competing images of agency governance: a dyadic and a compound approach. The second step suggests how moderator variables might bias agency governance toward either of the two. Studying public governance and how domestic central administrative institutions adapt to EU affairs is important for several reasons. First, domestic public administration are key institutions to the implementation and practice of EU jurisprudence. Because the transposition of EU law remains an administrative process relatively isolated from political actors, it is essential to study agencies in order to understand the likelihood of uniform implementation of (EU) law. Domestic decision-making processes are also essential since they are intertwined with the decision-­making processes of the EU, notably the Commission and EU agencies. Such interactions might affect power relationships between the ministry and subordinate agencies, between elected politicians and expert officials, as well as between domestic governance systems and the EU.

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

109

Step I: Competing Images on Agency Governance This section expands on two ideas from public administration literature that make distinctive forecasts for agency governance.  Dyadic Image I A This idea builds on the simple conjecture that public administration runs in dyadic and dichotomous domains, such as politics versus administration and national versus international (e.g., Ansell & Trondal, 2018; Rosenau, 1990; Waldo, 1952; Wilson, 1887: 221). Turbulence is thus understood to be dysfunctional—that is, exceptional, dangerous, or contradictory (Ansell et al., 2017). In this light, turbulence is seen to push organizations and institutions to their limits and mobilize surprising cascading dynamics that may undermine the sustainability of existing governance arrangements (Vigoda-Gadot & Mizrahi, 2014). Or it might produce maladaptive behaviors that trap governance into suboptimal outcomes. From this perspective, the emphasis is generally not how governing institutions manage turbulence, but how they withstand it and/or ultimately solve it (Ansell & Trondal, 2018). This image builds on the ‘coordination dilemma’ portrayed by contending governing dynamics undermining one another (Egeberg & Trondal, 2016). The ambition of strong coordination of governance processes at one level of government is arguably incompatible with strong coordination of governance processes across levels. For example, strong steering of public agencies from the domestic political leadership is arguably adversely related to strong steering from the EU level. Strong coordination by the Commission vis-à-vis domestic government agencies is thus expected to undermine ministerial political control. Image I explains both how turbulence generates ambiguities when assessing and suggesting ‘good’ governance solutions (Grindle, 2017) as well as challenges attached to ‘nationally embedding a supranational project’ (Bulmer & Joseph, 2016: 738). In the same vein, Image I underscores how ‘better coordination’ becomes difficult when public administration becomes embedded in multilevel governing structures. The quest for ‘order’ has become a shared goal in contemporary politico-administrative systems (e.g., Kassim et al., 2017; MacCarthaigh & Molenveld, 2018). Recent administrative doctrines have emphasized ‘better coordination’ and centralization of executive institutions (e.g., Lægreid et al., 2014). Image I, however, assumes

110 

J. TRONDAL

that it is impossible to combine strong coordination processes at one level of government with simultaneous strong coordination across levels. To illustrate, the multilevel EU polity is a case in point: The EU borrowed ‘indirect administration’ from IOs. Decisions assumed by IOs were thus implemented by member states and not by the IO itself (Hofmann, 2008). This structure, characterized by relatively weak coordination across levels, is thus compatible with strong coordination at the national level. Similarly, Wolfgang Wessels (1997) showed that the lack of coordination in Bonn may have been a prerequisite for advancing the wider ‘European cause.’ Poor national coordination is thus not merely a ‘management deficit’ (Metcalfe, 1994) but rather a requirement for coordination and steering across levels. This tension between policy coordination across levels of government and regional/local coordination has been well known in federal states, as well as within central-local relations in unitary states (e.g., Fossum & Jachtenfuchs, 2017). It has, however, been less recognized in the study of public administration. In this context, national agencies organized at arm’s length from parent ministerial departments and enjoying de facto autonomy vis-à-vis these departments, thus constituting the very administrative infrastructure for ‘agency capture’ by EU bodies (Trondal & Peters, 2013). Consequently, national agencies become building blocks of a multilevel EU administration (Egeberg & Trondal, 2018). One empirical proposition follows: P1: Strong administrative integration across levels of governance is orthogonal to strong national governmental steering, and so we expect significant negative correlations between EU-level and national-level steering.  Compound Image II A By contrast, Image II contends that public administration is complex by involving multiple actors, ill-structured causal relations, fluid resources, governing logics, and dynamics (Olsen, 2017; Trondal et al., 2010). In line with Chaps. 1 and 2 in this volume, Image II sees turbulence as a condition and inherent for public governance, not as a dysfunction (Ansell & Trondal, 2018; Howlett & Mukherjee, 2018). This image also sees the national versus international relations and institutions as a continuum. The general observation made by Ansell et al. (2017: 8) entails that ‘turbulence’ can be ‘almost a constitutive part of the institutional fabric’ in organizational structures and cultures. This argument was made earlier in

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

111

a different context by Ljiphart (1968: 104) when reading of Dutch politics as consociational resting on actors and institutions being ‘willing and capable of bridging the gaps between the mutually isolated blocs and of resolving serious disputes in a largely nonconsensual context.’ Similar ideas linger in contemporary literature on differentiated European (dis) integration, in which differentiation is seen as a condition for European Union governance (Fossum, 2019a) and as a systemic feature of the EU (Leruth et al., 2022). If embedded complexity and contestation are understood to be a condition of contemporary governance, one necessary implication is that efficient and effective governing institutions must manage turbulence as a condition for the policy process and tolerate ambiguity as part of the process (Orren & Skowronek, 2017: 91). Public governance, consequently, must be analyzed on the basis of continuous rather than dichotomous variables (Ansell et  al., 2017). In this light, public administration has been pictured as hybrid and compound (Emery & Giauque, 2014; Olsen, 2018). The idea of the compound administration responds to calls for going beyond ‘the tyranny of dichotomies’ and study ‘mixed political orders blending different forms of governance and organization’ (Olsen, 2008: 5–6). Public administration is engaged in several co-evolving worlds of executive governance, for example, when national agencies that are practicing EU law take on multiple roles and hats (Egeberg & Trondal, 2018). Similarly, a vast body of literature has pictured executive governance as characterized by the co-existence of institutions, decision-making dynamics, and levels of authority (Christensen & Lægreid, 2008; Hooghe & Marks, 2016; Mathieu & Rangoni, 2019; Olsen, 2007). Public administration is seen as facing complex and intertwined problems, solutions, institutions, and decision-making arenas (Olsen, 2007; Shapiro et  al., 2006). One effect might be that domestic agencies’ handling of EU affairs is best conceptualized as compound processes in which they mobilize a multidimensional repertoire. This idea follows a classical tradition in the study of public administration which argues that robust administrative systems tend to balance several competing concerns (e.g., Olsen, 2007) and that public governance rests on the mobilization of multiple sets of institutions, resources, interests, values, norms, and cleavages of conflict (Pollitt, 2016). One empirical proposition follows: P2: Strong administrative integration across levels of governance complements strong national governmental steering, and so we expect significant positive correlations between EU-level and national-level ­ coordination and steering.

112 

J. TRONDAL

Step II: Interaction Effects Agency governance in integrated administrative systems calls upon agency officials to balance competing concerns from different institutions. Integrated administrative systems produce ambiguity about what problems, solutions, and consequences to attend to at any time, and what actors are deemed legitimate and efficient (Ansell et al., 2017). This is the case under two particular conditions: First, political organizations, such as government agencies, are of a different kind than private organizations by being ‘created in order to handle conflicts …’ (Jacobsson et al., 2015: 35) and are thereby turbulent by design. Organized democracies have an embedded partisan responsiveness to numerous different cleavages of conflict (Rokkan, 1999), which private organizations clearly have not. As such, governing politically embedded agencies entails a systemic component of turbulence. Secondly, turbulence is conditional for sustained public rule in government systems characterized by multilevel layering of authority (Landau, 1969; Olsen, 2018). More specifically, in political systems with layered authority, when action capacity is distributed across levels, the involved organizations are likely to experience turbulence of scale in their exercising of tasks (Fossum & Jachtenfuchs, 2017; Moravcsik, 2018; Schmidt, 2018). For example, the more the EU is engaged in policy implementation, the more important become issues of administrative coordination across levels—if only because the EU lacks requisite administrative capacities to conduct ‘supranational’ policies independently of member states (Bauer & Becker, 2014; Heidbreder, 2011). Moreover, attempts from one actor to manage or limit turbulence (i.e., by the Commission) may have turbulence-increasing consequences for others (i.e., member-state agencies). Cyert and March (1963) suggested three mechanisms for how the firm may cope with such situations of turbulence: (1) through local rationalities, (2) through acceptable-level decision rules, and (3) through sequential attention to goals. However, public organizations rarely solve public problems but rather cope with them through organizational design and ‘choice-architectures’ that bias issue attention and prioritization (Bark & Bell, 2019; Egeberg & Trondal, 2018; Jones & Baumgartner, 2005; Thaler & Sunstein, 2009). These architectures are materialized in the

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

113

organizational fabric of government agencies through the vertical and horizontal specialization of departments, procedures for the recruitment of staff, the geographical location of offices, time rules for budgeting, and so on (Egeberg & Trondal, 2018; Jann & Wegrich, 2019). This section suggests that the relative importance of either Image I and II might be conditioned by intervening variables in which third variables affect the relationship between dependent and independent variables. The task is then to causally isolate conditions under which each image is likely to unfold, as well as to test their robustness. The method section (see below) specifies seven variables that might intervene and bias agency governance in the direction of either Image I or Image II. Building on recent empirical studies in public governance literature (for a recent review see Egeberg & Trondal, 2018), the following moderators have been selected from the dataset: (1) the degree to which policy fields are affected by EU/ EEA/Schengen, (2) administrative capacity to ensure coordination and collaboration between relevant actors, (3) the degree to which policy fields require cooperation across and between levels of governance, (4) the degree to which there are clear and established formal rules on how to conduct tasks, (5) the degree to which there are organizational duplication, (6) perceived trust between own agency and parent ministry, and (7) the degree of politicization within policy fields.

Empirical Strategy The chapter benefits from a unique large-N questionnaire survey, as outlined in Chap. 3, that was recently (2016) completed at the agency level in the Norwegian central administration (N = 1963). The survey represents the most thorough screening of the Norwegian central administration, and probably one of the most comprehensive datasets on public governance in national government administrations worldwide. Table 5.1 presents the independent and dependent variables of the study, including the moderators. The empirical analysis applies importance of (various) institutions as measurement of the alleged steering dilemma. As specified, two proxies are applied as dependent variables and two as independent variables. To empirically explore the theoretical propositions explicated above, we use a twostep procedure. First, we explore pairwise correlations within and across various variables clustered at both the national level and the EU level. Interpreting bivariate correlations singularly does not suffice in

114 

J. TRONDAL

Table 5.1  Descriptive statistics: dependent, independent, and moderator variablesa Nb

Dependent variables Importance of parent ministry when central decisions within your policy field are being made Importance of government when central decisions in your policy field are made

Value—Max

Value—Min

Value— Do not know

Meanc St. devc

1178 1—Very (1233) important

5—Not important

6—55

1.7

1

1035 1—Very (1235) important

5—Not important

6—200

2.1

1.2

Independent variables 5—Not 6—485 important

3.2

1.3

6—564

3.5

1.3

–

3.2

1.5

1536 1—Very good 5—Very poor (1686)

6—150

2.6

0.9

1719 1—To a very (1732) large extent

6—13

2.8

1.2

Importance of the 751 1—Very Commission when (1236) important central decisions in your policy field are made Importance of EU 670 1—Very agencies when central (1234) important decisions in your policy field are made Degree to which own policy field is affected by EU/EEA/ Schengen Administrative capacity to ensure coordination and collaboration between relevant actors Degree to which own policy field requires cooperation across and between levels of governance

1374

1—To a very large extent

5—Not important

Moderators 5—Not affected

5—To a very small extent

(continued)

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

115

Table 5.1  (continued) Nb

Value—Max

Degree to which there are clear and established formal rules on how to conduct tasksd Organizational duplicatione

1722 (–)

1462

Perceived trust between own agency and parent ministry Degree of politicization of own policy field

1169 1—Very high (1240) 1232

Value— Do not know

Meanc St. devc

1—Very clear 5—Rely on own rules judgement to a very large extent

–

2.6

1.1

1— Departments

4—No organizational duplication 5—Very poor

–

2.2

1.1

6—71

1.9

0.8

5—To a very small extent

–

2.7

1.2

1—To a very large extent

Value—Min

Missing values not included Includes only values 1–5 (includes value 6 in parenthesis) c Includes only value 1–5 d This variable is 5-scaled: (1) very large degree of formal rules (2) large degree of formal rules (3) neutral (4) rely on own judgement to a fairly large extent (5) rely on own judgement to a very large extent e This variable is 4-scaled: (1) departments (2) sections (3) positions (4) no organizational duplication a

b

evaluating whether agency governance is mainly dyadic or compound. In addition, one needs to look at the broader picture and assess patterns of correlations within and between levels simultaneously. As outlined in the theoretical section, strong positive bivariate correlation across national and EU level may indicate compound agency governance. However, this inference will be weakened, or appear arbitrary, if similar correlation patterns are not consistent for other pairs of variables across levels, or even if variables within levels correlate negatively. Likewise, while strong negative correlations between variables across national and EU levels correspond with a dyadic pattern, such an inference is more convincing if these correlations coexist with positive within-level correlations. If, on the other hand, correlations both within and across levels are strongly negative, the results would be indeterminate in supporting the regularity of a dyadic pattern. Table 5.2 illustrates this complexity and thus the need to look at broader patterns of correlations. More particularly, the table maps out four patterns of pairwise correlations between variables within and across

116 

J. TRONDAL

Table 5.2  Patterns of pairwise correlations

levels, of which two accentuate the ideal patterns for compound and dyadic governance, respectively. A clear compound governance pattern thus emerges where positive correlations are visible both across and within levels. In a similar vein, a clear dyadic pattern becomes visible when correlations across levels are consistently negative whilst correlations within levels are predominantly positive. Unfortunately, the terrain does not always fit the map. The complexity of interpreting correlational patterns thus increases with the emergence of low or non-significant correlations, within and/or across levels. In such instances, it becomes difficult to determine if a hypothesized strong correlation combined with low or no correlation are pointing toward compound or dyadic tendencies. One way of taking this complexity into account is to assess the dynamic and conditional nature of agency governance within the confines of a regression analysis. The next step is thus to explore the relationship between national-level and EU-level variables using a classic multiplicative interaction model:

Y   0   X   Z    XZ   



In this model the outcome variable Y is the perceived importance of institutions at the national level while the key independent variable X is perceived importance of EU-level institutions. The moderator variable Z is included to investigate whether, or to what extent, the strength and/or direction of the relationship between X and Y is conditional, that is, varies with the level of Z. In other words, a multiplicative interaction model

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

117

enables us to explore whether compound or dyadic tendencies may emerge under different institutional contexts, or, alternatively, whether a relationship is stable across various conditions or contexts. As such, the aim of the analysis is not to explain as much variance as possible but rather to investigate and reveal conditions under which a relationship between an outcome and an independent variable changes. Interactions effects will then be plotted graphically for illustrative and interpretative purposes.

Empirical Demonstration Table 5.3 makes a test of whether agency governance is mainly dyadic or compound by measuring bivariate correlations between ascribed importance to various institutions at the national and at the EU level. Three observations are made: First, we find strong, positive, and significant correlations between ascribed importance of EU-level institutions, with all the correlations being well above .700. Not surprisingly, the highest Table 5.3  Inter-correlations matrix on the importance of institutions (Pearson’s r) 1. 1. Own organization 2. Parent ministry 3. Other ministries 4. National Parliament 5. Government 6. EU Commission 7. EU agencies 8. EU Council 9. EU Parliament

2.

3.

4.

5.

6.

7.

8.

9.

1 .458** 1 .218** .329** 1 .238** .494** .309** 1 .238** .533** .335** .841** 1 .030

.093*

.124** .196** .142** 1

.015

.079

.103** .151** .116** .770** 1

.037

.176** .159** .264** .214** .801** .799** 1

.026

.160** .156** .245** .182** .767** .773** .918** 1

*Correlation is significant at the 0.01 level (2-tailed) **Correlation is significant at the 0.05 level (2-tailed)

118 

J. TRONDAL

correlation in this group is between political institutions at EU level, namely the European Parliament and the EU Council (correlation of .918). Secondly, we also find strong, positive, and significant correlations between ascribed importance of institutions at the national level, albeit these are not as high as in the EU-level cluster. Among national-level institutions, correlation between the government and the national parliament stands out as highest (correlation of .841). This is followed by correlation between the government and parent ministry at .533. The remaining correlations in this group are also fairly high, ranging from .218 to .494. Third and most importantly, the last set of correlations are those across levels of governance. Overall, significant correlations are lower than in the two previous groups, yet most are positive and significant. The highest correlation found is between ascribed importance of the national parliament and the EU council (.264). Most of the remaining correlations are significant and positive, yet moderate. In sum, we may draw the following two conclusions: First, there are relatively strong positive correlations between ascribing importance to institutions at the same level of governance. This relationship is strongest when it comes to EU-level institutions. Moreover, at the national level, agency officials who ascribe importance to own organization are most likely to ascribe importance to own parent ministry, while other ministries and political institutions are deemed less important. This observation is largely reflecting the fact that government agencies are organized at arm’s length from their parent ministries, entrusting agencies with room of maneuver from political steering. Such vertical specialization ensures that agencies are situated at relative distance from political institutions. Moreover, horizontal specialization ensures that agencies operating in different policy domains operate in relative isolation from one another. Consequently, the parent ministry becomes the important institution for a majority of agency officials. At the EU level, the distinction between administrative and political institutions is less clear to domestic agency officials, although the correlation between the EU council and European parliament stands out (.918). The second conclusion we can draw is that while we observe relatively strong correlations within each level of government, we observe relatively lower cross-level correlations. Nonetheless, contrary to the ideal-model Image I, the data does not demonstrate an adverse relationship between governance dynamics across levels. Agency officials may regard both national and European institutions as being of importance, yet it is more likely that they favor one set of institutions over

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

119

the other. Translated into theoretical terms, Table 5.3 is overall in accordance with the Image II, albeit insignificant and low cross-level correlation may arguably indicate moderate dyadic tendencies. While Table 5.3 serves as a starting point in the analysis of the alleged steering dilemma, the succeeding tables (Tables 5.4, 5.5, 5.6, and 5.7) examine conditions that might push agency officials toward dyadic or compound directions. Models 3 and 4 test the relationship between ascribed importance of own parent ministry (dependent variable 1) and the importance of three independent variables: the importance ascribed to the Commission, the importance ascribed to EU agencies, and the degree to which own policy area is affected by EEA/EU/Schengen (affectedness), with the list of moderators. Tables 5.5 and 5.6 test the relationship between the ascribed importance of the government (dependent variable 2) and the same list of independent variables and moderators. Each table contains six models in which we regress the independent variables on seven different moderators. A first observation is that ‘affectedness’ significantly moderates the impact of the independent variable, suggesting that the effect of the supranational institution on agency governance is contingent on the degree to which officials are affected by EU/EEA/Schengen. Table 5.3 shows two significant interaction effects: Administrative capacity is significant and close to the 0.05 level (1-tailed). Moreover, organizational duplication is significant at the 0.05 level (2-tailed). From this we learn that the effect of the Commission on the parent ministry is dependent on administrative capacity and organizational duplication. Similarly, the data reveals that organizational duplication renders significant interaction effects also in Table 5.4, suggesting that the same holds true for the effect of EU agencies on the parent ministry. Table 5.5 tests the relationship between the Commission and the national government revealing significant interaction effects of trust. This suggests that the effect of the Commission on the national government depends on significant trust relationships between agencies and their parent ministry. The five remaining moderator variables show no significant effects. Lastly, Table 5.6 reveals that the effect of EU agencies on the national government depends on formal rules and trust, albeit with the latter being merely significant at the 0.05 level in a 1-tailed test. Table 5.8 summarizes selected results by highlighting the significant moderating variables as found across the four main models, where two different dependent variables are regressed on two independent variables, respectively. As illustrated in Table  5.8, Tables 5.4, 5.5, 5.6, and 5.7

Constant Importance of the Commission Degree to which own policy area is affected by EEA/EU/Schengen (affectedness) Interaction importance of the Commission affectedness Degree to which own policy field requires cooperation across and between levels of governance (cooperation across levels) Interaction importance of the Commission cooperation across levels Administrative capacity to ensure coordination and collaboration between relevant actors (admin capacity) Interaction importance of the Commission administrative capacity Degree to which there are formal rules on how to conduct tasks (rules) 0.035

–0.059a

0.211

0.077**

0.071** 0.087

1.439 0.081 –0.310**

Model 2: Administrative capacity

1.697 –0.117 –0.318**

Model 1: Cross-level cooperation

–0.078

0.076**

2.163 –0.081 –0.324**

Model 3: Formal rules

0.066**

1.767 –0.143 –0.284**

Model 4: Organizational duplication

0.077**

1.852 –0.051 0.0325**

Model 5: Trust

0.065**

1.608 –0.095 –0.295**

Model 6: Politicization

Table 5.4  Relationship between perceived importance of the Commission and the perceived importance of own parent ministry (dependent variable), with interaction effects: Multivariate regression models

120  J. TRONDAL

a

Significant at the 0.05 level (one-tailed)

** Correlation is significant at the 0.05 level (2-tailed)

* Correlation is significant at the 0.01 level (2-tailed)

Interaction importance of EU rules Organizational duplication Interaction importance of EU organizational duplication Perceived trust between own agency and parent ministry (trust) Interaction importance of EU Commission trust Degree of politicization of own policy field Interaction importance of the Commission politicization

0.015 0.074 0.046*

–0.008

0.080

0.027

0.127

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

121

Constant Importance of EU agencies Degree to which own policy area is affected by EEA/EU/Schengen (affectedness) Interaction importance of EU agencies affectedness Degree to which own policy field requires cooperation across and between levels of governance (cooperation across levels) Interaction importance of EU agencies cooperation across levels Administrative capacity to ensure coordination and collaboration between relevant actors (admin capacity) Interaction importance of EU agencies administrative capacity Degree to which there are formal rules on how to conduct tasks (rules) 0.029

–0.056

0.210

0.069**

0.058** 0.121

1.413 0.072 –0.277**

Model 2: Administrative capacity

1.496 –0.079 –0.268**

Model 1: Cross-level cooperation

–0.069

0.065**

2.080 –0.082 –0.281**

Model 3: Formal rules

0.057**

1.868 –0.157 –0.256**

Model 4: Organizational duplication

0.060**

1.863 –0.058 –0.263**

Model 5: Trust

0.058**

1.405 –0.045 –0.0264**

Model 6: Politicization

Table 5.5  Relationship between perceived importance of EU agencies and the perceived importance of own parent ministry (dependent variable), with interaction effects: Multivariate regression models

122  J. TRONDAL

* Correlation is significant at the 0.01 level (2-tailed) ** Correlation is significant at the 0.05 level (2-tailed)

Interaction importance of EU agencies rules Organizational duplication Interaction importance of EU agencies organizational duplication Perceived trust between own agency and parent ministry (trust) Interaction importance of EU agencies trust Degree of politicization of own policy field Interaction importance of EU agencies politicization

0.018 0.013 0.056*

0.013

0.010

0.008

0.194

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

123

Constant Importance of the Commission Degree to which own policy area is affected by EEA/EU/Schengen (affectedness) Interaction importance of the Commission affectedness Degree to which own policy field requires cooperation across and between levels of governance (cooperation across levels) Interaction importance of the Commission cooperation across levels Administrative capacity to ensure coordination and collaboration between relevant actors (admin capacity) Interaction importance of the Commission administrative capacity Degree to which there are formal rules on how to conduct tasks (rules) 0.016

–0.022

0.121

0.125**

0.120** 0.199

2.305 0.000 –0.560**

Model 2: Administrative capacity

2.2021 –0.048 –0.565**

Model 1: Cross-level cooperation

–0.063

0.119**

2.720 –0.051 –0.560**

Model 3: Formal rules

0.107**

1.934 –0.015 –0.504**

Model 4: Organizational duplication

0.127**

1.843 0.157 –0.576**

Model 5: Trust

0.108**

1.636 0.021 –0.525**

Model 6: Politicization

Table 5.6  Relationship between perceived importance of the Commission and the perceived importance of the government (dependent variable), with interaction effects: Multivariate regression models

124  J. TRONDAL

* Correlation is significant at the 0.01 level (2-tailed) ** Correlation is significant at the 0.05 level (2-tailed)

Interaction importance of EU rules Organizational duplication Interaction importance of EU organizational duplication Perceived trust between own agency and parent ministry (trust) Interaction importance of the Commission trust Degree of politicization of policy field Interaction importance of the Commission politicization

0.014 0.258 0.003

–0.102*

0.398

–0.005

0.343

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

125

Constant Importance of EU agencies Degree to which own policy area is affected by EEA/EU/Schengen (affectedness) Interaction importance of EU agencies affectedness Degree to which own policy field requires cooperation across and between levels of governance (cooperation across levels) Interaction importance of EU agencies cooperation across levels Administrative capacity to ensure coordination and collaboration between relevant actors (admin capacity) Interaction importance of EU agencies administrative capacity Degree to which there are formal rules on how to conduct tasks (rules) –0.025

–0.041

0.223

0.118**

0.106** 0.372

2.077 0.032 –0.522**

Model 2: Administrative capacity

1.475 0.076 –0.503**

Model 1: Cross-level cooperation

–0.372

0.111**

3.519 –0.277 –0.518**

Model 3: Formal rules

0.097**

1.591 0.079 –0.454**

Model 4: Organizational duplication

0.121**

1.911 0.103 –0.538**

Model 5: Trust

0.103**

1.676 –0.018 –0.489**

Model 6: Politicization

Table 5.7  Relationship between perceived importance of EU agencies and perceived importance of the government (dependent variable), with interaction effects: Multivariate regression models

126  J. TRONDAL

* Correlation is significant at the 0.01 level (2-tailed) ** Correlation is significant at the 0.05 level (2-tailed) *** Correlation is significant at the 0.05 level (1-tailed)

Interaction importance of EU agencies rules Organizational duplication Interaction importance of EU agencies organizational duplication Perceived trust between own agency and parent ministry (trust) Interaction importance of EU agencies trust Degree of politicization of own policy field Interaction importance of EU agencies politicization

0.097** 0.415 –0.043

–0.085

0.382

–0.003

0.357

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

127

128 

J. TRONDAL

Table 5.8  Summary of findingsa

Includes only 2-tailed significant values

a

revealed four significant moderators. The moderating effect of affectedness was consistent across all models. In addition, organizational duplication, perceived trust, and formal rules were also significant moderators in one or more models. Graphical plots are needed for further interpretation of the interaction effects. To illustrate and examine the conditional nature of agency governance, Figs. 5.1, 5.2, 5.3, and 5.4 (in the Annex of the chapter) plot the interaction effects from Tables 5.4, 5.5, 5.6, and 5.7. As the plotted lines in Figs. 5.1, 5.2, 5.3, and 5.4 suggest, the relationship between the dependent variables and the independent variables varies across categories of the moderating variables. Each line represents the relationship within a category. Interactions occur whenever lines are not parallel. Unparallel lines may indicate significant and thus important interactions in which a compound or dyadic tendency is amplified or muted. Moreover, if lines are cross-cutting to the extent that they point in different directions, a relationship may change diametrically, from compound to dyadic (or vice versa) within the categories of the moderating variable. To be specific, a compound pattern is recognized by lines pointing upward, which indicates a positive relationship, while lines pointing downward reflect a negative relationship conforming to a dyadic pattern. Figure  5.1a, b in the Annex are graphical illustrations of the moderating effects of affectedness and organizational duplication, respectively. Figure  5.1a illustrates how

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

129

the relationship between importance ascribed to the parent ministry and the Commission varies across the range of degrees in affectedness. Crossing lines illustrate a significant interaction effect and, as the lines demonstrate, a more compound relationship emerges as affectedness decreases. Likewise, the compound image decreases as affectedness increases. From this we may infer that the effect of EU-level institutions is conditioned by the degree of affectedness. Figure 5.1b demonstrates the moderating effect of organizational duplication. While the interaction is less prominent, the lines illustrate that a compound pattern emerges as duplication increases. This indicates that increasing degree of organizational duplication also increases agencies’ ability to act compound. Figure 5.2a, b in the Annex demonstrates that the previous findings are robust also on the perceived importance of EU agencies as dependent variable. Still, the interaction effects remain almost identical. As perceived affectedness increases, the relationship becomes less compound. And likewise, organizational duplication reinforces the compound pattern. Although less evident, the moderating effect of affectedness remains identical when the dependent variable is changed to perceived importance of government: The compound pattern still decreases as affectedness increases. Another moderator that becomes significant is perceived trust between own agency and parent ministry. As the lines illustrate, a compound tendency becomes more evident as trust increases. Among those reporting lower levels of trust, the relationship weakens. This indicates that higher levels of trust between agencies and ministries increase the stronghold of Image II. Finally, when plotting the relationship between perceived importance of EU agencies and the perceived importance of the government, the moderating effect of affectedness is still robust. The compound pattern becomes more evident as affectedness decreases. The fourth moderator that turns out significant is the degree to which agency officials report clear and formal rules on how to conduct tasks. In this context, a compound pattern is evident when clear and established rules are present. Where rules are established in the relationship, a compound tendency tends to diminish.

130 

J. TRONDAL

Conclusions This chapter has outlined a framework for analyzing agency governance in integrated administrative systems. Empirically, data conveys a largely compound (Image II) image of agency governance processes under these conditions. Moreover, the compound image of agency governance is shown to be robust since the moderator variables are not able to fundamentally change relationships from one governance type to another. These findings align with recent studies that have shown the compound roles of public agencies and their staff (e.g., Bach & Ruffing, 2013; Egeberg, 2006; Trondal, 2011; van Dorp & ’t Hart, 2019). Our findings, however, challenge the alleged ‘coordination dilemma’ (Image I) facing agencies embedded in multilevel structures. The chapter also suggests that four significant moderators influence agency governance consistently across all models (Figs. 5.1, 5.2, 5.3, and 5.4). In addition, organizational duplication, perceived trust, and formal rules were also significant moderators in one or more models. However, whereas these moderators bias agency governance toward Image I and II, they do not change relationships diametrically from one governance type to another. One implication of this finding is that agency governance is fairly robust. The affiliated status grants the Norwegian central administration privileged access to most parts of the EU administration, which in turn paves the way for deep administrative integration at agency level. Agency-driven administrative integration across levels of governance mobilizes an administrative bias toward agency bodies in strongly affiliated nonmember states, which in effect fuels an ‘administrative state’ (Waldo, 1952) much more than a ‘policy state’ (Orren & Skowronek, 2017). Nonetheless, rather than a binary understanding of agency governance, this chapter supports Image II characterized by agencies making compromises and displaying skills to navigate conflicting concerns.

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

131

Annex

Fig. 5.1  (a, b) Relationship between perceived importance of the Commission and the perceived importance of own parent ministry (dependent variable), with interaction effects

132 

J. TRONDAL

Fig. 5.2  (a, b) Relationship between perceived importance of EU agencies and the perceived importance of own parent ministry (dependent variable), with interaction effects

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

133

Fig. 5.3  (a, b) Relationship between perceived importance of the Commission and the perceived importance of the government (dependent variable), with interaction effects

134 

J. TRONDAL

Fig. 5.4  (a, b) Relationship between perceived importance of EU agencies and perceived importance of the government (dependent variable), with interaction effects

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

135

References Ansell, C., & Trondal, J. (2018). Governing turbulence: An organizational-­ institutional agenda. Perspectives on Public Management and Governance, 1(1), 43–57. Ansell, C., Trondal, J., & Ogard, M. (Eds.). (2017). Governance in turbulent times. Oxford University Press. Bach, T., & Ruffing, E. (2013). Networking for autonomy? National agencies in European networks. Public Administration, 91(3), 712–726. Bach, T., & Ruffing, E. (2018). The Transformative effects of transnational administrative coordination in the European multi-level system. In E. Ongaro & S. Van Thiel (Eds.), The Palgrave of public administration and management in Europe. Palgrave Macmillan. Bark, T., & Bell, E. (2019). Issue prioritization by bureaucratic leaders: The influence of institutional structure. Administration & Society, 51(6), 915–950. Bauer, M.  W., & Becker, S. (2014). The unexpected winner of the crisis: The European Commission’s strengthened role in economic governance. Journal of European Integration, 36(3), 213–229. Bauer, M. W., & Trondal, J. (Eds.). (2015). The Palgrave handbook of the European administrative system. Palgrave Macmillan. Bulmer, S., & Joseph, J. (2016). European integration in crisis? Of supranational integration, hegemonic projects and domestic politics. European Journal of International Relations, 22(4), 725–748. Christensen, T., & Lægreid, P. (2008). The challenge of coordination in central government organizations: The Norwegian case. Public Organization Review, 68, 97–116. Curtin, D., & Egeberg, M. (2008). Tradition and innovation: Europe’s accumulated executive order. West European Politics, 31(4), 639–661. Cyert, R. M., & March, J. G. (1963). A behavioral theory of the firm. Prentice-Hall. Dahl, R. (2000). On democracy. Yale University Press. Danielsen, O.  A., & Yesilkagit, K. (2014). The effects of European regulatory networks on the bureaucratic autonomy of national regulatory authorities. Public Organization Review, 14(3), 353–371. Dehousse, R. (2008). Delegation of powers in the European Union: the need for a multi-principals model. West European Politics, 31(4), 789–805. Easton, D. (1965). A systems analysis of political life. Wiley. Egeberg, M. (Ed.). (2006). Multilevel union administration. Palgrave Macmillan. Egeberg, M., & Trondal, J. (1999). Differentiated integration in Europe: The case of the EEA country Norway. Journal of Common Market Studies, 37(1), 133–142. Egeberg, M., & Trondal, J. (2015). National administrative sovereignty – under pressure. In E. O. Eriksen & J. E. Fossum (Eds.), The European Union’s non-­ members. Routledge.

136 

J. TRONDAL

Egeberg, M., & Trondal, J. (2016). Why strong coordination at one level of government is incompatible with strong coordination across levels (and how to live with it). The case of the European Union. Public Administration, 94(3), 579–592. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies: What have we learnt (and where do we go from here). Journal of Common Market Studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Emery, Y., & Giauque, D. (2014). The hybrid universe of public administration in the 21st century. International Review of Administrative Sciences, 80, 23–32. Fossum, J.  E. (2019a). The unbearable lightness of differentiated integration. Manuscript, ARENA Centre for European Studies, University of Oslo. Fossum, J. E. (2019b). What is the Norway model? Mode of affiliation or political compromise? The Political Quarterly. https://doi.org/10.1111/1467 -­923X.12684 Fossum, J.  E., & Graver, H.  P. (2018). Squaring the circle on Brexit. Bristol University Press. Fossum, J. E., & Jachtenfuchs, M. (2017). Federal challenges and challenges to federalism. Insights from the EU and federal states. Journal of European Public Policy, 24(4), 467–485. Genschel, P., & Jachtenfuchs, M. (Eds.). (2014). Beyond the regulatory polity? Oxford University Press. Goetz, K. H., & Meyer-Sahling, J. -H. (2008). The Europeanization of national political systems: parliaments and executives. Living Reviews in European Governance. http://www.europeangovernance-­livingreviews.org/Articles/ lreg-­2008-­2/. Grindle, M.  S. (2017). Good governance, R.I.P.: A critique and an alternative. Governance, 30, 17–22. Gunnell, J. G. (2011). David Easton and the road to system. Paper presented at the ECPR General Conference, Reykjavik, Iceland. Heidbreder, E. (2011). Structuring the European administrative space: policy instruments of multi-level administration. Journal of European Public Policy, 18(5), 709–727. Hofmann, H.  C. H. (2008). Mapping the European administrative space. West European Politics, 31(4), 662–676. Hofmann, H. C. H., & Turk, A. (Eds.). (2006). EU administrative governance. Edward Elgar. Hooghe, L., & Marks, G. (2016). Community, scale, and regional governance. Oxford University Press.

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

137

Howlett, M., & Mukherjee, I. (2018). The contribution of comparative policy analysis to policy design: Articulating principles of effectiveness and clarifying design spaces. Journal of Comparative Policy Analysis, 20(1), 72–87. Jacobsson, B., Pierre, J., & Sundstrom, G. (2015). Governing the embedded state. Oxford University Press. Jann, W., & Wegrich, K. (2019). Generalists and specialists in executive politics: Why ambitious meta-politics so often fail. Public Administration, 97(4), 845–860. Jones, B. D., & Baumgartner, F. R. (2005). The politics of attention. The University of Chicago Press. Kassim, H., Connolly, S., Dehousse, R., Rozenberg, O., & Benjaballah, S. (2017). Managing the house: The presidency, agenda control and policy activism. Journal of European Public Policy, 24(5), 653–674. Kühn, N. S., & Trondal, J. (2018). European integration and the administrative state a longitudinal study on self-reinforcing administrative bias. Journal of European Public Policy, 26(9), 1373–1394. Lægreid, P., Sarapuu, K., Rykkja, L., & Randma-Liiv, T. (Eds.). (2014). Organizing for coordination in the public sector. Palgrave Macmillan. Landau, M. (1969). Redundancy, rationality, and the problem of duplication and overlap. Public Administration Review, 29(4), 346–358. Lavenex, S. (2009). EU rules beyond EU borders: theorizing external governance in European politics. Journal of European Public Policy, 16(6), 791–812. Leruth, B., Gänzle, S., & Trondal, J. (2019). Exploring differentiated integration in a Post-Brexit European Union. Journal of Common Market Studies, 57(5), 1013–1030. Leruth, B., Gänzle, S., & Trondal, J. (Eds.). (2022). The Routledge handbook of differentiation in the European Union. Routledge. Levi-Faur, D. (2011). Regulatory networks and regulatory agencification: toward a single European regulatory space. Journal of European Public Policy, 18(6), 810–829. Lijphart, A. (1968). The politics of accommodation: Pluralism and democracy in the Netherlands. University of California Press. MacCarthaigh, M., & Molenveld, A. (2018). Coordination in Europe. In E. Ongaro & S. Van Thiel (Eds.), The Palgrave of public administration and management in Europe. Palgrave Macmillan. Mathieu, E., & Rangoni, B. (2019). Balancing experimentalist and hierarchical governance in European Union electricity and telecommunications regulation: A matter of degrees. Regulation & Governance, 13(4), 577–592. Metcalfe, L. (1994). International policy co-ordination and public management reform. International Review of Administrative Sciences, 60, 271–290.

138 

J. TRONDAL

Miller, E. F. (1971). David Easton’s political theory. The Political Science Reviewer, 1, 184–235. Moravcsik, A. (2018). Preferences, power and institutions in 21st-century Europe. Journal of Common Market Studies, 56(7), 1648–1674. Olsen, J. P. (2007). Europe in search of political order. Oxford Univ. Press. Olsen, J.  P. (2008). The ups and downs of bureaucratic organization. Annual Review of Political Science, 11, 13–37. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford University Press. Olsen, J. P. (2018). Democratic accountability and the changing European political order. European Law Journal, 24(1), 77–98. Orren, K., & Skowronek, S. (2017). The policy state. Harvard University Press. Pollitt, C. (2016). Advanced introduction to public management and administration. Edward Elgar. Rittberger, B., & Wonka, A. (2011). Introduction: agency governance in the European Union. Journal of European Public Policy, 18(6), 780–789. Rokkan, S. (1999). State formation, nation-building, and mass politics in Europe. Oxford University Press. Rosenau, J. N. (1990). Turbulence in world politics. Princeton University Press. Schmidt, V. A. (2018). Rethinking EU governance: From ‘old’ to ‘new’ approaches to who steers integration. Journal of Common Market Studies, 56(7), 1544–1561. Shapiro, I., Skowronek, S., & Galvin, D. (Eds.). (2006). Rethinking political institutions. New York University Press. Thaler, R. H., & Sunstein, C. R. (2009). Nudge. Improving decisions about health, wealth and happiness. Penguin Books. Trein, P., & Maggetti, M. (2018). Reasserting the center of the nation state: How supranational institutions reshape the public sector. Manuscript, University of Lausanne. Trondal, J. (2010). An emergent European Executive Order. Oxford University Press. Trondal, J. (2011). Domestic agencies in an emergent European Executive Order. Journal of European Integration, 33(1), 55–74. Trondal, J., Marcussen, M., Larsson, T., & Veggeland, F. (2010). Unpacking international organisations. Manchester University Press. Trondal, J., & Peters, B. G. (2013). The rise of European administrative space. Lessons learned. Journal of European Public Policy, 20(2), 295–307. van Dorp, E.-J., & ’t Hart, P. (2019). Navigating the dichotomy: The top public servant’s Role. Public Administration, 97(4), 877–891. Vantaggiato, F. P. (2020). Regulatory relationships across level of multilevel governance systems: From collaboration to competition. Governance, 33(1), 173–189.

5  AGENCY GOVERNANCE IN INTEGRATED ADMINISTRATIVE SYSTEMS 

139

Verhoest, K., van Thiel, S., Bouckaert, G., & Lægreid, P. (Eds.). (2012). Government agencies. Palgrave. Vibert, F. (2007). The rise of the unelected. Cambridge University Press. Vigoda-Gadot, E., & Mizrahi, S. (2014). Managing democracies in turbulent times. Springer. Waldo, D. (1952). Development of theory of democratic administration. American Political Science Review, 46(1), 81–103. Wessels, W. (1997). An ever closer fusion? A dynamic macropolitical view on integration processes. Journal of Common Market Studies, 35(2), 267–299. Wilson, W. (1887). The study of administration. Political Science Quarterly, 2(2), 197–222.

CHAPTER 6

Restructuring Public Governance in Integrated Administrative Systems

Government ministries and agencies serve as vital components of the core executive of states and play a fundamental role in the democratic governing of modern societies (Dunleavy & Rhodes, 1990; Orren & Skowronek, 2017; Vibert, 2007). In recent decades, however, architecture of government has sustained significant transformation, notably by embedding government institutions and governing processes of EU member states in multilevel federal structures (Wilson, 1887: 221). In effect, government civil servants carry out a ‘double-hatted’ role in the multilevel executive order of the EU (EMEO) (Egeberg, 2006; Trondal, 2010). Serving as key actors in implementing and enforcing EU rules, government officials personalize the EMEO by working in national ministries and agencies while also partaking in European administrative networks, dealing with EU agencies and the Commission and implementing EU rules on national ground. Facing choice-architectures that are multiple, overlapping, and sometimes incompatible, it is crucial to study the way civil servants cope with potentially conflicting demands (Ansell et al., 2017: 1; Easton, 1965; Gunnell, 2011; Miller, 1971; van Dorp and ’t Hart 2019). Departing from the observations of Chap. 5, this chapter takes the analysis a step further by asking, if being part of an integrated multilevel administrative system can create systematic biases by mobilizing actors’ discretionary attention toward certain problems and solutions, structuring patterns of

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_6

141

142 

J. TRONDAL

conflict and cooperation in certain ways, and enabling coordination and steering along certain dimensions rather than others? Complex choice-­ architectures may challenge conventional wisdom concerning the conditions for public governance in situations where events, demands, and support interact and change in highly variable, inconsistent, unexpected, or unpredictable ways (Ansell & Trondal, 2018; Rosenau, 1990). The chapter examines what choices government officials make when subject to contending influences and conflictual premises on how to maneuver, and tests how such choices are organizationally contingent. Parallel to Chap. 5, this chapter shows that being embedded in multiple institutional structures mobilizes pragmatist compound behavior among government officials characterized by compromises and abilities to navigate conflicting concerns in this multilevel structure. The study theoretically argues that primary and secondary organizational structures shape actors’ behavior in complex ways, but that secondary structures are far less significant. Public administration is conventionally portrayed as founded on a series of dichotomies (cf. Image I governance) (Ebinger et al., 2018; Egeberg & Trondal, 2016; Olsen, 2008; Orton & Weick, 1990; Trein et al., 2020). As an alternative, this chapter conceptualizes and empirically demonstrates how government ministries and agencies are driven by a pragmatist compound dynamic characterized by the co-existence of multiple premises for decision-making (cf. Image II governance). The chapter makes two particular contributions to extant literature: –– First, it benefits from an organizational approach to public governance and derives distinct predictions on how government officials are likely to maneuver when embedded in integrated multilevel administrative orders. The chapter argues that organizational factors are structuring elements in the policymaking process, and that these elements are powerful tools available to deliberate design. The chapter concludes by suggesting that hybrid structures may be particularly important in addressing challenges to the implementation of joint policies as addressed in this special issue. This approach suggests that public governance is positive sum in which officials evoke multiple premises for decision-­making. Coping with future policy challenges, including improving implementation and enforcement of common policy agendas, calls for knowledge about the possibilities for organizational design.

