Implementation arrangements as institutions: Assessments and prospects

In 2007, the German government adopted the parental leave benefit scheme. Although this was a federal-level social policy, it was the responsibility of the German states (Länder) to establish the structures for its implementation. The Länder opted for quite different implementation arrangements, resulting in varied outcomes. In the state of Baden-Württemberg, the implementation was entrusted to the state bank, which established a dedicated infrastructure specifically for this purpose. In the city-state of Bremen, the administration of the program was managed by the social welfare offices. In the former case, the implementation proceeded smoothly and efficiently. However, in the latter, the implementation of the policy scheme was characterized by significant delays, as the responsibilities were added to an already overburdened administration. This led to a waiting period of up to 4 months after the birth of their child before receiving their first financial payment (Adam et al., 2019). Essentially, this left young parents struggling with a single income. This example vividly demonstrates that the choice and design of implementation arrangements are actually “fundamental empirical elements of public policy” (Schneider & Sidney, 2009, p. 104), especially in times when administrative resources already seem overstretched by an ever-increasing number of tasks but limited capacities (Fernández-i-Marín et al., 2024). Governments worldwide are grappling with significant challenges such as climate change, migration, and growing inequality. Public policies serve as the primary tools for addressing these complex issues. However, the production of these policies often introduces administrative burdens, stemming from the need for thorough implementation and enforcement. To alleviate the strain on public administration during such demanding times, it is thus crucial to ensure that implementation processes function efficiently. By optimizing these structures, governments can avoid the proliferation of unnecessary policies, thereby streamlining efforts to tackle the prevailing issues effectively. Despite the critical importance of implementation structures and arrangements, or “implementation polity” as termed by Sager and Gofen (2022), further progress in this research area is impeded by several shortcomings in the literature. The current state of research on implementation arrangements has several blind spots that hinder the development of suitable solutions. While the literature offers some insights into the inter-temporal dynamics of implementation arrangements, we know much less about the factors influencing the selection of different arrangements. Specifically, why do governments in different contexts choose various implementation arrangements, even for similar policy measures? What determines the selection of specific arrangements? Additionally, what impact does this choice have on the effectiveness of public policies? Existing scholarship provides little to no answers to these critical questions. In this introductory section of the special issue on “Implementation Arrangements in the Era of Grand Challenges,” we offer a brief and systematic overview of the literature on implementation structures, identifying certain gaps and underlying biases. Notably, there is a strong focus on individual cases, environmental and climate policy, and certain world regions such as the United States and Europe. We conclude this introduction by showing how the contributions in this Special Issue address some of these shortcomings. Although some gaps still remain, we are convinced the collective efforts underlying this special issue have advanced the study of implementation structures and arrangements substantially. Implementation arrangements and structures have long been a concern in the fields of public policy and public administration research. The central debate has revolved around whether public policies are typically carried out by a “single lonely organization” (Peters, 2014, p. 132) or if a “multiorganizational analytic perspective” (Hjern & Porter, 1981, p. 201) is necessary for a comprehensive understanding of the implementation process (Sætren & Hupe, 2018). Additionally, some scholarly contributions have examined how the types of actors involved in these implementation arrangements—such as governmental, business, and non-profit organizations—have changed over time (Hall & O'Toole, 2000). A substantial body of literature indicates that implementation arrangements increasingly involve multiple actors from both the public and private sectors (Steinebach, 2021; Thomann et al., 2018). We start from the assumption that all policies require some institutional arrangements to convert abstract policy ideas and desires into real-world actions designed to alter the policy addressees' behavior (May, 2015). In line with the definition provided by Hjern and Porter (1981), we understand implementation structures and arrangements as the “administrative entity that program implementors use to achieve objectives within programs” (p. 211). This entity can encompass “clusters of components from both public and private organizations” (Hjern and Porter (1981, p. 211)) that may be located at the same or different levels of government (see also Casula, 2022; Steinebach, 2021). This