Local Policy Networks Research Articles

ConcaveQ: Non-monotonic Value Function Factorization via Concave Representations in Deep Multi-Agent Reinforcement Learning

Value function factorization has achieved great success in multi-agent reinforcement learning by optimizing joint action-value functions through the maximization of factorized per-agent utilities. To ensure Individual-Global-Maximum property, existing works often focus on value factorization using monotonic functions, which are known to result in restricted representation expressiveness. In this paper, we analyze the limitations of monotonic factorization and present ConcaveQ, a novel non-monotonic value function factorization approach that goes beyond monotonic mixing functions and employs neural network representations of concave mixing functions. Leveraging the concave property in factorization, an iterative action selection scheme is developed to obtain optimal joint actions during training. It is used to update agents’ local policy networks, enabling fully decentralized execution. The effectiveness of the proposed ConcaveQ is validated across scenarios involving multi-agent predator-prey environment and StarCraft II micromanagement tasks. Empirical results exhibit significant improvement of ConcaveQ over state-of-the-art multi-agent reinforcement learning approaches.

CCGN: Centralized collaborative graphical transformer multi-agent reinforcement learning for multi-intersection signal free-corridor

Tackling traffic signal control through multi-agent reinforcement learning is a widely-employed approach. However, current state-of-the-art models have drawbacks: intersections optimize their own local rewards and cause traffic to waste time and fuel with a start-stop mode at each intersection. They also lack information sharing among intersections and their specialized policy hinders the ability to adapt to new traffic scenarios. To overcome these limitations, This work presents a centralized collaborative graph network (CCGN) with the core objective of a signal-free corridor once the traffic flows have waited at the entry intersection of the traffic intersection network on either side, the subsequent intersection gives the open signal as the traffic flows arrive. CCGN combines local policy networks (LPN) and global policy networks, where LPN employed at each intersection predicts actions based on Transformer and Graph Convolutional Network (GCN). In contrast, GPN is based on GCN and Q-network that receives the LPN states, traffic flow and road information to manage intersections to provide a signal-free corridor. We developed the Deep Graph Convolution Q-Network (DGCQ) by combining Deep Q-Network (DQN) and GCN to achieve a signal-free corridor. DGCQ leverages GCN’s intersection collaboration and DQN’s information aggregation for traffic control decisions Proposed CCGN model is trained on the robust synthetic traffic network and evaluated on the real-world traffic networks that outperform the other state-of-the-art models.

Policy beliefs, policy learning, and risk perception: Exploring the formation of local creative Placemaking‐catalyzed policy network

AbstractThis article studies the formation of a local Creative Placemaking (CPM) policy network based on the Advocacy Coalition Framework (ACF) and social capital perspective of policy networks. This article hypothesizes that policy beliefs, policy learning, social capital, and the perceived risks induced by defections, as well as macro‐level changes in the broader political and socio‐economic system, influence partner selection in the local CPM policy networks. The study collected survey data from policy actors participating in the Franklinton CPM‐catalyzed revitalization project in Columbus, Ohio. An exponential random graph model (ERGM) was applied to test the hypotheses. The findings partially support the hypotheses: policy learning, certain secondary policy beliefs, and risk perceptions of changes in economic and political factors are correlated with tie formation. The reciprocity‐driven bonding structure underlies the Franklinton CPM policy network, suggesting that policy actors perceive the Franklinton CPM policy network as a high‐defection‐risk network. This study has implications for policy makers in designing engagement strategies to better involve stakeholders holding different beliefs and occupying different network positions.

Global value chains participation to enhance export: Evidence from Indonesian apparel SMEs

After recording remarkable growth and gaining more prominence within the domestic economy, Indonesian small and medium enterprises (SMEs) are now being pushed by the government to go beyond the domestic market. Indonesian government has introduced global value chains (GVCs) to SMEs to boost their export in the last recent years. Apparel SMEs normally can join GVCs through backward or forward participation. But in this case, the GVCs’ impact on apparel SMEs export remains sluggish because they are highly concentrated in forward participation. This research, therefore, provides empirical cases on how Indonesian apparel SMEs can engage in backward participation to raise their competitiveness in global market using the evidence of apparel SMEs. Utilizing three dimensions of GVCs analysis: governance, upgrading, and local policy network, this research finds that SMEs can easily engage in backward participation to obtain foreign inputs by using the internet. Moreover, the decision to venture into GVCs and global market is heavily influenced by e-commerce. Findings of this research can provide guidance for Indonesian policymakers about the importance of GVCs on SMEs’ export performance, creating a better environment for manufacturing and cross-border e-commerce that can be exploited by SMEs as the largest economic driver.JEL Classification F59; F63; L67