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

143

–– Second, a large-scale study probes the empirical robustness of the organizational basis for public governance. The study applies the large-N survey dataset on ministry and agency officials as outlined in Chap. 3 which enables the study of agency officials embedded in two parallel politico-administrative systems: the central administration of a unitary state (Norway) and the administrative system of a quasifederal order (the EU). The comprehensive dataset consists of officials employed in 16 ministries and 47 government agencies tasked with, among others, regulating and implementing public policy. The analysis demonstrates that ministry and agency officials feature pragmatist compound behavior characterized by compromises and abilities to navigate conflicting concerns—such as those of domestic governments and EU-level institutions. Essentially, the data suggests that this behavior is systematically structured by organizational factors. Moreover, to illustrate how the interaction effects unfold beyond the analysis provided in Chap. 5, this chapter probes the effects of the independent variables on the dependent variable at various levels of the moderating variable. The conditional and contextual nature of public governance is thus explored within the framework of a classic multiplicative interaction model. Overall, these models find that no moderators make dramatic effects by profoundly weakening causal relationships. Thus, the findings support the observations made in Chap. 5, demonstrating the robustness of public governance. Public governance in integrated administrative systems calls upon government officials to choose or balance competing concerns from different institutions. Integrated administrative systems may produce ambiguity about what problems, solutions, and consequences to attend to at any time, and what actors are deemed legitimate and efficient (Ansell et al., 2017; Bauer & Trondal, 2015; Olsen, 2008). Government officials who are engaged in two parallel domains of executive governance may experience opportunities and constraints because different institutions send diverse information, signals, and mandates (Dehousse, 2008; Egeberg & Trondal, 2018). The chapter thus contributes to a mounting literature on the changing role of public governance in an integrated multilevel European executive order (Bauer & Trondal, 2015; Goetz & Meyer-­ Sahling, 2008; Trondal, 2010). This literature has identified the varied emergence and design of EU-level agencies (Christensen & Nielsen, 2010;

144 

J. TRONDAL

Egeberg & Trondal, 2017; Groenleer, 2009; Levi-Faur, 2011; Rittberger & Wonka, 2011), the interconnected nature of EU-level and national-­ level ministries and agencies (Bach & Ruffing, 2018; Curtin & Egeberg, 2008; Egeberg & Trondal, 2018; Maggetti, 2014; Trein & Maggetti, 2018), and implications for the autonomy of the national policymaking process (Bach et al., 2015; Bach & Ruffing, 2018; Egeberg & Trondal, 2009). This literature has so far described how the European administrative system represents a multilevel and nested network administration (e.g., Bach & Ruffing, 2018) with administrative bodies at different levels of government being linked together in the performance of tasks (Hofmann & Turk, 2006). However, we lack large-N studies that demonstrate how administrative integration affect national public governance processes generally, and the organizational basis of robust governing processes in particular. Studying how domestic ministries and agencies are influenced by the EU is important for two central reasons. First, domestic public administration is crucial to the implementation and enforcement of EU politics (Egeberg & Trondal, 2009). Because the transposition of EU law remains an administrative process relatively isolated from political actors, it is essential to understand the prospects of implementation of federal (EU) law by domestic government bodies. Second, domestic decision-making processes are crucial parts of federal policymaking since they are intertwined with the multilevel choice-architectures of the EU (e.g., Groenleer, 2009). However, one challenge plaguing contemporary scholarship is how to understand the role of national ministries and agencies in the governing of a deeply integrated EU multilevel system (Danielsen & Yesilkagit, 2014; Egeberg, 2006; Egeberg & Trondal, 2016: 1017; Vantaggiato, 2019; Verhoest et  al., 2012). It has proven challenging to study and understand the complex role of national ministries and agencies in the multilevel policymaking process in the EU.  This chapter offers one distinct contribution to this diverse literature by examining how domestic government officials maneuver in a multilevel European administrative order. The study proceeds as follows: The next section outlines a twostep framework for analysis and propositions for empirical enquiry. The subsequent sections include data and methodology as well as an empirical analysis. The concluding discussion summarizes key findings and reflects on their wider implications to this body of literature.

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

145

Theorizing Image II Governance in Two Steps This section is brought in two steps: The first step outlines an organizational approach and derives a set of independent variables. The second step outlines a multiplicative interaction model and a set of moderator variables. This section departs from one key observation of Chap. 5 that public governance is positive sum by involving multiple actors and co-evolving institutional resources (Olsen, 2017). It perceives this as a condition and an inherent trait of public governance and thus that governance processes tend to run in continuums rather than dichotomies (Ansell & Trondal, 2018; Howlett & Mukherjee, 2018). One necessary implication is that governing institutions tend to manage conflicts and tolerate ambiguities as part of the governing process (Orren & Skowronek, 2017: 91). In light of this, public administration has been pictured as hybrid and compound (Emery & Giauque, 2014; Olsen, 2018). One example is when public administration is engaged in coevolving worlds of executive governance, for example when national agencies take on multiple roles or ‘hats’ when practicing EU law (Egeberg & Trondal, 2018). Similarly, a vast body of literature has pictured executive governance as characterized by the coexistence of decision-making dynamics (Christensen & Lægreid, 2008; Hooghe & Marks, 2016; Mathieu & Rangoni, 2019; Olsen, 2007). Accordingly, public administration faces complex and intertwined problems, solutions, actors, and decision-making arenas (Olsen, 2007; Shapiro et  al., 2006). The next subsection outlines the organizational basis for compound public governance. Theorizing Compound Governance An organizational approach is grounded in the assumption that endogenous organizational characteristics may explain how institutions work. Consequently, organizational factors help explain the compound behavior of government officials at the micro level, which in turn may have implications for public governance at macro level (Egeberg & Trondal, 2018; March & Olsen, 1989; Olsen, 2008; Trondal & Bauer, 2017). Contemporary literature hosts competing ideas on the role of organizations in politics and human behavior (Meier & Capers, 2012; Meyer & Rowan, 1977; Niskanen, 1971). While some observers ascribe lack of government action to political leaders’ lack of will, this chapter advocates that

146 

J. TRONDAL

political will is endogenously shaped by organizational positions which also enable (and constrain) action. This chapter follows what Johan Olsen (2018) coined the ‘Bergen approach’ in political science, emphasizing the organizational dimension of politics. An organizational structure regulates actors’ access to decision processes, broadly defines the interests and goals that are to be pursued, delimits the types of considerations and alternatives that should be treated as relevant, and establishes action capacity by assigning certain tasks to certain roles. It influences decision-making behavior by providing individuals with ‘a systematic and predictable selection of problems, solutions and choice opportunities’ (March & Olsen, 1976). The organizational structure thus makes some choices become more likely than others (e.g., Egeberg & Trondal, 2018). This happens by regulating actors’ access to decision situations, mobilizing attention to certain problems and solutions, structuring patterns of conflict and cooperation (and thus influencing power relationships), and enabling coordination and steering along certain dimensions rather than others. Essential to this chapter is examining how different structural characteristics of core executive institutions matter in this respect. Two structural variables are unpacked in the following: vertical specialization and organizational affiliations. Vertical Specialization Vertical specialization denotes division of responsibility and labor within and between levels of authority. This chapter focuses on vertical specialization between ministries and agencies and within each organization between officials at different positional levels. Hierarchy between organizations provides certain decision biases, for example, by offering a more general view in hierarchical superior units compared to lower ranked units. Vertical specialization between organizations denotes installing an organizational boundary between a superior and a subordinate unit. Agencification, the process whereby regulatory tasks are hived off from ministerial departments into semi-independent regulatory bodies, is a topical example. The New Public Management (NPM) wave that swept across OECD states during the 1980s and 1990s made pleas for greater autonomy, fragmentation, and proliferation of public administration institutions and systems. As one result, vertical specialization in the form of structural devolution led to semi-autonomous agencies enjoying ever more degrees of autonomy at both the national and the EU levels (e.g., Bezes et al., 2013; Verhoest et al., 2012). Government

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

147

agencies represent vertical fragmentation of the polity and a source of administrative capacities to solve regulatory challenges (Bach et al., 2015). They are organizational compromises between political steering, professional autonomy, and technical regulation (Christensen & Nielsen, 2010), allowing them to operate relatively insulated from political intervention, yet it also makes them more exposed to ‘capture’ from external organizations such as EU-level institutions (Egeberg & Trondal, 2017). The tension between policy coordination across levels of government and regional/local coordination is also a well-known pattern in federal states and central-local relations within unitary states (e.g., Fossum & Jachtenfuchs, 2017). It has been less recognized in the study of public administration. In this context, national agencies organized at arm’s length from parent ministerial departments and enjoying de facto autonomy vis-à-vis from these departments constitute an administrative infrastructure for ‘agency capture’ by EU bodies (Trondal & Peters, 2013). Consequently, national agencies become building blocks of a multilevel EU administration (Egeberg & Trondal, 2018). Hierarchy within ministries and agencies—or intraorganizational vertical specialization—is measured by officials’ position or rank (see below). Studies have demonstrated that leaders tend to identify with a larger part of the organization than those at lower levels. Leaders also interact more frequently across organizational units and are exposed to broader flows of information than their subordinates and thus may be more attentive to broader organizational perspectives than lower ranked personnel (Egeberg & Trondal, 2018). Higher ranked staff in both ministries and subordinated agencies are shown to be more attentive to political signals than lower ranked personnel (e.g., Christensen & Lægreid, 2008; Egeberg & Sætren, 1999). This implies that leaders are better equipped to take into consideration a wider set of goals, alternatives, and consequences when making choices (Egeberg & Trondal, 2018). It follows that government officials with lower ranks are more loosely coupled to the political leadership and have more local perspectives on task execution compared to higher ranked staff. This has one important implication: a relative degree of insulation from political leadership makes lower ranked officials more eligible receivers of impulses from EU-level institutions and processes. Two propositions are derived from the above discussion (and based on the choice of dependent variable as outlined in the section on methods):

148 

J. TRONDAL

P1: Ministerial officials are more likely to emphasize the importance of own government and/or ministry and less and less influenced by EU institutions (the Commission and/or EU agencies) compared to agency officials. P2: Officials at higher ranks are more likely to emphasize the importance of own government and/or ministry and less likely to be influenced by EU institutions (the Commission and/or EU agencies) compared to officials at lower ranks. Organizational Affiliation Top public servants often serve in multiple structures and have multiple affiliations (van Dorp & ’t Hart, 2019). In conceptual terms, we may conceive of organizational affiliations as consisting of primary (‘strong’) and secondary (‘weak’) structures—of both the macrostructure of the government and the organizational substructures (Bertels & Schulze-Gabrechten, 2021). Government officials that operate within these structures are likely to be exposed to several behavioral premises from both primary and secondary affiliations. A ‘primary structure’ is the macrostructure of government, defined as the structure to which participants are expected to devote most of their loyalty, time, and energy. A typical example would be a bureaucratic unit like a ministry or agency. A ‘secondary structure’ is defined as the structure to which participants are expected to be part-­ timers. These are thus ‘weak’ institutions and are unlikely to shape actors’ decision behavior to the same extent as primary structures (Egeberg & Trondal, 2018). Secondary structures are the substructures of government, including collegial bodies, committees, expert groups, and networks. The EU multilevel administrative system is comprised of a set of interconnected secondary structures—for example, European Administrative Networks and EU agencies. However, being secondary structures, we should not assume European Administrative Networks and EU agencies to shape actors’ decision behavior to the same extent as primary structures. Empirical studies show that interdepartmental committees, public-private governing arrangements, regional councils, Commission expert committees, comitology committees, and Council working groups improve interaction and coordination among actors, and erect trust relationships among the participants. However, the effects are moderate (Ansell & Gash, 2008; Egeberg et  al., 2003; Lægreid et  al., 2014). Moreover, a logic of primacy suggests that the primary affiliation is

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

149

likely to affect behavior more extensively than the secondary (March, 1994). One general proposition is derived from the above discussion: P3: When taking part in both primary and secondary structures, government officials are likely to be most influenced by primary structures. Moreover, separating between different sets of secondary structures, one additional proposition may be suggested: P4: Whereas government officials attending secondary structures within secondary structures at national level either within own ministry/ agency (‘within’) or with subnational authorities (‘below’) are likely to emphasize the importance of own government and/or ministry, government officials attending EU-level secondary structures (‘beyond’) are more likely to emphasize the importance of EU-level institutions (the Commission and/or EU agencies). Interaction Effects To test the robustness of the above causal propositions, the analysis also enters interaction effects. It is assumed that the effects of each independent variable as specified above might be conditioned by intervening variables. Building on contemporary studies in public governance literature (see Egeberg & Trondal, 2018), eight moderator variables are specified. (i) The degree to which policy fields are generally affected by the EU: Across different policy domains, there is substantial variation in the degree to which national ministries and agencies are generally affected by EU decision-making processes (Kühn & Trondal, 2018). Moderator one assumes that exposure to EU-level decision-making increases the likelihood that government officials will emphasize and pay heed to EU-level institutions. (ii) The degree to which there are clear and established formal rules on how to conduct tasks: Less formalized, loosely coupled decision-­ making premises generally increase the room for discretionary behavior (March & Olsen, 1976; Orton & Weick, 1990). By contrast, government officials subject to clear and established formal rules are likely to face smaller room for discretion. Moderator two suggests that government officials who are subject to weak and less

150 

J. TRONDAL

established formal rules are likely to emphasize the importance of EU-level institutions. (iii) Administrative capacity to ensure coordination and collaboration between relevant actors: Sufficient administrative capacity provides resources to direct attention to public problems, solutions, actors, and decision-making arenas within and across levels of government (Egeberg & Trondal, 2018). Moderator three assumes that limited administrative capacities are likely to bias attention toward the immediate environment of officials (Simon, 1957), which in this study is the national government apparatus. Ample administrative capacities, by contrast, are assumed to enhance the emphasis on a broader range of institutions, for example, EU-level institutions. (iv) Perceived trust between own agency and parent ministry: Trust is ‘central to legitimate democratic government, to the formation of public policy, and to its implementation’ (Ruscio, 1996: 461). Trust ensures compliance without coercion, in particular in cases where enforcement mechanisms are limited, and where risks thus are involved (Chanley et al., 2000; Scholz & Pinney, 1995). Moderator four suggests that when trust relationships are high between agencies and ministry, government officials are likely to emphasize the importance of own government and/or own ministry/agency, and relatively less by EU-level institutions. (v) The degree to which own policy area is subject to politicization: Studies show that government officials whose issue area is characterized by public debate and conflict are inclined to attach importance to a political steer from the top (Ebinger et al., 2018; Egeberg & Trondal, 2009). Moderator five suggests that politicization is positively associated with emphasizing the importance of own government and/or own ministry/agency, and negatively associated with emphasizing the importance of EU-level institutions. (vi) Seniority in the central administration: Studies suggest that seniority in office is positively associated with taking own organization into account when decisions are made (Ebinger et al., 2018; Egeberg & Trondal, 2018). Moderator six suggests that whereas tenured staff are more likely to emphasize the importance of own ministry/ agency as well as own government, staff with less seniority are more likely to emphasize the importance of EU-level institutions. (vii) Exit plans: Studies show that contracted temporary officials who have permanent exit plans tend to be more loosely affiliated to

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

151

­ rganizations than permanent staff with few or no exit plans (e.g., o Murdoch & Trondal, 2013). Moderator seven suggests that officials with plans of exiting current position are less likely to emphasize the importance of own government and/or ministry/agency and put relatively more emphasis on EU-level institutions. ( viii) Identification with national central administration: Studies demonstrate that top ranked officials tend to identify with a larger part of the organization than those at lower levels (Egeberg & Trondal, 2018). Studies also show that officials who identify with national central administrations tend to assign weight to national politico-­ administrative institutions and much less toward EU-level institutions (Egeberg & Trondal, 2016). Moderator eight suggests that officials’ identification with the central administration as a whole is positively associated with emphasizing the importance of own government and/or own ministry/agency, and correspondingly negatively associated with emphasizing the importance of EU-level institutions.

Empirical Strategy The dataset covers by far the most integrated nonmember of the EU (Norway). Like most core executives, the Norwegian central administration is organized into core-executive ministries and subordinated agencies. It thereby mirrors the structure of the Commission consisting of horizontally specialized Directorates-General with subordinate EU agencies. The Norwegian central administration is characterized by ministerial primacy where subordinated agencies are subject to political control and administrative accountability from the responsible minister and not primarily from a government collegium. Whilst Norwegian ministries are secretariats for the political leadership with planning and coordinating functions (Christensen & Lægreid, 2008), agencies are mainly responsible for advising ministries and being technical helpers but are also essential ingredients in the political processes of preparing policies and implementing and administering policies. Norway is not a formal EU member but is closely affiliated to the EU through more than 100 agreements. These agreements grant Norway privileged access to EU aquis and most parts of the EU administration, which in turn opens for administrative integration on par with EU member states (Fossum, 2019). Consequently, despite a lack of political representation in the (EU) Council and the European

152 

J. TRONDAL

Parliament, the Norwegian executive branch of government is tightly integrated with and influenced by the EU-level administrative institutions (Egeberg, 2006; Kühn & Trondal, 2018). This associated status with the EU may be regarded as territorially disintegrated but sectorally integrated system, making Norway a generalizable case of administrative integration. The study benefits from the large-N questionnaire survey that was outlined in Chap. 3, in which this chapter uses the most recent survey from 2016. This survey was completed at the ministry and agency level in the Norwegian central administration (a total of N  =  4285 which includes N = 2322 respondents in 16 ministries and N = 1963 among 47 agencies). To empirically evaluate and explore the theoretical propositions explicated above, a three-step procedure is applied. First, we use OLS regression to establish baseline models where four dependent variables are separately specified as a function of the same set of independent variables. Second, interaction terms are added to capture the complexity of variegated institutional contexts. Finally, and to illustrate how the interaction effects unfold, we probe the effects of the independent variables on the dependent variable at various levels of the moderating variable. The conditional and contextual nature of public governance are thus explored within the framework of a classic multiplicative interaction model: Y   0   X   Z    XZ   



Following the advice of Brambor et al. (2006), all constitutive terms are included in the model alongside the interaction term. The outcome variable Y is perceived importance of institutions at either the national level or the EU level, while the key independent variable X is organizational affiliation, rank, and participation in secondary structures. Various moderator variables Z are included to investigate whether, or to what extent, the strength and/or direction of the relationship between X and Y is conditional, that is, varies with the level of Z. A multiplicative interaction model thereby enables us to explore how robust the causal relationships exposed by the baseline model are under different institutional contexts. As such, the aim of adding moderating variables to the analysis is not to explain as much variance as possible but rather to investigate and reveal contextual conditions under which a causal relationship between an outcome and an independent variable unfolds. Caution is warranted when interpreting the coefficients of the independent variables, both in terms of

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

153

effect sizes and statistical significance. That statistical significance does not equate substantive significance (Bernardi et al., 2017; Ziliak & McCloskey, 2008), is especially the case when evaluating conditional hypotheses. A coefficient, significant or not, only captures the effect of X on Y when Z is zero. As the effect size and its significance vary across levels of the moderator Z, the coefficients must not be misinterpreted as unconditional marginal effects. To remedy this shortcoming, we estimate conditional effects at various levels of the moderators: at one standard deviation below the mean (low), at the mean (medium), and at one standard deviation above the mean (high), respectively. This enables us to track the effects along changing conditions, and thereby determine whether effects are strengthened, attenuated, or even flipped. To examine the conditional nature of the effects, and to get an overview of the multitude of interactions due to regressing four dependent variables on six independent variables and eight moderating variables, all relationships are presented and summarized to reveal various aspects of interest: Which relationships are robust and stable (or not) across varying moderating variables, that is, institutional contexts? Which dependent variables are affected by the complexity that moderating variables provide? And, which moderating variables are most frequently influencing the relationship between the dependent and independent variables? When using survey questionnaire as a method, there is the risk that common method variance (CMV) will bias the results of empirical analysis (Jakobsen & Jensen, 2015). However, this is less of an issue in more complex regression models involving multiple independent variables and particularly interaction terms. The presence of CMV cannot create interaction effects, only attenuate them (Siemsen et al., 2010). As such, CMV is less of an issue if the purpose is to examine interaction effects. If anything, finding significant interaction effects despite the influence of CMV should be taken as strong evidence that an interaction effect exists (Siemsen et al., 2010: 470). Although not being conclusive, a Harman single-factor test was conducted to detect whether, or to what extent, variance in the data can be attributed to a single factor (Tehsen et al., 2017). The test suggests that CMV is not a pervasive issue. Violating the homoskedasticity assumption of regression may invalidate statistical inferences, particularly by increasing the risk of making a Type 1 Error (Hayes & Cai, 2007). Since a Breusch-Pagan test revealed the presence of heteroskedasticity in the error terms, a heteroskedasticity consistent

154 

J. TRONDAL

standard error (HCSE) estimator was used in estimating the OLS regression parameters. In practice, this makes it more difficult for the coefficients to pass the significance test.

Empirical Demonstration This section provides two sets of analysis: Four OLS regression models are provided on each of the independent variables (Table 6.4) and multiple interaction models are summarized in Table 6.5, highlighting the extent to which the regression models are significantly moderated or not, and in which direction they are eventually moderated. First, however, Table 6.1 presents the independent and dependent variables of the study, including moderators. Two proxies are applied as dependent variables and two as independent variables. Whereas Table 6.1 predictively shows that national government institutions are deemed far more important than EU-level institutions, Table 6.2 provides a bivariate correlation matrix on the extent to which officials tend to mobilize compound behavior—that is emphasizing the importance of multiple institutions simultaneously. Table 6.2 indeed demonstrates compound behavioral patterns—particularly at each level of governance: Table 6.1  Descriptive statistics: Dependent, independent, and moderator variablesa

Missing values not included

a

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

155

Table 6.2  Intercorrelation matrix (Pearson’s r) 1 1. Own ministry/agency 2. Government 3. EU Commission 4. EU agencies

2

3

4

1 .213** .125**

1 .794**

1

1 .283** .002 –.025

** Significant at the 0.01 level (2-tailed)

Officials emphasizing the importance of own ministry/agency also tend to emphasize the role of the government, and those emphasizing the Commission tend to emphasize the role of EU agencies. Finally, whereas those emphasizing the government also tend to emphasize the role of the Commission, indicating ‘double-hatted’ officials (Egeberg & Trondal, 2016), we see no parallel correlations on the role of EU agencies. Next, Table 6.3 reveals how a set of independent variables influence the four independent variables, measuring ascribed importance to two national institutions (own ministry/own agency and the government) and two EU-level institutions (the Commission and EU agencies) respectively. Supporting P1, Table 6.3 first demonstrates that the independent variable measuring vertical specialization (ministry/agency) significantly affects all four dependent variables in a systematic and significant way. Ministry officials are generally more likely to ascribe importance to national governmental institutions than agency officials. By contrast, agency officials are more likely to ascribe importance to the EU level. Second, in support of P2, higher ranked officials are slightly more likely to ascribe importance to own ministry/agency than lower ranked officials. However, rank does not influence how officials weigh the importance of EU institutions. In sum, interorganizational vertical specialization (ministry/agency) has a significantly stronger effect on public governance than intraorganizational vertical specialization (rank). Second, P3 is also supported since the effects of secondary structures are generally weaker, varying and also conditioned by moderating variables (see Table  6.4). In short, the effects of secondary structures vary across different types of secondary structures. Moreover, as predicted by P4, different secondary structures (‘within,’ ‘below,’ and ‘beyond’) have different effects. In short, secondary structures ‘within’ the government structure tend to strengthen and fill in the role of primary structures, since participation in intraorganizational secondary structures (within) tends to

156 

J. TRONDAL

Table 6.3  Multivariate regression models (unstandardized coefficients) Independent variables

Dependent variables Model 1: Importance— Own ministry/ agency

Model 2: Importance— Government

Model 3: Model 4: Importance—EU Importance— Commission EU agencies

Constant Ministry— Agency (dummy) Rank Secondary structure— within Secondary structure— below Secondary structure— beyond (EU) R Square

4.325 0.402***

3.771 0.593***

2.905 –0.247***

0.029* 0.087**

0.068***

0.063**

0.206***

–0.184**

–0.119***

0.092

2.618 –0.525***

0.122

1.096***

0.744***

0.098

0.082

* Significant at the 0.05 level (1-tailed) ** Significant at the 0.05 level (2-tailed) *** Significant at the 0.01 level (2-tailed)

emphasize the influence of their own ministry/agency, and the influence of EU agencies as less important. However, participation in such structures does not affect how much officials ascribe the importance of their government or the Commission. Moreover, participating in interorganizational secondary structures with subordinate administrative bodies, such as working groups and project groups (below), increases ascribed importance to both own ministry/agency and the government, while the weight given to the EU institutions is not significantly affected. How officials ascribe importance to both the Commission and EU agencies is however influenced as predicted by participation in secondary structure at EU level (beyond), such as Commission expert committees. Officials participating in these EU-level committees tend to ascribe more importance to these institutions, as well as their own ministry/agency. However, participation beyond does not seem to affect importance to their government. In short and supporting P3, EU-level secondary structures do not seem to undermine the relevance of national-level primary structures.

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

157

Table 6.4  Multiple interaction models

* Significant at the 0.05 level (1-tailed) ** Significant at the 0.05 level (2-tailed) *** Significant at the 0.01 level (2-tailed)

Finally, while Table 6.3 serves as a departure point, Table 6.4 provides a bird’s eye view that allows us to examine patterns of complexity and/ or stability across and within the various regression models. In Table 6.4, while numbers within the parentheses are the unstandardized regression coefficients, numbers outside the parentheses are interaction effects.

158 

J. TRONDAL

Table 6.4 sheds light on several aspects. First, one can assess which, and to what degree, independent variables are robust or prone to moderation. Second, by comparing columns, one can examine which dependent variables are more frequently subject to moderated relationships. Finally, one can examine whether some moderating variables are more or less influential. To ease interpretation of the table, green cells indicate causal relationships that are reinforced or weakened by a particular moderating variable. White cells are also of interest. They indicate robust relationships that are unaffected by moderating variables, and thus robust. Grey cells correspond with nonsignificant causal relationships of the baseline model. One general observation is that moderators show significant effects in different regards, but we find that no moderators make dramatic effects by profoundly weakening causal relationships. One minor observation is that the moderator ‘seniority’ in office is never significant across the models. Thus, public governance is relatively structured and robust. Moreover, the statistically significant interactions of Table  6.5 are excerpted, and then expanded in Tables 6.5, 6.6, 6.7, and 6.8. While Table 6.4 highlighted statistically significant interaction terms to paint a broader picture of stability or complexity, the aim of Tables 6.5, 6.6, 6.7, and 6.8 is to shed further light on whether moderation matters, or, put differently, on whether different contextual conditions matter. As pointed out earlier, the extent to which moderation matters is not reducible to a question of statistical significance. Instead, we need to dig deeper and

Table 6.5  Conditional effects of the importance ascribed to own ministry/ agency

* Significant at the 0.05 level (1-tailed) ** Significant at the 0.05 level (2-tailed) *** Significant at the 0.01 level (2-tailed)

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

159

Table 6.6  Conditional effects on the importance ascribed to own government

** Significant at the 0.05 level (2-tailed) *** Significant at the 0.01 level (2-tailed)

Table 6.7  Conditional effect on the importance ascribed to the Commission

*** Significant at the 0.01 level (2-tailed)

Table 6.8  Conditional effects on the importance ascribed to EU agencies

*** Significant at the 0.01 level (2-tailed)

explore how the interactions unfold. For this purpose, conditional effects of the independent variables are estimated at three levels of the moderators. Of particular interest will be those interactions where the coefficients at some level of the moderator are either zero or change direction (red cells). Using that as a yardstick contributes to illuminate which relationships are moderated substantially rather than slightly. Hence, the more

160 

J. TRONDAL

cells colored red, the more relationships are moderated substantially, which in turn entails more complex causal trajectories. The red cells in Table 6.5 illustrate that the ascribed importance to own ministry/agency is most prone to substantial moderation. As evident, the effect of rank on ascribed importance vanishes at higher levels of both administrative capacity and trust. Also, only one moderating variable alters each of the otherwise stable relationships between participating in secondary structures—within, below, and beyond—and ascribing more importance to own ministry agency. However, for participation in secondary structures within, this effect only appears at low levels of identification and vanishes as identification reaches medium level. A similar pattern is evident regarding participation in secondary structures below. Here, the effect occurs at low and medium levels, but flips at high levels, of administrative capacity. Under the latter conditions, participation is associated with less ascribed importance to own ministry/agency. The effect of participating in secondary structures beyond is also affected by only one moderating variable. At low levels of trust those participating do not ascribe more importance to own ministry/agency, only at higher levels. Table 6.6 shows that ascribed importance to own government is the dependent variable most prone to moderated causal trajectories. Even so, the moderated effects show quite stable patterns, but we see no substantial moderating effects. The moderating variables either strengthen or attenuate already apparent causal effects. Next, Table  6.7 shows that ascribed importance to the Commission is the dependent variable least prone to moderated relationships, where only one relationship is significantly yet only slightly moderated due to identification with the national central administration. Finally, Table 6.8 shows that the ascribed importance of EU agencies is exposed to only one substantially moderated relationship. The tendency of ministry officials to ascribe less importance to EU agencies than what agency officials do appears only when the level of ‘affectedness’ increases.

Conclusions As a consequence of emerging EU-level institutional architectures that may influence public governance in national public administration, this chapter has outlined a framework for analyzing public governance in integrated administrative systems. It shows that organizational factors—both primary and secondary—are essential premises for behavioral processes in

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

161

government institutions. These findings align with recent studies that have shown the compound roles of public civil servants (e.g., Chap. 5 in this volume; Egeberg, 2006; Trondal, 2011; Van Dorp & ’t Hart, 2019). Moreover, the models presented in this study find that no moderators make dramatic effects by profoundly weakening causal relationships. Thus, public governance is relatively structured and robust. The affiliated status as an integrated EU nonmember state grants the Norwegian central administration privileged access to parts of the EU administration, which in turn paves the way for administrative integration of the national central administration. Previous studies have suggested that administrative integration across levels of governance mobilizes an administrative bias toward expert bodies (Kühn & Trondal, 2018), which may fuel the administrative state (Waldo, 1952). Nonetheless, this study demonstrates the hybrid nature of compound public governance characterized by ministries and agencies making compromises and displaying skills to navigate conflicting concerns between the national government, national ministries and agencies, the Commission, and EU-level agencies. An organizational approach derives design implications useful to tackle the agendas of governments (Egeberg & Trondal, 2018; Self, 1972: 290). Given the findings of this study, certain organizational factors are available for redesigning public policy, notably by reforming the vertical specialization of government systems. Governments face times of turbulence with complex organizational tradeoffs. Governments may choose a strategy of managing turbulence through vertical integration, such as incorporation of agencies into government ministries, or they may opt for differentiation by for example decentralizing power to agencies. At the extreme of integration, we may expect vertically integrated government organizations and as we move to the other extreme of the differentiation, we may expect a proliferation of smaller specialized agencies. Moreover, policy dilemmas might be addressed through hybrid organizational solutions that combine components from various organizational forms, for example, choosing both integration and differentiation in different governmental domains (Ansell & Trondal, 2017; Battilana & Lee, 2014; Brandsen & Karré, 2011; Minkoff, 2002; Skelcher & Smith, 2015). Importantly, hybridity helps to create structural flexibility necessary to respond to competing and varied demands, thus designing dynamic resilience (Ansell & Trondal, 2018). It takes advantage of the heterogeneity and pluralism of institutional environments (Kraatz & Block, 2008) by treating organizational forms as building blocks that can be combined in various ways (Battilana

162 

J. TRONDAL

& Lee, 2014). Hybrid institutional solutions may be particularly important institutional architectures for improving the implementation and enforcement of joint EU policies, notably by establishing and institutionalizing secondary structures that complement the primary structures of government (P3 and P4).

Appendix: Moderating Variables Variable

Original question in the survey

Affectedness Formal rules

To what extent do EU, EFTA, and/or Schengen affect your portfolio? Are there clear rules or well-established practices regarding the handling of your daily work? How would you assess the administrative capacity in the central administration within your portfolio as regards getting actors to collaborate How would you characterize the level of mutual trust between your own ministry and the subordinate agencies and bodies within your portfolio? How would you characterize the level of mutual trust between your own agency and the parent ministry? To what extent is your portfolio subject to public debate? For how long have you been employed in the central administration as a whole? Do you currently have plans or wishes of leaving current position and transfer to another profession? How strong or weak is your identification with the central administration as a whole?

Administrative capacity Trust

Politicization Seniority Exit plans Identification

References Ansell, C., & Gash, A. (2008). Collaborative governance in theory and practice. Journal of Public Administration Research and Theory, 18(4), 543–571. Ansell, C., & Trondal, J. (2017). Coping with turbulence. In C. Ansell, J. Trondal, & M. Ogard (Eds.), Governance in turbulent times. Oxford University Press. Ansell, C., & Trondal, J. (2018). Governing turbulence: An organizational-­ institutional agenda. Perspectives on Public Management and Governance, 1(1), 43–57. Ansell, C., Trondal, J., & Ogard, M. (Eds.). (2017). Governance in turbulent times. Oxford University Press. Bach, T., & Ruffing, E. (2018). The Transformative effects of transnational administrative coordination in the European multi-level system. In E. Ongaro

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

163

& S. Van Thiel (Eds.), The Palgrave of public administration and management in Europe. Palgrave Macmillan. Bach, T., Ruffing, E., & Yesilkagit, K. (2015). The differential empowering effects of Europeanization on the autonomy of national agencies. Governance, 28(3), 285–304. Battilana, J., & Lee, M. (2014). Advancing research on hybrid organizing— Insights from the study of social enterprises. Academy of Management Annals, 8(1), 397–441. Bauer, M. W., & Trondal, J. (Eds.). (2015). The Palgrave handbook of the European administrative system. Palgrave Macmillan. Bernardi, F., Chakaia, L., & Leopold, L. (2017). “Sing me a song with social significance”: The (mis)use of statistical significance testing in European sociological research. European Sociological Review, 33(1), 1–15. Bertels, J., & Schulze-Gabrechten, L. (2021). Mapping the black box of intraministerial organization: An analytical approach to explore structural diversity below the portfolio level. Governance, 43(1), 171–189. Bezes, P., Fimreite, A. L., Lidec, P. L., & Lægreid, P. (2013). Understanding organizational reforms in the modern state: Specialization and integration in Norway and France. Governance, 26(1), 147–175. Brambor, T., Clark, W.  R., & Golder, M. (2006). Understanding Interaction Models: Improving Empirical Analyses. Political Analysis, 14, 63–82. Brandsen, T., & Karré, P. M. (2011). Hybrid organizations: No cause for concern? International Journal of Public Administration, 34(13), 827–836. Chanley, V. A., Rudolph, T. J., & Rahn, W. M. (2000). The origins and consequences of public trust in government. A time series analysis. Public Opinion Quarterly, 64(3), 239–256. Christensen, J.  G., & Nielsen, V.  L. (2010). Administrative capacity, structural choice and the creation of EU agencies. Journal of European Public Policy, 17(2), 176–204. Christensen, T., & Lægreid, P. (2008). The challenge of coordination in central government organizations: The Norwegian case. Public Organization Review, 8(2), 97–116. Curtin, D., & Egeberg, M. (2008). Tradition and innovation: Europe’s accumulated executive order. West European Politics, 31(4), 639–661. Danielsen, O.  A., & Yesilkagit, K. (2014). The effects of European regulatory networks on the bureaucratic autonomy of national regulatory authorities. Public Organization Review, 14(3), 353–371. Dehousse, R. (2008). Delegation of powers in the European Union: the need for a multiprincipals model. West European Politics, 31(4), 789–805. Dunleavy, P., & Rhodes, R. A. W. (1990). Core executive studies in Britain. Public Administration, 68, 3–28. Easton, D. (1965). A systems analysis of political life. Wiley.

164 

J. TRONDAL

Ebinger, F., Sylvia, V., & Fromm, N. (2018). The partisan-professional dichotomy revisited: Politicization and decision-making of senior civil servants. Public Administration, 97(4), 861–876. Egeberg, M. (Ed.). (2006). Multilevel Union Administration. Palgrave Macmillan. Egeberg, M., & Sætren, H. (1999). Identities in complex organizations: A study of ministerial bureaucrats. In M.  Egeberg & P.  Lægreid (Eds.), Organizing political institutions. Scandinavian University Press. Egeberg, M., Schaefer, G. F., & Trondal, J. (2003). The many faces of EU committee governance. West European Politics, 26(3), 19–40. Egeberg, M., & Trondal, J. (2009). Political leadership and bureaucratic autonomy. Effect of agencification. Governance, 22(4), 673–688. Egeberg, M., & Trondal, J. (2016). Why strong coordination at one level of government is incompatible with strong coordination across levels (and how to live with it). The case of the European Union. Public Administration, 94(3), 579–592. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies: What have we learnt (and where do we go from here). Journal of Common Market Studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Emery, Y., & Giauque, D. (2014). The hybrid universe of public administration in the 21st century. International Review of Administrative Sciences, 80(1), 23–32. Fossum, J. E. (2019). What is the Norway Model? Mode of affiliation or political compromise? The Political Quarterly. https://doi.org/10.1111/1467-­ 923X.12684 Fossum, J. E., & Jachtenfuchs, M. (2017). Federal challenges and challenges to federalism. Insights from the EU and federal states. Journal of European Public Policy, 24(4), 467–485. Goetz, K. H., & Meyer-Sahling, J.-H. (2008). The Europeanization of national political systems: parliaments and executives. Living Reviews in European Governance, 3(2). Groenleer, M. (2009). The autonomy of European Union Agencies. Eburon. Gunnell, J. G. (2011). David Easton and the road to system. Paper presented at the ECPR General Conference, Reykjavik, Iceland. Hayes, A. F., & Cai, L. (2007). Using heteroscedasticity- consistent standard error estimators in OLS regression: An introduction and software implementation. Behavior Research Methods, 39(4), 709–722. Hofmann, H. C. H., & Turk, A. (Eds.). (2006). EU administrative governance. Edward Elgar. Hooghe, L., & Marks, G. (2016). Community, scale, and regional governance. Oxford University Press.

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

165

Howlett, M., & Mukherjee, I. (2018). The contribution of comparative policy analysis to policy design: Articulating principles of effectiveness and clarifying design spaces. Journal of Comparative Policy Analysis, 20(1), 72–87. Jakobsen, M., & Jensen, R. (2015). Common method bias in public management studies. International Public Management Journal, 18(1), 3–30. Kraatz, M. S., & Block, E. S. (2008). Organizational implications of institutional pluralism. In R. Greenwood, C. Oliver, T. B. Lawrence, & R. E. Meyer (Eds.), The Sage handbook of organizational institutionalism. SAGE. Kühn, N. S., & Trondal, J. (2018). European integration and the administrative state a longitudinal study on self-reinforcing administrative bias. Journal of European Public Policy, 26(9), 1373–1394. Lægreid, P., Sarapuu, K., Rykkja, L., & Randma-Liiv, T. (Eds.). (2014). Organizing for coordination in the public sector. Palgrave Macmillan. Levi-Faur, D. (2011). Regulatory networks and regulatory agencification: toward a single European regulatory space. Journal of European Public Policy, 18(6), 810–829. Maggetti, M. (2014). The rewards of cooperation: The effects of membership in European regulatory networks. European Journal of Political Research, 53(3), 480–499. March, J. G. (1994). A primer on decision-making. Free Press. March, J.  G., & Olsen, J.  P. (1976). Ambiguity and choice in organizations. Universitetsforlaget. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions: The organizational basis of politics. Free Press. Mathieu, E., & Rangoni, B. (2019). Balancing experimentalist and hierarchical governance in European Union electricity and telecommunications regulation: A matter of degrees. Regulation & Governance, 13(4), 577–592. Meier, K. J., & Capers, K. J. (2012). Representative bureaucracy: Four questions. In B. G. Peters & J. Pierre (Eds.), The SAGE handbook of public administration. SAGE. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Miller, E. F. (1971). David Easton’s political theory. The Political Science Reviewer, 1, 184–235. Minkoff, D. C. (2002). The emergence of hybrid organizational forms: Combining identity-based service provision and political action. Nonprofit and Voluntary Sector Quarterly, 31, 377–401. Murdoch, Z., & Trondal, J. (2013). Contracted government. Unveiling the European Commission’s contracted staff. West European Politics, 36(1), 1–21. Niskanen, W. (1971). Bureaucracy and representative government. Aldine. Olsen, J. P. (2007). Europe in search of political order. Oxford Univ. Press.

166 

J. TRONDAL

Olsen, J.  P. (2008). The ups and downs of bureaucratic organization. Annual Review of Political Science, 11, 13–37. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford University Press. Olsen, J. P. (2018). Democratic accountability and the changing European political order. European Law Journal, 24(1), 77–98. Orren, K., & Skowronek, S. (2017). The policy state. Harvard University Press. Orton, J. D., & Weick, K. E. (1990). Loosely coupled systems: a reconceptualization. Academy of Management Review, 15, 203–223. Rittberger, B., & Wonka, A. (2011). Introduction: agency governance in the European Union. Journal of European Public Policy, 18(6), 780–789. Rosenau, J. N. (1990). Turbulence in world politics. Princeton University Press. Ruscio, K.  P. (1996). Trust, democracy, and public management: A theoretical argument. Journal of Public Administration Research and Theory, 6(3), 461–477. Scholz, J. T., & Pinney, N. (1995). Duty, fear, and tax compliance: The heuristic basis of citizenship behavior. American Journal of Political Science, 39(2), 490–512. Self, P. (1972). Administrative theories and policies. George Allen & Unwin. Shapiro, I., Skowronek, S., & Galvin, D. (Eds.). (2006). Rethinking political institutions. New York University Press. Siemsen, E., Roth, A., & Oliveira, P. (2010). Common method bias in regression models with linear, quadratic, and interaction effects. Organizational Research Methods, 13(3), 456–476. Simon, S. A. (1957). Administrative behavior. The Free Press. Skelcher, C., & Smith, S.  R. (2015). Theorizing hybridity: Institutional logics, complex organizations, and actor identities: The case of nonprofits. Public Administration, 92, 433–448. Tehsen, S., Ramayah, T., & Sajilan, S. (2017). Testing and controlling for common method variance: A review of available methods. Journal of Management Sciences, 4(2), 146–175. Trein, P., Biesbroek, R., Bolognesi, T., Cejudo, G. M., Duffy, R., Hustedt, T., & Meyer, I. (2020). Policy coordination and integration. A research agenda. Public Administration Review. https://doi.org/10.1111/puar.13180 Trein, P., & Maggetti, M. (2018). Reasserting the center of the nation state: How supranational institutions reshape the public sector. Unpublished manuscript, University of Lausanne. Trondal, J. (2010). An emergent European Executive Order. Oxford University Press. Trondal, J. (2011). Domestic agencies in an emergent European Executive Order. Journal of European Integration, 33(1), 55–74. Trondal, J., & Bauer, M.  W. (2017). Conceptualizing the European multilevel administrative order. Capturing variation in the European administrative system. European Political Science Review, 9(1), 73–94.