way, implementation structures do form institutions in their own right (Peters, 2014). To evaluate how the concept of implementation arrangements and structures has been addressed in public policy and administration literature, we conducted a systematic literature review. We extracted all articles from the Web of Science that included the keywords: implementation AND policy AND structure*, as well as implementation AND policy AND arrangement* in their abstracts. Our search targeted journals ranked in the First and Second Quartile (Q1 and Q2) of the “Public Administration” and “Political Science” categories. In total, we retrieved approximately 1361 articles between the year 1990 and 2022. From this collection, we selected a subset using a random number generator. We carefully reviewed the selected articles to determine whether they genuinely addressed implementation structures, as opposed to coincidentally including the keywords for unrelated reasons. This iterative process continued until we compiled a final set of 210 articles, representing approximately 15 percent of the original pool. For each article in our sample, we systematically coded a range of elements with the objective of analyzing the application of implementation arrangements over the past few decades. Our analysis focused on five key dimensions: (1) the level of analysis; (2) the methodology employed; (3) the geographical region; (4) the policy sector; and (5) whether the concept was used as an explanans (a factor explaining something) or an explanandum (a phenomenon to be explained). In analyzing the first dimension, we focused on coding the primary level of government addressed in each study. Our analysis distinguished between several tiers of governance: the supranational level, involving entities such as the European Union or United Nations, where policies operate beyond national borders; the national level, where studies examine how national governments adopt and implement policies; the subnational, or state level, which involves analyses of regional governments managing policy implementation within their jurisdictions; and the local level, concerning municipal or community governance to understand how local authorities and institutions implement policies and projects. Moreover, we noted the number of governmental levels covered in each analysis. While some studies primarily concentrate on one level, others also consider interactions or comparisons across multiple levels. In examining the methodologies employed, we identified the research methods utilized in each study, initially distinguishing between qualitative, quantitative, and mixed methods approaches. However, our analysis revealed that the papers typically leaned heavily toward one methodological approach, either qualitative or quantitative, and rarely integrated both. To code the geographical areas, we concentrated on broad distinctions between continents. Specifically, we identified and categorized the regions as North America, Latin America, Europe, Africa, Asia, and Australia. To capture the policy areas covered, we adopted a somewhat more inductive methodology. First, we recorded the descriptions of the policy areas as outlined in the text. Then, we grouped them into broader categories, including climate/environmental/water/energy policy; welfare/social policy; infrastructure/urban/housing policy; health policy; financial/fiscal/economic policy; law and order policy; education policy; technology and competition policy; development policy; fishery/agriculture/food policy; security/defense policy; and migration policy. For the last dimension, we coded whether implementation structures were used as explanatory factors or rather treated as dependent variables. This approach helped us understand whether the study's focus was on using these structures to explain various outcomes or on analyzing the factors that shape the structures themselves. In the first case, we looked at how implementation structures clarify certain phenomena, and which specific aspects of these structures are used for that explanation. In the second case, we explored which aspects of the implementation structures are being explained and what factors account for the characteristics of these structures. Some papers do not fall into any of these two categories but instead, discuss general considerations for evaluating implementation structures. We have labeled these as “conceptual” contributions. Figure 1 illustrates the central insights gained from our literature review. It shows that a significant number of papers focus on the local and national levels of governance. In contrast, the supranational and international levels are far less examined concerning implementation structures and arrangements. This aligns well with Steinebach et al.'s (2024) observation that international organizations often lack the capacity and structures for implementation, rendering studies of supranational implementation structures relatively rare. Our analysis also uncovers that most studies of implementation structures concentrate on a single governance level. This finding is surprising, considering the widespread general understanding that modern policy implementation frequently involves multi-level governance systems. Additionally, our analysis highlights that qualitative approaches dominate the study of implementation