The politics and policy of science and technology: Past and future

The Review of Policy Research (RPR) is a long-existing and traditional journal of policy research and part of the journal family published under the aegis of the Policy Studies Organization (PSO) and in cooperation with Wiley. Originally given the name “Policy Studies Review” and as a sister journal of the Policy Studies Journal, which currently is the policy process journal with the highest impact factor, the RPR developed its own standing and profile. While generally being concerned with the politics and policy of science and technology, the recently published articles address the most pressing topics of our time. The July issue includes important articles on climate change leadership (Glasgow et al., 2021), innovation policy in China (Feng & Jiang, 2021), technology transfer and the use of digital technologies to boost productivity local policy networks (Möck, 2021), and digitalization in health care (Bogumil-Uçan & Klenk, 2021). With this editorial introduction, we are more than happy to announce that effective July 1, 2021, we are the new editorial team of this outstanding journal. Because of the great and effortful work of our predecessors, Nita Farahany, Tim Profeta, Kenneth Rogerson, and Sarah Rispin Sedlak, which we could not be more thankful for, we are able to take over a journal that is at the top of the academic community. Our deepest and most sincere thanks for making this transition so smooth and uncomplicated, and for your stellar past work! The current issue still is the result of the excellent work of this previous editorial team. All articles have been managed by them, and it is awesome to see what great compilation of articles it is that constitutes this issue. We are pleased to stay in contact with the previous editors, especially by having Ken on (editorial) board. In the months to come, we will continue to broaden the editorial board and to extend it to so far underrepresented regions. To those who are already on it: Thank you for supporting the journal in the last years, and we hope that you will hold faith with us. A huge thank you also to Paul Rich, PSO President, and Daniel Gutierrez-Sandoval, for their persistent and tireless support with all questions and requests. The trust and support you have shown us is more than any editor could ever dream of. The new editorial team consists of Nils C. Bandelow, Johanna Hornung, and Colette S. Vogeler, all based at TU Braunschweig, Germany. All of us have an international background, by having studied or having stayed at other academic institutions across Europe. All of us are academically active at various international conferences, in academic organizations and self-governance, and through publications in diverse journals (https://www.tu-braunschweig.de/coppp). Together with European colleagues, Nils and Johanna co-edit the European Policy Analysis (EPA) journal that recently published special issues on COVID-19 (Colfer, 2020) and the state of the art of theoretical frameworks of public policy, such as the Programmatic Action Framework (Bandelow et al., 2021; Hassenteufel & Genieys, 2021) and the Narrative Policy Framework (e.g., Tosun & Schaub, 2021). Therefore, we bring in some useful experience in the cooperation with the PSO and Wiley and in editing a journal. We want to pursue a long-term strategy for the RPR over the next few years. In doing so, we want to build on the existing and important focal points and the outstanding networks in the academic world and develop them further. Based on our experience with the EPA journal, we first want to strengthen the areas of communication and special issues in the short term. This means that the RPR will have a Twitter account from now on (@RPR_Journal) and will use it to promote and inform as well as network with supporters of the journal. We also want to enable guest editors to consider the RPR for high-quality Special Issue publication projects. Such special issue projects should usually consist of six articles. If you are interested in guest-editing an issue: We welcome any first or already completed ideas on special issue, e.g., from panels at international conferences or workshops! In the long run, we might think about enabling possible sister projects of RPR and EPA, which will consist of publishing new theoretical perspectives first in the EPA journal and presenting empirical applications to pressing current issues afterwards in the RPR. For any questions, comments or ideas regarding publications in RPR, feel free to always contact us via email: rpr@ipsonet.org. We'd love to connect with you!

How does Teach For America engage its alumni politically? A case study in Detroit

We describe the alumni engagement efforts by Teach For America (TFA) in Detroit as a case study of the specific ways that the organization works to influence its alumni’s involvement in educational politics and disposition towards particular types of educational reform. During the 2019-20 school year, TFA Detroit facilitated a series of “policy workshops” for its alumni, intended to inspire TFA corps members and alumni to engage in political and policy advocacy. Combining field notes and other artifacts from the policy workshops with a social network analysis of the featured participants and central organizations, we show that TFA Detroit drew upon its local, state, and national policy networks to construct workshops that in turn would politically mobilize alumni to support their networks’ preferred city and state policies and reforms.

Challenges, opportunities and solutions for local physical activity stakeholders: an implementation case study from a cross-sectoral physical activity network in Northeast England