6  RESTRUCTURING PUBLIC GOVERNANCE IN INTEGRATED… 

167

Trondal, J., & Peters, B. G. (2013). The rise of European administrative space. Lessons learned. Journal of European Public Policy, 20(2), 295–307. Van Dorp, E.-J., & ’t Hart, P. (2019). Navigating the dichotomy: The top public servant’s craft. Public Administration, 97(4), 877–891. Vantaggiato, F. P. (2019). Regulatory relationships across level of multilevel governance systems: From collaboration to competition. Governance, 31(1), 173–189. Verhoest, K., Van Thiel, S., Bouckaert, G., & Lægreid, P. (Eds.). (2012). Government agencies. Palgrave. Vibert, F. (2007). The rise of the unelected. Cambridge University Press. Waldo, D. (1952). Development of theory of democratic administration. American Political Science Review, 46(1), 81–103. Wilson, W. (1887). The study of administration. Political Science Quarterly, 2(2), 197–222. Ziliak, S.  T., & McCloskey, D.  N. (2008). The cult of statistical significance. University of Michigan Press.

CHAPTER 7

Designing the Administrative State

Times of turbulence call upon public organizations to adapt, anticipate, reform, and innovate. Public innovation has subsequently climbed to the top of government agendas with ambitions to make public administration flexible and agile (e.g., Ansell & Torfing, 2014: 3). Consequently, there is a growing body of literature on institutional absorptive capacity which tries to identify how institutions and systems respond to surprises, uncertainty, and errors (Hermus et  al., 2020; Schulman, 2022). Studies also provide insights on how different institutional conditions enable individuals and organizations to respond to uncertainty (e.g., Castellacci & Natera, 2013). Despite this, organizational factors are seen as a frequent barrier to administrative reform (Cinar et al., 2019; De Vries et al., 2015: 157). By contrast, this chapter argues that organization theory might help to build a theory of meta-governance, which may also serve as a bridge between theory and practice (Selznick, 1996; Stoker, 2013; Trein et al., 2021). An organizational approach to meta-governance serves to link scholarship to the realities of practice, concerned not just with how things are, but how things might be (Gulick, 1937; Meier, 2010: 284). Given certain goals, such as innovation in public organizations, organization designers would be capable of recommending structural solutions. Examining conditions for meta-governance is pertinent since governments experience frequent criticism of existing organizational arrangements and calls for major reforms of the state (Emery & Giauque, 2014; Lodge & Wegrich, 2012; © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_7

169

170 

J. TRONDAL

Torfing et al., 2012). Responding to a call to make the design orientation in public administration and policy studies more ‘designerly’ (Hermus et al., 2020: 36), this chapter outlines the contours of an organizational approach to meta-governance (Cinar et al., 2019; De Vries et al., 2015; Goodin, 1996; Lewis et al., 2020; Peters, 2018; Romme & Meijer, 2020). The research question motivating this chapter is how the organizational structuring of ongoing reform processes (meta-governance) may affect decision-making behavior within reform processes. Meta-governance encompasses deliberate attempts at governing by (re-)organizing governing structures (Egeberg & Trondal, 2018; Jacobsson et al., 2015). In this chapter, we focus on meta-governance as centered on the (re)structuring of ongoing reform processes. The illustrative example provided, based on existing literature, is on processes of policy innovation. Two arguments are made: (i) public meta-governance is an available tool to policy change and (ii) meta-governance may ultimately be shaped by organizational structuring. Aiming to make public policy and administration scholarship available to practitioners (Gulick, 1937; Pollitt, 2016; Raadschelders, 2013; Self, 1972; Stoker, 2013), this chapter offers an organization theory approach to meta-governance and illustrates how public organizations may organize policy change by (re-)designing organizational choice-architectures. Despite the rich scholarship on governance and organization theory, contemporary literature has lacked a theory-based analysis of the organizational dimension of meta-governance (Peters and Pierre, 2017). Whereas organizational studies have revealed the effects of organizational structures on decision-making behavior and processes, less focus has been given to how organizational structures of reforms may affect the actual reform processes. Despite decades of studying governance and organizations, these literatures have remained largely disconnected, as illustrated by the Oxford Handbook of Governance, which lacks a chapter on organization theory (Levi-Faur, 2012). This is also reflected by two decades of core-executive governance scholarship that largely has abandoned the organizational dimension (Elgie, 2011). Moving beyond a ‘policy tools’ literature (Howlett, 2011; Howlett et al., 2018), this chapter outlines the contours of a theory of how organizational factors may shape meta-­ governance processes. In his book, The Tools of Government, Hood (1983: 4–7) distinguished among four basic government ‘resources’: ‘Nodality,’ ‘treasure,’ ‘authority,’ and ‘organization.’ An organizational approach to meta-governance zooms in on the organizational dimension. It is thus argued that the relevance of organizational theory may range from understanding organizational conditions for mundane administrative reforms

7  DESIGNING THE ADMINISTRATIVE STATE 

171

(Durant, 2019) to organizational conditions for large-scale organizational transformations (Ansell & Lindvall, 2021). Yet, our analytical model is limited to mesolevel/organizational-level variables and excludes macrolevel/contextual/regime-level variables which have been extensively covered elsewhere (Ansell & Lindvall, 2021; Clarke & Craft, 2019; Lindvall, 2017). Nor does this model emphasize how meta-governance is affected by policy accumulation due to ambitious executive leaders and evolving responsiveness traps (Adam et al., 2021; Meier et al., 2019). In short, by emphasizing organizational factors, this chapter does not exhaust causal identification of meta-governance (Lewis et  al., 2020; Timeus & Gasco, 2018). Since organizational variables may be subject to direct manipulation by organizational designers, they are useful tools to change the direction of organizations by deliberate intervention (Egeberg & Trondal, 2018). Yet, organizational designers are also constrained by requisite power since multiple factors might influence ordered organizational reform—such as ambiguous, shifting and competing goals, shortage of attention and capacity to monitor processes, previous conflicts that could be reopened, the stickiness of existing organizational arrangements, and the influence of shifting institutional fads and fashions (Hood & Jackson, 1991; March, 2008; March & Olsen, 1983; Pierson, 2004; Simon, 1983: 83; Weick & Suttcliffe, 2001). Two contributions are thus offered. First, the chapter outlines the contours of an organizational approach to meta-governance— that is problem-solving through organizational engineering. Meta-­ governance encompasses activities directed toward deliberate reforms of the infrastructure of organizations—that is, the deliberate changing of organizational structures, demography, and/or location. Jacobsson et al. (2015: 2) refer to meta-governance as ‘governance by organizing, i.e., steering through organizational design.’ Focus is thus on ‘manipulable variables’ in which research results can be systematically used as an intervention (Bennis, 1966: 100). Taken together, this paves the way for an organizational approach to meta-governance, making it less vulnerable to shifting dogmas and ideologies (Hood & Jackson, 1991), as well as showcasing the applicability of organization theory to organizational design (Evan, 1980). Second, merely to illustrate the theoretical argument, the chapter empirically examines how public innovation processes may be organizationally biased by (i) routinization of reform, (ii) organizational loose coupling, (iii) organized complexity, and (iv) temporal sorting. The next section outlines an organizational approach to meta-­ governance, and this section in particular outlines what we call a

172 

J. TRONDAL

short-­term reform-optimistic approach. This section unpacks the framework for analysis, focusing on four elements: organizational specialization and duplication, organizational capacity, organizational ecology, and organizational temporality. The final section empirically illustrates how public innovation processes might be organizationally shaped by (i) routinization of reform, (ii) organizational loose coupling, (iii) organized complexity, and (iv) temporal sorting. The conclusion suggests how an organizational approach to meta-­ governance might both explain meta-governance and make it practically relevant for solving societal challenges.

Theorizing Meta-governance The role of organizations and organizational factors in meta-governance is contested in the literature. March and Olsen (1983) contrasted long-term fiascos and short-term successes. Given that institutional scholarship has emphasized the study of long-term reform-pessimism (Pierson, 2004), this section in particular advances the short-term reform-optimistic approach. Long-Term Reform-Pessimism March and Olsen (1989) draw the lesson, from efforts in US political institutions, that long-term courses of action tend to be characterized by incremental adaptation to changing problems in which pre-existing solutions tend to be applied to new problems. Organizational designers who seek to design a polity in accordance with an architectonic design and envisage themselves as polity engineers may find themselves in situations they have not foreseen when facing the many factors that constrain organizational engineering and orderly reform (Pierson, 2004): the stickiness of existing organizational arrangements, institutional fads and fashions, shifting, ambiguous and competing goals, short attention spans, limited capacity to monitor processes, and a history of previous conflict that could at any time reemerge (e.g., March, 2008; Pierson, 2004). Reformers sometimes experience that reorganizations mainly formalize developments that have already taken place (March & Olsen, 1989: 114) or that reform processes mainly involve sense-making and meaning formation based on senses of identity (March & Olsen, 1989). The long-term development of segmented political orders fit into this picture (Bátora and Fossum, 2019).

7  DESIGNING THE ADMINISTRATIVE STATE 

173

It should thus be apparent that the development of political orders within a world that is dominated by nation states is often not deliberately designed processes. The EU was originally established to deal with deficiencies in a European system of states; how far it would develop and which challenges and deficiencies it would grapple with was a matter of profound contestation and has remained so until this day. Nevertheless, the lack of an overall design does not necessarily mean that a political order develops willy-nilly. Local and stepwise reforms, each of which is sensible, may be driven by local rationality and yet may add up to form a type of polity that nobody envisaged. Short-Term Reform Optimism March and Olsen (1989) also observed in the US political system that short-term courses of action tend to be influenced by the organization of attention, linked to power and pre-existing resources. This chapter takes as a point of departure a reform-optimistic approach that focuses on deliberate intervention and change through redesign of organizational structures. This approach, in particular, draws on an organizational approach which emphasizes how decision processes and human behavior respond to a set of fairly stable organizational routines (Cyert & March, 1963). Essentially, stable premises for behavioral choices are past experiences encoded in rules and expressed in the organizational structure of a government apparatus (Olsen, 2017). Organizational characteristics of meta-­ governance processes are thus likely to variously enable and constrain such processes, making some organizational choices more likely than others (Waldo, 1952). Organizational factors mobilize attention and action capacity around certain problems and solutions while ignoring others (Simon, 1983: 21). By redesigning organizational structures, the attention structure or the choice-architecture of decision-makers is systematically redirected. An organizational approach thus departs from the short-term politics of (re)organizing. Organization theory may be helpful in understanding how different ways of organizing reform processes may give different reform trajectories (and outcomes). Organizational factors may influence meta-governance in two ways: first, existing organizational structures may affect reform processes, and second, reform processes themselves may be deliberately organized on a temporary basis to achieve particular goals. Our discussion will privilege the latter since this has greatest design

174 

J. TRONDAL

implications. Our focus is thus on how to organize meta-governing processes. The argument can be illustrated as follows: Concerning the enabling and constraining role of existing organizational structures, whether a reform process is anchored within a ministry or an agency, at the level of political executives or at the level of administrative civil servants, is likely to influence what kind of actors that are deemed legitimate participants, which arguments are seen as appropriate, and how problems are framed. Studies show that stakeholder groups, such as external interest groups and internal trade unions, tend to have stronger influence at the agency level than at the ministry level where a political (hierarchical) control is more acceptable (Egeberg & Trondal, 2018). Moreover, reform processes that encompass not one, but several ministries, are likely to become more complex since they usually rely on horizontal coordination mechanisms. Finally, since the ordinary governing apparatus must run everyday business while engaging in reorganization, one additional question is also how much requisite capacity is left for the reform process (March & Olsen, 1983). An organizational approach sees reform processes as decision-making processes that allocate attention, resources, capabilities, roles, and identities. Second, reform organizations have structures, demographics, and locations that distribute rights, obligations, power and resources, and normally do so unevenly. The organizational structure of a reform is thus likely to enable and constrain the set of actors that are likely to be mobilized, the number of sequences involved, the access of affected stakeholder groups, the role of political executives, the degree of leeway for profound reform, and so on. Therefore, effects of reform organizations on the governing of reform processes are, in principle, parallel to effects of organizational structure on substantial decision-making processes (Lægreid & Roness, 1999: 302). One conventional reason for examining the effects of organizational variables in studies of public governance is that we may learn how public governance processes are organizationally orchestrated. One less conventional reason is that such variables may serve as instrumental toolkits for how public governance processes may be deliberately shaped through organizational design. This is so since organizational variables are comparatively malleable and thus subject to design. The framework for analysis is thus not only a framework for understanding meta-governance, but also for restructuring such processes (see Fig. 7.1). Departing from contemporary organizational literature, this section selects four analytical

7  DESIGNING THE ADMINISTRATIVE STATE 

175

Organizational specialization Organizational capacity

Meta-governance

Organizational ecology Organizational temporality

Fig. 7.1  Framework for analysis

dimensions of meta-governance and their design implications: organizational specialization, organizational capacity, organizational ecology, and organizational temporality. Organizational Specialization Even though political organizations tend to engage in continuous reforms, partly due to the initiative of political executives (Brunsson & Olsen, 1993), public organizations are also characterized as not very ‘changeful’ (Brunsson, 2000: 163) since they involve a complex set of political and administrative actors with multiple preferences, resources, and goals. The short-term reform-optimistic approach assumes that despite embedded constraints, organizational characteristics of meta-governance processes are thus likely to variously enable and constrain such processes, making some organizational choices more likely than others. Reform processes may, for example, be organized into specialized or nonspecialized structures—both vertically and horizontally. Organizational specialization is a way of decomposing and isolating problems and solutions into autonomous parts, ‘to reduce large problems into their component parts’ (March, 1994: 12) and to add systematic attention to reform. While specialized reform structures are likely to limit the access to actors and premises for choice, nonspecialized structures are more likely to open reform processes to a broader spectrum of actors and choice premises (Simon, 1983: 88). One way to mobilize reform is thus to organizationally specialize

176 

J. TRONDAL

meta-­governance by separating ‘thinking’ and ‘acting’ into separate organizational reform subunits. This subsection examines three such principles of organizational specialization that also may serve as design tools: vertical and horizontal (de-)specialization and organizational duplication.  ertical Specialization and Duplication V First, vertically specialized reform organizations are likely to limit the access of actors from lower levels of the hierarchical chain and favor the access of organizational leaders such as political executives. This organizational structure establishes communication barriers across organizational levels and reduces information flows across the chains. By organizing meta-governance processes close to the executive leadership, the potential for strategic policy-connected outputs is increased (Hustedt, 2013: 209). Trondal and Kiland (2010) illustrate such a vertically specialized reform organization in the case of the geographical reform/relocation of government agencies. This study shows that reform trajectories reflect the organizational structuring of meta-governance processes. The organizational setup of the particular reform was vertically specialized through a structured inclusion (and exclusion) of actors, and a careful design of the reform schedule. The reform organization was thus vertically specialized and temporally sequenced. However, this study also shows how the reform structure was complemented with a vertically nonspecialized structure at a second stage of the reform cycle. This stage was characterized by a de-­ specialized structure that selectively invited stakeholder groups that were considered supportive to the reform. In sum, this study illustrates how variation in the vertical specialization of the reform process mobilized two different governance patterns at different points in time. The first stage of the meta-governing processes was vertically specialized, privileging the responsible political executive and some few top officials, as well as limiting the number of decision sequences. In effect, the reform organization evaluated few alternatives and few consequences, did not examine causal relationships between alternatives solutions, and minimized the inclusion of actors. The second stage of the reform process was organizationally de-­ specialized and subsequently open to a vast number of shifting actors, problems, and solutions. In effect, the garbage-can-like characteristics of the second stage was a consequence of the organizational design of the reform process. As will be illustrated below, garbage-can processes (and the general idea of garbage-can) are sensitive to organizational structures and thus to organizational design thinking.

7  DESIGNING THE ADMINISTRATIVE STATE 

177

Vertical specialization is an organizational solution that also tend to fuel subunit autonomy in reform organizations (Leavitt, 1965: 1147). Yet, subunit autonomy might hinder reform implementation. This organizational dilemma has been illustrated in how governments have organized public sector innovation (PSI) labs across a spectrum of countries (Clarke & Craft, 2019; Lewis et al., 2020). To illustrate, Timeus and Gasco (2018) show how the city of Barcelona has established innovation labs that are organizationally isolated from their parent organizations. This vertically specialized organizational solution has increased subunit reform activities in the labs and increased their capacity to explore new solutions (see below), yet it has also limited the innovation capacity in city government at large. The tendency to ‘silo,’ or cause expertise to form in small and self-contained groups that do not communicate adequately with one another, was augmented in this case by the creation of vertically specialized reform structures such as PSIs. One organizational solution in which parent organizations may mitigate such isolation and increase the migration of innovation from organizational subunits such as labs is organizational duplication, in which the parent organization installs structures which are compatible with the lab. Studies show that organizational duplication between ministries and agencies tend to increase mutual communication, coordination, and influence between them (Egeberg & Trondal, 2018). One ‘takeaway’ implication of this discussion is that vertical specialization of reform organizations tends to limit the set of actors and conflicts. In short, vertical specialization reduces reform resistance and increases reform speed in meta-governance processes. Horizontal Specialization Second, horizontal specialization tends to mobilize decision-making myopia and blind spots, where problems and solutions outside one’s own organizational ‘turfs’ are neglected or seen as illegitimate, fuzzy, and distant. Horizontal specialization is therefore likely to increase a logic of exploitation in meta-governance processes. ‘Exploitation includes such things as refinement, choice, production, efficiency, selection, implementation, execution’ (March, 1994: 127). One consequence is reduced inconsistencies inside reform organizations, where the ‘global’ outlook across organizational borders is sacrificed for ‘local attention’ and siloization. A similar argument goes for the specialization of expertise in reform organizations. Christensen (2017) shows that organizations that are

178 

J. TRONDAL

dominated by economists are likely to focus on marked-conforming efficiency policies to a larger extent than organizations staffed with a blend of expertise. Processes of exploration within meta-governing processes are thus likely to thrive if multiple sets of expertise and skill sets are systematically organized in. This argument suggests that horizontally de-specialized reform organizations are likely to mobilize multiple problems and solutions. Horizontally de-specialized structures are also likely to mobilize a logic of exploration among actors by breaking up organizational silos, blind spots, and turfs. ‘Exploration includes things captured by such terms as search, variation, risk taking, experimentation, play, flexibility, discovery, and innovation’ (March, 1994: 237). Exploration increases the inclusion of actors and arguments in reform processes. One effect of such organizational formats might be search for novel solutions and thus increased innovation capacity. The flip side might be lack of control and oversight in decision-making processes, and thus a lack of exploitation (see the next section). As argued, meta-governance processes tend to combine different principles of organizational specialization and thereby different logics of behavior at different stages of reform processes. One ‘takeaway’ design implication is that exploration in meta-governance thrives on horizontal specialization. By contrast, increasing exploitation, and thus the likelihood of getting reform ideas through, is increased in de-specialized reform organizations. Organizational Capacity Meta-governance arguably requires organizational capacity. Yet, literature harbors competing ideas on the effects of capacity in meta-governance processes. One idea sees meta-governance as ad-hoc events. Organizations are pictured as stable equilibrium-seeking bodies, only interrupted by sudden occasional changes. A reform-pessimistic literature as outlined above advocates that institutions are ‘settled, stable, and integrated,’ (Selznick, 2015: 15) making them robust vis-à-vis their task environments. Change happens incrementally and gradually, and less as a consequence of deliberate reforms (e.g., Mahoney & Thelen, 2010; March & Olsen, 1989; Streek & Thelen, 2005). A competing idea, as argued in this study, sees meta-governance as routine activity and subject to deliberate intervention (Brunsson & Olsen, 1993). A pragmatist and reform-optimistic literature thus offer a middle ground in which organizations may both defend core values through processes of exploitation and, at the same time, adapt to

7  DESIGNING THE ADMINISTRATIVE STATE 

179

ongoing problems through processes of exploration (Ansell, 2011; Ansell et al., 2015). As also argued, this is likely to happen when meta-­governance processes combine different principles of organizational specialization and thereby different logics of behavior. Following a pragmatist argument, stability and change as well as exploration and exploitation might thus coexist in meta-governance processes, contingent on its organizational design (March, 1981: 563). Selznick (1992: 321) suggested that mindful institutional leadership would imply that the meta-governance should combine elements of change and continuity. It was essential for leaders to draw on what he calls the ‘institutional character’ as a source of direction. Meta-governance is thus not profoundly challenging the deeper ‘personality’ of the organization, while at the same time adjusting to environmental demands of the day (Krygier, 2012: 77). For example, reforms in certain parts of an organization may occur parallel to stability in other parts. As argued above, this is likely to happen when reform organizations are horizontally specialized, enabling solid borders between those parts of the organizations that are subject to reform and those focusing on business-as-usual. Another way of crafting meta-­ governance is by routinizing reform by allocating continuous attention to such processes through permanent organizational capacities (Brunsson & Olsen, 1993). Meta-governance is thus likely to become a routine activity in the organization and would serve as an organized and routinized ‘stability’ (Brunsson & Olsen, 1993: 33). The routinization of reform through organizational capacity-building may moreover safeguard the institutional character in the long term, while routinely updating the organization (Ansell et al., 2015). The routinizing of attention to meta-governance may be crafted by establishing permanent organizational attention and recruiting permanent staff earmarked to reform. Less routinization would be the result if reforms are structured as temporary ad-hoc organizations staffed with contracted short-term personnel (Bakker et al., 2016). Timeus and Gasco (2018), for example, show how organizations may mobilize innovation capacity in meta-governance processes by establishing permanent attention to innovation through PSIs. Such labs serve as a permanent organizational capacity to initiate reforms. Timeus and Gasco (2018) show how the city government of Barcelona has overall strengthened its innovation capacity through PSIs. Routinizing meta-governance through permanent structures has made it easier to initiate innovation reforms, yet not necessarily

180 

J. TRONDAL

to implement them. Organizational capacities do indeed routinise exploration, yet do not necessarily safeguard its execution. The routinization of administrative reform in nation states are typically done by the creation of ministries and agencies for government reform (Brunsson & Olsen, 1993). At the international level, the Puma committee of the OECD was originally established as a vehicle to routinize organizational reforms in the member states (e.g., OECD, 2001). Moreover, the OECD also installed implementation structures that enable reform ideas to be exploited and transposed by member states, such as the Observatory of Public Sector Innovation (OPSI), established in 2014. The EU has also established their own implementation structure, notably by EU-level agencies, which has enhanced EU-level control of how member states implement reforms (Egeberg & Trondal, 2017). On a grand scale, democracies may be seen as an organizational capacity that routinizes reforms by systematically organizing partisan opposition and policy contestation into parliaments and governments. Democracies have embedded competing claims for how the state should be structured and governed, and thereby increase states’ capacity for routinized and continuous calls for reforms (Ansell, 2011; Dahl, 2000). As such, democracies are indeed ‘changeful’ due to their organizational capacity to mobilize exploration from within. One ‘takeaway’ design implication may thus be that meta-governance is likely to become routine activity if requisite standby organizational capacities are installed. Organizational Ecology Reform processes often do not live alone. They are sometimes organized together into ecologies of reforms. They may thus intersect and feed on one another (Brunsson & Olsen, 1993: 33). Small and incremental reforms are sometimes elements in large-scale reforms (Ansell, 2011: 44). Moreover, reform processes at one place, or in the future, may be affected by reform processes elsewhere, or in the past. Meta-governance should thus be analyzed as parts of ecologies of nested and co-evolving reforms (Olsen, 2010). According to an organization theory approach, reform organizations may be designed in ways that make them mutually interdependent. Two such designs are discussed in the following. First, isolated reforms might be tied or designed into reform ecologies through organizational linkages or bridges. We may therefore think of small reforms as parts of larger ecologies and thus as ‘co-evolving processes’ that are likely to mobilize interactive effects (March, 1994: 97;

7  DESIGNING THE ADMINISTRATIVE STATE 

181

Olsen, 2010: 14). We may also consider large-scale reforms as not typically ‘designed’ per se, but as the product of smaller reforms through bridges converging into transformative change. Incremental change often has the advantage of being politically feasible to launch. Linking small reforms into wider ecologies of ‘meta-reform’ makes it more likely that decisions made in smaller reform processes feed into one another, generating aggregated transformation. One organizational design that is likely to mobilize reform ecologies is secondary structures that are temporary in nature— notably collegial structures such as committees and project organizations (Bakker et al., 2016: 1705), as well as contracted personnel that serve as boundary spanners across reform organizations. Secondary structures are temporary, niche-filling structures supplying additional problem-solving capacity to primary structures. A recent example is the Panama Canal Extension Program established to widen the capacity of the existing canal, comprising more than 10,000 employees. This reform organization coordinated ‘multiple interrelated sub-projects’ (van Marrewijk et al., 2016). Profound reform may thus be less challenging than assumed by reform-­ pessimistic literature, through the accumulated effects of multiple interrelated reforms coordinated through secondary structures (e.g., Brunsson, 2000: 180). Concomitantly, co-evolving reforms might combine elements of exploration and exploitation by the organizational flexibility that secondary structures allow for. In sum, secondary organizational structures serve as a design factor that may be used to create reform ecologies (Bolman & Deal, 1997; Olsen, 2010). Second, reform ecologies are likely to emerge when meta-governance processes are organized as nonspecialized (open) structures that increase available time and energy among reformers to be used elsewhere—that is, in adjacent reform organizations. This argument suggests that anarchic organizational designs are likely to increase the mutual learning potential and information exchange across reform processes, which in sum is likely to foster reform ecologies. Perceptions of legitimate and efficient solutions to problems might thus be transferred between different reform organizations. The flip side of organized exploration might, however, be loss of control over the overall reform trajectories and outcomes. Reforms in one organization might kick into neighboring reforms to the surprise of reform leaders. ‘Good’ solutions in one reform might cause problems in adjacent reforms, causing turbulence of scale (Ansell & Trondal, 2017). The likelihood of this happening arguably also increases if decision opportunities open in several reforms at the same time.

182 

J. TRONDAL

Organizational Temporality Meta-governance processes require time to be initiated and executed. Time concerns factors such as speed, the number and types of tempi, and the sequencing of reforms (e.g., Ansell & Trondal, 2017; Goetz, 2014). Essential to our argument is that the temporal logic of meta-governance is subject to organizational design. First, organizational factors might affect the speed of reform, for example, by setting tight deadlines. With increased speed comes a tendency for repetition and exploitation of past choices (March, 2010). When the tempo increases in meta-governing processes, established ideas and practices are likely to be subject to testing, yet pre-existing solutions are likely to be selected due to the sheer lack of time. High-speed meta-governance processes are thus likely to experience a tendency of repeating past successes, or what are perceived as past successes (March, 2010: 16). By repeating in this way, organizations may be victims of trained incapacity to improvise—merely due to the high speed of the process. In effect, up-­ tempo meta-governance may reduce the likelihood of exploration or improvisation—and the leeway for profound reform. Yet, this might subsequently increase the likelihood of reform implementation. Meta-­ governance may thus thrive if reform organizations operate at medium speed and balance elements of exploitation and exploration. Moreover, this argument also suggests a need for temporal diversity in meta-­ governance processes, in which meta-reforms include secondary structures (e.g., project organizations such as PSI labs) that proceed at different speeds in their schedules. Perhaps even more critical, reforms may come too late. Thus, the timing of reforms may be crucial. This speaks to the need to organize requisite proactive capacities in organizations to anticipate the unforeseen (Weick & Suttcliffe, 2001). Yet, this organizational design is most helpful when organizational challenges are on the horizon and not urgent. Following the above discussion on organizational capacity, organizing for future surprises may be achieved by designing permanent organizational capacities for meta-governance that are able to provide foresight activities that may subsequently pre-empt unanticipated shocks. Put together, these arguments highlight how organized capacity for temporal diversity may influence meta-governing processes. If multiple speeds are organized into reform programs, this may enable a combination of exploitation and exploration (March, 2010), continuity and change (Ansell et  al., 2017), and conservative and dynamic elements (Selznick,

7  DESIGNING THE ADMINISTRATIVE STATE 

183

1958). Multi-tempo meta-governance was illustrated in a two-staged reform process on the issue of geographical relocation of Norwegian government agencies (Trondal & Kiland, 2010): A high-speed organizational design effectively excluded critics to the reform and kept the reform on track without too much resistance. Yet, due to the high speed of the reform process, this organizational structure favored exploitation at the expense of exploration. This case also illustrates how the sheer number of reform sequences was subject to organizational (re)design. Different reform tempos were established at different stages of the reform process. This case thus illustrates how large-scale reform processes incorporated multiple smaller reforms scheduled with different speeds, tempi, and sequences. This was achieved through organizational specialization reform, which established buffers between different temporal logics in different parts of the reform organization. The level of resistance and the mobilization of opposition to a reform may also be influenced by the temporal structuring of the reform. Arguably, the mobilization of opposition to reforms is likely to be reduced if the meta-governing processes have a limited number of decision situations or sequences. Assuming that each decision situation is likely to mobilize a set number of actors, limiting the sheer number of decision situations is likely to reduce the number of actors that may potentially mobilize against a reform (Ansell & Trondal, 2017). Temporal sorting might thus be organizationally engineered across reform trajectories. As illustrated in the case above, whereas one part of the reform was tightly structured and dominated by logics of exploitation, the second sequence was organizationally de-specialized, loosely coupled, and open to exploration among the involved actors. In sum, the above discussion suggests four analytical dimensions of meta-governance and their design implications: organizational specialization, organizational capacity, organizational ecology, and organizational temporality. The next section aims to illustrate the argument on public sector innovation processes.

Public Sector Innovation To illustrate the organizational approach to meta-governance, this section examines how processes of public innovation might be structurally designed by (i) organizational routinization, (ii) organizational loose coupling, (iii) organized complexity, and (iv) temporal sorting. Benefiting from

184 

J. TRONDAL

empirical findings in a vast literature on PSI in general and observations from PSI labs and project groups in particular, this section suggests how innovation processes are organizationally contingent (see Al-Noaimi et al., 2022). PSI labs and project groups are secondary, collegial, and often nascent structures at arm’s length distance from the parent (primary) organizations that provide physical spaces for actors to meet and interact. Following Ansell and Torfing (2014: 5), PSI processes include change ‘that either disrupt established practices or challenge the common wisdom in a certain field’. Innovation thus involves exploration and playfulness (March, 1991), a logic of experimentation (Ansell & Trondal, 2017), and improvisation (Weick & Suttcliffe, 2001). PSI, in short, involves processes that explore re-combinations of established elements in new and pragmatic ways (Ansell & Torfing, 2014) as well as introducing new elements into organizations (OECD, 2017). Following Clarke and Craft (2019), this section argues against optimal design literature advocating one-size-fits-all recipes for public innovation processes and that PSI requires particular organizational designs—such as collaborative and interactive structures and processes (Ansell & Torfing, 2014; Torfing et al., 2012), contracting out and getting incentives right (e.g., Christensen & Lægreid, 2006; Friedman, 2008; Osborn & Gaebler, 1993; Torfing & Sorensen, 2016), and entrepreneurial leadership styles (Bason, 2014). By contrast, this section suggests that various organizational characteristics may spur public innovation processes. Moreover, as argued above, it is not assumed that organizational structures are subject to ‘choice.’ Organizational designers face constraints of resources, institutional histories, environmental chocks and fashions (Margetts & Duneavy, 2013). Their ‘relative autonomy’ and their ‘power to command’ is therefore circumscribed (Goodin, 1996: 13). Organizational Routinization Innovation is associated with the ability of organizations to launch and implement reforms. The PSI literature has advocated that the traditional characteristics of public organizations favor stability and predictability over innovation capacity. This section provides empirical examples of how meta-governance in general, and innovation processes in particular, may be routinized through a variety of organizational designs. First, routinizing reforms in organizations is an organizational design that fosters the allocation of continuous attention to innovation.

7  DESIGNING THE ADMINISTRATIVE STATE 

185

Organizational capacities translate to slack resources in which reform becomes ‘the new normal’ that represents some kind of organizational ‘stability’ (Brunsson & Olsen, 1993: 33). One way of routinizing attention to innovation processes is to design organizational capacities and personnel earmarked to organizational reform. One often used measure of routinizing administrative reform is to establish permanent ministries and agencies responsible for government reform. Innovation processes may also be fueled by organizing reforms into ecologies. As such, innovative ideas developed within one reform might spill over to adjacent reform processes and increase the sum of innovation capacity in the organization at large. As outlined above, reform ecologies may be organizationally designed by establishing secondary structures such as networks among organizations to achieve what is commonly coined ‘collaborative governance’ (Ansell & Torfing, 2014). Studies suggest that governments see the establishment of separate administrative units (secondary structures such as public sector innovation (PSI) labs) as a tool to introduce innovation capacity without disrupting established bureaucratic structures (Bason & Carstensen, 2002; Karo & Kattel, 2015). A recent study by Tõnurist et al. (2017) illustrates that innovation labs have become a preferred organizational platform to launch innovation processes, in part because their organizational structure at arm’s length distance from the ‘mother’ organization allows them to circumvent certain bureaucratic routines that are often seen as barriers to innovation. Yet, Tõnurist et al. (2017) also acknowledge that innovation labs often operate in relative isolation from the primary (‘parent’) structure of the organization. Similarly, Timeus and Gasco (2018) show how secondary structures establish organized capacity for innovation processes. They show that city governments increasingly strengthen capacity to meta-govern by installing secondary structures (PSI labs) to mobilize ‘smart cities’. Semi-­ autonomous innovation labs generate autonomous ideas, experience freedom to develop unconventional solutions, report greater risk-taking, and so on. However, their study also shows that innovation labs risk becoming isolated from their parent organization, limiting their overall impact on innovation capacity, and thus the sustainability of innovation in public organizations writ large (Lewis et al., 2020). Timeus and Gasco (2018: 1001) show that innovation labs tend to work in isolation with weak abilities to disseminate their knowledge throughout their parent organizations, and thus limiting the overall organizational innovation capacity.

186 

J. TRONDAL

Yet, innovation processes may be supported and fueled if designed as reform ecologies. According to the argument outlined above, one organizational design is to establish organizational bridges between reform processes through secondary structures, such as reform committees and ad-hoc project organizations (Bakker et al., 2016: 1705), as well as the recruitment of contracted personnel that are involved across several reform processes. In addition to routinizing attention to innovation through permanent and earmarked staff, routinized attention to innovation may be organizationally installed through the routinized inclusion of adjacent ideas by staff turnover. This, in turn, may be achieved by offering temporary contracts that increase overall staff mobility. Arguably, staff turnover is likely to induce new ideas into meta-governing processes by routinizing the inclusion of personnel with different educational backgrounds and different experiences. Secondary structures such as innovation labs are likely to increase staff mobility and educational differentiation. For example, hiring staff with a variety of skillsets is likely to reduce ‘group-thinking’ and increase the possibility of varied perspectives and outlooks on policy solutions. A parallel actor-oriented tool is the mobility of staff across organizational subunits, thereby circulating learning and innovation throughout the reform organization (Hood, 1999: 63). Routinizing mobility may thus induce organizational learning capacities from one organizational unit to another, or from one organization to the next. Secondary structures such as committees, boards, and labs are likely to mobilize stakeholder groups who may provide inputs to their primary structures (Hood, 1999: 64). In a recent study, Agger and Sorensen (2014) show how the establishment of committees of Majors, politicians, and citizens resulted in a learning process in which political executives, administrators, and citizens came to share similar views on city innovation. In sum, secondary structures may be designed alongside primary structures to mobilize innovation processes alongside the mundane routines of everyday organizational life. Moreover, the empirical examples suggest how innovation processes are routinized through a variety of organizational designs. Organizational Loose Coupling Innovation processes may also be paired with ideas from garbage-can theory, which emphasizes fluidity and randomness in decision-making processes (Cohen et al., 1972). Following Hood (1999: 62), ‘elements of the

7  DESIGNING THE ADMINISTRATIVE STATE 

187

garbage-can model might be deliberately introduced into organizational design.’ Innovation processes resemble garbage-can processes by involving exploration and playfulness (March, 1991), a logic of experimentation (Ansell & Trondal, 2017), and improvisation (Weick & Suttcliffe, 2001). An organizational approach suggests that such processes may be organizationally designed in such a way that decision situations become deliberately ambiguous by organizing a structurally loose coupling of actors, problems, and solutions. A tightly coupled organized system, by contrast, is likely to increase actors’ risk-averse behavior. Loose coupling of meta-governing processes is likely to increase opportunities for discretionary behavior and leeway for actors to choose novel policy solutions. Loosely coupled structures are therefore likely to enable a logic of exploration in meta-­ governance, and thereby fuel processes of innovation (March, 1991). In this way, organizational structures might be designed to ‘encourage disharmony and hence dynamics, to force us to reconsider and perhaps to change the way we are doing things from time to time’ (Goodin, 1996: 38–39). The abovementioned secondary structures are but one way to deliberately increase garbage-can elements in innovation processes— essentially by encouraging fluid participation and mobilizing unclear objectives in the primary structures. Moreover, the primary structure might also be designed to increase ambiguity and discovery—that is through loose coupling of those structures. Such structures are likely to mobilize a logic of exploration and innovative search among actors, to reduce the consistency and collective nature of preferences, increase the fluidity of participation, and increase trial-and-error learning (Lomi & Harrison, 2012: 10). Organized Complexity Third, innovation processes may be facilitated by deliberately designing complex organizational structures which are likely to mobilize complex sets of actors, preferences, values, and normative standards. The organizational approach suggests that innovation processes are likely to result from organizational structures characterized by conflicting organizational principles—in short, organized complexity. Organizations structured by competing principles provide complex choice-architecture for the actors, making them aware of multiple preferences, concerns, and considerations, such as political loyalty, due processes, rechtsstaat values, openness, transparency, participation, resilience, predictability, service quality,

188 

J. TRONDAL

responsiveness, efficiency, and effectiveness. Typical examples are hybrid organizational designs (Trondal, 2017). Hybrid structures characteristically combine components from various organizational forms and induce ‘chaos’ in the governing process where different parts of the reform organizations are likely to mobilize rival means-ends thinking, competing concerns, and adjacent normative standards. Hybrid organizations are thus rigged for innovation to a larger extent than nonhybrid structures (Ansell & Trondal, 2017). Innovation labs, as outlined above, add hybridity to primary structures and induces complexity to the organizational choice-­ structure. Organizational complexity is thus likely to mobilize a variety of skills, ideas, and visions for public innovation. In the same vein, a mix of professional skills might trigger creative thinking in a similar way. A recent study of innovation in Danish municipalities illustrates how hybrid organizational structures boosted intraorganizational conflict and dilemmas, and thus the emergence of a variety of behavioral frames. In sum, the hybrid municipal organization provided innovation capacity to the municipality at large (Thorup, 2017). Temporal Sorting Finally, innovation processes may be facilitated by (re)designing the temporality of reform processes (Ansell & Trondal, 2017). Following the example above, organizing multiple temporalities, such as speeds, into innovation processes, is likely to increase the possibility of ambiguity as to when actors are legitimate participants as well as what concerns are appropriate to mobilize at what point in time. Multispeed or poly-rhythmic decision-making processes are arguably more likely to amplify discovery and innovation than mono-temporal processes. In addition, innovation processes might be facilitated by organizing meta-governance processes in slow tempo. Slow tempo decision processes increase the likelihood that multiple actors have time to mobilize and that potential critics are activated throughout the innovation process. Slow tempo decision-making leaves more time for each actor to attend to multiple problems and solutions and thus to question pre-existing solutions. To establish dynamic resilience and permanent capacity to innovate, organizational structures such as bureaucracies may also facilitate innovation processes, due to their long-term horizons, their permanence, as well as their ability to break down problems into their component parts through organizational specialization. Permanent organizations with

7  DESIGNING THE ADMINISTRATIVE STATE 

189

permanent staff recruited for life would thus arguably be more explorative and innovative than temporary organizations staffed with temporary agents. Elected political executives with short-time horizons based on electoral cycles are thus likely to have shorter time for exploration compared to nonelected office holders (Meier et al., 2019). Taken together, multispeed, slow tempo, and permanent organizational capacities are likely to mobilize contrasting ideas and consequently the overall likelihood of innovation in the public domain. Innovation processes may thus be organizationally designed by structuring the temporality of such processes.

Conclusions Responding to a call to theorize the design orientation in studies of public policy and administration, this chapter outlines an organization theory approach. This approach accounts for how public organizations may organize policy change and reform by (re-)designing organizational choice-­ architectures. First, the chapter has outlined an organizational approach to meta-governance and, second, has offered an illustrative case of meta-­ governance by examining how public innovation processes are shaped by organizational designs. Two arguments are proposed: (i) first, that public meta-governance is an accessible tool for facilitating policy change, and (ii) second, that meta-governance may be systematically biased by organizational structuring. Meta-governance arguably encompasses activities directed toward deliberate reformation of the infrastructure of organizations. This refers to governance by organizing, that is, steering through organizational design. Moreover, by examining ‘manipulable variables,’ the chapter has introduced a theory that may serve to build bridges between theory and practice. Moving forward, the chapter paves the way for an organizational research agenda to meta-governance and for building the theory-practice divide. Whereas science and craft have often been mutually disregarded (Galbraith, 1980), an organizational approach advocates that organization theory as ‘craft’ in fact requires organization theory as ‘science.’ As advocated already by Luther Gulick (1937), ‘craft’ and ‘science’ are complementary joint ventures, not opposing endeavors. Taken together, knowledge about how organizational variables affect public governance, as well as knowledge about conditions for organizational change, are indeed necessary preconditions for applying organization theory for practical use. An organizational approach to meta-governance, it is argued, might both

190 

J. TRONDAL

explain meta-governance and make it practically relevant for solving societal challenges (Bennis, 1966: 97; Clarke & Craft, 2019; Egeberg & Trondal, 2018; Stoker, 2013).

References Adam, C., Hurka, S., Knill, C., & Steinebach, Y. (2021). Policy accumulation and the democratic responsiveness trap. Cambridge University Press. Agger, A., & Sorensen, E. (2014). Designing collaborative policy innovation: Lessons from a Danish municipality. In C. Ansell & J. Torfing (Eds.), Public innovation through collaboration and design. Routledge. Al-Noaimi, H., Durugbo, C. M., & Al-Jayyusi, O. R. (2022). Between dogma and doubt: A meta-synthesis of innovation in the public sector. Australian Journal of Public Administration, 81(1), 201–223. Ansell, C. (2011). Pragmatist democracy. Oxford University Press. Ansell, C., Boin, A., & Farjoun, M. (2015). Dynamic conservatism: How institutions change to remain the same. In M. S. Kraatz (Ed.), Institutions and ideals. Bingeley. Ansell, B. W., & Lindvall, J. (2021). Inward Conquest. Cambridge University Press. Ansell, C., & Torfing, J. (Eds.). (2014). Public innovation through collaboration and design. Routledge. Ansell, C., & Trondal, J. (2017). Governing turbulence. An organizational-­ institutional agenda. Perspectives on Public Management and Governance, 1(1), 43–57. Ansell, C., Trondal, J., & Ogard, M. (Eds.). (2017). Governance in turbulent times. Oxford University Press. Bakker, R. M., DeFillippi, R. J., Schwab, A., & Sydow, J. (2016). Temporary organizing: Promises, processes, problems. Organization Studies, 37(2), 1703–1719. Bason, C. (2014). Design attitude as an innovation catalyst. In C.  Ansell & J. Torfing (Eds.), Public innovation through collaboration and design. Routledge. Bason, C., & Carstensen, H. V. (2002). Powering collaborative policy innovation: Can innovation labs help? The Public Sector Innovation Journal, 17, 2–26. Bátora, J., & Fossum, J. E. (Eds.). (2019). Towards a segmented political order. Palgrave Macmillan. Bennis, W. G. (1966). Changing Organizations. McGraw-Hill Book Company. Bolman, L. G., & Deal, T. E. (1997). Reframing organizations. Jossey-Bass. Brunsson, N. (2000). The irrational organization. Fagbokforlaget. Brunsson, N., & Olsen, J. P. (1993). The reforming organization. Fagbokforlaget. Castellacci, F., & Natera, J. M. (2013). The dynamics of national innovation systems: A panel cointegration analysis of the coevolution between innovative capability and absorptive capacity. Research Policy, 42(3), 579–594.