structures. Considering that these qualitative studies are often more detailed but naturally focus on fewer cases, it is not surprising that the study of implementation arrangements is also dominated by single case studies rather than comparative ones. Geographically, studies on implementation arrangements are predominantly situated in Europe and North America, which collectively account for approximately three-quarters of all papers in the field. This distribution suggests a regional bias that may overlook insights from other parts of the world. In the realm of scholarly inquiry into policy implementation structures, a limited number of papers delve into conceptual discussions, offering theoretical insights into their roles and significance within policy frameworks. Typically, implementation structures are examined as independent variables, used to explain various outcomes related to policy implementation, including effectiveness, efficiency, legitimacy, trust, fairness, and quality, such as management and accountability (Adam et al., 2019). Most studies focus on organizational-level characteristics, like internal structures, specialization, mandates, and autonomy, as well as sector-level attributes such as inter-organizational dependencies, the need for coordination, and fragmentation levels. There is overall less emphasis on the creation of agencies or authorities and emerging governance, cooperation, or collaboration forms. Few studies treat implementation structures as dependent variables, exploring why certain governments establish independent agencies and what factors motivate these decisions. Research on decentralization, devolution, or multi-level governance remains sparse. In studies that investigate implementation structures, political factors—including partisanship, interest groups, electoral structures, and power—are predominant, followed by functional explanations such as operational, financial, or personal capacity and the unique aspects of specific policies. Path dependency or legacy factors are less frequently considered. Figure 2 builds on these earlier insights by illustrating the policy areas covered in various studies. It reveals that the majority of publications focus on implementation structures in climate and environmental policy, followed by welfare state and social policy, as well as infrastructure and housing, health, and financial and economic policy. The prominence of the first two policy areas is consistent with the overall distribution of policy implementation. While social policy was particularly significant in the early contributions to policy implementation research (Fernández-i-Marín et al., 2025), environmental policy has been pivotal in studies on EU compliance (Bondarouk & Mastenbroek, 2018). These insights derived from our literature analysis have enabled us to differentiate between “typical” and more “atypical” perspectives on implementation structures and arrangements. A “typical” study in this domain generally concentrates on a (1) single level of governance, often focusing on either the local or national level. Methodologically, such studies are (2) predominantly qualitative, often utilizing a single case study approach to derive insights. These investigations are frequently centered geographically on regions such as (3) Europe and North America. The policy areas most commonly scrutinized are (4) climate and environmental policies. Additionally, within these studies, implementation structures are typically regarded as (5) independent variables, used to explain the success or failure of specific policies. In contrast, studies that adopt quantitative, comparative methodologies, addressing multiple levels of government and extending their focus beyond Europe, the United States, and traditional fields like environmental and climate policy, are relatively scarce. Additionally, existing analyses seldom delve into the underlying reasons why certain policies are structured in specific ways, leaving gaps in our understanding of the decision-making processes behind the selection and design of certain implementation structures. In the previous section, we identified several dominant trends as well as shortcomings in research on implementation structures. In this section, we offer a detailed exploration of how the various contributions in this special issue advance our understanding of policy implementation structures and arrangements. Table 1 presents the characteristics of the papers across the five dimensions discussed earlier. Health policy/ Covid-19 pandemic The Evolution of Mandated Coordination in Policy Implementation: Evidence from China's The Special Issue features several studies that focus on the subnational (state) level and explore policy implementation arrangements across all levels of governments. For instance, Schröder and Watson (2024) systematically map the organizations involved in implementing the water framework in Germany. They illustrate how these structures, which differ by German states (Länder), impact the participatory potential within these regions. Similarly, the Special Issue includes several quantitative studies on implementation structures, one of which combines both quantitative and qualitative insights. Green and Koebele (2024) examine how variations in state policy implementation approaches influence the structure and mode of coordination within