7  DESIGNING THE ADMINISTRATIVE STATE 

191

Christensen, J. (2017). The power of economists within the state. Stanford University Press. Christensen, T., & Lægreid, P. (2006). Agencification and regulatory reform. In T. Christensen & P. Lægreid (Eds.), Autonomy and regulation. Edward Elgar. Cinar, E., Trott, P., & Simms, C. (2019). A systematic review of barriers to public sector innovation process. Public Management Review, 21(2), 264–290. Clarke, A., & Craft, J. (2019). The twin faces of public sector design. Governance, 32, 5–21. Cohen, M. D., March, J. G., & Olsen, J. P. (1972). A garbage can model of organizational choice. Administrative Science Quarterly, 17(1), 1–25. Cyert, R.  M., & March, J.  G. (1963). A Behavioral theory of the firm. Wiley-Blackwell. Dahl, R. (2000). On democracy. Yale University Press. De Vries, H., Bekkers, V., & Tummers, L. (2015). Innovation in the public sector: A systematic review and future research agenda. Public Administration, 94(1), 146–166. Durant, R. F. (2019). Building the compensatory state. Routledge. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies: What have we learnt (and where do we go from here). Journal of Common Market Studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Elgie, R. (2011). Presidentialism, parliamentarism and semi-presidentialism: Bringing parties back In. Government & Opposition, 46(3), 392–409. Emery, Y., & Giauque, D. (2014). The hybrid universe of public administration in the 21st century. International Review of Administrative Sciences, 80(1), 23–32. Evan, W. M. (1980). Paradigms, theories, and frontiers in organization theory: An introduction. In W. M. Evan (Ed.), Frontiers in organization and management. Praeger Publishers. Friedman, B. L. (2008). Policy analysis as organizational analysis. In M. Moran, M. Rein, & R. E. Goodin (Eds.), The Oxford handbook of public policy. Oxford University Press. Galbraith, J. R. (1980). Applying theory to the management of organizations. In W.  M. Evan (Ed.), Frontiers in organization and management. Praeger Publishers. Goetz, K. H. (2014). Time and power in the European Commission. International Review of Administrative Sciences, 80(3), 577–596. Goodin, R. E. (1996). Institutions and their design. In R. E. Goodin (Ed.), The theory of institutional design. Cambridge University Press. Gulick, L. (1937). Notes on the theory of organizations. With special references to government in the United States. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration, Columbia University.

192 

J. TRONDAL

Hermus, M., van Buuren, A., & Bekkers, V. (2020). Applying design in public administration: A literature review to explore the state of the art. Policy & Politics, 48(1), 21–48. Hood, C. (1983). The tools of government. Palgrave. Hood, C. (1999). The garbage can model of organization: Describing a condition or a prescriptive design principle? In M.  Egeberg & P.  Lægreid (Eds.), Organizing political institutions. Scandinavian Political Press. Hood, C., & Jackson, J. (1991). Administrative Argument. Dartmouth. Howlett, M. (2011). Designing public policies: Principles and instruments. Routledge. Howlett, M., Capano, G., & Ramesh, M. (2018). Designing for robustness: Surprise, agility and improvisation in policy design. Policy and Society, 37(4), 405–421. Hustedt, T. (2013). Ministerwervaltung in Wandel. Nomos. Jacobsson, B., Pierre, J., & Sundstrom, G. (2015). Governing the embedded state. Oxford University Press. Karo, E., & Kattel, R. (2015). Innovation bureaucracy: Does the organization of government matter when promoting innovation?. Manuscript. Lund University. Krygier, M. (2012). Philip Selznick. Ideals in the world. Stanford Law Books. Lægreid, P., & Roness, P. G. (1999). Administrative reform as organized attention. In M.  Egeberg & P.  Lægreid (Eds.), Organizing political institutions. Scandinavian University Press. Leavitt, H. J. (1965). Applied organizational change in industry: Structural, technological, and humanistic approaches. In J.  G. March (Ed.), Handbook of organizations. Rand McNally. Levi-Faur, D. (2012). From “big government” to “big governance”. In D. Levi-Faur (Ed.), The Oxford Handbook of Governance. Oxford University Press. Lewis, J. M., McGann, M., & Blomkamp, E. (2020). When design meets power: Design thinking, public sector innovation and the politics of policymaking. Policy & Politics, 48(1), 111–130. Lindvall, J. (2017). Reform capacity. Oxford University Press. Lodge, M., & Wegrich, K. (Eds.). (2012). Executive politics in times of crisis. Palgrave Macmillan. Lomi, A., & Harrison, J.  R. (2012). The garbage can model of organizational choice: Looking forward at forty. In A.  Lomi & J.  R. Harrison (Eds.), The garbage can model of organizational choice: Looking forward at forty. Bingley. Mahoney, J., & Thelen, K. (Eds.). (2010). Explaining institutional change. Cambridge University Press. March, J. G. (1981). Footnotes on organizational change. Administrative Science Quarterly, 26(4), 563–577.

7  DESIGNING THE ADMINISTRATIVE STATE 

193

March, J.  G. (1991). Exploration and exploitation in organizational learning. Organization Science, 2(1), 71–87. March, J. G. (1994). A primer on decision making. The Free Press. March, J. G. (2008). The myth of rationality. In U. Sverdrup & J. Trondal (Eds.), The organizational dimension of politics. Bergen. March, J. G. (2010). The ambiguities of experience. Cornell University Press. March, J. G., & Olsen, J. P. (1983). Organizing political life: What administrative reorganization tells us about government. American Political Science Review, 77(2), 281–297. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions. The Free Press. Margetts, P., & Duneavy, P. (2013). The second wave of digital-era governance: A quasi-paradigm for government on the web. Royal Society Publishing. http:// rsta.royalsocietypublishing.org/ Meier, K.  J. (2010). Governance, structure, and democracy: Luther Gulick and the future of public administration. Public Administration Review, 70(1), 284–291. Meier, K. J., Compton, M., Polga-Hecimovich, J., Song, M., & Wimpy, C. (2019). Politics, bureaucracy, and successful governance: The problem of political failure. Administration and Society, 51(10), 1576–1605. OECD. (2001). Governance in the 21st century. OECD Publishing. OECD. (2017). Government at a glance 2017. OECD Publishing. Olsen, J.  P. (2010). Governing through institutional building. Oxford University Press. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford University Press. Osborn, D., & Gaebler, T. (1993). Reinventing government. Middlesex. Peters, B. G. (2018). Policy problems and policy design. Edward Elgar. Peters, B. G., & Pierred, J. (Eds.). (2017). The next public administration. SAGE. Pierson, P. (2004). Politics in time. Cambridge University Press. Pollitt, C. (2016). Advanced introduction to public management and administration. Edward Elgar. Raadschelders, J. C. N. (2013). Public administration. Oxford University Press. Romme, A. G. L., & Meijer, A. (2020). Applying design science in public policy and administration research. Policy & Politics, 48(1), 149–165. Schulman, P.  R. (2022). Reliability, uncertainty and the management of error: New perspectives on the Covid-19 era. Journal of Contingencies and Crisis Management, 30(1), 92–101. Self, P. (1972). Administrative theories and politics. George Allen & Unwin. Selznick, P. (1958). Leadership in administration. University of California Press. Selznick, P. (1992). The moral commonwealth. The University of California Press. Selznick, P. (1996). Institutionalism “old” and “new”. Administrative Science Quarterly, 41(2), 270–277.

194 

J. TRONDAL

Selznick, P. (2015). On sustaining research agendas: Their moral and scientific basis. In M. S. Kraatz (Ed.), Institutions and ideals: Philip Selznick’s legacy for organizational studies. Bingley. Simon, H. A. (1983). Reason in human affairs. Stanford University Press. Stoker, G. (2013). Designing politics: A neglected justification for political science. Political Studies Review, 11, 174–181. Streek, W., & Thelen, K. (Eds.). (2005). Beyond continuity. Oxford University Press. Thorup, M. L. K. (2017). Hybride organiseringer i danske kommuner. PhD thesis. University of Roskilde. Timeus, K., & Gasco, M. (2018). Increasing innovation capacity in city governments: Do innovation labs make a difference. Journal of Urban Affairs, 40(7), 992–1008. Tõnurist, P., Kattel, R., & Lember, V. (2017). Innovation labs in the public sector: What they are and what they do? Public Management Review, 19, 1455–1479. Torfing, J., Peters, B.  G., Pierre, J., & Sorensen, E. (2012). Interactive governance. Oxford University Press. Torfing, J., & Sorensen, E. (2016). Metagoverning collaborative innovation in governance networks. American Review of Public Administration, 47(7), 826–839. Trein, P., Biesbroek, R., Bolognesi, T., Gejudo, G. M., Duffy, R., Hustedt, T., & Meyer, I. (2021). Policy coordination and integration: A research agenda. Public Administration Review, 81(5), 973–977. Trondal, J. (2017). Organized turbulence. In C. Ansell, J. Trondal, & M. Ogard (Eds.), Governance in turbulent times. Oxford University Press. Trondal, J., & Kiland, C. (2010). Bureaucracy and geography. Geographic relocation of the Norwegian central administration. World Political Science Review, 6(1), 1–28. van Marrewijk, A., Ybema, S., Smiths, K., Clegg, S., & Pitsis, T. (2016). Clash of the titans: Temporal organizing and collaborative dynamics in the Panama Canal megaproject. Organization Studies, 37(2), 1745–1769. Waldo, D. (1952). Development of theory of democratic administration. American Political Science Review, 46(1), 81–103. Weick, K. E., & Suttcliffe, K. M. (2001). Managing the unexpected. Jossey Bass.

PART III

Outlooks on the Multilevel and International Administrative State

CHAPTER 8

Administrative Sciences and the Multilevel Administrative State

The challenge of understanding social and political order is enduring in the social sciences (Elster, 2007; Waldo, 1992) with continuous disputes over ‘the legitimate role of democratic politics in society’ and ‘forms of political association’ (Olsen, 2016: 1–5). This chapter makes a plea for public administration scholarship and organization theory to understand the multilevel administrative state. Government institutions are essential components of contemporary democracies and play fundamental buildingblocks in the democratic governing of modern societies (Dunleavy & Rhodes, 1990; Orren & Skowronek, 2017; Wilson, 1887: 221). In recent decades, however, architecture of government has faced profound challenges by being increasingly embedded into multilevel federal structures. Yet, political order formation above and beyond the nation state is less studied and still poorly understood (Benz et al., 2021). This chapter discusses administrative integration in the EU and examines implications for political order. Political order consists of a relatively stable arrangement of institutions that are formalized and institutionalized. One implication is that government civil servants carry dual roles in a multilevel executive order that span levels of authority. Serving as key actors in implementing and enforcing EU rules on national ground, government officials personalize multilevel executive orders by working within national ministries and agencies while simultaneously partaking in European administrative networks and interacting with EU agencies and the Commission (Egeberg, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_8

197

198 

J. TRONDAL

2006; Trondal, 2010). The chapter outlines a conceptual scheme of administrative integration and outlines an organizational theory approach to account for dynamics of administrative (dis)integration. Crisis and disintegration have faced ‘grand theories’ of European integration with conceptual and explanatory difficulties, while midrange theories have fared comparatively better (Riddervold et al., 2021). Yet, whereas attempts at theoretical rehabilitation have prioritized grand theories, midrange accounts have been comparatively less discussed in contemporary literature. One reason might be that big theoretical ideas can explain general trends more adequately than particularities of cases (Boin & Lodge, 2016). This chapter aims to fill the void of midrange theorizing by applying an organizational theory approach to administrative integration. With an ambition to understand the administrative component of the EU administrative system, as well as to advance studies of political order, this chapter outlines ways to theoretically conceptualize administrative integration by the use of organization theory. We thus conceive of political order as consisting of a relatively stable arrangement of institutions that are both formalized and institutionalized. A common political order arguably requires that relevant institutions are (i) fairly independent of preexisting institutions; (ii) relatively integrated and internally cohesive; and (iii) reasonably able to influence governance processes within other institutions. A political order thus consists of independent, integrated, and influential institutions that allocate ‘authority, power, information, responsibility, and accountability’ (Olsen, 2016: 3). However, an ‘order’ does not suggest institutions that are perfectly integrated, profoundly influential, and impeccably independent. They are often imperfectly so (Trondal & Bauer, 2017). The chapter will first outline a conceptual framework that highlights the administrative dimension of multilevel governance; Second, it offers empirical illustrations of administrative integration in the EU; Thirdly, it advances an organization theory approach to the study of the multilevel administrative state. The chapter is presented as follows: The next section reviews the study of administrative integration and outlines a research agenda; the subsequent sections suggest conceptual dimensions that might be used for the empirical study of administrative integration. The final section outlines the contours of an organizational approach to administrative integration.

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

199

A Review of Literature Studies of state-building demonstrate that the emergence of political orders involves balancing acts between generating centralized and decentralized administrative capacities (Rokkan, 1999). Similar tensions and dilemmas are also involved in the EU where the rise of politico-­ administrative capacities at EU level challenge policy processes within member states. For example, studies suggest that the Commission profoundly biases power distributions across levels of government and circumvents national democratic governance processes (Egeberg & Trondal, 2009; Trondal, 2016). This section briefly reviews existing studies of administrative integration and recognizes three waves of scholarship. A comprehensive review of the study of political order involves examining both why such orders emerge and disappear (Bartolini, 2005; Fukuyama, 2013; March & Olsen, 1995; Padgett & Powell, 2012)—their policy outcomes (Olsen, 2007; Orren & Skowronek, 2004)—and how political order may be theoretically conceptualized (Stinchcombe, 1987). This chapter is particularly interested in the latter. Since the classics in administrative sciences (Gulick, 1937) and up to recent studies of public administration (Emery & Giauque, 2014; McDonald III et  al., 2022; Olsen, 2016), studies have prioritized established political orders. One recent exception has been a vibrant literature on international and transnational governance (Barnett & Finnemore, 2004; Bauer et  al., 2019; Bauer & Trondal, 2015; Trondal, 2016; Trondal et al., 2010; Trondal & Bauer, 2017). International bureaucracies constitute a distinct and increasingly important feature of both global governance studies and public administration scholarship. This entails that the study of international organizations has been somehow ‘normalized’ in both public administration and international relations scholarship (Trondal, 2007). Research shows that international bureaucracies not only profoundly influence global governance (Biermann & Siebenhüner, 2009) but also transform power distributions across levels of government and the conduct of domestic public governance (Egeberg & Trondal, 2009; Keohane et al., 2009). Moreover, international institutions are called upon to cope with ever more severe and unruly public problems. Yet, public administration scholarship has largely deserted the comparative study of international bureaucracies, including its multilevel character (Benz, 2015). This lacuna, we believe, reflects a more general gulf between most social science subdisciplines where the empirical foci of several

200 

J. TRONDAL

subdisciplines often poorly intersect. For instance, despite vast scholarship on both (public sector) governance and organization theory, respectively, these strands of research have been in mutual disregard (Olsen, 2010). Moreover, whereas research on public sector organizations has largely focused on domestic ministerial departments and subordinate agencies (Verhoest et al., 2012), IO scholarship has paid scant attention to their bureaucratic interior (Hawkins et  al., 2006; Karns & Mingst, 2004). Besides, EU studies have primarily been preoccupied with studying the Commission and subordinated regulatory agencies and largely neglected systematic comparative assessments (the N  =  1 fallacy) (Bauer & Trondal, 2015). We may identify three waves of studies of European political order and administrative integration. This chapter serves particularly as part of the third wave. Briefly identified, the first wave saw domestic political orders as sealed systems of governance in which exogenous variables to the nation state were conceptually seen as epiphenomena (Wilson, 1989). Studies of public administration were largely circumscribed to the study of domestic governing systems, including studies of decision-making and reform (Christensen & Lægreid, 2007). A second wave of studies directed attention to how international institutions—such as international bureaucracies—contribute to the transformation of national political order. These studies illuminated processes of transformation from a European Westphalian order to a post-Westphalian order characterized by a restructuring of political authority toward institutionalized multilevel governance (Ansell & Di Palma, 2004; Bartolini, 2005; Egeberg, 2006; Hooghe & Marks, 2001). This second surge of literature included the research program on multilevel governance (MLG) by Gary Marks and Liesbet Hooghe as well as the first wave of literature on multilevel administration. From the 1960s onward, a separate literature on administrative integration emerged with a focus on the integration of domestic and international administrative institutions and processes. Studies of administrative integration argued that the domestic-international distinction was conceptually and empirically fuzzy (Rosenau, 1966). The ‘descriptions of the [EU] Community as “above”, “alongside” or “outside” the member states were seen as oversimplifications’ (Pag, 1987: 446), stressing ‘bureaucratic interpenetration’ between member-state and EU administrative bodies (Cassese, 1987; Rosenau, 1966). More recently, the interdependencies of political orders have perhaps been most successfully captured by the MLG approach (Hooghe & Marks, 2001). The European political

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

201

order is thus seen as a nested order that integrates governance levels into a marble cake consisting of a patchwork of interconnected political institutions at different levels of authority (Bulmer, 2008: 173). This chapter subscribes to a third wave of literature focusing on the administrative basis for political order and the organizational dimension of public governance. The European administrative system is pictured as a ‘multilevel and nested network administration, where administrative bodies at different levels of government are linked together in the performance of tasks’ (Hofmann & Turk, 2006: 583). In short, this third wave of literature represents a ‘public administration turn’ in the study of political order and squarely put administrative integration center-stage in the study of the EU (Trondal, 2007). The empirical focus of this third wave has been to study European administrative institution building (Egeberg, 2006; Rittberger & Wonka, 2011) and the organizational foundation of a multilevel European administrative system (Curtin & Egeberg, 2008; Egeberg, 2010; Egeberg & Trondal, 2009). Here we see studies of regulatory agencies and the administrative state, European administrative networks, and how vertical and horizontal interconnections of regulatory bodies give rise to double-hatted national agencies. Studies of EU agencies have documented their varied origins, how they are situated closely to the European Commission, and how they contribute to autonomize national regulatory agencies, making them double-hatted vis-à-vis both the Commission and domestic ministerial departments (Bach et al., 2015; Egeberg & Trondal, 2017). This literature has examined the role of EU-level agencies in policy uploading and policy implementation (Egeberg & Trondal, 2009; Groenleer, 2009; Levi-Faur, 2011; Rittberger & Wonka, 2011), the interconnected nature of policymaking between EU-level and national-level ministries and agencies (Bach & Ruffing, 2018; Curtin & Egeberg, 2008; Egeberg & Trondal, 2009; Maggetti, 2014; Maggetti et al., 2021; Trein & Maggetti, 2018), and the consequences for administrative autonomy (Bach et  al., 2015; Bach & Ruffing, 2018; Trondal, 2010). Studies of European administrative network also show how they emerge largely from preexisting network formations; how internally connected they have become, that is, in bridging the gap between levels of governance; and how they manage to influence regulatory governance, for example, by providing knowledge exchange and facilitating learning (Maestenbroek et  al., 2022; Polman, 2020; Trondal & Peters, 2013). With an ambition to capture essential aspects of political order (Painter &

202 

J. TRONDAL

Peters, 2010: 6; Waldo, 1992: 37), the next section breaks this third wave into some conceptual dimensions.

The Administrative Dimension of Political Order This section outlines how administrative integration may be seen as one crucial element of political order. An enduring challenge in conceptualizing social phenomena is to establish theoretically powerful and empirically helpful categories that capture essential and enduring characteristics (Fukuyama, 2016: 222). Our starting assumption is that adequate analytical categories should both offer more empirical variation between than within categories as well as be generalizable across time and space. Examples of less meaningful categories are those treating social phenomena as sui generis and thus noncomparable. Effectively, sui generis phenomena are not generalizable and easily prevent the accumulation of general knowledge. One recent example of sui generis theorizing is when new intergovernmentalists treats EU agencies as ‘de novo’ bodies (Bickerton, 2012). Our approach puts the administrative dimension of political order center-­stage, and there are two reasons for doing so. The first is that administrations represent the action capacities of political orders. Without administrative institutions, public policies will not be initiated, drafted, nor implemented (Fukuyama, 2013). Moreover, public administration is essential for good governance since the wellbeing of citizens is shown to be nurtured by societies administered by ‘impartial’ public bureaucracies (Rothstein, 2012). Independent administrative capacities are central for making ‘good’ and ‘living’ political orders (March & Olsen, 1989). Second, the administrative dimension is by and large neglected in studies of political order (Rokkan, 1999). The administrative dimension has been largely neglected by leading theoretical approaches in EU studies, such as social constructivism (Checkel, 2005) and intergovernmentalism (Moravcsik, 1998). Advancing a public administration approach to the study of political order, this chapter supports a thread of thought from neo-functionalist literature which have argued that bureaucratic integration of administrative elites is vital for European integration in particular, as well as political order formation in general (Haas, 1958: 16; Niemann, 2006: 280). Notwithstanding developments in the study of the EU administrative system (Bauer & Trondal, 2015; Trondal & Bauer, 2017), contemporary

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

203

scholarship faces one major challenge. Public administration literature has devoted limited attention to the broader discussion on how administrative structures may affect the rise and decline of political order: Studies have for instance paid scant attention to how changing structuring of the state—for example, agencification and networking of agencies—coincide with forms of multilevel administrative governance (Danielsen & Yesilkagit, 2014; Egeberg, 2006; Fukuyama, 2013; Raadschelders, 2011; Trondal, 2014). This literature has not sufficiently studied how organizing public policy at one level of governance may influence ways of making public policy across levels of governance (Egeberg & Trondal, 2016). Consequently, the macrolevel consequences of administrative structures are poorly understood. Focusing on the administrative dimensions of political order, arguably, should not imply that the role of public administration is analyzed in isolation—as an ‘intellectual wasteland’ (Bobrow et al., 1977: 421)—but as requisite capacities that mobilize ‘bias’ in the making of public policy (Schattschneider, 1975). A public administration approach to political order formation is, and should be, also a theory of political organization (Olsen, 2016). The next section suggests how administrative science may provide a conceptual toolkit to the study of political order.

Structural Elements of Administrative Integration The EU serves as a living research laboratory for understanding political order (Olsen, 2007). One main thread in this literature has been studying the multilevel nature of the EU polity (Marks, 1993; Piattoni, 2010). According to Trondal and Bauer (2017), a ‘level’ refers to a distinct and independent sets of institutions, rules, procedures, and personnel. A multilevel order thus encompasses distinct and independent platforms that connect these elements at national level with parallel elements at levels above. This platform of elements consists of a puzzling mix of institutional autonomy and interdependence across levels of governance. It contains institutions that act relatively independently from domestic governments as well as enjoy institutional interdependencies or ties between the very same institutions (March, 1999). Recognized by the MLG literature (Hooghe & Marks, 2001), studying this mix of institutional autonomy and interdependencies is vital in order to adequately capture the multilevel character of political order.

204 

J. TRONDAL

Moreover, contemporary literature harbors two complementary approach to the study of administrative integration (Trondal & Peters, 2013) (‘MLA I’ and ‘MLA II’). • ‘MLA I’: The first surge of studies of multilevel administration (MLA) was particularly interested in the convergence of administrative systems and the convergences of public policies across EU member states (Olsen, 2007). MLA was consequently measured by its outcome—that is, the convergence of administrative forms, practices, and ways of doing things across the member states. This field of study emerged from comparative government and comparative public administration literature, examining the roots of common administrative systems (Knill, 2001; Meyer-Sahling & Yesilkagit, 2011) and practices (Christensen & Lægreid, 2011). Administrative integration was thus conceived of as the ‘convergence on a common European model’ (Olsen, 2003: 506). Lægreid and Rykkja (2020) reflect this approach in a study of administrative cooperation among Nordic ministries and agencies, focusing on the emergence of convergent administrative policies among Nordic states. • ‘MLA II’: A second surge of research conceives of MLA as featuring particular institutional constellations and configurations that span levels of governance. This literature has been particularly interested in understanding patterns or processes of (dis)integration of public administration institutions—not their outcomes (Benz, 2015). Administrative integration is thus conceived of as processes of European administrative capacity building (Rittberger & Wonka, 2011) and processes of multilevel administrative governance of ministerial departments and regulatory agencies (Bach et  al., 2015; Curtin & Egeberg, 2008; Egeberg, 2010; Egeberg & Trondal, 2009). Ultimately, administrative integration is gauged by the transformation of administrative governance, measured by patterns of administrative independence, influence, and integration (see below). This chapter draws attention to this second MLA II literature: from an organizational theory point of view, the question of how administrative institutions are organized, and how this in turn shape how they evolve and work (Egeberg & Trondal, 2018). The horizontal specialization of administrative institutions may, for example, affect how they interact with and influence other institutions. Studies show that the sector-specialization of

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

205

EU agencies fuels networking behavior vis-à-vis corresponding agencies at national level (Egeberg & Trondal, 2017). The way organizational structure shapes interaction, loyalty, cooperation, and information-processing are more adequately recognized in the organization theory literature than in most other social science literatures—for example, the IO literature (Cox & Jacobson, 1973), the governance literature (Levi-Faur, 2012), as well as brother political science literatures (Olsen, 2010). Within public administration literature, we may assume that bureaucracies possess internal capacities to shape staff through mechanisms such as socialization (behavioral internalization through established bureaucratic cultures), discipline (behavioral adaptation through incentive systems), and control (behavioral adaptation through hierarchical control and supervision) (Page, 1992; Weber, 1983). These mechanisms ensure that bureaucracies perform their tasks relatively independently from outside pressure, but within boundaries set by the legal authority and (political) leadership of which they serve. Explanatory emphasis is thus put on the endogenous organizational structures of bureaucracy, picturing bureaucratic organizations as architects of ‘organizational man’ (Simon, 1965) and as stabilizing elements in politics (Olsen, 2010). According to this model, bureaucracies develop their own nuts and bolts quite independently of society and that civil servants may act upon roles that are shaped by the organization in which they are embedded. By conceptually unpacking MLA II, three analytical dimensions have been suggested for analysis in the literature: institutional independence, influence, and integration (Trondal & Bauer, 2017). Administrative integration is recognized by the extent to which governing institutions are independent, integrated, and able to influence other institutions. Each of these items are shortly introduced below. Independence Huntington (1968) saw autonomy as a necessary requirement of state formation. One necessary factor in building common political order was thus the establishment of common institutions, including a permanent congress independent of national governments serving the common interest (Skowronek, 1982). Saint-Simon (1964: 35–38) argued that one necessary factor in building political order is the making of administrative systems that consist of common sets of bureaucratic bodies. In an international context, it necessitates the rise of separate international institutions

206 

J. TRONDAL

that are able to act relatively independently. Administrative integration thus involves the rise of relatively independent administrative capacities; that is, the permanent governing institutions that operate relatively independently of preexisting political institutions at national level—which is the bureaucracy of nation states (Matthews, 2012). The argument here is that the autonomy of bureaucratic organizations is supplied within these organizations and not merely due to cost-benefit analyses (Lipsky, 1980; Wilson, 1989) or socialization processes outside bureaucracy (Hooghe, 2007). When focusing on order formation in a European context, what matters is the extent to which a European political order is in practice autonomous from key components of an intergovernmental order, not whether it is autonomous in general. Empirically, political order at EU level would require that the core-executive institutions of the Union—that is, the Commission and EU-level agencies—act relatively independent of member-­state core-executive institutions. Contemporary studies of the Commission (Kassim et  al., 2013) and EU-level agencies (Egeberg & Trondal, 2017) indeed suggest fairly high degrees of institutional independence of member-state policy processes. In practice, however, administrative integration is likely to involve continuous and coevolving tensions between administrative dependencies and independencies within and between administrative systems (Trondal, 2017). Influence Political order also requires that institutions are able to mutually influence decision-making processes within and across institutions and governing systems. The independence and integration of administrative capacities may not only influence how public policy is formulated and implemented, but it may also affect the capacity to influence and challenge other institutions (Ahrne & Brunsson, 2019). Within the EU, the Commission has requisite capacities to influence governance processes within EU agencies as well as national agencies, making national agencies ‘double-hatted’ serving both as national and EU-level regulatory bodies. Studies demonstrate that the Commission has administrative capacities to influence everyday governing activities of domestic agencies—making them in practice partly European and partly national (Egeberg & Trondal, 2017) as well as making them semi-autonomous vis-à-vis their parent ministries (Bach et al., 2015).

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

207

Integration A final feature of administrative integration is the internal integration of administrative institutions and administrative systems. The question is thus how competing patterns of administrative integration and disintegration, or fragmentation and siloization, coexist within and among administrative institutions. Studies suggest that the emergence of common political orders does not necessarily lead to the rise of coherent and integrated institutions. Instead, Orren and Skowronek (2004) suggest that different elements of administrative bodies tend to overlap, counteract, and be poorly coordinated rather than organized and well ‘ordered.’ Patterns of integration and disintegration tend to be cumulatively layered over time (Curtin & Egeberg, 2008). Studies have established that parallel to the vertical specialization of administrative systems, there is a push for administrative coordination and centralization within national governments, strengthening the executive branch of government (Poguntke & Webb, 2005). This pattern is also observed within the Commission. Contemporary studies show that the Commission has become increasingly coordinated and contracted internally—both between the different layers of the Commission administration and between the Commission administration and the political level (College of Commissioners and their cabinets; Bauer et al., 2021; Kassim et al., 2017). These observations are also reflected throughout the history of the Commission with periods of internal integration and periods of internal disintegration. This was illustrated in the Jaques Delors Commission (1985–1994) which was characterized by presidential power and a neglect of ordinary administrative rules and routines (Christiansen, 2008: 63). Kassim et  al. (2013) show how the president of the Commission has gained more organizational capacities, notably by strengthening the secretariat-general into a political secretariat for the president. However, Trondal (2012) also illustrates parallel processes of administrative siloization and fragmentation between departments (Directorate-Generals) of the Commission. New studies also show how executive contraction detraction within the Commission tend to coexist and coevolve (Birkeland & Trondal, 2022).

208 

J. TRONDAL

Toward an Organization Theory Approach of Administrative Integration This section outlines the contours of an organizational approach to the study of administrative integration. In doing so, we systematically compare MLG and MLA along three dimensions: (i) units of analysis, (ii) organizational differentiation, and (iii) causal mechanisms. The discussion suggests how both literatures may complement each other in the study of administrative integration. Yet, the MLA literature in particular highlights the organizational dimension of administrative integration (Table 8.1). Units of Analysis MLG literature outlined subnational authorities as unit of analysis, stressing the complex actorhood that spans levels of authority in contemporary public governance processes (Börzel, 2021; Marks et  al., 2008: 113; Marks et  al., 1996; Tatham, 2021). This literature focuses on how the authority of subnational institutions provides them with requisite capacities to bypass national governments in their interface with EU institutions, thus making the different stages of the policy cycle multilevel (Tatham, 2021). Due to the interconnectedness of political authority across levels of governance, this literature has showed how the distinction between domestic and international relations, as well as between domestic and foreign policy administration, is obsolete and as a consequence, subnational actors mobilize independent policy processes beyond the state (Hooghe & Marks, 2001: 4). By contrast, the MLA literature focuses on the administrative structures of political order at different levels of government, notably on the mutual interaction of bureaucratic subunits across levels of Table 8.1  Conceptual elements MLA Units of analysis

Administrative institutions (public sector organizations and their sub-components) Organizational Degrees of differentiation differentiation (administrative systems are organizationally specialized systems) Causal mechanisms Supply of organizational capacities

MLG Sub-national arenas and actors (e.g., regions, cities, municipalities) No differentiation (regions treated as black boxes) Supply of (regional) authority

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

209

governance. Empirical studies have paid attention to the interaction of administrative subunits across levels of governance and how this shapes behavioral logics, bureaucratic autonomy, administrative coordination, and administrative styles among office holders (Bauer et  al., 2019; Egeberg, 2006; Trondal & Peters, 2013). These studies’ focus is on how administrative structures ‘mobilize biases’ in governance processes, systematically shaping administrative behavior among civil servants (Schattschneider, 1975; Simon, 1965). The MLA approach assumes that how bureaucracies and their subunits are organized at all levels of government is likely to systematically shape the administrative behavior evoked by bureaucratic staff, ultimately influencing processes of administrative integration. Organizational Differentiation Although the MLG literature successfully challenged the unitary conception of the nation state, it has simultaneously preserved subnational authorities as conceptual black boxes. In short, the operationalization of regional authority has improved measurements of multilevel governance architectures, but these endeavors do not include the organizational structuring of subnational institutions (Marks et al., 2008: 115). Consequently, Ongaro (2015) criticized the MLG literature for being a loose umbrella concept rather than a clear-cut theory. To illustrate, Marks et al. (2008) carefully measured local authority by the use of nine dimensions. However, neither of these dimensions suggest how organizational characteristics of subnational institutions should be conceptualized, measured, or applied to a causal model. Due to an absence of an organizational theory to multilevel governance, the MLG literature has focused on ‘the allocation of authority across general-purpose jurisdictions’ (Marks et al., 2008: 111). In short, a soft rationalist ontology of MLG conceives organizational variables as epiphenomena to the interaction and negotiation among subnational actors across levels of authority (see below). MLA literature, by contrast, conceives administrative institutions and systems as subject to organizational differentiation. Echoing the ideas of institutional polycentrism that studied systems of interconnected subunits (Ostrom, 2009), the MLA literature assumes that politico-administrative institutions may be organizationally differentiated and that patterns of differentiation are likely to systematically bias patterns of administrative integration. Viewing political orders as organizationally differentiated,

210 

J. TRONDAL

moreover, implies that organizational structuring is being conceptually and causally significant (Olsen, 2016: 8; see next paragraph). Egeberg and Trondal (2018) suggest how organizational variables can be applied to unpack organizational differentiations—such as organizational capacity (high/low), organizational departmentalization (horizontal/vertical), organizational association (primary/secondary), geographical locus (integrated/disintegrated), organizational demography (e.g., educational background of staff), and organizational coupling (tight/loose). To illustrate the argument, organizational differentiation might take the form of internal and/or external differentiation. The primary organizational affiliation of decision-makers is the unit to which they are likely to dedicate most attention, time, and resources. Examples are a ministerial department, a Commission DG, or a regulatory agency. Committees, collegial bodies, and networks, on the other hand, constitute secondary structures, meaning that participants are expected to be part-timers, having their primary affiliation somewhere else, for example, in a department or agency. Secondary structures—such as administrative networks of regulatory authorities—organize transactions by bringing together part-time participants in mutual exchange of information and expertise (Wood, 2019). Studies show that such bodies facilitate interaction, coordination, and trust among primary structures (Billis & Rochester, 2020; Klijn & Koppenjan, 2012; Lægreid et al., 2016; Maestenbroek et al., 2022). For example, central administrations that complement hierarchies with horizontal devices (e.g., Commission expert committees, comitology committees) seem to be better coordinated than those relying entirely on hierarchies (Wegrich & Stimac, 2014). Causal Mechanisms Finally, MLG and MLA literatures diverge on basic ontological assumptions. Although criticized for being theoretically descriptive (Ongaro, 2015), one essential explanatory component of the MLG literature is actor-centered institutionalism that focuses on the actorness and actorhood of subnational institutions (Scharpf, 1997; Tatham, 2021). A rich literature shows how a variety of actors are mobilized to engage in multilevel governance processes (Benz et  al., 2021). Based on a rationalist ontology, the MLG literature has successfully studied the role of subnational authority as a local push-factor for multilevel governance (Marks et al., 2008). To the extent that organizational variables are included in

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

211

the equation, however, they are largely applied as aggregative items (Marks et  al., 1996: 170). Organizational variables are thus akin to a thin ‘exchange-based’ idea of institutions and human choice (March & Olsen, 1995). According to an exchange-based theory of politics, March and Olsen (1995: 7) argue that ‘politics can be seen as aggregating individual preferences into collective action by procedures of rational bargaining, negotiation, coalition formation, and exchange.’ Institutional variables do not figure prominently as independent variables in the MLG literature, but as intervening variables that constrain human choices and policy processes. MLA literature, by contrast, argues that organizations should be treated as independent variables in the analytical scheme. The MLA literature thus fundamentally rests on an organizational approach to political science. An organizational approach is grounded on the assumption that organizational characteristics may explain both how organizations act and how they change. An organizational approach emphasizes how decision processes and human behavior respond to a set of fairly stable organizational routines (Cyert & March, 1963). Essentially, stable premises for behavioral choices are past experiences encoded in rules and expressed in the organizational structure of a government apparatus (Frederickson et al., 2012; Olsen, 2017; Waldo, 1952). Organizational characteristics of the governmental apparatus systematically enable and constrain public governance processes, making some policy choices more likely than others. The administrative structures of government systems are thus fundamentally organized choice-architectures. Contemporary studies in organization theory focus particularly on the explanatory role of organizational structure (Egeberg, 2012; Egeberg & Trondal, 2018). An organizational structure is a normative structure; it is a decided order, composed of rules and roles specifying who is expected to do what, when, and how (Ahrne & Brunsson, 2019; Egeberg & Trondal, 2018; Scott & Davis, 2016). As such, organization theory builds on decision theory with a focus on explaining decision-making behavior (Simon, 1965). This entails that organizational factors do not impact directly on society; rather, they have an indirect effect by influencing the policy process and the decisions made within and outside organizations (Ahrne & Brunsson, 2019; Arellano-Gault et al., 2013; Egeberg & Trondal, 2018). MLA literature builds on an organizational approach of public governance and thus establishes the organizational foundation of administrative integration. One such variable is the organization of administrative capacities

212 

J. TRONDAL

supplied at each level of government. How organizational capacities are distributed in subunits is likely to influence human choices and processes of administrative integration. Moreover, organizations at different levels that are structured by competing principles may provide multiple and competing choice-architectures for decision-makers, making them aware of multiple preferences, concerns, and considerations during the decision-­ making cycle. In sum, organizational capacities provide administrative institutions leverage to act independently, enabling them capacities for internal integration and providing structures for influence.

Conclusions This chapter draws attention to an organizational approach to the study of the multilevel administrative state. It has moreover accentuated how the organizational dimension of administrative integration assume relevance to the study of political order. The chapter thus suggests that an organizational approach is a necessary, yet often neglected, element in a general theory of political order. An organizational approach to administrative integration is generally helpful for studying the consolidation of embryonic political orders. According to March and Olsen (1995), organizations and organized systems are temporary systems of rules and roles, and according to Rokkan (1999), they are provisional resolutions of societal conflicts. An organizational approach to the study of political order helps to capture how emergent administrative systems that span multiple tiers of authoritative decision-making are dependent on organizational resources. Yet, proponents of an organization theory approach (Egeberg & Trondal, 2018) do not claim to provide a complete or comprehensive explanation of policy processes and policy contents. Rather, the argument is that organizational factors (independent variables) might intervene in actors’ behavioral perceptions (dependent variable) and create a systematic bias, thus making some process characteristics and outputs more likely than others (Gulick, 1937; March & Olsen, 1984; Simon, 1965). Organizations also discriminate between which conflicts should be attended to and which should be deemphasized (Egeberg, 2006). The empirical laboratories that are available to social science bias our theoretical lessons. As the discipline of public administration has been largely locked in national laboratories, the theoretical apparatus available to understand processes of administrative (dis)integration has been limited. Consequently, the subdiscipline of public administration in political

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

213

science has paid scarce attention to how administrative systems are essential ingredients to political order above and beyond the state. This chapter has aimed to capture the administrative dimension of political order in general, as well as multilevel administrative integration in particular, along three conceptual dimensions: institutional independence, integration, and influence. These dimensions have also served to accentuate what makes the organizational dimension of such processes important. Moving forward, two avenues for future studies of the multilevel administrative state seems apparent. First, there is need to invest in collaborative research infrastructures. Empirical studies of administrative integration are often confined in isolated datasets. Future research ambitions should be to establish collaborative data infrastructures that allow for large-scale studies of how and under what conditions administrative (dis) integration may emerge, evolve, and prosper. Moving beyond methodological nationalism, this would require the development of common infrastructures for joint collection large-scale data on a global scale. Following from this, future research should go beyond the EU in the study of administrative integration, for example, studying international bureaucracies (cf. Chaps. 11 and 12 in this volume) and administrative networks (see Börzel, 2021 for one such initiative). Studies may focus on the extent to which and how international bureaucracies operate as hubs for global agency networks, for example, contributing to interagency coordination to solve transnational societal problems (Mele & Cappellaro, 2018). This agenda is what Chaps. 11 and 12 examine at the end of this volume.

References Ahrne, G., & Brunsson, N. (Eds.). (2019). Organization outside organizations. Cambridge University Press. Ansell, C., & Di Palma, G. (2004). Restructuring territoriality. Cambridge University Press. Arellano-Gault, D., Demortain, D., Rouillard, C., & Thoenig, J.-C. (2013). Bringing public organizations and organizing back in. Organization Studies, 34(2), 145–167. Bach, T., & Ruffing, E. (2018). The transformative effects of transnational administrative coordination in the European multi-level system. In E.  Ongaro & S. Van Thiel (Eds.), The Palgrave of public administration and Management in Europe. Palgrave Macmillan.

214 

J. TRONDAL

Bach, T., Ruffing, E., & Yesilkagit, K. (2015). The differential empowering effects of Europeanization on the autonomy of national agencies. Governance, 28(3), 285–304. Barnett, M., & Finnemore, M. (2004). Rules for the world. Cornell University Press. Bartolini, S. (2005). Re-structuring Europe. Oxford University Press. Bauer, M. W., Beyerlein, L., Ege, J., Knill, C., & Trondal, J. (2019). Perspectives on international public administration research: A rejoiner to Johan Christensen and Kutsal Yesilkagit. ARENA Working Paper 1. University of Oslo. Bauer, M. W., Kassim, H., & Connolly, S. (2021). The quiet transformation of the EU Commission cabinet system. Journal of European Public Policy, 30(2), 354–374. https://doi.org/10.1080/13501763.2021.2003423 Bauer, M. W., & Trondal, J. (2015). The administrative system of the European Union. In M.  W. Bauer & J.  Trondal (Eds.), The Palgrave handbook of the European administrative system. Palgrave Macmillan. Benz, A. (2015). European public administration as a multilevel system administration: A conceptual framework. In M.  W. Bauer & J.  Trondal (Eds.), The Palgrave handbook on the European administrative system. Palgrave Macmillan. Benz, A., Broschek, J., & Lederer, M. (Eds.). (2021). A research agenda for multilevel governance. Edward Elgar. Bickerton, C. J. (2012). European integration. Oxford University Press. Biermann, F., & Siebenhüner, B. (Eds.). (2009). Managers of global change. MIT Press. Billis, D., & Rochester, C. (Eds.). (2020). Handbook on hybrid organisations. Edward Elgar. Birkeland, T., & Trondal, J. (2022). The rift between executive contraction and executive detraction: The case of European Commission battery policy-­making. Journal of European Public Policy. https://doi.org/10.1080/1350176 3.2022.2118356 Bobrow, D. B., Eulau, H., Landau, M., Jones, C. O., & Axelrod, R. (1977). The place of policy analysis in political science: Five perspectives. American Journal of Political Science, 21(2), 415–433. Boin, A., & Lodge, M. (2016). Designing resilient institutions for transboundary crisis management: A time for public administration. Public Administration, 94(2), 289–298. Börzel, T.  A. (2021). Regional and global multilevel governance between and beyond states. In A. Benz, J. Broschek, & M. Lederer (Eds.), A research agenda for multilevel governance. Edward Elgar. Bulmer, S. (2008). Building a multi-level polity in Europe. In U.  Sverdrup & J. Trondal (Eds.), The organizational dimension of politics. Bergen. Cassese, S. (1987). Divided powers: European administration and national bureaucracies. In D. Berlin, S. Pag, & C. Bourtembourg (Eds.), The European administration. Maastricht.

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

215

Checkel, J.  T. (2005). International institutions and socialization in Europe. International Organization, 59(4), 801–826. Christensen, T., & Lægreid, P. (2007). Transcending new public management. Ashgate. Christensen, T., & Lægreid, P. (2011). Beyond NPM? Some development features. In T. Christensen & P. Lægreid (Eds.), The Ashgate research companion to new public management. Ashgate. Christiansen, T. (2008). The institutional politics of the European Union. (Unpublished doctoral dissertation). University of Maastricht, The Netherlands. Cox, R.  W., & Jacobson, H.  K. (1973). The anatomy of influence. Yale University Press. Curtin, D., & Egeberg, M. (2008). Tradition and innovation: Europe’s accumulated executive order. West European Politics, 31(4), 639–661. Cyert, R. M., & March, J. G. (1963). A Behavioral theory of the firm. Prentice Hall. Danielsen, O.  A., & Yesilkagit, K. (2014). The effects of European regulatory networks on the bureaucratic autonomy of national regulatory authorities. Public Organization Review, 14, 353–371. Dunleavy, P., & Rhodes, R. A. W. (1990). Core executive studies in Britain. Public Administration, 68, 3–28. Egeberg, M. (Ed.). (2006). Multilevel union administration. Palgrave Macmillan. Egeberg, M. (2010). L’administration de l’Union Europeenne: Niveaux multiples et construction d’un centre. Revue Francaise d’Administration Publique, 133, 17–26. Egeberg, M. (2012). How bureaucratic structure matters: An organizational perspective. In B. G. Peters & J. Pierre (Eds.), The Sage handbook of public administration. SAGE. Egeberg, M., & Trondal, J. (2009). National agencies in the European administrative space: Government driven, commission driven, or networked? Public Administration, 87(4), 779–790. Egeberg, M., & Trondal, J. (2016). Why strong coordination at one level of government is incompatible with strong coordination across levels (and how to live with it): The case of the European Union. Public Administration, 94(3), 579–592. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies: What have we learned (and where to go from here). Journal of Common Market Studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Understanding and design. Oxford University Press. Elster, J. (2007). Explaining social behavior. Cambridge University Press. Emery, Y., & Giauque, D. (2014). The hybrid universe of public administration in the 21st century. International Review of Administrative Sciences, 80(1), 23–32.

216 

J. TRONDAL

Frederickson, H. G., Smith, K. B., Larimer, C. W., & Licari, M. J. (2012). The public administration theory primer. Westview Press. Fukuyama, F. (2013). What is governance? Governance, 26(3), 347–368. Fukuyama, F. (2016). Macro theory and the study of political development. Scandinavian Political Studies, 39(3), 207–225. Groenleer, M. (2009). The autonomy of European Union agencies. Eburon. Gulick, L. (1937). Notes on the theory of organizations: With special references to government in the United States. In L. Gulick & L. F. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration, Columbia University. Haas, E. B. (1958). The uniting of Europe. Notre Dame Press. Hawkins, D. G., Lake, D. A., Nielson, D. L., & Tierney, M. J. (2006). Delegation and Agency in International Organizations. Cambridge University Press. Hofmann, H.  C. H., & Turk, A.  H. (Eds.). (2006). EU administrative governance. Edward Elgar. Hooghe, L. (2007). Several roads lead to international norms, but few via international socialization. A case study of the European Commission. In J. T. Checkel (Ed.), International institutions and socialization in Europe. Cambridge University Press. Hooghe, L., & Marks, G. (2001). Multilevel governance and European integration. Rowman & Littlefield Publishers. Huntington, S. P. (1968). Political order in changing societies. Yale University Press. Karns, M. P., & Mingst, K. A. (2004). International organizations. Lynne Rienner Publishers. Kassim, H., Connolly, S., Dehousse, R., Rozenberg, O., & Benjamballah, S. (2017). Managing the house: The presidency, agenda control and policy activism in the European Commission. Journal of European Public Policy, 24(5), 653–674. Kassim, H., Peterson, J., Bauer, M. W., Connolly, S., Dehousse, R., Hooghe, L., & Thompson, A. (2013). The European Commission of the twenty-first century. Oxford University Press. Keohane, R.  O., Macedo, S., & Moravcsik, A. (2009). Democracy-enhancing multilateralism. International Organization, 63(1), 1–31. Klijn, E. H., & Koppenjan, J. F. M. (2012). Governance network theory: Past, present and future. Policy and Politics, 40(4), 187–206. Knill, C. (2001). The Europeanisation of National Administrations. Cambridge University Press. Lægreid, P., & Rykkja, L. (2020). Nordic administrative collaboration: Scope, predictors and effects on policy design and administrative reform measures. Politics and Governance, 8(4), 21–32. Lægreid, P., Randma-Liiv, T., Rykja, L., & Sarapuu, K. (2016). Coordination challenges and administrative reforms. In G. Hammerschmid, S. Van de Walle, R. Andrews, & P. Bezes (Eds.), Public administrative reforms in Europe. Edward Elgar.

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

217

Levi-Faur, D. (2011). Regulatory networks and regulatory agencification: Towards a single European regulatory space. Journal of European Public Policy, 18(6), 810–829. Levi-Faur, D. (Ed.). (2012). The Oxford handbook of governance. Oxford University Press. Lipsky, M. (1980). Street-level bureaucracy. Russel Sage Foundation. Maestenbroek, E., Schrama, R., & Martinsen, D.  S. (2022). The politics of European administrative interaction. Explaining knowledge exchange in the European migration network. Journal of European Public Policy, 29(10), 1656–1676. Maggetti, M. (2014). The rewards of cooperation: The effects of membership in European regulatory networks. European Journal of Political Research, 53(3), 480–499. Maggetti, M., Di Mascio, F., & Nattalini, A. (2021). National regulators, regulatory networks, and European agencies. Connecting the dots. Public Policy and Administration, 36(3), 275–280. March, J. G. (1999). A learning perspective on the network dynamics of institutional integration. In M.  Egeberg & P.  Lægreid (Eds.), Organizing political institutions. Scandinavian University Press. March, J. G., & Olsen, J. P. (1984). The new institutionalism: Organizational factors in political life. American Political Science Review, 78, 734–749. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions. The Free Press. March, J. G., & Olsen, J. P. (1995). Democratic governance. The Free Press. Marks, G. (1993). Structural policy and multilevel governance in the EC.  In A. W. Cafruny & G. G. Rosenthal (Eds.), The state of the European Community. Boulder, CO. Marks, G., Hooghe, L., & Schekel, A. H. (2008). Measuring regional authority. Regional and Federal Studies, 18(2/3), 111–121. Marks, G., Nielsen, F., & Ray, L. (1996). Competencies, cracks, and conflicts: Regional mobilization in the European Union. Comparative Political Studies, 29(2), 164–192. Matthews, F. (2012). Governance and state capacity. In D.  L. Faur (Ed.), The Oxford handbook of governance. Oxford University Press. McDonald, B. D., III, Hall, J. L., O’Flynn, J., & Van Thiel, S. (2022). The future of public administration research: An editor’s perspective. Public Administration, 100(1), 597–591. Mele, V., & Cappellaro, G. (2018). Cross-level coordination among international organizations: Dilemmas and practices. Public Administration, 96, 736–752. Meyer-Sahling, J. H., & Yesilkagit, K. (2011). Differential legacy effects: Three propositions on the impact of administrative traditions on public management reform in Europe east and west. Journal of European Public Policy, 18(2), 311–322.

218 

J. TRONDAL

Moravcsik, A. (1998). The choice for Europe. Cornell University Press. Niemann, A. (2006). Explaining decisions in the European Union. Cambridge University Press. Olsen, J. P. (2003). Towards a European administrative space? Journal of European Public Policy, 10(4), 506–531. Olsen, J. P. (2007). Europe in search of political order. Oxford University Press. Olsen, J. P. (2010). Governing through institution building. Oxford University Press. Olsen, J.  P. (2016). Democratic accountability and the terms of political order. European Political Science Review, 9(4), 519–537. Olsen, J. P. (2017). Democratic accountability, political order, and change. Oxford University Press. Ongaro, E. (2015). Multi-level governance: The missing linkages. In E. Ongaro (Ed.), Multi-level Governance. Bingley. Orren, K., & Skowronek, S. (2004). The search for American political development. Cambridge University Press. Orren, K., & Skowronek, S. (2017). The policy state. Harvard University Press. Ostrom, E. (2009). A general theory framework for analyzing sustainability of social-ecological systems. Science, 325(5939), 419–422. Padgett, J. F., & Powell, W. W. (2012). The emergence of organizations and markets. Princeton University Press. Pag, S. (1987). The relations between the commission and national bureaucracies. In S. Cassese (Ed.), The European administration. Maastricht. Page, E.  C. (1992). Political authority and bureaucratic power. Harvester Wheatsheaf. Painter, M., & Peters, B. G. (Eds.). (2010). Tradition and public administration. Palgrave Macmillan. Piattoni, S. (2010). The theory of multilevel governance. Oxford University Press. Poguntke, T., & Webb, P. (Eds.). (2005). The Presidentialization of politics. Oxford University Press. Polman, D. (2020). Participation of implementing agencies in European administrative networks. Journal of Common Market Studies, 58(4), 818–835. Raadschelders, J. C. N. (2011). Public administration. Oxford University Press. Riddervold, M., Trondal, J., & Newsome, A. (Eds.). (2021). The Palgrave handbook of EU crises. Palgrave Macmillan. Rittberger, B., & Wonka, A. (2011). Introduction: Agency governance in the European Union. Journal of European Public Policy, 18(6), 780–789. Rokkan, S. (1999). State formation, nation building, and mass politics in Europe. Oxford University Press. Rosenau, J. N. (1966). Pre-theories and theories of foreign policy. In R. B. Farrell (Ed.), Approaches to comparative and international politics. Northwestern University Press.

8  ADMINISTRATIVE SCIENCES AND THE MULTILEVEL ADMINISTRATIVE… 

219

Rothstein, B. (2012). Good governance. In D. L. Faur (Ed.), The Oxford handbook on governance. Oxford University Press. Saint-Simon, H. (1964). Social organization, the science of man and other writings. Harper Torchbooks. Scharpf, F. W. (1997). Games real actors play. Westview Press. Schattschneider, E. E. (1975). The Semisovereign people. Harcourt Brace Jovanovich College Publishers. Scott, W. R., & Davis, G. F. (2016). Organizations and organizing. Routledge. Simon, H. A. (1965). Administrative behavior. The Free Press. Skowronek, S. (1982). Building a new American state. The expansion of national administrative capacities 1877–1920. Cambridge University Press. Stinchcombe, A.  L. (1987). Constructing social theories. The University of Chicago Press. Tatham, M. (2021). Regions and cities in multilevel governance. In A.  Benz, J. Broschek, & M. Lederer (Eds.), A research agenda for multilevel governance. Edward Elgar. Trein, P., & Maggetti, M. (2018). Reasserting the center of the nation state: How supranational institutions reshape the public sector. University of Lausanne. Trondal, J. (2007). The public administration turn in integration research. Journal of European Public Policy, 14(6), 960–972. Trondal, J. (2010). An Emergent European Executive Order. Oxford University Press. Trondal, J. (2012). On bureaucratic Centre formation in government institutions: Lessons from the European Commission. International Review of Administrative Sciences, 78(3), 425–446. Trondal, J. (2014). Agencification. Public Administration Review, 74(4), 545–549. Trondal, J. (2016). Advances to the study of international public administration. Journal of European Public Policy, 23(7), 1097–1108. Trondal, J. (2017). The rise of independent international administration: The case of the International Atomic Energy Agency (IAEA). In J. Trondal (Ed.), The rise of common political order. Edward Elgar. Trondal, J., & Bauer, M.  E. (2017). Conceptualizing the European multilevel administrative order: Capturing variation in the European administrative system. European Political Science Review, 9(1), 73–94. Trondal, J., Marcussen, M., Larsson, T., & Veggeland, F. (2010). Unpacking international organisations. Manchester University Press. Trondal, J., & Peters, B. G. (2013). The rise of European administrative space: Lessons learned. Journal of European Public Policy, 20(2), 295–307. Verhoest, K., Van Thiel, S., Bouckaert, G., & Lægreid, P. (Eds.). (2012). Government agencies. Palgrave Macmillan. Waldo, D. (1952). Development of theory of democratic administration. American Political Science Review, 46(1), 81–103.

220 

J. TRONDAL

Waldo, D. (1992). The Enterprise of Public Administration. Chandler & Sharp Publishers. Weber, M. (1983). On capitalism, bureaucracy and religion. Harper Collins Publishers. Wegrich, K., & Stimac, V. (2014). Coordination capacity. In M.  Lodge & K.  Wegrich (Eds.), The problem-solving capacity of the modern state. Oxford University Press. Wilson, W. (1887). The study of administration. Political Science Quarterly, 2(2), 197–222. Wilson, J. Q. (1989). Bureaucracy. Basic Books. Wood, M. (2019). Hyper-active governance. Cambridge University Press.

CHAPTER 9

European Integration and the Administrative State

This chapter presents an empirical study that shows the profound and rising role of public administration in the multilevel governing system of the EU. Chapter 8 outlined the structural elements of a European multilevel administrative system (MLA) consisting of strongly interconnected administrative bodies across levels of governance (cf. Bauer & Trondal, 2015). Administrative capacity building by stealth at the EU level is seen as challenging administrative autonomy among member-state governments (e.g., Genschel & Jachtenfuchs, 2014; Trondal, 2010). Studies also suggest how an organizationally heterogeneous EU sends a plethora of differentiated ‘signals’ to the member states, for example, concerning how the Council fuels strong member-state coordination and perceptions of national preferences, whereas the Commission fuels a circumvention of domestic political control and privileges non-majoritarian bodies (e.g., Egeberg et al., 2003; Trondal & Veggeland, 2003). This chapter makes two contributions to this volume: • Theoretically, it examines the role of organizational factors in administrative integration which this chapter argues is nudged by ‘favorable’ organizational conditions at the domestic level of government. • Empirically, the chapter demonstrates how the EU contributes to a self-reinforcing administrative bias due to domestic-level organizational factors. Strong European integration without membership © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_9

221

222 

J. TRONDAL

reinforces a politico-administrative gap, and this gap expands over time. The chapter applies an extreme case of high integration without formal EU membership represented by Norway. The data suggests that the EU contributes to reinforce the administrative state through strong unintended assimilation effects. To demonstrate how essential parameters of domestic public governance are profoundly influenced by the EU, even in a non-EU member-state, the chapter presents a novel longitudinal dataset. This dataset consists of a large-N (N = 3562) questionnaire study among government officials at three points in the Norwegian central administration: 1996, 2006, and 2016 (see Chap. 3 in this volume). The self-reinforcing administrative bias is illuminated by Norwegian government officials dealing with the EU being increasingly loosely coupled to the political level (low-ranked staff) and (surprisingly) tightly interwoven and influenced by EU institutions. The chapter thus also contributes to a ‘public administration turn’ in EU studies (e.g., Bauer & Trondal, 2015; Egeberg, 2006; Trondal, 2007). Essential to this turn has been to understand the role of administrative actors in political-administrative life of the EU (Olsen, 2018). The inherent state prerogative of preparing policymaking and getting things done has been challenged by the rise of independent and integrated administrative capacities at the EU level. The supply of organizational capacities inside the Commission—together with the rise of EU agencies—has enabled an emergent EU-level executive order to act independently of domestic government institutions (Egeberg & Trondal, 2017). Moreover, organizational capacities of EU-level administrative actors supply them with a capacity to influence non-majoritarian institutions (agencies) at domestic level (e.g., Egeberg, 2006). This chapter shows how the impact of the EU is mediated by organizational variables at the national level. Moreover, the chapter provides novel data from the core-executive level of government (ministerial departments), and thereby adds to a literature that empirically has been dominated by studies of agencies. Methodologically, the empirical strategy is to examine the government administration of a formally non-EU-member state—the Norwegian central administration. As such, the chapter contributes to studies of external differentiation of the EU and the external effects of EU norms and rules beyond EU borders (see Rittberger & Blauberger, 2018). Still, compared to contemporary instances of EU external governance, Norway is by far the most strongly integrated EU nonmember through a dense web of

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

223

institutionalized relations (Egeberg & Trondal, 1999; Lavenex, 2011; Fossum & Graver, 2018). This affiliated status grants the Norwegian central administration privileged access to most parts of the EU administration, which in turn paves the way for deep administrative integration. Moreover, administrative integration might go even further within affiliated nonmember states than in member states due to their exclusion from political representation at EU level. This chapter shows how a lack of political representation in the Council (and the European Parliament) mobilizes an administrative bias in the core-executive of government. Consequently, integration may happen more easily by stealth in affiliated states such as Norway—even though the official position is not to become a political member (Majone, 2005). The chapter is outlined as follows: The next section outlines an organizational approach to administrative integration beyond membership. The next section presents the dataset and methodology succeeded by a presentation and discussion of empirical findings. The chapter concludes with key findings and some suggestions for future research.

An Organizational Approach to Administrative Integration Public administration is not a technical apparatus or tool in the hands of shifting governments. An organizational approach emphasizes that public administration might be an important agent in its own right. This theoretical framework is grounded on the assumption that internal organizational characteristics of public administration may explain how it works and how it changes. Consequently, organizational factors help explaining just how domestic ministries ‘adopt’ the influence of the EU—and thus how integration may happen even without formal EU membership and affect the pursuit of domestic public governance (Egeberg et al., 2016; Egeberg & Trondal, 2015; March & Olsen, 1989; Olsen, 2009: 24; Trondal & Bauer, 2017). Extant literature harbor competing ideas on the extent to which actual decision behavior reflects the organizational structure (see Meier & Capers, 2012; Meyer & Rowan, 1977; Niskanen, 1971). While some observers ascribe lack of government action to political leaders’ lack of will, this chapter advocates that political will is profoundly shaped endogenously by organizational positions which also enable (and constrain)

224 

J. TRONDAL

action. While organizational structure does not necessarily predict nor determine actual decision-making behavior, it does make some choices become more likely than others (e.g., Egeberg & Trondal, 2018). This happens by regulating actors’ access to decision situations, mobilizing attention to certain problems and solutions, structuring patterns of conflict and cooperation (and thus influencing power relationships), and enabling coordination and steering along certain dimensions rather than others. This entails that organizational factors do not impact directly on society; rather, societal consequences can normally only occur via governance processes and public policies. Bounded rationality (Simon, 1957) is one of three key mechanisms that connect role expectations to actual behavior; the organizational structure helps simplify actors’ cognitive worlds by directing attention toward a selection of possible problems and solutions and ways to connect them. This concept holds that decision-makers operate under three limitations: limited information about possible solutions and alternatives, limited cognitive capacity to evaluate and process information, and limited time to make decisions. Consequently, actors opt for a selection of satisfactory alternatives instead of optimal ones, and often turn to their immediate surroundings and readily available data and knowledge to find suitable choices (Simon, 1957). The second mechanism, the logic of appropriateness, views human action as driven by internalized perceptions of what is perceived as appropriate (March & Olsen, 1989). Lastly, actors may find that rule and role compliance is in accordance with their own self-interest. Organizations are thus seen as incentive systems that administer rewards and punishments (e.g., Ostrom & Ostrom, 2015). In sum, these mechanisms may explain why structural characteristics within central administrations bias how the EU ‘hits’ domestic governmental institutions. Essential to this chapter is examining how different structural characteristics of core-executive institutions matter in this respect. Three such structural variables are outlined in the following: horizontal specialization, vertical specialization, and organizational affiliations. Horizontal Specialization Horizontal specialization refers to how tasks or portfolios are divided horizontally within and between organizations. Those policy areas that are encompassed by the same organizational unit are supposed to be more coordinated than those that belong to different units. Luther Gulick

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

225

(1937) contemplated four fundamental principles of horizontal specialization, namely specialization according to (i) territory, (ii) sector/purpose, (iii) function/process, and (iv) clientele. He also emphasized the mutual relationship between specialization and coordination; dossiers that are encompassed by the same organizational unit are more likely to be coordinated than those belonging to different units. Accordingly, empirical studies show that while an overwhelming majority of officials finds coordination to work effectively within their own unit, this holds only for a minority regarding coordination between departments (Kassim et  al., 2013: 188–89). Therefore, the departmental affiliation of various policy units could make a significant difference. In the same vein, cleavages of conflict were assumed to reflect these principles of specialization. For example, specialization according to purpose is likely to mobilize sectoral lines of cooperation and conflict and thereby foster policy consistency within its respected field. We thus expect that officials that are embedded in organizations that are primarily specialized by purpose would coordinate more strongly within their respective policy domains than across these domains. In the case of government ministries, we thus expected that officials in sector ministries report stronger intra-ministerial than inter-ministerial coordination. Yet, domestic public administration in the EU consists within a larger politico-administrative order. Consequently, organizational compatibilities within such an order might matter. Though transnational cooperation on issue-specific tasks and practices has existed for decades, the EU executive center has emerged as a more challenging actor within regulatory networks (e.g., Dehousse, 1997; Egeberg & Trondal, 2015; Joosen & Brandsma, 2017; Levi-Faur, 2011; Majone, 1996). Essentially, the executive center at EU level (concentrated at the Commission and EU agencies (e.g., Bauer & Trondal, 2015; Egeberg, 2006)) is mainly specialized according to purpose (sector) and function (process), largely compatible to national central administrations. Arguably, organizational compatibility across levels of governance is likely to facilitate sectoral allegiances and cooperation across levels of governance. Studies have shown that organizations that are specialized according to similar organizational principles tend to align more easily across levels of governance than those institutions that are organized according to different principles (e.g., Martens, 2010). Hence, the effects of organizational principles at one level of governance may be conditioned by the degree of organizational compatibility across levels (e.g., Cowles et al., 2001; Knill & Lehmkuhl, 1999; March

226 

J. TRONDAL

& Olsen, 1995). Commission DGs and their system of expert committees are largely organized similarly as domestic sector ministries (purpose). By contrast, domestic ministries of foreign affairs (MFA) are mainly specialized based on territory and thus organizationally compatible to the core structure of the Council of Ministers (Council). This suggests that domestic sector ministries are likely to align with the Commission DGs more than with the Council structure. Moreover, the continuous expansion of scope and content in the EEA agreement (from 1994 onward) gives reasons to believe that this pattern, if anything, has expanded over time. The following propositions are derived: P1: Due to organizational compatibility, officials in sector ministries are more likely to interact with the Commission than with other EU institutions. P2: The coordination of EU-related work is relatively lower between than within ministries. Vertical Specialization Vertical specialization denotes division of responsibility and labor within and between levels of authority. This chapter focuses empirically on the effect of hierarchy—or vertical specialization—within ministerial departments. Hierarchy between organizations provides certain decisional inputs, for example, a more general view on hierarchical superior units compared to lower ranked units, that cannot easily be achieved through purely horizontal arrangements. Vertical specialization between organizations means to modify hierarchy by installing an organizational boundary between a superior and a subordinate unit. Agencification, the process whereby regulatory tasks are hived off from ministerial departments into semi-independent regulatory bodies, is a topical example. The New Public Management (NPM) wave that swept across OECD states during the 1980s and 1990s made pleas for greater autonomy, fragmentation, and proliferation of public administrative institutions and systems. As one result, vertical specialization in the form of structural devolution became a major reform trend across Europe (in public administration terms: agencification; in organizational terms: interorganizational vertical specialization). This reform thread led to semi-autonomous agencies enjoying ever more degrees of autonomy at both the national and the EU level (e.g., Bezes et al., 2013; Lægreid et al., 2010).

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

227

Hierarchy within ministries—or intraorganizational vertical specialization—is measured in this chapter by officials’ ranks within their respective ministry (see below). Within organizations, it has been demonstrated that leaders identify with a larger part of the organization than those at lower levels. Leaders also interact more frequently across organizational units and are exposed to broader flows of information than their subordinates. Higher ranked staff in both ministries and subordinated agencies is shown to be more attentive to political signals than lower ranked personnel (e.g., Christensen & Lægreid, 2009; Egeberg & Sætren, 1999). This implies that leaders are better equipped to take into consideration a wider set of goals, alternatives, and consequences when making choices (Egeberg & Trondal, 2018). Studies show that top-ranked staff tend to identify more frequently with organizations as a whole than staff located at lower echelons (Egeberg & Sætren, 1999). Additionally, these officials are exposed to a broader range of information than lower-level staff and thus may be more attentive to broader organizational perspectives than lower ranked personnel (Egeberg & Trondal, 2018). It follows that government officials with lower ranks are more loosely coupled with the political leadership and have a more local perspective on task execution compared to higher ranked staff. This has one important implication: A relative degree of insulation from political leadership makes lower ranked officials more eligible receivers of impulses from EU-level institutions and processes. The following proposition is derived: P3: Ministerial officials at higher ranks are less likely to be strongly involved in EU-related work and thereby less likely to interact with EU institutions compared to ministerial officials at lower ranks. Organizational Affiliation Organizational affiliations consist of primary and secondary structures. A ‘primary structure’ is defined as the structure to which participants are expected to devote most of their loyalty, time, and energy. A typical example would be a bureaucratic unit like a ministry. A ‘secondary structure’ is defined as the structure to which participants are expected to be part-­ timers. It follows that secondary structures are unlikely to shape actors’ decision behavior to the same extent as primary structures. Secondary structures include collegial bodies, committees, and networks (Egeberg, 2012). The EU multilevel administrative system is comprised of a set of

228 

J. TRONDAL

Table 9.1  Summary of propositions Horizontal specialization

Vertical specialization Organizational affiliation

P1: Due to organizational compatibility, officials in sector ministries are more likely to interact with the Commission than with other EU institutions. P2: Due to organizational compatibility, officials in sector ministries are more likely to interact with the Commission than with other EU institutions. P3: Ministerial officials at higher ranks are less likely to be strongly involved in EU-related work and thereby less likely to interact with EU institutions compared to ministerial officials at lower ranks. P4: When taking part in both primary and secondary structures, ministerial officials are likely to attend to and emphasize signals from both structures, albeit most strongly from primary structures. P5: When in conflict, ministerial officials are more likely to emphasize signals from primary than from secondary affiliations.

interconnected organizations at different levels of authority. Ministerial officials that operate within this frame are likely to be exposed to several behavioral premises from their primary and secondary affiliations. Empirical studies show that interdepartmental committees, regional councils, and public-private governing arrangements enhance interaction and coordination and erect trust relationships among the participants; however, the effects are moderate (Ansell & Gash, 2008; Jacobsen, 2015; Lægreid et al., 2016; Stigen, 1991). Moreover, a logic of primacy suggests that the primary affiliation is likely to affect behavior more extensively than the secondary (March, 1994). The following propositions are derived: P4: When taking part in both primary and secondary structures, ministerial officials are likely to attend to and emphasize signals from both structures, albeit most strongly from primary structures. P5: When in conflict, ministerial officials are more likely to emphasize signals from primary than from secondary affiliations (Table 9.1).

Empirical Strategy The study benefits from the unique survey data on the role of central administration in the public governance process as outlined in Chap. 3. This chapter contains data from 1996, 2006, and 2016, giving an overall N of 3562. At all periods, the survey was sent to all officials at the level equivalent to the ‘A-level’ with a minimum of one year in office.

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

229

Table 9.2  Sample size and response rates in the ministry surveys, by year

N Response rates

1996

2006

2016

1497 72%

1874 67%

2322 60%

Appointment at this level usually requires a university degree. Hence, the sample of this survey is the total universe of ‘A-level’ civil servants in Norwegian ministries (see the Appendix for an overview of ministries included). To allow for comparisons, the questionnaires are kept nearly identical over time with just few necessary adjustments. Two caveats should be mentioned: First, studies that rely on respondents’ own behavioral perceptions do not directly observe public governance processes or behavior. Respondents may exaggerate or downplay own role or the role of others. However, the use of three large-N surveys does, however, substantially increase the likelihood of robust conclusions. Moreover, the use of multiple proxies increases the validity of measurement. Second, the use of cross-sectional data may be biased by interindividual variability in perceptions or that it fails to take into account concurring factors that may influence outcome. Ideally, to draw robust conclusions on causal effects, research on developments over time should benefit from longitudinal panel data. Nonetheless, this does not mean that cross-sectional data cannot provide useful and interesting insights regarding continuity and change (Table 9.2).

Administrative Integration Without Membership? This section demonstrates how the EU contributes to a self-reinforcing administrative bias in extreme cases of high integration without membership. Over time, this administrative bias shows a self-reinforcing effect: Norwegian officials dealing with the EU are increasingly far from the political level and are strongly ‘Europeanized’ by being tightly interwoven and influenced by EU institutions. This partly reflects the dynamic character of the EEA agreement, which requires Norwegian law to continuously adjust to new EU legislation. The agreement is based on the premise of dynamic homogeneity, and currently more than 14,000 EU acts have been incorporated into the agreement. The agreement covers the Single Market acquis and a number of

230 

J. TRONDAL

additional policy fields—making it the most extensive form of agreement between the EU and a nonmember as regards regulatory scope and the legal obligations resulting from the contractual relations (Fossum & Graver, 2018; Lavenex, 2011). At the same time, the agreement blocks Norwegian governments qua state from political representation in the Council. Nonetheless, whereas Norway’s associated status does not grant formal access to EU’s decision-making institutions, it provides for administrative participation at various stages of EU’s legislative process. Norwegian ministries and agencies are represented in Commission expert committees and comitology committees, sit on most EU agencies’ boards and committees, and are entitled to second national experts to the Commission. Norwegian civil servants are thus granted privileged access to EU-related work and are largely responsible for handling everyday relationships with EU institutions—in upstream processes (the agenda-setting processes) and particularly in downstream processes (the implementation and practice of EU law). Thus, whereas Norway is politically sidelined in EU decision-making processes, the Norwegian national central administration is granted favored access to the EU bureaucratic apparatus, quite similar to that of EU member states (see below). The launch of the EEA agreement in 1994 marked the beginning of a new era characterized by dense administrative integration between EU institutions and the Norwegian central administration. At the time, the EEA agreement was viewed as a prelude and an interim period toward full EU membership as the prospects of such were both open as well as vigorously pursued by the then-government. However, the following rejection of EU membership in a heated national referendum reintroduced the EEA agreement as Norway’s foundational connection to the EU in the years to come. Additionally, close historical and cultural ties, as well as common interests in a host of policy areas, led to subsequent agreements in areas outside the framework of the EEA agreement (Meld.St 5 (, 2012–13)). At present there are approximately 100 agreements between the Norway and the EU, with the EEA and Schengen being the most encompassing. In effect, while the form of affiliation has remained stable during almost three decades, the scope of the affiliation has undergone significant expansion— partly in response to policy progressions in the EU, partly in response to the dynamic character of the EEA agreement, and partly because of Norway signing evermore sectoral agreements with the EU. In sum, this has increased the subsequent likelihood of Norwegian government

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

231

institutions and governance processes being affected by EU-level institutions and processes. Quite similar to the Commission and EU agencies, the Norwegian central administration is organized according to the principles of purpose and function. From an organizational perspective, this has two substantial implications: First, it is likely to encourage sectoral allegiances and integration across levels of governance, and second to underpin national inter-­ ministerial fragmentation. Moreover, as far as policy harmonization is concerned, the form of affiliation does in fact warrant EEA countries the same level of integration as full member states. Since Norway is not subjected to political representation in the Council, it has been argued that Norwegian sector ministries are likely to be even more strongly ‘hit’ by the Commission than member states’ ministries (Egeberg & Trondal, 1999; Trondal, 2002). This assumption, however, is beyond the scope of this chapter. Like most core executives, the Norwegian central administration is organized into core-executive ministries and subordinated agencies. The system is characterized by a ministerial primacy where subordinated agencies are subject to political control and administrative accountability from the responsible minister and not primarily from a government collegium. The total number of employees within the central administration1 has increased from around 13,000 in 1994 to 21,000 in 2015 of which 4600 are employed in the ministries and the remaining 16,400 are employed in subordinate agencies (DiFi, 2015). While Norwegian ministries are secretariats for the political leadership, mostly partaking planning and coordinating functions (Christensen & Lægreid, 2009), agencies however are mainly responsible for advising ministries and being technical helpers but are also essential ingredients in the political processes of preparing policies and implementing and administering policies. Our survey asked ministerial officials how affected they are, in general, by the EU/EEA/Schengen in their daily work. Table 9.2 displays an overall stability in this regard over time, yet with a small increase during the last decade. In the remaining, only those officials who report being affected (to a fairly small degree, or more) by the EU/EEA/Schengen are included in the analysis.

 Includes all employees in ministries and subordinate units. For full list, see https://www. difi.no/rapporter-og-statistikk/nokkeltall-og-statistikk/ansatte-i-staten#4817 1

232 

J. TRONDAL

Ministry officials were asked about their contacts and participation in EU-level institutions. Table 9.3 reveals two main patterns: first, stability over time in multilevel participation and second, how organizational affiliation matter in this regard (P1). First, while we observe increased contact between sector ministries, the Commission, expert committees and comitology committees from 1996 to 2006, these patterns remain stable during the last decade. This suggests that Norwegian ministries have experienced a threshold in its access to the EU administration. Correspondingly, the gap between the sector ministries and the MFA shows a notable increase between 1996 and 2006, and then remains stable from 2006 to 2016. This gap reflects organizational compatibility between sector ministries and the EU administration (P2). One additional explanation is that the EEA agreement makes the Commission the main interaction partner for sector ministries, thus biasing the access structure for Norwegian ministries vis-à-vis EU institutions. Also, worth noting is that whereas contacts toward the Commission have remained stable during the past decade, sector ministries’ contact toward EU agencies have decreased. This reflects the fact that national agencies have acquired a larger position as access points for EU agencies (see Egeberg & Trondal, 2015). Second, the data shows that ministry affiliation matters (P1) as Table 9.3 reveals a substantial difference between staff affiliated to sector ministries and staff affiliated to the MFA.  Except for the European Parliament (2016 data only), sector ministries are consistently more involved with EU institutions, even the Council. Moreover, reflecting compatible principles of organization, officials from sector ministries tend to concentrate attention toward the Commission, EU agencies, and Commission expert committees, whereas officials from the MFA give more attention to the Council.

Table 9.3  Percentage of officials who report being affected by the EU/EEA/ Schengen, by year

To a very large extent To a fairly large extent To some extent To a fairly small extent Not affected N

1996

2006

2016

11 12 23 27 27 100 (1463)

13 12 18 21 37 100 (1704)

14 12 23 31 20 100 (1773)

233

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

As mentioned above, this patterns also reflect the access structure embedded in the EEA agreement. Next, respondents were asked about their coordination behavior. Table 9.4 reveals an overall low level of coordination regarding EU-related work and an increasing reliance on intra-ministerial coordination over time. In support of P2, coordination is slightly higher within ministries than between ministries at all three points in time. Moreover, the gap between intra-ministerial and low-level inter-ministerial coordination appears to increase over time. Since the Foreign Office chairs the high-­ level inter-ministerial coordination committee, these data also testify the declining role of the FO over time. This in sum shows rising reliance on sector ministries and sector specialists and less on the generalists from the FO. Also worth noting is the general decrease in this type of coordination behavior, particularly from 2006 to 2016. One important caveat should Table 9.4  Percentage of officials who report contact with* and/or participation in** EU-level institutions, by year and ministerial affiliation (sector ministries (SM)/Ministry of Foreign Affairs (MFA)) (percentages) 1996

2006 SM

2016

SM

MFA

MFA

Contact with the Commission

68

32

Contact with the Council

–

–

–

–

57 43

Contact with the European Parliament

–

–

–

–

22 78

Contact with EU agencies

–

–

90 10

76 24

Participation in expert committee(s)

82

18

95

5

89 11

Participation in comitology committee(s) 86

14

100

0

Participation in the Council

–

–

–

–

60 40

Participation in committees, boards, etc. in EU agencies

–

–

–

–

94

83 17

SM

MFA N

83 17

100

0

6

100 (115) 100 (21) 100 (9) 100 (39) 100 (94) 100 (21) 100 (10) 100 (18)

* The ‘contact’ variables combine values 1 and 2 on the following four-point scale: (1) App. every week, (2) app. every month, (3) a few times, (4) never ** The ‘participation’ variables apply value 1 on the following three-point scale: (1) several times, (2) once, (3) never

234 

J. TRONDAL

be remarked though: Although a common assumption is that the primary objective of coordination committees is to coordinate, studies have shown low levels of substantial coordination in inter-ministerial coordination committees (Trondal, 2001). These committees are characterized by mutual inter-ministerial information sharing but low levels of coordination concerning tasks, policies, and positions. Moreover, bearing in mind that EU affectedness has remained stable over time (Table 9.2), one may contemplate that the decrease in coordination behavior reflects EU-related work becoming more routinized, which in turn diminishes needs for formal dialogue through committees. As proposed (P3), Table 9.5 confirms that officials’ ranks are negatively associated with interaction with EU-level institutions. This finding is robust since the pattern unfolds consistently over time both with regard to contact with various EU institutions and participation in EU committees—including those under EU agencies. In other words, coordination is consistently exercised by low-level bureaucrats, thus reinforcing a politico-­ administrative gap. Interestingly, Table  9.5 shows a 10 to 15 percent increase of interaction with EU-level institutions for low-level officials from 2006 to 2016 and a corresponding decrease in interaction for medium and high-level officials on all variables. This provides further support for the assumption of a self-reinforcing politico-administrative gap. Moreover, the sheer increase in interaction is also fostered by an increasing

Table 9.5  Percentage of officials who have met in ministerial coordination committees, by year*

Intra-ministerial coordination committees** Low-level inter-ministerial coordination committees*** High-level inter-ministerial coordination committees**** N

1996

2006

2016

18 17 5 100 (1038)

18 15 7 100 (1056)

13 9 5 100 (1402)

* This table applies value 1 on a three-point scale: (1) multiple times, (2) once, (3) never ** These committees are established to pursue intra-ministerial coordination of EU-related work *** These committees are headed by the responsible sector ministry to coordinate across affected sector ministries **** This committee is headed by the Ministry of Foreign Affairs to coordinate those dossiers that the low-level inter-ministerial coordination committees do not solve

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

235

number of agreements between Norway and the EU, affecting ever more policy areas and government officials. A similar analysis (Table  9.6) on the effect of rank on coordination behavior reveals a similar pattern: a negative correlation between rank and participation in coordination committees and a parallel negative correlation between rank and those reporting that their ministry’s work has been subject to coordination from the Prime Minister’s Office, the MFA, other ministries, the parliament, and/or interest groups. Officials at low ranks score consistently higher on these variables than officials at medium and high ranks. This finding is also consistent over time, supporting the observation of a self-reinforcing politico-administrative gap. This behavior, we argue, measures degrees of involvement in EU-work, not merely coordination as such. It follows that lower level officials will enjoy a more comprehensive view of their ministry’s EU-related work compared to higher Table 9.6  Percentage of officials who report contact with* or participation in** the following EU-level institutions, by year and rank (lower-level officials (L)***/ medium and higher-level officials (MH)****) (percentage) 1996 L

2006

MH L

2016

MH L

MH N

Contact with the Commission

74 26

63 37

76

24

Contact with the Council Contact with the European Parliament Contact with EU agencies Participation in comitology committee(s) Participation in the Council Participation in committees, boards, etc. in EU agencies

– – – 71 – –

– – 73 70 – –

90 100 83 86 80 72

10 0 17 13 20 28

– – – 29 – –

– – 27 30 – –

100 (115) 100 (8) 100 (20) 100 (39) 100 (21) 100 (10) 100 (18)

* The ‘contact’ variables combine values 1 and 2 on the following four-point scale: (1) App. every week, (2) app. every month, (3) a few times, (4) never ** The ‘participation’ variables apply value 1 on the following three-point scale: (1) several times, (2) once, (3) never *** Lower-level officials include the following ranks: 1996: executive officer, higher executive officer, principal officer, assistant director. 2006: executive officer/adviser, principal officer/adviser, assistant director/adviser. 2016: adviser/senior adviser, specialist director/special adviser or equivalent **** Medium and higher-level officials include the following ranks: 1996: director, director general, positions higher than director general. 2006: director/adviser, director general/adviser, positions higher than director general. 2016: director or equivalent, director general or equivalent

236 

J. TRONDAL

level officials. Supporting these findings in greater detail, Table 9.6 displays increased participation in high-level coordination committees from around 55 percent in 1996/2006 to almost 70 percent in 2016. Moreover, from 2006 to 2016 our data shows an average of 11 percent increase in participation of lower level officials in the three types of coordination committees. A similar pattern can be found when examining modification/change due to coordination with other actors: From 2006 to 2016 we see an average of 16 percent increase in lower level officials reporting that their ministry’s work had been modified or changed due to the actors listed above. Table 9.7 examines whether various patterns of coordination behavior are mutually supplementary or contradictory, and thus deemed important by the same officials. To merely probe patterns, Table 9.7 applies two sets of variables from the 2016 survey: the degree to which officials participate in intra- and inter-ministerial coordination committees (variables 1 to 3) and the extent to which they report that own ministry’s position in EU-related work is modified due to coordination with various other institutions (variables 4 to 8). The findings suggest that different coordination behavior tend to be mutually supplementary, but also that substantive coordination is weakly associated with participation in coordination committees. From the latter, we infer that coordination committees, albeit meeting actively, are not instrumental coordinating devices in EU-related work. Next, Table  9.8 illustrates the perceived importance of both primary and secondary structures (P4). Two main findings are displayed: as proposed, primary structures are deemed significantly more important than secondary structures. Moreover, this pattern is robust over time. The most noteworthy change is the increased importance ascribed to national agencies from 1996 to 2016. This finding reflects the ‘agencification’ wave in Norway, as in most OECD countries, during the same time period (e.g., Verhoest et al., 2010) and the parallel ‘agencification’ of the EU administration (Egeberg & Trondal, 2017). While Table 9.3 revealed decreasing contacts between sector ministries and EU agencies, Table  9.7 shows a corresponding decrease in perceived importance of EU agencies. This finding reflects the increased importance of national agencies, and not ministerial departments, as national access points for EU agencies. Studies have demonstrated a tendency for EU agencies to bypass the ministerial level and interact directly with the national agencies contributing to a ‘direct’ multilevel administrative structure (e.g., Egeberg & Trondal,

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

237

Table 9.7  Percentage of officials who report the following coordination behavior*, by year and rank (lower-level officials (L)**/medium and higher-level officials (MH)***) 1996 L Participation in intra-ministerial coordination committee Participation in low-level inter-ministerial coordination committee Participation in high-level inter-ministerial coordination committee Modified/changed own ministry’s position in EU-related work due to coordination with the Prime Ministers’ Office Modified/changed own ministry’s position in EU-related work due to coordination with the MFA Modified/changed own ministry’s position in EU-related work due to coordination with other ministries Modified/changed own ministry’s position in EU-related work due to coordination with the National Parliament Modified/changed own ministry’s position in EU-related work due to coordination with the Interest groups

2006

MH L

2016

MH L

MH N

71 29

62 38

73 27

100 (181) 100 (116) 100 (61) 100 (35)

83 17

75 25

84 16

56 44

55 45

69 31

63 38

48 52

70 30

63 37

54 46

73 27

100 (60)

65 35

64 37

69 31

100 (82)

83 17

60 40

77 23

100 (20)

76 24

58 42

77 23

100 (27)

*This table applies value 1 on the following three-point scale: (1) multiple times, (2) once, (3) never ** Lower level officials include the following ranks: 1996: executive officer, higher executive officer, principal officer, assistant director. 2006: executive officer/adviser, principal officer/adviser, assistant director/adviser. 2016: adviser/senior adviser, specialist director/special adviser or equivalent *** Medium and higher level officials include the following ranks: 1996: director, director general, positions higher than director general. 2006: director/adviser, director general/adviser, positions higher than director general. 2016: director or equivalent, director general or equivalent

2009a). This may have taken off some EU workload from the ministerial level. Finally, to probe P5, respondents were asked to reflect on conflict behavior: They were questioned how they prioritize when conflicts appear between the wishes of their political leadership and the requirements of EU law in their EU-related work. Table 9.9 demonstrates that, when in conflict, most ministerial officials seek to compromise between the wishes

238 

J. TRONDAL

Table 9.8  Inter-correlation matrix on coordination behavior (Pearson’s r, 2016 data) 1) 2) 1. Participation in intra-ministerial coordination committee 2. Participation in low-level inter-ministerial coordination committee 3. Participation in high-level inter-ministerial coordination committee 4. Modified/ changed own ministry’s position in EU-related work due to coordination with the Prime Ministers’ Office 5. Modified/ changed own ministry’s position in EU-related work due to coordination with the MFA 6. Modified/ changed own ministry’s position in EU-related work due to coordination with other ministries

3)

4)

5)

6)

7)

8)

N

0.49** 0.45** 0.16** 0.32** 0.26** 0.17** 0.16** 684

41**

0.07 0.22** 0.25** 0.10

0.15** 568

0.17** 0.31** 0.33** 0.17** 0.19** 396

0.71** 0.62** 0.62** 0.36** 359

0.73** 0.49** 0.49** 362

0.53** 0.60** 359

(continued)

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

239

Table 9.8  (continued) 1) 2)

3)

4)

5)

6)

7)

7. Modified/ changed own ministry’s position in EU-related work due to coordination with the National Parliament 8. Modified/ changed own ministry’s position in EU-related work due to coordination with interest groups

8)

N

0.69** 351

Table 9.9  Percentage of officials who report that the following institutions are important* when making decisions on their own policy area, by year

Primary structures

Secondary structures

N

Own ministry Subordinated agencies Other ministries National Parliament National government Commission Council European Parliament EU agencies

1996

2006

2016

95 53 64 78 86 23 – – – 100 (1043)

96 65 64 77 90 29 – – 11 100 (1007)

98 69 72 80 91 22 14 12 9 100 (1340)

*This table combines values 1 and 2 on a five-point scale: (1) very important, (2) fairly important, (3) somewhat, (4) fairly unimportant, (5) very unimportant, (6) Do not know/not relevant

of their political leadership and the requirements of EU law. This suggests that ministry officials, not only agency officials as studied by Egeberg and Trondal (2009b), are ‘double-hatted’ in their EU-related work. Moreover, Table  9.9 illuminates the intrusiveness of the ‘double-hatted’ national

240 

J. TRONDAL

Table 9.10  Percentage of officials who report the following priorities if conflicts occur between the wishes of their national political leadership and the requirements of EU law*

Primarily follow the wishes of own political leadership Try to combine the wishes of own political leadership and EU law Primarily follow the requirement of EU law N

Ministry

Agency

11 84 5 608 (100)

15 68 17 476 (100)

* The table includes those officials that incorporates and/or practices EU legislation within their own policy field. The category ‘not relevant/not occupied with such tasks’ are coded missing

central administrations since ministry officials seem to serve ‘two masters’. This observation is an even stronger probe of the idea of a national ‘double-­hatted’ central administration, given that ministry officials are less likely to ‘go to Brussels’ than agency officials simply by being more strongly bound by national political steering. Among the few respondents reporting being ‘single hatted,’ their primary structures (the wishes of own political leadership) are deemed slightly more important than secondary structures (the requirement of EU law) (Table 9.10).