multilevel governance systems, and they assess the impact these differences have on policy implementation processes. The authors blend a quantitative mapping of the implementation structures with qualitative information on the resulting coordination modes. The Special Issue also includes several studies that explore implementation structures in a comparative manner. For instance, Ottone and Barbieri (2023) examine the implementation arrangements for research and innovation policies in four European countries—Italy, Spain, France, and Germany. Meanwhile, Dallara and Lacchei (2024) investigate the implementation arrangements for asylum appeals in Italy, France, and Greece. Nesti et al. (2024) further analyze and compare pilot implementation schemes for minimum income programs at the city level, examining and comparing on Barcelona and Utrecht. In terms of geographical coverage, this Special Issue not only includes studies focusing on “commons” subjects (Europe and the North Americas) but also explores other world regions. For example, Waheed (2024) investigates the concept of absorptive capacity as a distinctive approach to understanding implementation arrangements within resource-scarce public sector organizations, using Bangladesh as a case study. Similarly, both Ma et al. (2025) and Peng et al. (2025) examine implementation arrangements within the context of China. Just as with world regions, the Special Issue also extends beyond the “usual” policy fields. Di Mascio et al. (2024) explore the dual-level implementation structures in the context of economic policy and economic recovery organizations. Dallara and Lacchei (2024) focus on the implementation of migration (asylum) policies. Lastly, our special issue offers several contributions focusing on implementation as the dependent variable. Di Mascio, Natalini, and Profeti examine the interplay between EU timetable pressures and internal political dynamics in shaping and reshaping implementation arrangements. Nesti and Graziano (2024), in turn, explore how policy legacies and feedback mechanisms influence implementation structures and practices. In conclusion, this special issue contributes significantly to the field of policy implementation by addressing identified gaps and offering fresh perspectives on implementation structures and arrangements. Through an array of studies focusing on various levels of governance and utilizing diverse methodologies, our contributors have expanded the geographical and thematic scope of existing research. By exploring lesser-studied regions like Asia and policy fields beyond the common focus areas, these papers offer valuable insights into the multifaceted nature of implementation structures. Whether examining the dynamics within the European Union or delving into the intricacies of economic and migration policies, the studies shed light on how different contexts and pressures influence implementation arrangements. Additionally, the special issue highlights the importance of treating implementation structures as both explanatory and dependent variables, thereby enhancing our understanding of the factors shaping their design and functioning. While some research gaps remain, the collective efforts within this issue advance the study of policy implementation, encouraging future research to further explore these complex and critical elements of public policy. Mattia Casula (University of Bologna, [email protected]) Yves Steinebach (University of Oslo & LMU Munich, [email protected]) 2007年，德国政府通过了育儿假福利计划。尽管这是一项联邦级社会政策,但建立其实施结构却是德国各州(Länder)的责任。各州选择了截然不同的实施安排,结果也各不相同。在巴登-符腾堡州，实施工作委托给州银行,该银行专门为此建立了专用基础设施。在不来梅市,该计划由社会福利办公室管理。在前一种情况下,实施工作进展顺利、高效。然而,在后一种情况下,政策计划的实施出现了严重的延误,因为这些责任被加到了本已不堪重负的行政部门上。这导致父母在孩子出生后要等待长达四个月才能收到第一笔财务付款(Adam et al. 2019)。从本质上讲,这让年轻的父母只能靠一份收入挣扎度日。 这个例子生动地表明,实施安排的选择和设计实际上是“公共政策的基本经验要素”(Schneider and Sidney 2009, p. 104),尤其是在行政资源似乎已经因任务数量不断增加、自身能力有限而捉襟见肘的时候(Fernández-i-Marín, et al. 2014, p. and Porter, 1981, p. and Hupe, and O'Toole, et al., Steinebach, also Steinebach 2021; Casula of et al., et al., & Mastenbroek, Health policy/ Covid-19 pandemic The Evolution of Mandated Coordination in Policy Implementation: Evidence from China's Mattia Casula (University of Bologna, [email protected]) Yves Steinebach (University of Oslo & LMU Munich, [email protected]) 2007, social a a Länder Baden-Württemberg, a Bremen, a a a a (Adam et al. 2019). a (Schneider Sidney, 2009, p. 104), et al., 2024). a a crucial A Sager Gofen (2022), ideas a a debate central a (Peters, 2014, p. 132) (Hjern Porter, 1981, p. 201) (Sætren Hupe, 2018). (Hall O'Toole, 2000). a sector et al., Steinebach, 2021). ideas (May, 2015). Hjern Porter (1981), (p. 211). Steinebach, 2021; Casula, (Peters, 2014). Web of Science (Q1 Q2) total, 1990 2022. no final 210 15 (1) (2) (3) (4) sector (5) explanans factor explanandum municipal Asia Australia. a a no 1 local a Steinebach et al. (2024) a general no regional (Adam et al., 2019). a a personal 2 general social a (Fernández-i-Marín et al., 2025), (Bondarouk Mastenbroek, 2018). (1) a local (2) a (3) (4) (5) 1 Schröder Watson (2024) (Länder), Green Koebele (2024) Ottone Barbieri (2023) Dallara Lacchei (2024) (2024) a a Barcelona Utrecht. no Waheed (2024) sector Bangladesh Ma et al. (2025) Peng et al. (2025) China. Di Mascio, Profeti (2024) Dallara Lacchei (2024) Di Mascio, Profeti a Nesti Graziano (2024), a A Asia a