Conclusions The chapter has made two main contributions to the volume: Theoretically, it has examined the role of organizational factors in public governance in general, as well as in administrative integration in particular. The data suggests that administrative integration is promoted and nudged by ‘favorable’ organizational conditions at the domestic level of government. Moreover, the chapter highlights how strong integration without political membership in the EU fuels the administrative state. Following the public administration turn in EU studies, one might also envisage that while a member state withdraws from political membership in the EU, domestic agencies might find themselves somehow integrated with EU administrative networks. Moreover, the observed mechanism may play out more generally in other (member) states, that is, the observed general EU coordination techniques could also be expected to have similar effects in member states, especially the higher involvement of lower ranks than involvement of higher (more political) levels. The general theoretical idea is captured by the public administration approach to European integration

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

241

that sees the EU as consisting of interconnected sets of agencies, ministries, and regulatory networks. Empirically, this study demonstrates how the EU contributes to a self-­ reinforcing administrative bias in domestic core executives. To do so, the chapter applies an extreme case of high integration without formal EU membership. Over time, this administrative bias gets a self-reinforcing effect: Norwegian officials dealing with the EU are increasingly far from the political level (low-ranked staff) and are (surprisingly) strongly ‘Europeanized’ by being tightly interwoven and influenced by EU institutions. In greater detail, the data consistently displays a relatively high degree of interaction between sector ministries and EU-level executive institutions (P1). It also affirms a higher level of coordination within than between ministries (P2) and that officials of lower ranks are more extensively involved at the EU level than officials of higher ranks (P4). Finally, the importance of primary organizational structures is affirmed (P4), and the data demonstrates a strong tendency of ‘double-hattedness’ amongst ministerial officials (P5). Future longitudinal studies are needed that document continuity and change in patterns of administrative integration.

Appendix List of Ministries in Survey (2016) Office of the Prime Minister Ministry of Agriculture and Food Ministry of Children and Equality Ministry of Climate and Environment Ministry of Culture Ministry of Defense Ministry of Education and Research Ministry of Finance Ministry of Foreign Affairs Ministry of Health and Care Services Ministry of Justice and Public Security Ministry of Labour and Social Affairs Ministry of Local Government and Modernization Ministry of Petroleum and Energy Ministry of Trade, Industry and Fisheries Ministry of Transport and Communications

242 

J. TRONDAL

References Ansell, C., & Gash, A. (2008). Collaborative governance in theory and practice. Journal of Public Administration Research and Theory, 18(4), 543–571. Bauer, M., & Trondal, J. (Eds.). (2015). The Palgrave handbook of the European administrative system. Palgrave Macmillan. Bezes, P., Fimreite, A. L., Lidec, P. L., & Lægreid, P. (2013). Understanding organizational reforms in the modern state: Specialization and integration in Norway and France. Governance, 26(1), 147–175. Christensen, T., & Lægreid, P. (2009). Living in the past? Change and continuity in the Norwegian central civil service. Public Administration Review, 69(5), 951–961. Cowles, M. G., Caporaso, J. A., & Risse-Kappen, T. (2001). Transforming Europe: Europeanization and domestic change. Cornell University Press. Dehousse, R. (1997). Regulation by networks in the European Community: The role of European agencies. Journal of European Public Policy, 4(2), 246–261. DIFI. (2015). Ansatte i staten. Retrieved February 2017, from https://www.difi. no/rapporter-­og-­statistikk/nokkeltall-­og-­statistikk/ansatte-­i-­staten Egeberg, M. (Ed.). (2006). Multilevel union administration. Palgrave Macmillan. Egeberg, M. (2012). How bureaucratic structure matters: An organizational perspective. In B.  G. Peters & J.  Pierre (Eds.), The SAGE handbook of public administration. SAGE. Egeberg, M., Gornitzka, Å., & Trondal, J. (2016). Organization theory. In C. Ansell & J. Torfing (Eds.), Handbook on theories of governance. Edward Elgar. Egeberg, M., & Sætren, H. (1999). Dentities in complex organizations: A study of ministerial bureaucrats. In M.  Egeberg & P.  Lægreid (Eds.), Organizing political institutions. Scandinavian University Press. Egeberg, M., & Trondal, J. (1999). Differentiated integration in Europe: The case of EEA country, Norway. Journal of Common Market Studies, 37(1), 133–142. Egeberg, M., & Trondal, J. (2009a). National agencies in the European administrative space: Government driven, commission driven or networked? Public Administration, 87(4), 779–790. Egeberg, M., & Trondal, J. (2009b). Political leadership and bureaucratic autonomy: Effects of agencification. Governance, 22(4), 673–688. Egeberg, M., & Trondal, J. (2015). Hy strong coordination at one level of government is incompatible with strong coordination across levels (and how to live with it): The case of the European Union. Public Administration, 94(3), 579–592. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies. What have we learned (and where to go from here?). Journal of Common Market Studies, 55(4), 675–690.

9  EUROPEAN INTEGRATION AND THE ADMINISTRATIVE STATE 

243

Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Egeberg, M., Schaefer, G. F., & Trondal, J. (2003). The many faces of EU committee governance. West European Politics, 26(3), 19–40. Fossum, J.  E., & Graver, H.  P. (2018). Squaring the circle on Brexit. Bristol University Press. Genschel, P., & Jachtenfuchs, M. (Eds.). (2014). Beyond the regulatory polity. Oxford University Press. Gulick, L. (1937). Notes on the theory of organization. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration. Jacobsen, D.  I. (2015). Regional governance networks: Filling in or hollowing out? Scandinavian Political Studies, 38(2), 115–136. Joosen, R., & Brandsma, G. (2017). Transnational executive bodies: EU policy implementation between the EU and member state level. Public Administration, 95, 423–436. Kassim, H., Peterson, J., Bauer, M. W., Connolly, S., Dehousse, R., Hooghe, L., & Thompson, A. (2013). The European Commission of the twenty-first century. Oxford University Press. Knill, C., & Lehmkuhl, D. (1999). How Europe matters. Different mechanisms of Europeanization. European Integration Online Papers (EIoP), 3(7) Available at SSRN https://ssrn.com/abstract=302746 or. https://doi.org/10.2139/ ssrn.302746 Lægreid, P., Randma-Liiv, T., Rykkja, L. H., & Sarapuu, K. (2016). Coordination challenges and administrative reforms. In G. Hammerschmid, S. Van de Walle, R. Andrews, & P. Bezes (Eds.), Public administration reforms in Europe: The view from the top. Edward Elgar. Lægreid, P., Rolland, V. W., Roness, P. G., & Ågotnes, J. E. (2010). The structural anatomy of the Norwegian state: Increased specialization or a pendulum shift? In P. Lægreid & K. Verhoest (Eds.), Governance of public sector organizations. Palgrave Macmillan. Lavenex, S. (2011). Concentric circles of flexible “European” integration: A typology of EU external governance relations. Comparative European Politics, 9(4/5), 372–393. Levi-Faur, D. (2011). Regulatory networks and regulatory agencification: Towards a single European regulatory space. Journal of European Public Policy, 18(6), 810–829. Majone, G. (1996). Regulating Europe. Routledge. Majone, G. (2005). Dilemmas of European integration. Oxford University Press. March, J. G. (1994). A primer on decision-making. The Free Press. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions: The organizational basis of politics. Free Press. March, J. G., & Olsen, J. P. (1995). Democratic governance. The Free Press.

244 

J. TRONDAL

Martens, M. (2010). Organized administrative integration. ARENA Report, No. 2. Meier, K. J., & Capers, K. J. (2012). Representative bureaucracy: Four questions. In B. G. Peters & J. Pierre (Eds.), The SAGE handbook of public administration. SAGE. Meld. St. 5 (2012–13). The EEA Agreement and Norway’s other agreements with the EU. Report to the Storting (White Paper), Oslo: Ministry of Foreign Affairs, https://www.regjeringen.no/globalassets/upload/ud/vedlegg/europa/ nou/meldst5_ud_eng.pdf Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Niskanen, W. (1971). Bureaucracy and representative government. Aldine. Olsen, J. P. (2009). Change and continuity: An institutional approach to institutions of democratic government. European Political Science Review, 1(1), 3–32. Olsen, J. P. (2018). Democratic accountability and the changing European political order. European Law Journal, 24(1), 77–98. Ostrom, E., & Ostrom, V. (Eds.). (2015). Choice, rules and collective action. ECPR Press. Rittberger, B., & Blauberger, M. (2018). Introducing the debate section: “The EU in crisis: EU studies in crisis?”. Journal of European Public Policy, 25(3), 436–439. Simon, S. A. (1957). Administrative behavior. The Free Press. Stigen, I. (1991). Avbyråkratisering og Modifisert Forhandling? Om Bruk av Prosjektorganisasjon i Norsk Sentraladministrasjon. Norsk Statsvitenskapelig Tidsskrift, 7(3), 173–191. Trondal, J. (2001). Administrative integration across levels of governance. Doctoral dissertation,. Department of Political Science, University of Oslo. Trondal, J. (2002). Beyond the EU membership-non-membership dichotomy? Supranational identities among national EU decision-makers. Journal of European Public Policy, 9(3), 468–487. Trondal, J. (2007). The public administration turn in integration research. Journal of European Public Policy, 14(6), 960–972. Trondal, J. (2010). An emergent European executive order. Oxford University Press. Trondal, J., & Bauer, M.  W. (2017). Conceptualizing the European multilevel administrative order: Capturing variation in the European administrative system. European Political Science Review, 9(1), 73–94. Trondal, J., & Veggeland, F. (2003). Access, voice and loyalty: The representation of domestic civil servants in EU committees. Journal of European Public Policy, 10(1), 59–77. Verhoest, K., Roness, P.  G., Verschuere, B., Rubecksen, K., & MacCarthaigh, M. (2010). Autonomy and control of state agencies. Palgrave Macmillan.

CHAPTER 10

Incomplete Contracting and Policy Influence

This chapter contributes to studies of differentiated European integration, multilevel administrative integration, and the role of government agencies in policy uploading in the EU. European integration has woven EU institutions and processes into the political, administrative, and judicial life of European nation states. This observation led Olsen (1998: 2) to ask ‘what happens to organized political units when they become part of a larger unit?’. Studies show that different forms of integration contribute to differentiated institutional architectures in the EU and varieties of policy adaptation in EU member states and associated ‘third countries’ (Leruth et al., 2022). Whereas formal membership in the EU grants the state full access to EU decision-making processes, associated membership implies partial access without being fully integrated in the cycle of decision-­making in the EU. Association agreements and ‘quasi-memberships’ have led to varieties of administrative integration across levels of governance (Trondal & Bauer, 2017). Concomitantly, associated EU members are overly reliant on administrative interaction with the EU administrative system (Egeberg, 2006). Arguably, bureaucratic structures are essential providers of political premises in democratic policymaking (Peters, 2018; Waldo, 1952). This chapter illustrates how administrative participation and integration is likely to affect policy uploading toward EU executive institutions. Chapter 8 outlined the contours of a European multilevel administrative system (MLA) consisting of strongly interconnected administrative © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_10

245

246 

J. TRONDAL

bodies across levels of governance (cf. Bauer & Trondal, 2015; Trondal & Bauer, 2017). Administrative capacity building at the EU level is seen as challenging administrative autonomy among national governments (Egeberg & Trondal, 2016), yet also providing access to opportunities for domestic bureaucratic bodies toward EU institutions (Genschel & Jachtenfuchs, 2014; Trondal, 2010). This chapter demonstrates how an associated non-EU member state accesses EU institutions and contributes with what Benz et  al. (2016) call ‘policy-shaping MLA.’ The chapter examines what role ‘third countries’ outside the EU may enjoy with respect to policy uploading in the EU, and second how organizational factors may shape such processes. The chapter benefits from a large-N study (N=1031) among agency officials, supplemented with semi-structured interviews. The chapter offers two distinct contributions to the book: (i) First, it probes the effects of two distinctive drivers of policy influence: access to secondary structures, and available domestic administrative capacities. Contrary to the majority of literature in the field of differentiated integration which is dominated by studies of legal commitments and European regulatory transfer, this chapter offers an organizational approach to the study of EU-third country relations and differentiated administrative integration. (ii) Second, it presents novel large-N data on how domestic agency officials perceive their influence vis-à-vis the Commission and EU agencies. A vast literature on Europeanization established how Europe ‘hit home’ (Börzel & Risse, 2000; Maestenbroek, 2018), but also how member state institutions ‘hit back’ (Beyers & Trondal, 2004). Within this broader literature, an emergent organization theory literature has focused on how administrative institutions may bias processes of policy downloading and uploading (Egeberg, 2006; Bauer & Trondal, 2015). Administrative integration represents processes that enable policy transfer across levels of governance and thus also as a distinct administrative tool for national government institutions to influence EU decision-making processes (Trondal, 2010). This is particularly so for affiliated non-EU countries that are not granted full access to EU decision-making. During the course of EU integration, administrative integration has incrementally added new institutional layers to the EU executive order and added ever more administrative access points for ‘external’ actors (Trondal, 2010; Trondal &

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

247

Peters, 2013). As a consequence, Peters (1994: 21) has labeled the EU an ‘agenda-setter’s paradise’. Yet, despite its monopoly of agenda initiative and an ‘engine of integration’ (Pollack, 1993), the Commission has come under increased pressure due to the politicization of EU governance, international polycrises, and informalized decision-making processes (Deters & Falkner, 2021; Riddervold et al., 2021; White, 2022). As a more general point of departure, in a policymaking system where agenda items are put on the agenda with little effort, a heavy burden is put on executive institutions to structure, control, and filter what enters the policymaking machinery. In the EU, a large part of that responsibility rests with the Commission. Yet, the government may enjoy limited control over what items make it into its agenda. One of the most precious and scarce resources of governments is the availability of attention. Governments are usually only capable of managing a limited number of items at the same time. Executive institutions may thus be essentially reactive institutions in their agenda role (Mayntz & Scharpf, 1975). Since the Commission and EU agencies are intimately related to member state government institutions in its agenda role (Egeberg & Trondal, 2017), this chapter examines what role domestic agencies in an EU third-country play in accessing the agenda-setting processes of EU executive institutions. The chapter shows how such processes are channeled through interaction through a varied set of secondary administrative access structures within each of these institutions. Such consultative expert groups are under the Commission (Blauberger & Rittberger, 2014; Coen & Thatcher, 2008; Dehousse, 1997; Gornitzka & Sverdrup, 2008; Joerges & Neyer, 1997; Krick & Gornitzka, 2020; Levi-Faur, 2013; Mathieu, 2016; Vestlund, 2017), comitology committees who monitor implementation of EU regulation (Brandsma et al., 2008), European Regulatory Networks (ERNs) consisting of administrative networks of officials from national agencies and ministers (Maesternbroek & Martinsen, 2018), and management boards of EU agencies that often include representatives from both member states and affiliated states (Egeberg & Trondal, 2017). These secondary structures are populated by national officials from member states, but also regularly invites actors from associated countries who have been granted access through association agreements, as in the case of the Economic Area Agreement (EEA). By systematically organizing external institutions into EU secondary structures, the Commission and EU agencies may be made aware of, sensitive to, and reactive toward the problems, solutions, and consequences of concern to these external actors.

248 

J. TRONDAL

The chapter proceeds as follows: The next section outlines a theoretical argument that offers propositions as to what organizational factors are likely to make government institutions generally influential and suggests how to measure this empirically. The subsequent two sections outline the data and methodologies of the study and then present the main findings. The chapter closes by examining the implications of our argument and empirical findings.

What Makes Government Institutions Influential? Departing from the assumption that varieties of organizational characteristics may lead to varieties of institutional influence in political systems (March & Olsen, 1989), this section outlines two particular organizational variables: organizational affiliations and organizational capacity (Egeberg & Trondal, 2018). Organizational structures serve to systematically buffer the information and role expectations relevant for their members, simplifying their search for alternatives, their preference formation, and, ultimately, their biased and bounded choice of behavior (Beckman, 2021; Egeberg, 1999; March, 2008; Steinmo et al., 1992). Moreover, the local rationality of actors is thereby systematically aggregated by this buffer function into organizational rationality (Gulick, 1937; 1957). The organizational selection of relevant information, premises for decision-­ making and of role enactment, might systematically affect how actors think, feel, and act. Subsequently, administrative behavior is expected to systematically reflect organizational structures (Stinchcombe, 2001). Studies show that decision-makers’ position in the organization structure, when controlling for various demographic backgrounds of actors, serve as a key variable when it comes to accounting for decision behavior within the organization (Christensen & Lægreid, 2009; Olsen & Lægreid, 1984; Meier & Nigro, 1976). Even officials on short-term contracts, who therefore cannot become re-socialized in a deeper sense, tend to act primarily on behalf of their new organizational unit (Trondal et al., 2008). A persistent observation in literature on administrative integration in Europe is how organizational architectures shape decision-making processes across levels of governance (Bauer & Trondal, 2015; Egeberg, 2006). For example, studies show that administrative bodies that enjoy compatible principles of organizational specialization have a likelihood of interacting and coordinating activities (Egeberg, 1999; Egeberg & Trondal, 1999; Trondal, 2010). To illustrate this point, studies show that government

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

249

institutions that are specialized according to a territorial principle of specialization, such as the Council in the EU, tend to mobilize national ministries who have an inter-ministerial coordinating role across the government apparatus (e.g., foreign offices or prime minister’s offices) and insert an intergovernmental decision-making logic among government officials. By contrast, government institutions that are specialized according to a sectoral principle of specialization, as the Commission and EU agencies, tend to mobilize sector ministries and agencies and evoke sectoral behavioral logics among government officials (Beyers & Trondal, 2004; Egeberg et al., 2003; Trondal & Veggeland, 2003). This implies that government officials are primarily guided by sectoral agendas, priorities, problems, and solutions when conducting their tasks (Trondal, 2010). Organizational theory has identified four generic independent variables: organizational structure, organizational demography, organizational locus, and organizational culture (Egeberg, 2006). These variables represent generic dimensions in the organization literature that may allow for generalizations (Egeberg, 2012; Egeberg & Trondal, 2018). This chapter zooms in on two such variables: organizational affiliation and organizational capacity. Organizational Affiliations Organizational affiliations denote all institutional commitments, both full-­ time (e.g., an agency position) and part-time (e.g., participation in advisory boards and working groups). Organizational literature has so far largely treated this variable as a dichotomous distinction between primary and secondary organizational structures. As outlined in previous chapters in this volume, a primary structure refers to the organization that represents an official’s full-time commitment (their main employer) to which participants are expected to devote most of their loyalty, time, and energy. A typical example would be a bureaucratic unit like a Commission DG. Secondary structure is an umbrella term that denotes all other part-­ time commitments such as advisory boards, working groups, and expert committees both within and across primary structures. A ‘secondary structure’ is defined as the structure to which participants are expected to be part-timers (Egeberg, 2012). However, this stringent dichotomy between primary and secondary structures hides a potential variety of scales and forms of affiliation that are likely to exist in contemporary government systems (Ansell et al., 2016). Secondary structures are not likely to serve

250 

J. TRONDAL

as a coherent, equally invasive set of structures but rather a complex, hybrid, and multifaceted architecture of different degrees of involvement, constellations, mandates, and agendas. We thus suggest that organizational affiliation would benefit from escaping the ‘tyranny of dichotomies’ (Olsen, 2009) and that the implications of secondary structures are contingent on degrees of involvement among its members. To this end, we suggest a broader approach to secondary structures by introducing a scale of involvement from low to medium, to high. The EU multilevel administrative system is comprised of a set of interconnected secondary structures at different levels of authority. National agencies may access EU institutions through a variety of secondary affiliations. As a consequence, participants are likely to be exposed to multiple behavioral premises not only from primary structures, but from secondary structures too. Empirical studies show that interdepartmental committees, regional councils, and public-private governing arrangements enhance interaction and coordination, and foster trust relationships among the participants. However, the effects are moderate (Ansell & Gash, 2008; Lægreid et al., 2016). It follows that secondary structures are unlikely to shape actors’ decision behavior to the same extent as primary structures. We predict that secondary structures may render differentiated effects depending on the degree of actor involvement. Administrative Capacity Administrative capacity refers to the set of organizational resources assigned to a specific field or unit/department, and it thus denotes requisite organizational resources of a polity (Joaquin & Greitens, 2021). The concept of administrative capacity has been widely applied across a variety of fields within political science literature. Administrative capacity building has been central to literature on political development and state building (Almond, 1965; Carpenter, 2001; Mann, 1984). Mann (1984) contended that the ability of a state to act autonomously largely derives from strong infrastructural power that is vested in its administration. Such ideas are echoed by Carpenter (2001), who sees administrative capacity in agencies as instrumental in gaining de facto independence from political actors in forging bureaucratic autonomy in the policy process. Similar findings are observed by Verhoest et  al. (2010), who report a positive relationship between size of an agency (capacity) and autonomy. Within EU implementation research, varying levels of implementation effectiveness have

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

251

been attributed to the availability of requisite administrative capacities within member state administrative bodies (Adam et al., 2021). This line of scholarship has examined how varieties of administrative capacities serve as an intervening factor in policy implementation (Adam et  al., 2021; Hille & Knill, 2006) as well in shaping the agenda-setting role of core-­ executive institutions (Deters & Falkner, 2021; Kassim et  al., 2017). Studies of public policy and administration also show that the extent to which agency personnel assign weight to political signals from their respective ministries partly depends on the number of ministerial staff available for monitoring and steering agency activities (Egeberg & Trondal, 2009). Finally, studies of IOs show that the size of international secretariats is associated with the likelihood of IO survival (Debre & Dijkstra, 2021). We may thus argue that the sheer number of roles that are to be filled in an organization indicates an organization’s capacity to initiate policies, develop alternatives, and to implement final decisions. Correspondingly, organizational scholars argue that problem solving and problem attention are dependent on the degree to which such activities are underpinned by organizational capacity (Egeberg et al., 2016: 34). This argument builds on Herbert Simon’s (Simon, 1957) seminal concept of bounded rationality which maintains that decision-making behavior is limited by available information, and subsequently, that behavior is guided by local attention, rationalities, and experiences. In this chapter, we thus predict that administrative capacity is positively related to agency influence. Requisite administrative capacity increases the likelihood of directing attention to EU-related tasks and consequently to influence policy processes. Accordingly, it is assumed that available resources serve as a prerequisite for national officials to influence EU administrative institutions.

Empirical Strategy Our study benefits from a large-N survey (N=1031) administered amongst A-level officials in 45 Norwegian agencies (see Chap. 3 in this volume). Benefiting from the recent survey in 2016, this chapter employs a dataset with a response rate of 60 percent. Moreover, the survey data has been triangulated with the help of six supplementary semi-structured and in-­ person interviews with officials in the Norwegian Medicines Agency (NoMA) in 2019 for this chapter. The interview data aim to corroborate general patterns as observed in the survey and offer some in-depth observations of policy uploading within one government agency. The selection

252 

J. TRONDAL

criteria of survey and interview data were as follows: respondents should (i) hold an A-level position, (ii) have a minimum one year of experience in the agency, and (iii) be involved in preparing, implementing, and/or practicing EU legislation. Following the study by Vestlund (2015), NoMa was selected to be part of this study due to its extensive interaction with EU regulatory networks. It thereby represents a case of high integration. NoMA is a semi-autonomous agency subordinate to the Ministry of Health and Care Services. The mission of NoMA is to ‘[…] evolve and safeguard public and animal health by ensuring efficiency, quality and safety of medicines and to administer and enforce the medical devices regulation’ (NoMA, 2019a). The total number of employees in the agency is approximately 320, including full-time positions allocated to supranational policymaking. It is a highly specialized agency that actively participates in all seven scientific committees1 in the European Medicines Agency (EMA) (NoMA, 2019b). The European medicines regulatory cooperation pools resources conveyed through routine interaction and exchange of valuable assets such as knowledge, information, practices, and experiences among participants. Additionally, there is a routinized division of labor within the network where the participants mutually adapt and specialize in different tasks and fields of expertise (Vestlund, 2017: 63). The benefits of cooperation are also emphasized at the NoMA website: ‘[Through European medicines regulatory] cooperation, smaller countries such as Norway may actively contribute with expertise […] and influence decision-making. At the same time, this enables us to increase own competences by benefitting from the knowledge of others’2 (NoMa, 2019b). When measuring (perceived and reported) agency influence vis-à-vis the Commission and EU agencies (dependent variable) and probing the causal effect of secondary structures and administrative capacity (independent variables), we use a question from the survey in which agency officials are asked to assess agency influence vis-à-vis different institutions, including the Commission and EU agencies, on a five-point Likert scale from very good to very poor (see Table  10.1). Responding to Ansell et  al.’s (2016) argument that the dichotomous distinction between primary and 1  There are approximately 50 working groups in total under the seven scientific EMA committees. NoMA participates in approximately 25 of these. 2  Author’s own translation. Original quote: «Legemiddelsamarbeidet gjør at små land som Norge kan bidra med mye kunnskap og stille viktige spørsmål, og dermed påvirke beslutningen. Samtidig øker vår egen kompetanse på legemidler ved at vi drar nytte av. andres kunnskap, som også er helt nødvendig for den nasjonale legemiddelforvaltningen.»

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

253

secondary organizational affiliations may hide the potential variety of scales and forms of affiliation that is likely to exist in contemporary governments, we use multiple questions from the survey that arguably serve as indicators of low, medium, and high involvement in EU executive institutions and administrative structures (see Table 10.1 below): (i) By assuming that contact patterns serve as a fairly weak indicator of institutional involvement, we use ‘contact to EU executive institutions’ as an indicator of low involvement. Respondents were asked

Table 10.1 Operationalizations Variable

Dimensions

Question in survey

Scale

Influence

–

Five-point Likert scale

Secondary structure

Low

Please indicate the extent to which your organization has been successful in uploading viewpoints and agendas to the following organizations: [EU Commission] /[EU-agencies/ regulatory networks] Please indicate the extent to which you have been in contact with the following organizations: [EU Commission] / [EU-agencies/ regulatory networks] Please indicate the extent to which you have participated in the working groups, advisory boards, etc., with the following organizations: [EU Commission] /[EU-agencies/ regulatory networks] Have you, during the past year, been involved in any of the following: [EU-policy development]? Please indicate the extent to which your agency has sufficient administrative capacity for the following: [Facilitate cooperation and coordination between different actors] Please indicate the extent to which your agency has sufficient administrative capacity for the following: [Provide advice and information]

Medium

High

Administrative capacity

Cooperate and coordinate

Provide advice and information

Four-point Likert scale

Three-point Likert scale

Dichotomous

Five-point Likert scale

Five-point Likert scale

254 

J. TRONDAL

to indicate their contact level to different EU executive institutions on a four-point scale from every week to never. (ii) Participation in EU administrative structures serves as an indicator of medium involvement. We assume that participation will have a relatively stronger effect on influence; however, not all participation is likely to be equally effective or targeted at policy influence. Respondents were asked to indicate how often they have participated in different EU level institutions, applying a three-­ point scale ranging from several times to never. (iii) Finally, active involvement in the development of EU policy serves as an indicator of high involvement. Unlike the former two, this question includes a dichotomous scale (yes/no) and does not specify any particular institution in which policy development takes place. We thus use this question as an independent variable on both the Commission and EU agencies. Second, administrative capacity is measured on two different measures (see Table 10.1 above): (i) Capacity to initiate cooperation and coordination. (ii) Capacity to provide advice and information. The capacity to initiate cooperation and coordination serves as an indication of the degree to which an agency has sufficient resources to attend other actors (national and international). The capacity to provide advice and information serves as an indication of the level of competences which is a key tool for administrative cooperation at the EU level. Interview data is added to provide more in-depth insights. However, since NoMA is tightly integrated into EU structures, these data focus on policy influence through involvement in EU-level regulatory networks (EMAs). Since access to the administrative structures of the EU is largely similar for EU member states and associated nonmember states, the findings of this study are arguably generalizable beyond the case at hand. Compared to EU member states, Norway’s form of affiliation to the EU does not, however, provide access to political venues for decision-making in the Commission, the (Union) Council, or the European Parliament. Norway’s relationship with the EU may thus best be regarded as territorially disintegrated and sectorally integrated. Concomitantly, Norway is subjected to administrative integration within most policy domains, yet politically or

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

255

territorially disintegrated by not accessing formal policymaking venues within the EU.  This blend of sectoral integration and territorial dis-­ integration is why the EU is best described as an externally differentiated polity (Lavenex, 2014; Leruth et al., 2022) in which the dyadic membership/nonmembership dichotomy becomes blurred (Trondal, 2002).

Empirical Demonstration This section is presented in two stages: The first stage presents the case of Norway as an instance of an associated non-EU member state with incomplete and dynamic contracting, while the second stage outlines core findings from the survey and case study. Incomplete and Dynamic Contracting Examining government administration of a formally non-EU member state, the Norwegian case contributes to studies of external differentiation of the EU and the external effects of EU norms and rules beyond EU borders (Rittberger & Blauberger, 2018). Yet, the case also illustrates how incomplete contracting at T1, with the signing of the first agreements with the EU, has led to dynamic updating and amendments of more contracts or treaties over time. Compared to contemporary instances of EU external governance, Norway is by far the most strongly integrated EU nonmember through a dense web of institutionalized relations (Egeberg & Trondal, 1999; Lavenex, 2011; Fossum & Graver, 2018). This affiliated status grants the Norwegian central administration privileged access to most parts of the EU executive system, which in turn paves the way for deep administrative integration. Moreover, administrative integration might go even further in affiliated nonmember states than in member states due to their exclusion from political representation at EU level. The next section shows a subsequent mobilization of administrative bias in the national core-­ executive. Integration may thus evolve largely ‘by stealth’ in affiliated states such as Norway—even though the official position has been not to become a complete political member (Majone, 2005). Since Norway is not subjected to political representation in the Council, it has been argued that Norwegian sector ministries are perhaps and paradoxically more strongly ‘hit’ by the EU executive than member states’ ministries—who may buffer this effect through political participation (Egeberg & Trondal,

256 

J. TRONDAL

1999; Trondal, 2002). This assumption, however, is beyond the scope of this chapter. The launch of the EEA agreement in 1994 marked the beginning of an era of deep administrative integration between the EU executive and the Norwegian central administration. At the time, however, the EEA agreement was viewed as a temporal solution and an interim period toward full EU membership, as the prospects of this were both open and pursued by the then-government. However, the following rejection of EU membership in a heated national referendum reintroduced the EEA agreement as Norway’s foundational connection to the EU in the years to come. Yet, since this agreement has become increasingly incomplete in comparison to the dynamic integration of the EU, it has led to fears of losing out on the Norwegian side, and subsequent negotiations of new agreements in areas outside the framework of the EEA agreement. In effect, EFTA countries have negotiated approximately 100 treaty agreements with the EU, with the EEA and Schengen being the most encompassing ones (Figs. 10.1 and 10.2). While the form of affiliation between Norway and the EU has thus remained fairly stable during the past 30 years, the scope of the affiliation has undergone significant expansion—partly in response to policy progressions in the EU, partly in response to the dynamic character of the 9

Number of agreements

8 7 6 5 4 3 2 1 0 1973

1978

1983

1988

1993

1998

2003

2008

2013

2018

Year Fig. 10.1  Number of agreements between Norway and the EU, by year, 1973–2018. Source: Norwegian Government (2019)

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

257

120

Number of agreements

100

80

60

40

20

0

Year

Fig. 10.2  Accumulated number of agreements between Norway and the EU, 1973–2018. Source: Norwegian Government (2019)

EEA agreement, and partly because of Norway signing evermore sectoral agreements with the EU. One consequence is providing Norwegian central administration increased access to the EU executive apparatus This has increased the subsequent likelihood of Norwegian government institutions and governance processes being profoundly absorbed by EU-level decision-making processes. Figures 10.1 and 10.2 illuminates the dynamic character of the associated membership of Norway, and how incomplete contracting at T1 (1994) led to dynamic updating of the treaty base. Figure 10.1 provides an overview of the number of agreements decided between Norway and the EU each year and Fig. 10.2 shows how these agreements have accumulated over time. Overall, these figures suggest a profoundly dynamic character of the associated membership. Yet, the form of association has remained stable over time, being largely based on the EEA agreement. Illuminating one causal driver of dynamic contracting, Fig. 10.1 shows how treaty dynamics largely coincides with major treaty negotiations within the EU—notably, the Single European Act, the Maastricht Treaty, the Amsterdam Treaty, the Nice Treaty, and the Lisbon Treaty.

258 

J. TRONDAL

Because of the dynamic character of the EEA agreement, Norwegian law has been continuously adjusted to new EU legislation. The agreement is based on the premise of dynamic homogeneity. Consequently, since 1994 more than 14,000 EU legal acts have been incorporated into this agreement alone. This agreement covers mostly the single market acquis and also several adjacent policy fields—making it the most extensive form of agreement between the EU and a nonmember as regards regulatory scope and the legal obligations resulting from contractual relations (Fossum & Graver, 2018; Lavenex, 2011). At the same time, it blocks Norwegian governments qua state from political representation within EU’s legislative institutions and processes. Nonetheless, whereas Norway’s associated status does not grant Norway formal access to EU’s decision-making institutions, it provides for administrative participation at various stages of EU’s legislative process. Norwegian ministries and agencies are represented in Commission expert committees and comitology committees, are represented on most EU agencies’ boards and committees, are entitled to second national experts to the Commission, and participate in a host of European administrative networks (EANs). Norwegian civil servants are thus granted privileged access to EU-related decision-making processes and are tasked with handling everyday relationships with EU executive institutions—in upstream processes (agenda-setting processes) and particularly in downstream processes (implementation and practice of EU law). Thus, whereas Norway is politically sidelined in EU decision-making processes, the Norwegian central administration is granted favored access to the EU bureaucratic system, quite similar to that of EU member states. Results from the Study To what extent do treaty agreements, as outlined, ‘hit home’ and make actors ‘hit back’ by affecting government officials’ perception of influence vis-à-vis EU executive institutions? Table  10.2 shows surprisingly high scores on the dependent variable: 30 percent of the respondents (when combining values 1 and 2 on the variable) report high perceived influence vis-à-vis the Commission, while 37 percent report the same for EU agencies. From this, we may draw two distinct observations: First, this suggests high level of perceived influence by national agency officials. Despite being an affiliated country with no access to political decision-making, roughly one third of the agency officials report being highly influential vis-à-vis EU executive institutions. A second observation is that agency officials

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

Table 10.2 Frequency distribution, influence vis-à-vis the Commission and EU agencies. Percent (N)

To a very large extent To a large extent Somewhat To a small extent To a very small extent Total (N)

259

The Commission

EU agencies

5 (12) 25 (64) 51 (131) 11 (29) 8 (21) 100 (257)

6 (15) 31 (74) 49 (116) 8 (18) 6 (14) 100 (237)

* This variable applies the following questions in the survey: Please indicate the extent to which your organization has been successful in uploading viewpoints and agendas to the following organizations: [Commission] /[EU-agencies/regulatory networks]” with a five-­point scale: (1) To a very large extent, (2) to a large extent, (3) somewhat, (4) to a small extent, (5) to a very small extent

report slightly more influence vis-à-vis EU agencies than toward the Commission. This observation may arguably reflect how administrative capacities are organized across levels of governance (Benz et  al., 2016; Trondal, 2010). We may consider the Commission and the EU agencies as European core-executive institutions that are likely to interact with national-level agencies due to compatible organizational capacities. We may take this assumption a step further and differentiate between the Commission and EU agencies, where the latter is structurally compatible with national agencies primarily by enjoying similar roles as decentered administrative capacities within a government system with fairly similar task profiles. Compatible organizational capacities may boost sectoral allegiances and mutual influence between administrative bodies across levels of governance. Supporting this argument, we observe that interviewees are involved in the policy process in EU agencies, while playing a more consultative role vis-à-vis the Commission. We are able to influence the work [in EMA]. We may suggest matters that should be put on the agenda in the working groups. For instance, if we have a concern or a question […] we can address it […] and agree on a unified stance among the member states (Interviewee L1)

260 

J. TRONDAL

However, because national agency officials enjoy a fair amount of de facto autonomy from national principals (Trondal, 2014; Verhoest et al., 2010) and tend to participate as semi-autonomous experts in EU-level secondary structures (Egeberg et al., 2003), our observations suggest that national agencies do not exert ‘national’ influence on the EU. Due to the organizationally decentralized position in the national governmental apparatus, agency officials mobilize sectoral behavioral logics vis-à-vis EU executive institutions. The interviewees report that agency officials tend to assign substantial weight to professional considerations and concerns. According to NoMA officials, they receive few direct and detailed national steering signals from the ministry,3 apart from general guidelines on prioritized areas. Consequently, NoMA officials enjoy substantial discretion in their work and are not systematically subjected to direct ministerial control or intervention in their EU-related work. While this in large part is due to their highly specialized task profile, it concomitantly implies that agencies are not subject to national political control or oversight. Agency officials are strongly reliant on their professional discretion. For us, national interests signify to be updated on development and changes and potentially contribute to such. And, of course, we consider [national priority areas], [national] viewpoints and [national] course of actions. (Interviewee L6).

Our survey data, however, suggests that national agency officials may upload policy viewpoints to EU executive institutions that ultimately may materialize in common European policies. They thus also suggest how affiliated countries may exert influence through secondary structures. Table 10.3, however, suggests that agency officials have few daily contacts with the Commission and EU agencies. Considering those that do report regular contacts, agency officials are slightly more directed toward EU agencies than toward the Commission. Also, Tables 10.2 and 10.3 show that contacts with EU executive institutions are significantly lower than reported influence. From this, we may infer that contact patterns alone do not serve as valid indicators of agency influence. Similarly, Table 10.4 (below) shows that 12 percent report having participated in EU executive institutions, while a majority report no participation. 3  Applies to matters that are not strongly politicized. Note also that these interviews were conducted prior to the outbreak of the Covid-19 pandemic.

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

Table 10.3 Frequency distribution, contact to the Commission and EU agencies. Percent (N)

Table 10.4 Frequency distribution, participation in the Commission and EU agencies. Percent (N)

Table 10.5 Frequency distribution, policy development. Percent (N)

Every week or more Every month A few times Never Total (N)

Several times Once Never Total (N)

261

The Commission

EU agencies

1 (8) 4 (43) 15 (149) 80 (824) 100 (1024)

3 (26) 5 (50) 12 (126) 80 (811) 100 (1013)

The Commission

EU agencies

7 (67) 5 (53) 88 (882) 100 (1002)

7 (71) 5 (47) 88 (870) 100 (988)

YES Involved in EU policy-development

13 (130)

NO 87 (867)

Moreover, 13 percent report that they have been actively involved in EU policy development (Table 10.5). In sum, our data suggests that overall national agency involvement (contact, participation, and policy development) in EU executive institutions appears to be fairly low. For an affiliated nonmember state, however, these observations may not come as a surprise. However, as evidenced in Table 10.1, a significantly higher proportion of national agency officials report being influential vis-a-vis EU executive institutions than their low levels of contact and participation would suggest. At least two possible explanations may be suggested. First, agency officials may overestimate or misrepresent their level of influence vis-à-vis EU executive institutions. Second, despite the emergence of direct forms of administration across levels of governance in the EU (Egeberg & Trondal, 2016), agency officials may still consider indirect forms of influence as a normal route, for

262 

J. TRONDAL

example, influence via other structures in their own agency (via their representatives to EU executive institutions), via other institutions (for instance, their parent ministry), or via political masters (responsible minister). The interviews provide a more detailed account of these general patterns. It is reported that national agency officials have considerable clout in EMAs. While they see some formal barriers, such as, for instance, restricted voting rights, agency officials report exerting influence vis-à-vis EU executive institutions through writing assessment reports and voicing their opinions in working groups. Influence is deemed particularly strong if national agency officials are responsible for leading a particular evaluation and draft proposals. [W]e are very much able to influence through an assessment report. We may affect everyone’s opinion. So, we can exert influence even though we do not have voting rights. (Interviewee L3)

Another observation that is made is that regular participation in working groups at the EU level is positively associated with agency influence. A key factor here is the development of collegiality and informal relationships that substantially lower the threshold for putting forward ideas and having informal discussions. One participant described their EMA working group as a ‘family’ (Interviewee L4). Regular participation also allows officials to familiarize themselves with norms, procedures, and the EU system in general. This indicates that continuity and intensity of participation are important drivers of influence. Next, to what extent do agencies report having requisite administrative capacity for policy influence? Table 10.6 shows that agency officials report a general high level of available administrative capacity within their agency, with 48 percent reporting administrative capacities to initiate cooperation and coordination, while 59 percent report administrative capacities to provide advice and give information (values 1 and 2 on the variables). Overall, administrative capacity appears to be relatively high in the surveyed agencies. In interviews, we asked about administrative capacity related to EU tasks specifically. One observation is that the ability to initiate and lead evaluations and proposals at the EU level is dependent on agency competences and administrative capacities. Countries that have invested substantially in their national agencies are also deemed capable of taking on bigger

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

263

Table 10.6  Frequency distribution, administrative capacity. Percent (N)

Very good Good Somewhat Poor Very poor Total (N)

Capacity to initiate cooperation and coordination

Capacity to provide advice and information

8 (74) 40 (378) 41 (383) 9 (86) 2 (18) 100 (939)

13 (122) 46 (435) 29 (272) 10 (92) 2 (17) 100 (939)

workloads at the European level and would be able to exert policy influence. Yet, interviewees also report that due to limited resources, they often must decide on which portfolios and dossiers to prioritize. [The leadership] does attempt to distribute responsibilities more or less evenly between countries. They also look at the competences. Some countries have invested a lot in their medicine agencies, that is not really the case in Norway. Large medicine agencies such as the ones from Germany, England or Sweden [often take the lead]. They have a lot of resources. If we would have had more resources, we could have taken the lead a lot more. (Interviewee L3)

Next, by correlating the set of variables included in our survey, Table  10.7 shows significant and strong positive correlation between reported influence vis-à-vis the Commission and EU agencies (0.75), suggesting that a majority of agency officials view themselves as influential within both. A positive and significant relationship between participation in the Commission and EU agencies (0.32) suggests that a smaller proportion of these officials attend both institutions. Moreover, we find significant positive correlations between access to secondary structures and reported influence in these same institutions (including both ‘participation’ and ‘contact’): Respondents reporting influence vis-à-vis the Commission also report contact toward (0.18) and participation within (0.24) the Commission. Likewise, respondents that report influence vis-à-­ vis EU agencies also report contact toward (0.29) and participation within (0.18) EU agencies. Reflecting findings in Tables 10.1, 10.2, and 10.3 above, these correlations are not particularly strong. Furthermore, the data shows a positive relationship between influence and reported

264 

J. TRONDAL

Table 10.7  Inter-correlation matrix on influence, secondary structures, and administrative capacity 1. 1. Influence on the Commission 2. Influence on EU agencies 3. Contact to the Commission 4. Contact to EU agencies 5. Participation in the Commission 6. Participation in EU agencies 7. Development of EU policy 8. Capacity to initiate cooperation and coordination 9. Capacity to provide advice and information

2.

3.

4.

5.

6.

7.

8.

9.

1 0.75** 1 0.18** 0.13 0.01

0.29** 0.44** 1

0.24** 0.14* −0.01

1

0.67** 0.33** 1

0.18** 0.42** 0.66** 0.32** 1

0.20** 0.19** 0.42** 0.33** 0.44** 0.37** 1 0.13*

0.16*

−0.03

0.22** 0.25** 0.01

−0.01

−0.01

−0.03

0.02 1

0.04

0.01

0.05

0.04 0.55** 1

participation in the development of EU policy (0.20 and 0.19), albeit these correlations are also moderate. Lastly, we find that perceived influence is positively associated with administrative capacity, and specifically capacity to provide information and advice. Finally, to determine controlled effects of access to secondary structures and domestic administrative capacities on agency influence vis-à-vis the Commission and EU agencies, Tables 10.8 and 10.9 specify Ordinary Least Squares (OLS) regression models that probe the relative effects of various types of secondary structures (low, medium, high) and various administrative capacities (capacity to cooperate and coordinate, and capacity to give advice and information). As predicted, both models report positive and significant effects of secondary structures (involvement in policy development (0.14)) and domestic administrative capacity to provide information and advice (0.20 and 0.21). Yet, they differ in that Table 10.7 finds significant effects of participation (in the Commission), while Table  10.8 finds

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

265

Table 10.8  Controlled effects of participation in secondary structures and domestic administrative capacity on agency officials’ reported influence vis-à-vis the Commission (OLS regression analysis with standardized beta coefficients) Low involvement (contact) Moderate involvement (participation) High involvement (policy development) Capacity to cooperate and coordinate Capacity to give advice and information N R2 Adjusted R2 F Statistic/F Significance/F

0.02 0.15* 0.14** −0.01 0.20** 223 0.1078 0.0873 5.24 0.0000

Table 10.9  Controlled effects of participation in secondary structures and domestic administrative capacity on agency officials’ reported influence vis-à-vis EU agencies (OLS regression analysis with standardized beta coefficients) Low involvement (contact) Moderate involvement (participation) High involvement (policy development) Capacity to cooperate and coordinate Capacity to give advice and information N R2 Adjusted R2 F Statistic/F Significance/F

0.26** −0.07 0.14** −0.01 0.21** 210 0.1360 0.1149 6.42 0.0000

significant effects of contact (with EU agencies). In sum, these findings suggest that government agencies in associated non-EU member states may exert influence on the Commission and EU agencies through a variety of secondary structures. Second, these findings only provide significant effects on ‘administrative capacities to give advice and information’, suggesting that capacities for policy advice seem more important than capacities for administrative coordination within the domestic civil service.

266 

J. TRONDAL

Conclusions Departing from the invitation and lessons of Waldo (1952), this chapter suggests that the administrative state has transformed its basic role in democracy. However, public administration scholarship has paid scant attention to the effects of agencification of the EU system. Few studies have examined the actual role of domestic agencies in the policymaking process of the EU.  At a more general level, this literature has failed to examine how agencification at one level of authority may have profound consequences for public governance at another level. This chapter has demonstrated how the policy influence of domestic agencies vis-à-vis EU executive institutions are shaped by EU-level secondary structures and domestic administrative capacities. More specifically, the chapter has illustrated how administrative participation and integration affect policy uploading toward EU executive institutions. Administrative integration represents processes that enable policy transfer across levels of governance and thus also as a distinct administrative tool for national government institutions to influence EU decision-­ making processes. This is particularly so for affiliated non-EU countries that are not granted full access to EU decision-making. The chapter has examined what role ‘third countries’ outside the EU may enjoy with respect to policy uploading in the EU, and how organizational factors may shape such processes. The chapter offers two distinct observations. First, it probes the effects of two distinct drivers of policy influence: Access to secondary structures and available domestic administrative capacities. Second, the chapter shows how the EEA agreement has resulted in deeper administrative integration over time, thus illuminating how administrative integration may create avenues for EU influence among associated third countries. Specifically, we examined the effect of secondary structures and domestic administrative capacities. The data suggests that government agencies in associated non-EU member states may exert influence on the Commission and EU agencies through a variety of secondary structures. The data also suggests a positive association between administrative capacity and the ability to influence European executive institutions. Interview data in addition emphasizes the effect of continued and regular participation in EU-level executive institutions.

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

267

References Adam, C., Hurka, S., Knill, C., & Steinebach, Y. (2021). Policy accumulation and the democratic responsiveness trap. Cambridge University Press. Almond, G. A. (1965). Political development: Analytical and normative perspectives. Comparative Political Studies, 1(4), 447–467. Ansell, C., & Gash, A. (2008). Collaborative governance in theory and practice. Journal of Public Administration Research and Theory, 18(4), 543–571. Ansell, C., Trondal, J., & Ogard, M. (Eds.). (2016). Governance in turbulent times. Oxford University Press. Bauer, M., & Trondal, J. (Eds.). (2015). The Palgrave handbook of the European administrative system. Palgrave Macmillan. Beckman, C. M. (2021). Alternatives and complements to rationality. Research in the Sociology of Organizations, 76, 3–18. Benz, A., Corcaci, A., & Doser, J. W. (2016). Unreavelling multilevel administration. Patterns and dynamics of administrative co-ordination in European governance. Journal of European Public Policy, 23(7), 999–1018. Beyers, J., & Trondal, J. (2004). How nation-states ‘hit’ Europe. Ambiguity and representation in the European Union. West European Politics, 27(4), 919–942. Blauberger, M., & Rittberger, B. (2014). Conceptualizing and theorizing European regulatory network. Regulation & Governance, 9(4), 367–376. Börzel, T., & Risse, T. (2000). When Europe hits home: Europeanization and domestic change. European Integration Online Papers (EIoP), 4(15), 24. Brandsma, G., Curtin, D., & Meijer, A. (2008). How transparent are EU comitology committees in practice? European Law Journal, 14(6), 819–838. Carpenter, D. P. (2001). Forging bureaucratic autonomy. Princeton University Press. Christensen, T., & Lægreid, P. (2009). Living in the past? Change and continuity in the Norwegian central civil service. Public Administration Review, 69(5), 951–961. Coen, M., & Thatcher, M. (2008). Network governance and multi-level delegation: European networks of regulatory agencies. Journal of Public Policy, 28(1), 49–71. Debre, M., & Dijkstra, H. (2021). Institutional design for a post-liberal order: Why some international organizations live longer than others. European Journal of International Relations, 27(1), 311–339. Dehousse, R. (1997). Regulation by networks in the European Community: The role of European agencies. Journal of European Public Policy, 4(2), 246–261. Deters, H., & Falkner, G. (2021). Remapping the European agenda-setting landscape. Public Administration, 9(2), 290–303. Egeberg, M. (1999). The impact of bureaucratic structure on policy-making. Public Administration, 77(1), 155–170. Egeberg, M. (Ed.). (2006). Multilevel union administration. Palgrave Macmillan.

268 

J. TRONDAL

Egeberg, M. (2012). How bureaucratic structure matters: An organizational perspective. In B.  G. Peters & J.  Pierre (Eds.), The SAGE handbook of public administration. SAGE. Egeberg, M., Gornitzka, Å., & Trondal, J. (2016). Organization theory. In C. Ansell & J. Torfing (Eds.), Handbook on theories of governance. Edward Elgar. Egeberg, M., Schaefer, G.  F., & Trondal, J. (2003). The many faces of EU Committee governance. West European Politics, 26(3), 19–40. Egeberg, M., & Trondal, J. (1999). Differentiated integration in Europe: The case of EEA country, Norway. Journal of Common Market Studies, 37(1), 133–142. Egeberg, M., & Trondal, J. (2009). Political leadership and bureaucratic autonomy: Effects of agencification. Governance, 22(4), 673–688. Egeberg, M., & Trondal, J. (2016). Why strong coordination at one level of government is incompatible with strong coordination across levels (and how to live with it). The case of the European Union. Public Administration, 94(3), 579–592. Egeberg, M., & Trondal, J. (2017). Researching European Union agencies. What have we learned (and where to go from here?). Journal of Common Market Studies, 55(4), 675–690. Egeberg, M., & Trondal, J. (2018). An organizational approach to public governance. Oxford University Press. Fossum, J.  E., & Graver, H.  P. (2018). Squaring the circle on Brexit. Bristol University Press. Genschel, P., & Jachtenfuchs, M. (Eds.). (2014). Beyond the regulatory polity. Oxford University Press. Gornitzka, Å., & Sverdrup, U. (2008). Who consults? The configuration of expert groups in the European union. West European Politics, 31(4), 725–750. Gulick, L. (1937). Notes on the theory of organization. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration. Hille, P., & Knill, C. (2006). ‘It’s the bureaucracy, stupid’: The implementation of the acquis Communautaire in EU candidate countries, 1999-2003. European Union Politics, 7(4), 531–552. Joaquin, M.  E., & Greitens, T.  J. (2021). American administrative capacity. Springer. Joerges, C., & Neyer, J. (1997). From intergovernmental bargaining to deliberative political processes: The constitutionalisation of comitology. European Law Journal, 3(3), 273–299. Kassim, H., Connolly, S., Dehousse, R., Rozenberg, O., & Bendjaballah, S. (2017). Managing the house: The presidency, agenda control and policy activism in the European Commission. Journal of European Public Policy, 24(5), 653–674. Krick, E., & Gornitzka, Å. (2020). Tracing scientization in the EU Commission’s expert group system. Innovation: The European Journal of Social Science Research. https://doi.org/10.1080/13511610.2020.1811649

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

269

Lægreid, P., Randma-Liiv, T., Rykkja, L. H., & Sarapuu, K. (2016). Coordination challenges and administrative reforms. In G. Hammerschmid, S. Van de Walle, R.  Andrews, & P.  Bezes (Eds.), Public administration reforms in Europe. Edward Elgar. Lavenex, S. (2011). Concentric circles of flexible “European” integration: A typology of EU external governance relations. Comparative European Politics, 9(4/5), 372–393. Lavenex, S. (2014). The power of functionalist extension: How EU rules travel. Journal of European Public Policy, 21(6), 885–903. Leruth, B., Gänzle, S., & Trondal, J. (Eds.). (2022). The Routledge handbook of differentiation in the European Union. Routledge. Levi-Faur, D. (2013). The odyssey of the regulatory state: From a “thin” monomorphic concept to a “thick” and polymorphic concept. Law & Politics, 35(1–2), 29–50. Maestenbroek, E. (2018). Europeanization of policies and public administration. In E. Ongaro & S. Van Thiel (Eds.), The Palgrave handbook of public administration and Management in Europe. Palgrave Macmillan. Maesternbroek, E., & Martinsen, D. S. (2018). Filling the gap in the European administrative space: The role of administrative networks in EU implementation and enforcement. Journal of European Public Policy, 25(3), 422–435. Majone, G. (2005). Dilemmas of European integration. Oxford University Press. Mann, M. (1984). The autonomous power of the state: Its origins, mechanism and results. European Journal of Sociology, 25(2), 185–213. March, J. G. (2008). The myth of rationality. In U. Sverdrup & J. Trondal (Eds.), The organizational dimension of politics. Bergen. March, J. G., & Olsen, J. P. (1989). Rediscovering institutions. The Free Press. Mathieu, E. (2016). When Europeanization feeds back into EU governance: EU legislation, national regulatory agencies, and the EU regulatory networks. Public Administration, 94(1), 25–39. Mayntz, R., & Scharpf, F.  W. (1975). Policy-making in the German Federal Bureaucracy. Elsevier Scientific Publishing. Meier, K., & Nigro, L. G. (1976). Representative bureaucracy and policy preferences: A study of the attitudes of federal executives. Public Administration Review, 36(4), 458–469. NOMA. (2019a). Departments. Retrieved November 15, 2019, from https:// legemiddelverket.no/english/about-­us/departments NOMA. (2019b). About us. Retrieved February 26, 2021, from https://legemiddelverket.no/english/about-­us Norwegian Government. (2018–2019). Prop. 47 LS (2018–2019). Retrieved from https://www.regjeringen.no/no/dokumenter/prop.-­4 7-­l s-­2 0182019/ id2630679/?ch=15

270 

J. TRONDAL

Olsen, J.  P. (1998). The new European experiment in political organization’. Paper presented at the conference ‘Samples of the Future’. SCANCOR, Stanford University, 20–22 September. Olsen, J. P. (2009). Change and continuity: An institutional approach to institutions of democratic government. European Political Science Review, 1(1), 3–32. Olsen, J. P., & Lægreid, P. (1984). Top civil servants in Norway: Key players—on different teams. In E. N. Suleiman (Ed.), Bureaucrats and policy making. Holmes & Meier. Peters, B.  G. (1994). Agenda-setting in the European community. Journal of European Public Policy, 1(1), 9–26. Peters, B.  G. (2018). The challenge of policy coordination. Policy Design and Practice, 1(1), 1–11. Pollack, M. (1993). The Engines of European Integration. Oxford University Press. Riddervold, M., Trondal, J., & Newsome, A. (Eds.). (2021). The Palgrave handbook on EU crises. Palgrave Macmillan. Rittberger, B., & Blauberger, M. (2018). Introducing the debate section: “The EU in crisis: EU studies in crisis?”. Journal of European Public Policy, 25(3), 436–439. Simon, S. A. (1957). Administrative behavior. The Free Press. Steinmo, S., Thelen, K., & Longstreth, F. (1992). Structuring politics. Cambridge University Press. Stinchcombe, A. (2001). When formality works. Chicago University Press. Trondal, J. (2002). Beyond the EU membership-non-membership dichotomy? Supranational identities among national EU decision-makers. Journal of European Public Policy, 9(3), 468–487. Trondal, J. (2010). An Emergent European Executive Order. Oxford University Press. Trondal, J. (2014). Agencification. Public Administration Revie, 74(4), 545–549. Trondal, J., & Bauer, M.  W. (2017). Conceptualizing the European multilevel administrative order: Capturing variation in the European administrative system. European Political Science Review, 9(1), 73–94. Trondal, J., & Peters, B. G. (2013). The rise of European administrative space. Lessons learned. Journal of European Public Policy, 20(2), 295–307. Trondal, J., Suvarierol, S., & van den Berg, C. (2008). The compound machinery of government. The case of seconded officials in the European Commission. Governance, 21(2), 253–274. Trondal, J., & Veggeland, F. (2003). Access, voice and loyalty: The representation of domestic civil servants in EU committees. Journal of European Public Policy, 10(1), 59–77. Verhoest, K., Van Thiel, S., Bouckaert, G., & Lægreid, P. (Eds.). (2010). Government agencies. Palgrave Macmillan.

10  INCOMPLETE CONTRACTING AND POLICY INFLUENCE 

271

Vestlund, N.M. (2015). Between centralization and decentralization. Decision behaviour in the EU’s multilevel administrative system (PhD thesis), University of Oslo. Vestlund, N. M. (2017). Pooling administrative resources through EU regulatory networks. Journal of European Public Policy, 24(1), 61–80. Waldo, D. (1952). Development of theory of democratic administration. American Political Science Review, 46(1), 81–103. White, J. (2022). The de-institutionalisation of power beyond the state. European Journal of International Relations, 28(1), 187–208.

CHAPTER 11

The Autonomy of International Public Administration

The aim of this chapter is twofold: The first and most important ambition is to theorize conditions for autonomy of bureaucratic organizations. The chapter argues that the autonomy of bureaucratic organizations is supplied endogenously within these organizations and not merely conditioned by exogenous factors such as member states’ cost-benefit analyses (Lipsky, 1980; Wilson, 1989), domestic politicization (Hooghe et  al., 2019), or socialization processes outside bureaucracy (e.g., Hooghe, 2007), to mention a few. One secondary ambition is to offer some empirical illustrations or footnotes on the autonomy among office holders in international bureaucracies. With the gradually increasing role of international bureaucracies, one yet unresolved question is to what extent and under what conditions such institutions may formulate their own policies and thus transcend a mere intergovernmental role (e.g., Knill & Steinebach, 2022; Trondal et  al., 2010). The craft of IOs is to a large extent supplied by the autonomy of its bureaucratic arm, that is, by the ability of international bureaucracies— and their staff—to act relatively independently of mandates and decision premises from member-state governments (Barnett & Finnemore, 2004; Biermann & Siebenhuner, 2009, 2013; Cox & Jacobson, 1973; Reinalda, 2013; Trondal, 2013). ‘International rules are prepared by top-rank administrators’ (Papadopoulos, 2013: 84). It is thus essential to know how autonomous these administrators are and what can explain it. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_11

273

274 

J. TRONDAL

‘[A]utonomy is about discretion, or the extent to which [an organization] can decide itself about matters that it considers important’ (Verhoest et al., 2010: 18–19). As an area of research, the extent to which and the conditions under which international bureaucracies are independent of member-­ state governments have become increasingly vibrant. However, the findings offered are still inconclusive (e.g., Beyers, 2010; Checkel, 2007; Moravcsik, 1999). The empirical focus of this chapter is actor-level autonomy as enacted by international civil servants. There are at least two rationales for applying an actor-level focus. First, the discretion available to bureaucracies is made real by individual officeholders (Cox & Jacobson, 1973). Second, institutional transformation—as with the rise of relatively autonomous international bureaucracies—involves that international civil servants’ ‘preferences and conceptions of themselves and others ….’ are affected (Olsen, 2005: 13). Moreover, one often neglected proxy of actor-level autonomy is the extent to which they activate a supranational behavioral logic (hereby termed ‘actor-level supranationalism’). This chapter argues and empirically illustrates that IOs in general and their bureaucracies in particular, may possess considerable clout to form actor-level supranationalism among its personnel (Marcussen & Trondal, 2011). Actor-level supranationalism denotes the rise of some shared norms, values, goals and codes of conduct among international civil servants. A supranational logic entails that staff is loyal to the mission and vision of the IO and show this loyalty by guarding against attempts, either by member-state governments or other actors, to direct the organization in other directions. The civil servants are expected to become ‘defenders of the system’ and to acquire collective behavioral perceptions independent of particular national interests. The appearance of actor-level supranationalism denotes actors’ feelings of loyalty and allegiance toward the IO as a whole—or toward parts of it (Deutsch et al., 1957: 5–6; Haas, 1958: 16; Herrmann et al., 2004: 6). In classic theories of European integration—such as neo-­functionalism— it is assumed that one of the key driving forces of integration is the shift of individual loyalties from the national to the international level (Eilstrup-­ Sangiovanni, 2006). International institutions are assumed to have a capacity to create a sense of community and belonging beyond the nation state, that is, by socializing staff (Checkel, 2007). The enactment of a supranational role may thus imply that individuals report loyalty and a sense of belonging to an IO, as well as share, and act in accordance with,

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

275

some shared norms, ideas, beliefs and goals of the organization. This chapter thus poses two research questions: • To what extent are international bureaucracies ‘hothouses’ of actor-­ level supranationalism? • To what extent is actor-level supranationalism formed within international bureaucracies? Theoretically, two mechanisms of organizational studies are shown to matter in this regard: Organizational rule-following and ‘in-house’ organizational socialization. The observations reported benefit from a large and novel set of 121 interviews with civil servants working in three international bureaucracies: the secretariat of the World Trade Organization (WTO), the secretariat of the Organization for Economic Co-operation and Development (OECD), and the European Commission administration. Comparing three seemingly different international bureaucracies in three different IOs, this chapter suggests that bureaucratic autonomy may be fostered equally inside bureaucratic organizations if they supply fairly similar organizational capacities and in-house socialization processes. We show that actor-level supranationalism is present among civil servants in all three international bureaucracies studied. Interestingly, the Commission is not substantially different from the two other international bureaucracies in this regard. One theoretical lesson learned from this observation is that international bureaucracies may possess considerable capacity to shape essential behavioral perceptions among its staff through the two causal mechanisms: (i) behavioral and role adaptation through organizational rule-following, and (ii) behavioral and role internalization through ‘in-house’ socialization processes. The chapter proceeds in the following way: The next section outlines an organizational theory approach to public sector organizations to explain variation in actor-level supranationalism among international civil servants. The subsequent sections outline the methodology and data used to illuminate actor-level supranationalism, followed by an empirical section reporting key findings.

276 

J. TRONDAL

Theorizing Bureaucratic Autonomy A Weberian bureaucracy model assumes that bureaucracies possess internal capacities to shape staff behavior through socialization, discipline, and control (Page, 1992; Weber, 1983; Yi-Chong & Weller, 2004, 2008). These mechanisms ensure that bureaucracies perform their tasks relatively independently from outside influence (Olsen, 2010; Weber, 1924). The organizational structure of international bureaucracies consists of the bureaucratic structure, and how this structure is embedded in the wider IO structure. Organizational dynamics and decision-making behaviors are thus framed by ‘in-house’ organizational structures (Radin, 2012: 17). Organizations create elements of robustness, and concepts such as ‘historical inefficiency’ and ‘path dependence’ suggest that the match between environments, organizational structures, and decision-making behavior is not automatic and precise (Olsen, 2010). An organizational approach thus suggests that the supply of organizational capacities have certain implications for how organizations and incumbents act. This approach departs from the assumption that formal organizational structures mobilize biases in public policy because formal organizations supply cognitive and normative shortcuts and categories that simplify and guide decision-­ makers’ search for problems, solutions, and consequences (Ellis, 2011; Schattschneider, 1975; Simon, 1965). There may be several reasons why international civil servants enact a supranational behavioral logic. This chapter suggests two main mechanisms: Adaptation through organizational rule-following and internalization through ‘in-house’ socialization processes. The chapter thus makes an analytical distinction between actor-level supranationalism caused by the internalization of roles and behavioral perceptions, on the one hand (e.g., Checkel, 2007), and actor-level supranationalism caused by behavioral and role adaptation through control and discipline, on the other (e.g., Trondal et al., 2008). Whereas much existing literature argues that actor-­ level supranationalism originates from outside of international bureaucracies (e.g., Dehousse & Thompson, 2012; Hooghe, 2007, 2012), this chapter argues that actor-level supranationalism may largely emerge from within the structures of international bureaucracies. Lipsky (1980: 19) claimed that bureaucratic autonomy is driven by actors’ conspicuous desire for maximizing their own autonomy. By contrast, it is argued here that bureaucratic autonomy is organizationally contingent. It is the organizational rules and routines established in a

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

277

bureaucracy that regulate, constitute, and bias the decision-making behavior and role perceptions evoked by civil servants, ultimately advancing bureaucratic autonomy (Barnett & Finnemore, 2004: 3). Civil servants live with a constant overload of potential and inconsistent information that may be attended to in decision situations. Organizations guide the decision-making behavior of civil servants due to the computational limitations and the need for selective search. Organizations form collective order out of cognitive disorders by creating local rationalities among the organizational members (March & Shapira, 1992). Organizations are systematic devices for simplifying, classifying, routinizing, directing, and sequencing information toward particular decision situations (Schattschneider, 1975: 58). They ‘are collections of structures, rules and standard operating procedures that have a partly autonomous role in political life’ guiding officials to systematically de/emphasize certain aspects of organizational realities (March & Olsen, 2006: 4). Derived from this organizational approach, two organizational variables may systematically foster actor-level supranationalism: organizational rule-­ following and ‘in-house’ organizational socialization. Organizational Rule-Following An organizational approach suggests that the supply of organizational capacities have certain implications for how organizations and humans act. An organizational approach assumes that organizational capacity-building supplies government institutions with leverage to act independently (Trondal & Peters, 2013). This approach departs from the assumption that formal organizational structures mobilize biases in public policy because formal organizations supply cognitive and normative shortcuts and categories that simplify and guide decision-makers’ behavior (Schattschneider, 1975; Simon, 1965). The behavior role and identity perceptions evoked by international civil servants are expected to be primarily directed toward those administrative units that are the primary supplier of relevant decision premises. In this study, international bureaucracies are arguably primary suppliers of relevant decision premises for international civil servants. Organizations tend to accumulate conflicting organizational principles through horizontal and vertical specialization (Olsen, 2005). When specializing formal organizations horizontally, one (among several) important principle that has been suggested by Luther Gulick (1937) is

278 

J. TRONDAL

specialization by purpose—like research, health, food safety, etc. This principle of organization tends to activate patterns of co-operation and conflicts among incumbents along sectoral cleavages (Egeberg, 2006). Co-ordination and contact patterns tend to be channeled within sectoral portfolios rather than between them. Arguably, organizational specialization by purpose is likely to bias decision-making dynamics inward to the bureaucratic organization where preferences, contact patterns, roles, and loyalties are directed toward sectoral portfolios, divisions, and units. This mode of horizontal specialization results in less than adequate horizontal co-ordination across departmental units and better co-ordination within units (Ansell, 2004: 237). This principle of specialization is most pronounced among most international bureaucracies. For example, the Commission is a horizontally pillarized administration specialized by purpose. Similarly, the WTO and OECD secretariats are horizontally specialized administrations consisting of divisions or directorates responsible for different areas of co-operation (such as agriculture, environment, development, statistics, etc.). Essentially, international bureaucracies serve as the primary organizational affiliation for international civil servants, rendering them particularly sensitive to the organizational signals and selections provided by their organization. As argued, the horizontal specialization of international bureaucracies by major purpose is conducive to autonomization of the behavioral dynamics of incumbents. This argument derives the following proposition: H1: International civil servants embedded within international bureaucracies specialized by purpose are likely to evoke supranational ‘enthusiasm’ based on the stated purpose of the IO and the underlying and linked purpose of the specialized bureaucratic unit. In other words, we assume that formal organizational structures matter regarding the civil servants’ enactment of supranational roles.

‘In-House’ Organizational Socialization A vast literature reveals that the impact of pre-socialization of actors is modified by organizational re-socialization (e.g., Checkel, 2007). Officials entering international bureaucracies for the first time are subject to an organizational ‘exposure effect’ (Johnston, 2005: 1039) that may contribute to such re-socialization. Socialization processes are conducive to

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

279

‘autonomization’ of the socializees because the socializer may educate, indoctrinate, teach, or diffuse his or her norms and ideas to the socializee. Socialization is a dynamic process whereby individuals absorb the norms and rules of a given community. By this process individuals internalize some shared norms, rules, and interests of the community (Checkel, 2007). The socialization argument also claims that behavioral autonomy is conditioned by enduring experiences with institutions, accompanying perceptions of appropriate behavior (Herrmann & Brewer, 2004: 14). The potential for socialization to occur is assumed to be positively related to the duration and the intensity of interaction among the organizational members. According to the neo-functionalist approach, the potential for re-socialization to occur (‘shift of loyalty towards a new center’) is assumed to be positively related to the duration and the intensity of interaction among actors (Haas, 1958: 16). This claim rests on socialization theory that emphasizes a positive relationship between the intensity of participation within a collective group and the extent to which members of this group develop perceptions of group belonging and an esprit de corps. Intensive in-group contact and interaction are conducive to the emergence of relative stabile social, normative, and strategic networks that provide autonomous impact on the participants’ perceptions of strategic and appropriate behavior (Atkinson & Coleman, 1992: 161; Börzel, 1998: 259; Hay & Richards, 2000; Knox et al., 2006: 120). Often, such networks resemble ‘ego-networks’ ‘between a given individual and his or her ‘alters’ (Knox et al., 2006: 118). The network literature suggests that networks are transformative entities that considerably bias the behavior of the participants (Börzel, 1998: 258; Windhoff-Hèritier, 1993). However, as an explanatory toolkit, network approaches have to be supplemented by more generic causal mechanisms, such as socialization mechanisms, in order to explain behavioral implications (Marin & Mayntz, 1991: 44). In sum, the length of stay in international bureaucracies—or the individual seniority of incumbents—may foster socialization toward a supranational behavioral pattern. Hence, behavioral and role autonomy is fostered by the quantity and quality of actor interaction inside international bureaucracies. Ultimately, such actor interaction contributes to the surfacing of tight networks inside international bureaucracies rather insulated from member-state influence. This argument derives the following proposition: H2: International civil servants with long seniority within international bureaucracies are likely to evoke a ‘general’ or diffuse supranational

280 

J. TRONDAL

e­ nthusiasm. In other words, we assume that long tenure among employees and thus persistent interaction with the norms and values of the IO (formal and informal) increase the capacity of the organization to create defenders of the system and supranational enthusiasts.

Empirical Strategy The empirical illustrations benefit from synchronized comparative studies of permanent officials in the Commission administration, the WTO Secretariat, and the OECD Secretariat. The study is synchronized in the sense that the same interview guide has been applied to all three bureaucracies and with respect to the selection of administrative subunits within each bureaucracy. The 71 interviews were semi-directed, using a standardized interview guide that was applied flexibly during interviews. The interviews were carried out in Brussels, Geneva, and Paris. All interviews were recorded and fully transcribed. All interviewees are treated with full anonymity. Consequently, quotations from interviews are referred to as follows: Commission 2, WTO 15, and so on. The questions posed in the interviews were directed at measuring the behavioral perceptions among the civil servants. Key themes covered during interviews were the following: interaction patterns, the role of nationality, the development of esprit de corps, the emergence of shared perceptions of identity among staff toward different institutions and actors, and the role perceptions deemed important by staff when doing daily work. Interviewees were selected from similar administrative subunits in all three international bureaucracies in order to control for variation in policy sectors. These subunits were first trade units (such as DG Trade in the Commission and the numerous trade units in the OECD and WTO Secretariats) and second, the general secretariats (such as the Secretariat-­ General of the Commission, the offices of the Deputies of the Director General in the WTO Secretariat, and the general secretariat of the OECD). General secretariats represent the bureaucratic centers of international bureaucracies, and the trade units represent one among several policy sectors of international bureaucracies. Finally, interviewees were selected from different levels of rank in these subunits—from director generals to executive officers. However, by concentrating on officials at the A-level, we aim to study officials who are involved in policymaking activities. Two general caveats are warranted: First, the selected cases are illustrative devices to examine actor-level supranationalism within international

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

281

Table 11.1  List of interviewees among permanent officials, by formal rank

The Commission The OECD Secretariat The WTO Secretariat Total

Top managers (director generals, deputy director generals, or equivalent)

Middle managers (directors, heads of unit, deputies, or equivalent)

Desk officials Total (advisors, counselors, case handlers, analysts, officers, or equivalent)

1

9

14

24

0

10

18

28

2

4

13

19

3

23

45

71

bureaucracies. Second, while these cases can illuminate causal mechanisms, they cannot prove them. Table  11.1 provides an overview of the conducted interviews.

Comparative Observations from International Bureaucracies Organizational Rule-Following Socialization sometimes fails to occur fully if the mechanisms of internal and external control and discipline operate alone (Checkel, 2007; Gheciu, 2007). This section shows that control and discipline, through supervision, incentives, and rewards, shape both what we have termed supranational sector enthusiasts and general supranational enthusiasts among international civil servants. Reflecting the latter, one aspect of actor-level supranationalism among international civil servants is related to external representation of the IO. The enactment of a supranational role—as a representative of the IO as a whole—is evident in the following quote: It is obvious that when we are operating outside the WTO, in other intergovernmental organizations, then we are representing the WTO as an institution, and we have to be aware of that. (WTO 1)

The WTO official indicates in this quote that s/he has to be aware of the supranational role; it is considered mandatory to represent the WTO

282 

J. TRONDAL

as a whole. In their external representation, civil servants report that if they act in conflict with core rules of the organization, they may be subject to sanctions, internally from the bureaucratic leadership and externally from the member states. When asked about how to behave when representing the WTO externally, one WTO official responds: Yes, of course, you have to be careful not to say weird things and things that are totally just not acceptable, or contentious. To say things about the negotiations sort of … on some contentious issues … or to express a strong opinion that you support one view or another—that is dangerous, and it is not to be tolerated. But it’s not a question of asking permission. Now, of course, to speak at conferences you have to get permission, for obvious reasons. But it’s not so much that you send your statement to your boss to check. (WTO 13)

When asked about the possible sanctions for going against these norms, the same official says: ‘You are fired … or you are called in.’ (WTO 13) Q: Even though the assessments are made according to the WTO rules, do you, as a Secretariat official, have to be careful about making formulations such as “This is the best solution according to what I believe”? Because if that sentence is there, is there a risk that the paper will just be “shot down” by the member states? Oh yes, oh yes, there are things you have to be aware of and, you know, sometimes you get caught by surprise. There’s a sensitivity that you weren’t aware of and somebody reacts very strongly to something and you … “Oh, where did that come from?”’ (WTO 15)

This quote shows that civil servants have to ‘tread a careful path’ because of the awareness and control of member states. The following quotes from OECD officials are also illustrative: As an OECD person, you should be kind of neutral. I’m not working for the French or US government; I am working for the OECD. Period. (OECD 26) It is quite imperative not to be biased by your nationality. (OECD 15) So, I am aware of the OECD line and agreed position, and I know it is incumbent upon me to reflect that agreed line and the conclusions of the work that we have done. It’s not my position to bolster independent opin-

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

283

ions on some policy issues that we have not done any research on or where my research hasn’t been done as part of an agreed OECD process. (OECD 17)

When asked if they all consider themselves international civil servants, one OECD official responds: ‘Yes. You have to, in that job; you wouldn’t last long otherwise. We are not here to represent our own countries in any way’ (OECD 2). In sum, external and internal control and discipline seem to enhance the development of supranational enthusiasts among international civil servants. Some officials seem to enact a supranational role in the form of ‘guardians of the system.’ A quote from one WTO official illustrates this. S/he was asked whether the WTO agreements amount to a kind of constitution that the civil servants have to relate to at all times: Exactly! But I don’t really think … I don’t think I will ever come across someone who doesn’t really believe in that. But some people have different views about … you know, some people look at it more from a developing-­ country perspective and other people from other perspectives. Some of the people might think that some of the rules are more or less equitable. (WTO 1)

The observations from the interviews show that even though civil servants may become ‘guardians of the system’ and have to constantly relate to the organization’s rules, they do not necessarily believe in or agree with all rules. Internal control from the bureaucratic leadership and external control from the member states, in addition to discipline through career opportunities, may foster a supranational role among civil servants in the sense of appearing as ‘guardians of the system’. Civil servants gain authority and credibility through their expertise and impartiality against particular national interests, and through their emphasis on the aims and rules of the organization—what in this context appear to them as the ‘common good.’ In this way, the bureaucracy may gain autonomy within the boundaries set by the vision and mission of the IO. However, the degree to which the norms of the organization have been internalized among the civil servants may still vary. Some are true believers. For others the defense of the system is conditional. Still, both these groups are shaped and socialized into following the same basic codes of conduct: to represent, defend, and follow the logic of the system, not particular national interests.

284 

J. TRONDAL

‘In-House’ Organizational Socialization Civil servants may share the norms of IOs even before entering. These civil servants are predisposed to become loyal to the organization’s vision and mission quite quickly upon arrival. Moreover, if such pre-socialization is salient, there is a potential for a biased (self-) selection among the respondents. In line with the theory of representative bureaucracy, the international bureaucracy will in such cases be representative mainly of the enthusiasts and true believers in the organization. Some civil servants indeed start working for an IO because they truly believe in the organization. It is, however, less clear from our data how such beliefs affect actual decision-making behavior among staff. Thus, pre-socialization might indeed happen, but also be fairly unimportant in order to understand decision-­making behavior among international civil servants. The need to be dedicated to the organization is emphasized in the following quotes from two WTO officials: We are the guardians of the book. We have to believe in what is in here, because if we don’t believe, nobody believes. Then we might as well go home. (WTO 9) I think we have to be committed to what the WTO is as an institution, which basically is for trade liberalization, and so clearly you have to believe in that. Otherwise, it could be very difficult, personally, if you don’t believe in the goal of your organization, that the WTO is an institution which basically is a force for good—you know, the goals of the WTO… (WTO 3)

One OECD official indicates that s/he had been a general supranational enthusiast for a long time, not in relation to the OECD in particular but in relation to IOs in general. Another OECD official indicates that his or her enthusiasm toward the OECD comes from his or her prior experience in private sector: I think there was probably something philosophical in the beginning, because when I finished university, I was very attracted by the international organizations: The values, the mission, things like that. (OECD 27) I am the treasurer of the OECD, and I have a business card. And I am proud of the OECD. That is one of the things I like about working at the OECD: I like what the OECD does, it has a positive influence in the world. Coming

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

285

from the private sector as an American, I am very much in favour of a lot of the things the OECD does—like free trade; like intelligent government policy over business, over taxation; having environment regulations that work, that businesses and people can work with. That governments can promote better policies in areas such as taxation thanks to the work that the OECD does, is very positive. (OECD 23)

One WTO official stated that s/he believed in the GATT/WTO before starting to work there, but s/he emphasizes particular aspects of the organization’s mission: I believed that market access for products, and how countries become less dependent on money by helping them to sell abroad… I believed in that […] but across the border, random trade liberalization… when I came, no I didn’t think…But the GATT never stood for that either. The GATT was not about free trade, the GATT was about, the WTO is about, breaking down certain barriers and trade-distorted measures so that countries at least have more opportunity to sell abroad… (WTO 2)

These observations illustrate pre-socialization toward a general supranational enthusiastic orientation based on the vision and mission of an IO. As indicated above, socialization can take several forms. One important distinction runs between the ‘true believers’—those who believe in the overall mission of the organization, even as a ‘force for good’—and the ‘sector or portfolio enthusiasts’—those who believe in particular issues that the organization deals with and in the organization’s role in solving and handling these issues. Even though pre-socialization remains as a factor to be considered in relation to supranational enthusiasm, an important finding in this study is that the civil servants are further shaped by the organization setting in which they operate. Our data clearly shows that international civil servants are affected by internal factors, that is, they are shaped and socialized toward supranational enthusiasm through experience. This form of socialization is linked to the Weberian bureaucracy model. In the following, we look in particular at how experience from working within an international bureaucracy can affect civil servants’ commitment and loyalty to the organization. Pre-socialization, however, does not exclude the effect of re-­ socialization: Someone who shares the norms of the IO before working there may be further socialized within the organization. To illustrate this dual mechanism, one civil servant responds:

286 

J. TRONDAL

Oh yes, I am convinced. I saw… as a junior diplomat I participated in the making of this organization. I saw this organization being born. I was here in Geneva when the organization was created, and I was here in Geneva when these agreements were negotiated. So, I truly believe in the ideas. (WTO 9)

This official thus relates his or her beliefs in the organization to the close contact s/he had with the WTO in pervious jobs. One Commission official also mentions that his or her commitment to the EU began a long time before s/he joined the Commission: But I always bore in mind the possibility to work for the institutions, maybe not from 16 years old but certainly from 22 years old. (Commission 20)

When asked if it was a wish early on to come to the Commission, s/he replied: I was very much conscious of the project of building Europe. (…) And I knew about Jean Monnet […] and I thought “It’s a big project, an important project, and it is a project qui vient féderer les états”. It’s politically a very difficult project but it is certainly a project I want to work for with my very small means, my very small competencies and capacities. (Commission 20) Q: Is it easy to follow that vision—your European vision—in your day-to-­ day work? Yes, because the vision is very strong. My vision of what I want to do and of what the Commission wants to do is very coherent. They match one another. But also that vision is stronger than, let’s say, the everyday life and problems I can have. That is my view. Some other people are more concerned with their own career. (Commission 20)

These observations illustrate how pre-commitment to the vision and mission of an IO can enhance the subsequent enactment of general supranational enthusiasm after being hired. The following quote from an OECD official illustrates how long tenure in an international bureaucracy can nurture a socialization of staff toward a general supranational enthusiasm:

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

287

Fundamentally, my impression is that when people have joined the OECD, and they have worked here for a while, they no longer behave as nationals of any particular member country, but they serve the interests of the organization. And it doesn’t really matter whether they are Canadian, Australian or Belgian—they work toward the common aim of the organization. (OECD 11)

When asked about his or her general commitment to the goals of their organization, two officials respond as follows: Yes, I think so. I mean I’ve spent 30+ years of my life here, so it would be bizarre if I did not. I do feel commitment to the organization. [….] Yes, I feel a commitment. I think it would be difficult if you didn’t believe in an open-rules-based trading system, that it was in the basic interest of humanity. (WTO 1) (…) But, as I said, I think that being an international civil servant and the more years you do that type of job, the more you tend to represent the organization rather than individuals or divisions or whatever. (OECD 16)

Socialization into the norms of the organization can also be illustrated by the following response from a WTO official when asked what kind of advice s/he could offer the members: ‘Of course it has to be WTO-­ friendly, and then after a while you get… you cannot go against the philosophy of what this institution stands for’ (WTO 2). When asked to what extent the WTO Secretariat should be the ‘guardians of the treaties’, one WTO official confirmed that s/he had taken on the role of a ‘guardian’. However, s/he also includes a more proactive role, that is, as an agent for improving the system: I think this is what has been agreed, but I have my views, and I think there are things that should be changed in this agreement to make it fairer, to make it more effective, and I will be happy to defend my views. But I think it has to be changed by negotiation in this organization. You are not going to change it by destroying the WTO. That is the message. I am the guardian of the book. The book is not perfect. So, my task is to convince people that this book should be improved, here. That is the mission. The mission is to make a multilateral trading system which is fair, which is fair to the developing countries, and which is better than what it is now. (WTO 9)

288 

J. TRONDAL

This observation illustrates a sense of commitment to the organization while at the same time seeing the secretariat’s role as being more than a neutral facilitator. This WTO official emphasizes the role as a defender of the system, but s/he also includes in his or her role the task of suggesting needs for change. Socialization within international bureaucracies can also strengthen civil servants’ feeling of being part of a collective, being part of something ‘beyond the technical requirements of the task at hand’ (Selznick, 1957: 17). In-house socialization like this thus tends to foster general supranational enthusiasm among staff. Two Commission officials also report supranational enthusiasm by emphasizing that their feeling of belonging is aimed toward European integration generally and not toward particular EU institutions or member states: Definition-wise, I am working for the European Commission and, in a broader sense, for the Union. So, I wouldn’t have any ideological problems working for the Council Secretariat, for example. (Commission 11) Well, it is the Community interest in the matter which is prime, and this interest of the Community is not necessarily identical with the interests of any single member state—even if you take them all together. (Commission 14)

This last quote indicates that the civil servant sees the Community interest as something more than the aggregate of member states’ collective interests. It alludes to a general belief in a supranational interest—a collective EU interest relatively independent of member states. Another Commission official illustrates beliefs shaped by the organization. However, in this case socialization within the Commission seems to have made him or her less dedicated to the ‘EU project’: I am more focused on what I do. I am very happy with the job I have, with the colleagues I have, I couldn’t be happier. I have to struggle to remain faithful to Europe. I am still, but when I joined the Commission, I was for a very long time enthusiastic. I mean I was very proud working in the Commission. Not only proud, I thought we were going to take us very far. But today I am much more skeptical. (Commission 24)

This quote illustrates that socialization within organizations may sometimes result in less enthusiastic attitudes toward the organization. Socialization should thus not be conflated with the emergence of ‘pro-­ norm behavior’ (Zurn & Checkel, 2007) or ‘pro-social’ behavior (Lewis,

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

289

2007). Lack of organizational enthusiasm may more easily emerge when organizations face periods of enlargement or internal reforms, potentially challenging preexisting norms and long-cherished beliefs among the personnel (see Ban, 2013; Bauer, 2012: 469; Dehousse & Thompson, 2012: 126). ‘Socialization processes do not necessarily entail harmony and the absence of conflicts’ (Beyers, 2010: 912).

Conclusions This chapter shows that ‘supranational actors’ are present among civil servants both in the WTO Secretariat, the OECD Secretariat and the Commission administration. The Commission is not any different in this regard when viewed from the perspective of civil servants’ behavioral perceptions. This study thus substantiates that international bureaucracies can possess considerable capacity to act relatively independent of member state governments. This capacity is demonstrated when international civil servants develop a supranational mindset—through mechanisms of discipline and control as well as through behavioral and role internalization and adaptation. The most important ambition of this chapter was to theorize two conditions for bureaucratic autonomy. The empirical illustrations suggests that actor-level supranationalism tends to arise foremost through processes internal to IOs. As a consequence, a larger variation in actor-level supranationalism is observed within, rather than between, international bureaucracies. What is surprising is not the observation of a supranational behavioral logic, but the fact that the same behavioral logic among international civil servants is observed almost to the same extent within three international bureaucracies embedded in three seemingly different IOs. One theoretical lesson learned is thus that international bureaucracies may possess considerable capacity to shape essential behavioral perceptions among its staff in general, and foster behavioral autonomy in particular, through the two causal mechanisms: (i) behavioral and role adaptation through organizational rule-following and (ii) behavioral and role internalization through ‘in-house’ organizational socialization. International bureaucracies thus have capacity to create codes of conduct and senses of community and belonging that are relatively independent of constituent states. There is a growing literature on the ‘public administration turn’ to IO studies that support the findings reported here (Knill & Steinebach, 2022;

290 

J. TRONDAL

Trondal, 2007). Studies demonstrate that international bureaucracies, not only the Commission, may have authority and capacity to autonomize its staff. Barnett and Finnemore (2004: 3) demonstrates that the Secretariat of the IMF, the UN High Commissioner for Refugees (UNHCR), and the UN Secretariat ‘were not simply following the demands issued by states but instead acting like the bureaucracies that they are’. Similarly, Lewis (2007) observes organizational socialization inside the COREPER whereby national officials internalize new community roles. Concelmann (2008) shows that international secretariats may have some leeway in organizing and structuring debates and, moreover, that secretariats may gain influence and authority in IOs based on their expertise in technical complex areas. Others have also stressed the importance of the (legal) expertise of international bureaucracies (Marcussen, 2004; Mathiason, 2007; Schmeil, 2004; Yi-Chong & Weller, 2008). As Mathiason (2007:16) has phrased it: ‘The source of legitimate power is essential legal’. Yi-Chong and Weller (2004: 278-279) conclude their study of GATT/WTO by stating that secretariats ‘…provide the continuity and the cement, the credibility and the connections … the final decision may not be theirs, but the creativity surely is.’ Finally, Johnston (2005: 1037) observes ‘some evidence that those individuals most directly exposed to intensive social interaction … are more likely to have a positive attitude toward multilateralism…’ Thus, international bureaucracies tend to foster supranational defenders of the system ‘from within’.

References Ansell, C. K. (2004). Territoriality, authority, and democracy. In C. K. Ansell & G. Di Palma (Eds.), Restructuring territoriality. Cambridge University Press. Atkinson, M. M., & Coleman, W. D. (1992). Policy networks, policy communities and the problems of governance. Governance, 5(2), 154–180. Ban, C. (2013). Management and culture in an enlarged European Commission. Palgrave Macmillan. Barnett, M., & Finnemore, M. (2004). Rules for the world. Cornell University Press. Bauer, M.  W. (2012). Tolerant, if personal goals remain unharmed: Explaining supranational bureaucrats’ attitudes to organizational change. Governance, 25(3), 485–510. Beyers, J. (2010). Conceptual and methodological challenges in the study of European socialization. Journal of European Public Policy, 17(6), 909–920. Biermann, F., & Siebenhuner, B. (2009). Managers of global change. The MIT Press.

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

291

Biermann, F., & Siebenhuner, B. (2013). Problem solving by international bureaucracies: The influence of international secretariats on world politics. In B. Reinalda (Ed.), Routledge handbook of international organization. Routledge. Börzel, T. (1998). Organizing Babylon – On the different conceptions of policy networks. Public Administration, 76(2), 253–273. Checkel, J.  T. (2007). International institutions and socialization in Europe: Introduction and framework. In J. T. Checkel (Ed.), International institutions and socialization in Europe. Cambridge University Press. Concelmann, T. (2008). Beyond the carrot and the stick: State reporting procedures in the WTO and the OECD. In J. Joachim, B. Reinalda, & B. Verbeek (Eds.), International organizations and policy implementation. Routledge. Cox, R.  W., & Jacobson, H.  K. (1973). The anatomy of influence. Yale University Press. Dehousse, R., & Thompson, A. (2012). Intergovernmentalists in the commission: Foxes in the henhouse? Journal of European Integration, 34(2), 113–132. Deutsch, K., et  al. (1957). Political community and the North Atlantic area. Princeton University Press. Egeberg, M. (Ed.). (2006). Multilevel union administration. Palgrave Macmillan. Eilstrup-Sangiovanni, M. (Ed.). (2006). Debates on European integration. Palgrave Macmillan. Ellis, D. C. (2011). The organizational turn in international organization theory. Theorizing International Organizations. http://www.journal-­iostudies.org/ sites/journal-­iostudies.org/files/JIOS1012.pdf. Gheciu, A. (2007). Security institutions as agents of socialization? NATO and the “new Europe”. In J. T. Checkel (Ed.), International institutions and socialization in Europe. Cambridge University Press. Gulick, L. (1937). Notes on the theory of organizations. With special references to government in the United States. In L. Gulick & L. Urwick (Eds.), Papers on the science of administration. Institute of Public Administration, Columbia University. Haas, E. (1958). The uniting of Europe. Stanford University Press. Hay, C., & Richards, D. (2000). The tangled webs of Westminister and Whitehall: The discourse, strategy and practice of networking within the British core executive. Public Administration, 78(1), 1–28. Herrmann, R., & Brewer, M.  B. (2004). Identities and institutions: Becoming European in the EU. In R. K. Herrmann, T. Risse, & M. B. Brewer (Eds.), Transnational identities. Becoming European in the EU. Rowman & Littlefield Publishers. Herrmann, R. K., Risse, T., & Brewer, M. B. (Eds.). (2004). Transnational identities. Becoming European in the EU. Rowman & Littlefield Publishers. Hooghe, L. (2007). Several roads lead to international norms, but few via international socialization. A case study of the European Commission. In J. T. Checkel

292 

J. TRONDAL

(Ed.), International institutions and socialization in Europe. Cambridge University Press. Hooghe, L. (2012). Images of Europe: How commission officials conceive their institution’s role. Journal of Common Market Studies, 50(1), 87–111. Hooghe, L., Lenz, T., & Marks, G. (2019). Contested world order: The delegitimation of global governance. Review of International Organizations, 14(4), 731–743. Johnston, A. I. (2005). Conclusions and extensions: Towards mid-range theorizing and beyond Europe. International Organization (Special Issue), 59, 1013–1044. Knill, C., & Steinebach, Y. (Eds.). (2022). International public administrations in global public policy. Routledge. Knox, H., Savage, M., & Harvey, P. (2006). Social networks and the study of relations: Networks as method, metaphor and form. Economy and Society, 35(1), 113–140. Lewis, J. (2007). The janus face of Brussels: Socialization and everyday decision making in the European unio. In J. T. Checkel (Ed.), International institutions and socialization in Europe. Cambridge University Press. Lipsky, M. (1980). Street-level bureaucracy. Russel Sage Foundation. March, J. G., & Olsen, J. P. (2006). Elaborating the “new institutionalism”. In R. A. W. Rhodes, S. A. Binder, & B. A. Rockman (Eds.), The Oxford handbook of political institutions. Oxford University Press. March, J. G., & Shapira, Z. (1992). Behavioral decision making theory and organizational decision theory. In M. Zey (Ed.), Decision making. Newbury Park. Marcussen, M. (2004). Multilateral surveillance and the OECD: Playing the idea game. In K.  Armingeon & M.  Beyeler (Eds.), The OECD and the European welfare states. Edward Elgar. Marcussen, M., & Trondal, J. (2011). The OECD civil servant. Between Scylla and Charybdis. Review of International Political Economy, 18(5), 592–621. Marin, B., & Mayntz, R. (Eds.). (1991). Policy Networks. Campus Verlag. Mathiason, J. (2007). Invisible governance: International secretariats in global politics. Kumarian Press Inc. Moravcsik, A. (1999). A new statecraft? Supranational entrepreneurs and international cooperation. International Organization, 53(2), 267–306. Olsen, J.P. (2005). Unity and diversity – European style. ARENA working paper 24. Olsen, J.  P. (2010). Governing through institutional building. Oxford University Press. Page, E.  C. (1992). Political authority and bureaucratic power. Harvester Wheatsheaf. Papadopoulos, Y. (2013). Democracy in crisis? Politics, governance and policy. Palgrave MacMillan. Radin, B. A. (2012). Federal Management Reform. Georgetown University Press.

11  THE AUTONOMY OF INTERNATIONAL PUBLIC ADMINISTRATION 

293

Reinalda, B. (2013). International organization as a field of research since 1910. In B.  Reinalda (Ed.), Routledge handbook of international organization. Routledge. Schattschneider, E. E. (1975). The Semisovereign people. Harcourt Brace Jovanovich College Publishers. Schmeil, Y. (2004). Expertise and political competence: Consensus making within the World Trade Organization and the World Meteorological Organization. In B. Reinalda & B. Verbeek (Eds.), Decision making within international organizations. Routledge. Selznick, P. (1957). Leadership in administration. University of California Press. Simon, H. A. (1965). Administrative behavior. The Free Press. Trondal, J. (2007). Is the European Commission a hothouse for supranationalism? Exploring actor-level supranationalism. Journal of Common Market Studies, 45(5), 1111–1133. Trondal, J. (2013). International bureaucracy: Organizational structure and behavioural implications. In B. Reinalda (Ed.), Routledge handbook of international organization. Routledge. Trondal, J., & Peters, B. G. (2013). The rise of European administrative space. Lessons learned. Journal of European Public Policy, 20(2), 295–307. Trondal, J., van den Berg, C., & Suvarierol, S. (2008). The compound machinery of government. The case of seconded officials in the European Commission. Governance, 21(2), 253–274. Trondal, J., Marcussen, M., Larsson, T., & Veggeland, F. (2010). Unpacking international organizations. Manchester University Press. Verhoest, K., Roness, P.  G., Verschuere, B., Rubecksen, K., & MacCarthaigh, M. (2010). Autonomy and control of state agencies. Palgrave Macmillan. Weber, M. (1924). Legitimate authority and bureaucracy. In D.  S. Pugh (Ed.), Organization theory. Selected readings (3rd ed.). Penguin Books. Weber, M. (1983). On capitalism, bureaucracy and religion. Harper Collins Publishers. Wilson, J. Q. (1989). Bureaucracy. Basic Books. Windhoff-Hèritier, A. (1993). Policy network analysis: A tool for comparative political research. In H. Keman (Ed.), Comparative politics. VU University Press. Yi-Chong, X., & Weller, P. (2004). The governance of world trade. International civil servants and the GATT/WTO. Edward Elgar. Yi-Chong, X., & Weller, P. (2008). To be, but not to be seen: Exploring the impact of civil servants. Public Administration, 86(1), 35–51. Zurn, M., & Checkel, J.  T. (2007). Getting socialized to build bridges: Constructivism and rationalism, Europe and the nation state. In J. T. Checkel (Ed.), International institutions and socialization in Europe. Cambridge University Press.

CHAPTER 12

The Organizational Dimension of Global Governance

As outlined in Chap. 1, the rise of executive order through organizational capacity building and bureaucratic autonomization is seen as one key ingredient of state formation (Bartolini, 2005). Order formation above nation-state structures, however, is much less studied and poorly understood. Since the end of World War II, the rise of executive authority of IOs has increased capacity for global problem solving but has also challenged the sovereignty of nation states (Eilstrup-Sangiovanni & Hofmann, 2020; Haas, 1964; Trondal, 2010). Rising incongruence between collective action capacities, existing territorial boundaries, and the domains in which policy solutions are subject to executive authority obfuscate public problem-­solving at both the national and the international level (Hooghe & Marks, 2016). As public administration at the international level increasingly serves as a backbone to emerging orders beyond the nation state, scholars have become increasingly interested in understanding the scope autonomy of international bureaucracies (Barnett & Finnemore, 2004; Bauer et  al., 2019; Cox & Jacobson, 1973; Fleischer & Reiners, 2021; Haas, 1964; Heady, 1998; Herold et  al., 2021; Knill & Bauer, 2016;

Supplementary Information  The online version contains supplementary material available at https://doi.org/10.1007/978-­3-­031-­28008-­5_12. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. Trondal, Governing the Contemporary Administrative State, European Administrative Governance, https://doi.org/10.1007/978-3-031-28008-5_12

295

296 

J. TRONDAL

Marcussen & Trondal, 2011; Ness & Brechin, 1988; Stone & Moloney, 2019; Thorvaldsdottir et al., 2021). The ambition of this chapter is twofold: The first and most important ambition is to theorize conditions for autonomy of bureaucratic organizations. Following the observations of Chap. 11, this chapter argues that the autonomy of bureaucratic organizations is supplied by endogenous organizational properties. One secondary ambition is to offer some minor empirical illustrations on the autonomy among office holders in international bureaucracies of the ‘Global South’ (see below). The chapter poses two research questions: To what extent are international bureaucracies featured by actor-level autonomy? To what extent is actor-level autonomy forged endogenously within international bureaucracies? Theoretically, two mechanisms from organizational studies are shown to matter in this regard: organizational specialization and organizational affiliation. With the gradually increased role of international bureaucracies one unresolved question is to what extent and under what conditions such organizations may formulate own policies and thus transcend a mere intergovernmental role. As argued and shown in Chap. 11, the capacity of IOs is to a large extent supplied by the ability of international bureaucracies— and their staff—to act relatively independently of mandates and decision premises from member-state governments (Barnett & Finnemore, 2004; Biermann & Siebenhüner, 2009, 2013; Cox & Jacobson, 1973; Gänzle et al., 2018; Reinalda, 2013; Tieku et al., 2020; Trondal, 2013). Recent studies show that international bureaucracies are policy entrepreneurs and rule-makers, and sometimes rule-implementers (Barnett & Finnemore, 2004; Bauer & Ege, 2016). International bureaucracies thus have become policymaking bodies that transcend a mere passive supply chain for IO plenary assemblies (Bauer et  al., 2017; Tieku, 2021). Scholars have increasingly started to causally identify conditions under which international bureaucracies are autonomous bodies. Some studies have established how the organizational architectures of international bureaucracies shape how they interpret and process behavioral discretion during everyday decision-making processes (Barnett & Finnemore, 2004; Trondal, 2016; Trondal et  al., 2010) and how requisite administrative capacities serve as scope conditions for crisis responses (Van Hecke et  al., 2021). Others suggest that organizational characteristics of international bureaucracies matter for their survival in the long term (Debre & Dijkstra, 2021; Jones & Hameiri, 2022). Extant literature thus illustrates the extent to which and the conditions under which international bureaucracies are

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

297

autonomous of member-state governments. However, this literature both offers inconclusive findings (see Beyers, 2010; Checkel, 2007; Johnston, 2005; Moravcsik, 1999) and lacks broader empirical scope beyond ‘western cases.’ Whereas contemporary studies have profoundly advanced our knowledge on the independent role of international bureaucracies in the provision of public goods (Bauer et al., 2019; Beyerliein et al., 2018; Christensen & Yesilkagit, 2018), the organizational dimension of autonomy of international bureaucracies has received scant attention (Barnett & Finnemore, 2004; Hawkins et al., 2006; Heady, 1998; Weiss, 1982). The ambition of this chapter is to extend an emergent organizational literature on the autonomy of international bureaucracies (Egeberg & Trondal, 2018) and offer two hard cases on the executive commissions of the African Union (AU) and Economic Community of West African States (ECOWAS). These executive bodies are embedded in intergovernmental IOs and thus represent least likely cases of autonomy of international bureaucracies. By empirically establishing that organizational factors are structuring elements in the autonomy of least likely cases, and moreover that these elements are powerful tools available to deliberate design, the chapter draws lessons of practical value in which organizational variables may be ‘nudged’ to achieve desired outcomes. Empirical lessons from the study may thus have practical implications. The chapter makes three distinct contributions to the volume: • Theoretically, it draws on an organizational approach to public governance and applies it to the study of autonomy of international bureaucracies. It outlines organizational conditions that enable international bureaucracies to mobilize their own policy agendas quite separately from member-state governments. Establishing a hard case of autonomy, two cases are studied that are embedded within essentially intergovernmental IOs with international bureaucracies that are governed by ‘informality’ (Tieku, 2018). It derives distinct predictions on how international civil servants maneuver under different organizational conditions. The chapter suggests that organizational factors are structuring elements of autonomy and that these elements are powerful tools allowing for deliberate design. In this way, theoretically informed empirical research may serve as an instrumental device for global governance.

298 

J. TRONDAL

• Empirically, the empirical strategy of the study is to study actor-level autonomy as enacted by international civil servants. This study presents data from two novel surveys among civil servants in the headquarters of the AU and ECOWAS, Addis Ababa and Abuja, respectively. There are at least two rationales for applying an actor-­ level focus. First, the discretion available to bureaucracies is made real by individual officeholders (Cox & Jacobson, 1973). Secondly, the rise of relatively autonomous international bureaucracies requires that international civil servants’ ‘preferences and conceptions of themselves and others’ are affected (Olsen, 2005: 13). Moreover, actor-level autonomy is operationalized by two ‘behavioral profiles’: the task profile as a policymaker which is focused on drafting policy proposals, providing scientific, technical, and/or legal advice, giving political advice, providing background information, and the profile as an organizational manager which is oriented toward facilitating compromises—between departments and units, between organizations, and between member states. The empirical analysis benefits from what we term a multi-proxy probe based on the use of multiple variables: the task profile of staff (Tables 12.2, 12.3, 12.4, and 12.5), the considerations and concerns they emphasize (Tables 12.6, 12.7, and 12.8), whose arguments they deem important (Tables 12.9, 12.10, and 12.11), their role perceptions, and the importance of nationality (Table 12.12). Finally, to measure controlled effects, the two datasets were combined to enable an OLS regression model. The surveys unveil that international bureaucracies that are embedded within intergovernmental IOs serve as more than the neutral secretariats for member states by acting as policymakers. The data establishes that international civil servants within both international bureaucracies are biased toward internal affairs of international bureaucracies by leaning extensively toward their own organization, administrative units, policy sectors, and expertise rather than toward external affairs of member states and other IOs. These observations establish how international civil servants may leave independent marks on the policy processes of international bureaucracies. • Third, responding to a call for comparative studies of ‘non-Western’ administrative systems (Haque et  al., 2021), the chapter extends extant literature on international bureaucracies beyond ‘Western’ cases. While the AU has established itself at the very core of the

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

299

African political order today (Abdulqawi & Ouguergouz, 2012; Fagbayibo; Karbo & Murithi, 2018; Hardt, 2016; Tieku, 2018), ECOWAS has developed into a key IO of Western Africa both in economic and security terms (Coleman, 2007; Lokulo-Sodipe & Osuntogun, 2013; Plenk, 2014). The chapter proceeds in the following steps: The next section outlines the organizational approach and derives four testable propositions. Then, core features of the AU and ECOWAS Commissions are outlined in terms of functions and staff. The subsequent three sections discuss the data and methodology, outline main findings from the surveys, and draw cautious conclusions on the study of international bureaucracies in global governance.

Theorizing the Autonomy of International Bureaucracies An organizational approach is grounded on the assumption that endogenous organizational characteristics may explain how organizations make decisions. Stable structures of governance systems are assumed to supply stable infrastructures for public policymaking (Blom-Hansen et al., 2021; Trondal, 2021). Yet, organizations are continuously reformed. New organizational forms arise that for instance reshape public–private boundaries or create new links between existing organizations (Ahrne & Brunsson, 2005). Organizational births and deaths can go hand in hand with established organizations being split into new ones, merged, or, less frequently in the case of public organizations, simply terminated (Boin et al., 2010). Using the example of the AU and ECOWAS Commissions, this chapter argues that stable organizational structures are likely to shape the discretionary behavior of organizational members toward actor-level autonomy (as specified above). The general theoretical prediction is that variation in actor-level autonomy will reflect variation in the organizational embedding of staff. The role of organizing has often been neglected and contested in much social sciences literature (Olsen, 2010). Extant literature harbors competing ideas on the extent to which actual decision behavior reflects the organization structure within which actors are embedded. This chapter advocates that actor-level autonomy is shaped endogenously by

300 

J. TRONDAL

organizational structures (Egeberg & Trondal, 2018; March & Olsen, 1989; Olsen, 2008; Trondal & Bauer, 2017). An organizational approach thus posits that organizations systematically influence human behavior and decision-making processes by allocating actors’ attention toward certain problems and solutions (Gulick, 1937; Thaler & Sunstein, 2009). Contemporary organization theory literature focuses on the explanatory power of organizational factors for two reasons (Egeberg & Trondal, 2018): First, empirical studies demonstrate that organizational structures trigger systematic and significant effects on decision-making behavior (Christensen & Lægreid, 2008; Egeberg & Trondal, 2018; Trondal et al., 2008). Second, organizational structure allows for deliberate (re)design and may thus be applied as a design instrument of the context of choice in public governance (Egeberg & Trondal, 2018; Thaler & Sunstein, 2009). Organizational structures consist of sets of routines for attention allocation (Egeberg & Trondal, 2018). This occurs by allocating actors access to decision situations, mobilizing attention to certain problems and solutions, structuring patterns of conflict and cooperation (and thus influencing power relationships), and enabling coordination and steering along certain dimensions rather than others. An organizational approach also acknowledges the multiple causal mechanisms which may connect role expectations to actor-level behavior. Among these are notably rule and role compliance based on actors’ calculation of own self-interest in which organizations are incentive structures that administer rewards and punishments (Ostrom & Ostrom, 2015), bounded rationality which helps simplify actors’ cognitive overload by directing attention toward a selection of possible problems and solutions and ways to connect them (March, 1981; Simon, 1957), and a logic of appropriateness which views human action as driven by internalized normative perceptions (March & Olsen, 1989). To account for actor-level autonomy, two organizational variables are specified: vertical organizational specialization and organizational affiliation. Vertical Organizational Specialization Given the hierarchical structure of international bureaucracies, the attention of staff is likely to reflect their organizational belonging within the hierarchy. Hierarchy within organizations is likely to lay the foundations for decision biases, for example, by offering a more general view in hierarchically superior units compared to lower ranked units. Vertical

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

301

specialization within organizations installs organizational boundaries between hierarchically ‘superiors’ and ‘subordinates.’ Hierarchy within organizations is measured in this chapter by the respective official’s position or rank (see below). Vertical specialization denotes the division of responsibility and labor within and between organizations. This is both an organizational tool and an organizational signal to anchor relatively independent expertise in certain organizational units—such as government agencies—and thereby balance political loyalty and professional neutrality (e.g., Bach et al., 2015; Egeberg & Trondal, 2017; Lægreid & Verhoest, 2010; Pollitt & Talbot, 2004). Interorganizational vertical specialization enables agencies to operate relatively insulated from political steering while being relatively influenced by stakeholder groups and professional concerns, often organized into advisory systems (Egeberg & Trondal, 2009; Gornitzka & Sverdrup, 2008; Veit et al., 2017). Intraorganizational vertical specialization denotes division of responsibility and labor within levels of authority. In this chapter, we measure this variable by official’s rank within their respective international bureaucracy. Studies show that higher ranked staff in government organizations is more attentive to political signals than lower ranked personnel and that those in higher ranks usually face more and various ‘audiences’ compared to those at lower ranks (e.g., Christensen & Lægreid, 2009; Egeberg & Sætren, 1999). Studies also show that top-­ranked staff have a wider range of attention and identify more frequently with organizations as wholes than staff located at lower echelons. Executive heads of international bureaucracies are likely to interact more frequently across organizational units and are exposed to broader flows of information than their subordinates and thus may be more attentive to broader organizational perspectives than lower ranked personnel and are often shown to be more attentive to political signals (Egeberg & Sætren, 1999; Egeberg & Trondal, 2018; Christensen & Lægreid, 2008). It follows that officials of international bureaucracies with lower ranks are more loosely coupled to the political leadership and have more local perspectives on task execution compared to higher ranked staff. This has one important implication: It seems that some degree of insulation from political leadership makes lower ranked officials more eligible receivers of impulses from stakeholder groups (Egeberg & Trondal, 2018). Concomitantly, higher ranked staff are exposed to a broader range of information than lower level staff and thus may be more attentive to broader organizational perspectives than lower ranked personnel (Egeberg & Trondal, 2018). Focusing on vertical specialization within the AU and

302 

J. TRONDAL

ECOWAS Commissions, the role as a policymaker is more likely among higher ranked than lower ranked staff (March & Olsen, 1984). It thus follows that: P1: Actor-level autonomy in general, and the role as a ‘policymaker’ in particular, is positively associated with the professional rank of staff (as measured by ‘employment category’). Organizational Affiliation Top-level (international) public servants often serve in multiple structures and have multiple affiliations (Van Dorp & ’t Hart, 2019). Some of these affiliations are sequential when officials experience mobility between departments within their organizations, while others are simultaneous when officials occupy multiple affiliations at the same time. In conceptual terms, we may distinguish between primary and secondary organizational affiliations (Bertels & Schulze-Gabrechten, 2021). A ‘primary structure’ is the macrostructure of organizations, defined as the structure to which participants are expected to devote most of their loyalty, time, and energy. A typical example would be a bureaucratic unit like a ministry, agency, or an international bureaucracy. A ‘secondary structure’ is defined as the structure to which participants are expected to be ‘part-timers.’ Secondary structures are the substructures of organizations, such as collegial bodies, committees, think-tanks, expert groups, and administrative networks. These are ‘weak’ organizations that are unlikely to shape actors’ decision behavior to the same extent as primary structures (Egeberg & Trondal, 2018), yet they may contribute to flexible and informal international cooperation and coordination (Benz & Goetz, 2021). Empirical studies show that interdepartmental committees, public-private governing arrangements, and expert committees improve interaction and coordination among actors, and erect trust relationships among the participants, however, the effects are moderate (Ansell & Gash, 2008; Egeberg et al., 2003; Lægreid et  al., 2016). International civil servants who operate within such a multiaffiliational system are exposed to complex behavioral premises stemming from both primary and secondary structures. In general, a logic of primacy suggests that primary affiliations are likely to affect the behavior of staff more extensively than secondary affiliations (March, 1994). It thus follows that:

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

303

P2: When taking part in both primary and secondary structures, actor-level autonomy is likely to be mainly shaped by primary structures (as measured by ‘employment status’).

Some organizational affiliations, however, may be sequential with bureaucrats experiencing mobility between departments within their organization. Staff of international bureaucracies that experience such interservice mobility are exposed to a broader set of information and decision premises compared to staff with no such interservice mobility. Concomitantly, one may expect that having a broader outlook on the organization through organizational mobility is likely to contribute to actor-level autonomy and a ‘policymaking’ role compared to staff with no interorganizational mobility. The latter group is more likely to enact the role as an organizational manager. It thus follows that:

P3: International civil servants experiencing interservice mobility are likely to be involved in policymaking processes and thus evoke an actor-level autonomous ‘policymaker’ role as compared to international civil servants experiencing low interservice mobility (as measured by ‘number of departments’). Before we turn to the empirical section, the subsequent section presents our cases in general and explores in some detail the functioning and staff demography in the AU and ECOWAS Commissions.

Empirical Strategy This section outlines data and measurements and the study context of AU and ECOWAS.

304 

J. TRONDAL

Data and Measurements Most data on international bureaucracies have thus far been collected in either global (e.g., the United Nations) and/or ‘Western-based’ IOs (e.g., the EU and the OECD) (Bauer et al., 2019; Trondal et al., 2010). As one of our ambitions is to expand the empirical scope of extant literature, the dataset consists of data from the headquarters of two African international bureaucracies, the Commissions of ECOWAS and the AU. The first survey was distributed manually among ECOWAS officials in 2016, whereas the second dataset was collected electronically among AU officials in 2018. The surveys have been kept near to identical in terms of their questions. For the purpose of this study, both datasets have been combined. Table 12.1 presents an overview of the respondents in both organizations (N = 204) in which the AU represents 67 percent (N = 137) and ECOWAS 33 percent (N = 67) of the respondents.1 While surveys may offer great insights and are amenable to quantitative data analysis, they also come Table 12.1  Descriptive statistics, respondents, percent (N)

Respondents (N) Category of staff Employment status Gender Education

Highest academic degree

– High-level Low-level Permanent Temporary Male Female Law Economics/business Social science Humanities Technical/engineering PhD Master’s Bachelor None

ECOWAS

AU

Total

33 (67) 62 (38) 38 (23) 60 (30) 40 (20) 72 (48) 28 (19) 6 (4) 42 (28) 18 (27) 24 (16) 21 (14) 23 (15) 62 (40) 15 (10) 0 (0)

67 (137) 16 (21) 84 (114) 35 (48) 65 (89) 58 (48) 41 (34) 12 (10) 48 (39) 30 (25) 15 (12) 10 (8) 10 (8) 71 (56) 18 (14) 1 (1)

100 (204) 30 (59) 70 (137) 42 (78) 58 (109) 64 (96) 36 (53) 9 (14) 45 (67) 29 (43) 19 (28) 15 (22) 16 (23) 67 (96) 17 (24) 1 (1)

Source: own compilation

1  Majority of AU respondents are lower level employees because the AU is a ‘bottom-­ heavy’ organization. As Tieku et al. (2020) have noted, over 74 percent of AU staff are in the bottom half of the organization.

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

305

with significant challenges. First and foremost, response rates tend to be low in the study of IOs, in our cases hovering around 10 percent. Still, these numbers are comparable to other studies of international bureaucracies (Bauer et al., 2019). Second, respondents may be prone to misperceptions and perceptional bias. To make up for these challenges, we offer a multi-proxy probe based on multiple variables: the task profile of staff (Tables 12.2, 12.3, 12.4, and 12.5), the considerations and concerns they emphasize (Tables 12.6, 12.7, and 12.8), whose arguments they deem important (Tables 12.9, 12.10, and 12.11), and their role perceptions (Table  12.12). Each of these proxies will be outlined in greater detail throughout the empirical analysis. Moreover, when interpreting findings, biases on certain variables are weighted less in the overall analysis. Actor-­ level autonomy is measured by two ‘behavioral profiles’: the task profile as a policymaker which is focused on drafting policy proposals, providing scientific, technical, and/or legal advice, giving political advice, providing background information, and the profile as an organizational manager which is oriented toward facilitating compromises—between departments and units, between organizations, and between member states. A dichotomous variable was created to group higher and lower level staff. A total of 30 percent are higher level staff, whereas 70 percent are lower level staff. This variable is, however, unevenly distributed between the two international bureaucracies with 62 percent higher level staff in ECOWAS against 16 percent in AU. A similar pattern emerges regarding organizational affiliation (‘employments status’). Whereas 60 percent of ECOWAS officials report being permanently employed, only 35 percent of AU staff have permanent positions. As shown elsewhere, the AU Commission is rather ‘bottom-heavy’ (Tieku et  al., 2020) where most staff are placed at lower levels of the pay-scale. Both international bureaucracies share a similar educational profile among the staff in which a majority report having an educational background in economics or business, followed by social sciences. Surprisingly, only a small proportion of the respondents report a law degree. This may reflect a low regulatory competence and activity of these international bureaucracies, as compared to the European Commission (Kassim et  al., 2013). A vast majority of the respondents are male and hold a master’s degree as their highest achieved academic degree (Tieku et al., 2020). In addition to the organizational variables which measure post-­ recruitment factors, we control for the role of pre-recruitment factors by including the educational profiles of staff.

306 

J. TRONDAL

Extant literature on representative bureaucracy argues that the behavior of staff may be influenced by pre-recruitment factors—that is, their demographic profiles (Pfeffer, 1982: 277; Selden, 1997). According to the idea of individual pre-socialization outside organizations prior to recruitment, international civil servants may be ‘pre-packed’ already before entering organizations (Pfeffer, 1982: 277; Selden, 1997; Trondal et  al., 2018). However, studies suggest that background factors, except for educational background, seem to have only modest impact on officials’ actual behavior, also among international civil servants embedded in international bureaucracies (Christensen & Lægreid, 2009; Egeberg, 2003, 2006; Meier & Nigro, 1976; Olsen, 1983; Trondal, 2010; Trondal et al., 2010). This chapter uses educational background as proxy of individual pre-­ socialization in the post-recruitment phase and probes if international civil servants with different expertise are likely to perceive their bureaucratic role significantly different. In short, the length of education of staff is likely to be positively associated with taking on a ‘policymaking’ role. Research Context Despite being essentially intergovernmental IOs, the AU and ECOWAS are different in terms of membership (55 vs. 15 members) and in terms of the permanence of staff, in which ECOWAS staff primarily rely on permanent contract while the AU Commission is characterized by ‘short-­ termism’ (see Tieku et  al., 2020). With the intention of replacing its predecessor, the Organization of African Unity (OAU), the AU was founded in May 2001, and eventually launched in Durban, South Africa, in July 2002. As of 2017, it comprises all 55 African countries (after Morocco, having left the OAU in 1982, rejoined in 2017) and serves as the prime organization for continental economic and political cooperation and integration. Having adopted the statutes, the AUC became operational in 2002. The AUC is mandated to: • represent the AU and defend its interests under the guidance of, and as mandated by, the Assembly and Executive Council, • initiate proposals to be submitted to the AU’s organs, as well as implement decisions taken by them, • act as the custodian of the AU Constitutive Act and OAU/AU legal instruments, • provide operational support for all AU organs,

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

307

• assist member states in implementing the AU’s programs, • draft common African positions and coordinate member states’ actions in international negotiations, • manage the AU budget, resources, and strategic planning, • elaborate, promote, coordinate, and harmonize the AU’s programs and policies with those of the Regional Economic Communities (RECs), and • ensure gender mainstreaming in all AU programs and activities; take action as delegated by the Assembly and Executive Council (African Union 2014: 46). Functionally, the AUC acts as the ‘engine room’ of the AU and is responsible for the day-to-day management of AU affairs (Fagbayibo, 2012: 15). The AUC consists of a political and an administrative wing: Politically, the AUC comprises ten elected officials (Commissioners) and appointed corps of administrative and technical staff (African Union 2002). The elected officials include the Chairperson of the Commission (COC), the Deputy Chairperson of the Commission (DCP), and eight Commissioners. Each one of the five regions of Africa (i.e., Central Africa, East Africa, North Africa, Southern Africa, and West Africa) is entitled to two Commissioners (Tieku et al., 2020). The appointed staff of the AUC are made up of approximately 1700 at the headquarters in Addis Ababa, and at the representative missions around the world (African Union 2019: 232). They are categorized into two groups: Professional Staff (ranked from P1 to D1) and General Service Staff (GSS). The GSS people are grouped into General Service A (GSA) and General Service B (GSB). The GSA people are primarily administrative, clerical, maintenance, and paramedical personnel, while the GSB, or what the AUC calls the Auxiliary Staff, are mainly drivers and security personnel. ECOWAS, in turn, was established on 25 May 1975, and its founding treaty—the Treaty of Lagos—was signed by 15 member states. Although first and foremost concerned with matters of economic integration—pursuing the goal of a currency union—ECOWAS has gained considerable clout both as a proponent of free trade and labor migration and as a security actor in the region. During the 1990s and 2000s, ECOWAS eventually engaged on a track leading to several regional peacekeeping interventions (Coleman, 2007: 73–115). ECOWAS’ primary stated objective remained to promote ‘economic integration and the realization of the objectives of the African Economic Community’ (ECOWAS Treaty 1993: Article 2, para 1) in all fields of economic activity including labor and

308 

J. TRONDAL

capital. To achieve these goals, ECOWAS’ member states have not only established ‘Community Institutions […] with relevant and adequate powers’ but are also convinced ‘that the integration of the Member States into a viable regional Community may demand the partial and gradual pooling of national sovereignties to the Community within the context of a collective political will’ (Preamble to ECOWAS Treaty 1993: Article 1, emphasis added). At the head of the executive arm of ECOWAS is the Commission, responsible for providing policy drafts to both the Assembly and the Council and executing decisions and regulations adopted by the ECOWAS member states. Organizational reforms agreed by the Assembly, that is, the heads of state and government of ECOWAS in 2006, have also sought to strengthen the applicability of Community legal acts within the member states. Consequently, the Commission assumes a significant role in policy shaping and agenda setting. In addition, it is responsible for ‘Community development programs and projects, as well as regulating multinational enterprises of the region’ (ECOWAS Treaty 1993: Article 19, para 3). The Commission president is appointed by the Authority, the meeting of heads of state and government, for a nonrenewable term of four years (except for the transitional 2016–18 Commission). The ECOWAS president is assisted by a vice-president and 13 commissioners elected for a period of 4 nonrenewable years. The vice-president and commissioners are appointed by the Council of Ministers ‘following the evaluation of the three candidates nominated by their respective Member States to whom the posts have been allocated’ (ECOWAS Treaty 1993: Article 18, para 4). For the appointed ECOWAS staff, ‘due regard shall be had, subject to ensuring the highest standards of efficiency and technical competence, to maintaining equitable geographical distribution of posts among nationals of all Member States’ (ECOWAS Treaty 1993: Article 18, para 5). However, since 2008 ECOWAS has been affected by a hiring freeze. In terms of its administrative personnel, the ECOWAS Commission comprises approximately 700 staff, two thirds of the total number of staff in ECOWAS organizations overall. These include the 16 statutory employees of the political Commission which are appointed by the member states. At the administrative level, there are an equal ratio of professional staff (including 40 directors) and local staff (Financial Controller’s Annual Report, 2017). At the level of professional staff, ‘equitable geographical distribution of posts among nationals of all Member States’ is an explicit consideration (ECOWAS Treaty, 1993, Article 18(5)).

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

309

Empirical Findings This section offers a multi-proxy analysis of actor-level autonomy among AU and ECOWAS staff. As outlined above, the analysis benefits from several variables from the survey: The task profile of staff (Tables 12.2, 12.3, 12.4, and 12.5), the considerations and concerns they emphasize (Tables 12.6, 12.7, and 12.8), whose arguments they deem important (Tables 12.9, 12.10, and 12.11), and their role perceptions (Table 12.12). Tables 12.2, 12.3, 12.4, and 12.5 suggest an equally distributed task profile of policymaker and organizational manager among AU and ECOWAS Commission officials which is conducive to actor-level autonomy. Staff tend to be directly involved in political consultations as well as managers and coordinators ensuring the running of everyday business. This suggests prima facie that respondents are more than only organizational managers and coordinators, and also that they assume autonomous tasks within their bureaucracy. The data moreover subscribes to the thrust of recent studies on ‘Western’ international bureaucracies exhibiting significant policymaking roles (Bauer et al., 2019; Eckhard & Parizek, 2020;

Table 12.2  Time spent on the following tasks (mean (N) and standard deviation, min = 1 max = 5) Tasks: Drafting policy proposals Giving scientific, technical, and/or legal advice Giving political advice Providing background information Contacting and meeting people Facilitating compromises between departments and units Facilitating compromises between different bodies within the organization Facilitating compromises with member-states

ECOWAS

AU

Mean

Stdev

2.3 (64) 2.2 (62)

3.2 (91) 3.1 (92)

2.8 (155) 2.8 (154)

1.4 1.4

3.6 (61) 2.4 (62) 2.0 (64) 2.5 (63)

3.8 (88) 2.9 (89) 2.3 (91) 2.8 (91)

3.8 (149) 2.7 (151) 2.2 (155) 2.7 (154)

1.4 1.4 1.1 1.2

2.9 (60)

3.3 (89)

3.2 (149)

1.4

2.6 (63)

3.2 (89)

3.0 (152)

1.5

* Applies the following 5-value scale: (1) ‘Very much’, (2) ‘Much’, (3) ‘Somewhat’, (4) ‘Little’, (5) ‘Very little/none’

310 

J. TRONDAL

Mele & Cappellaro, 2018; Trondal et al., 2010, 2018). Next, compatible with an organizational approach, staff are generally preoccupied with internal tasks over external ones. Reflecting the organizational boundaries within international bureaucracies, officials report that they are less involved in interorganizational processes such as facilitating compromises between different bodies or between member states than intraorganizational processes. Table 12.2 also suggests that the blend of policymaking and organizational managing is reflected in patterns of networking, provision of background information, facilitating compromises between different departments and units of their respective international bureaucracy, and providing political advice. These observations in sum suggest a task profile that reflects actor-level autonomy. Moreover, ECOWAS officials are slightly more involved in providing policy advice and providing scientific and legal advice than AU staff. This may ostensibly reflect a higher proportion of high-level staff among ECOWAS respondents than among AU respondents (P1). Moreover, Tables 12.3, 12.4, and 12.5 reveal substantial association between items associated with the task profile of staff. Studying covariance by using factor analysis (principal component analysis), two factors display eigenvalues above 1. Items with factor loadings close to 0.6 (and above) were retained to measure the specific constructs. Items 1–4 comprise factor 1, whereas items 6–8 comprise factor 2. The former denotes an equally distributed task profile of policymakers and organizational managers in both AU and ECOWAS Commissions, which together is conducive to actor-level autonomy. Cronbach’s alpha for the four items in factor 1 is 0.79 and 0.83 for the three items in factor 2. An index variable was created for both factors. Next, respondents were asked which considerations and concerns they emphasize during everyday work. Tables 12.6, 12.7, and 12.8 map the compound sets of considerations and concerns emphasized by the staff and offer a factor analysis that brings these considerations into two main groupings—supranational/sectoral concerns that indicate actor-level autonomy and intergovernmental/ideological concerns to denote the reverse. In short, the tables suggest that intergovernmentalism is generally more strongly emphasized than supranationalism among AU and ECOWAS Commission staff. Yet, these data also suggest that respondents tend to emphasize a compound set of concerns and considerations that

311

12  THE ORGANIZATIONAL DIMENSION OF GLOBAL GOVERNANCE 

Table 12.3  Inter-correlation matrix on task profile (Pearson’s r) 1. 2. 1. Drafting policy proposals 2. Providing scientific, technical, and/or legal advice 3. Giving political advice 4. Providing background information 5. Contacting and meeting people 6. Facilitating compromises between departments and units 7. Facilitating compromises between different bodies within the organization 8. Facilitating compromises between member-states

3.

4.

5.

6.

0.59** 0.47** 0.53** 0.35** 0.19* 0.35** 0.46** 0.22** 0.17*

0.51** 0.43** 0.23* 0.44** 0.26*

7.

8.

0.32** 0.38** 0.15

0.24**

0.43** 0.46** 0.40** 0.44**

0.37** 0.35** 0.38** 0.66** 0.50**

0.68**

*p