BackgroundIncreasingly, national policy initiatives and programmes have been developed to increase physical activity (PA). However, challenges in implementing and translating these policies into effective local-level programmes have persisted, and change in population PA levels has been small. This may be due to insufficient attention given to the implementation context, and the limited interactions between local policy-makers, practitioners and researchers. In this paper we use a case study of a cross-sectoral network in Northeast England, to identify the local-level challenges and opportunities for implementing PA policies and programmes, particularly the updated 2019 UK PA guidelines.MethodsFive focus groups (n = 59) were conducted with practice partners, local policy-makers and researchers during an initial workshop in April 2018. Through facilitated discussion, participants considered regional priorities for research and practice, along with barriers to implementing this agenda and how these may be overcome. During a second workshop in December 2018, overarching findings from workshop one were fedback to a similar group of stakeholders, along with national policy-makers, to stimulate feedback from delegates on experiences that may support the implementation of the UK PA guidelines locally, focusing on specific considerations for research, evidence and knowledge exchange.ResultsIn workshop one, three overarching themes were developed to capture local challenges and needs: (i) understanding complexity and context; (ii) addressing the knowledge and skills gap; and (iii) mismatched timescales and practices. In workshop two, participants’ implementation plans encompassed: (i) exploring a systems approach to implementation; (ii) adapting policy to context; and (iii) local prioritising.ConclusionsOur findings suggest that academics, practitioners and policy-makers understand the complexities of implementing PA strategies, and the challenges of knowledge exchange. The updated UK PA guidelines policy presented an opportunity for multiple agencies to consider context-specific implementation and address enduring tensions between stakeholders. An organically derived implementation plan that prioritises PA, maps links to relevant local policies and supports a context-appropriate communication strategy, within local policy, practice and research networks, will help address these. We present 10 guiding principles to support transferable knowledge exchange activities within networks to facilitate implementation of national PA policy in local contexts.

A Multi-Agent Deep Reinforcement Learning Based Voltage Regulation Using Coordinated PV Inverters

This paper proposes a multi-agent deep reinforcement learning-based approach for distribution system voltage regulation with high penetration of photovoltaics (PVs). The designed agents can learn the coordinated control strategies from historical data through the counter-training of local policy networks and centric critic networks. The learned strategies allow us to perform online coordinated control. Comparative results with other methods show the enhanced control capability of the proposed method under various conditions.

Multilevel Environmental Governance: Vertical and Horizontal Influences in Local Policy Networks

In the context of multilevel governance, collaboration among governmental and non-governmental entities across different levels of government is increasingly popular in China’s environmental governance. Policy actors are engaged in two types of collaboration with other local governments: horizontal and vertical collaboration. Policy actors participate in horizontal collaboration when they work with entities at the same level, and in vertical collaboration when working with governments at different levels. This study examines multilevel environmental governance in China by studying how the decisions of policy actors to participate in local water governance networks are influenced by vertical pressures from higher level government and horizontal influences from other policy actors at the same level. We approach the research questions in the empirical context of local water governance in Dongguan city of Guangdong Province. With survey data collected from 31 municipal departments, 32 town governments, nine water-related private businesses and five NGOs, we tested the hypotheses with Ordinary Least Squares (OLS) regressions. The results indicate that vertical pressure and horizontal brokerage are both drivers for participation in the local water governance network.

Collaborative Environmental Governance, Inter-Agency Cooperation and Local Water Sustainability in China

Administrative fragmentation among government agencies has posed a significant challenge to environmental governance. However, few studies have theoretically examined and empirically measured how local governments in China address this with collaborative approaches. Informed by the ecology of games framework (EGF), this paper examined the status of inter-agency water governance in Dongguan City: the pilot city of aquatic ecologically friendly city of China. The research question we seek to address is: What are the main characteristics of the water governance network in local China? With empirical data collected via survey questionnaires and semi-structured in-depth interviews following the EGF, we find that local agencies have realized the need for cooperation in water resource management. However, they were not substantially familiar with each other in the local policy network, failing to reach their full potential in collaboration. Social network analysis was used to analyze the collaborative network. We found that the network has a weak to moderate density, allowing for collective action problems and leading to insufficient cooperative governance. The network has presented central actors with strong bridging capital to control information and resource flows. We also find strong bonding capital among some policy actors as shown in high reciprocity, clustering coefficient and transitivity. To strengthen the effectiveness of local water governance, efforts should be made to establish a functional department and extend the boundary of collaborative network.

The Structure of Policy Networks for Injury and Violence Prevention in 15 US Cities.

Changes in policy can reduce violence and injury; however, little is known about how partnerships among organizations influence policy development, adoption, and implementation. To understand partnerships among organizations working on injury and violence prevention (IVP) policy, we examined IVP policy networks in 15 large US cities. In summer 2014, we recruited 15 local health departments (LHDs) to participate in the study. They identified an average of 28.9 local partners (SD = 10.2) working on IVP policy. In late 2014, we sent survey questionnaires to 434 organizations, including the 15 LHDs and their local partners, about their partnerships and the importance of each organization to local IVP policy efforts; 319 participated. We used network methods to examine the composition and structure of the policy networks. Each IVP policy network included the LHD and an average of 21.3 (SD = 6.9) local partners. On average, nonprofit organizations constituted 50.7% of networks, followed by government agencies (26.3%), schools and universities (11.8%), coalitions (11.2%), voluntary organizations (9.6%), hospitals (8.5%), foundations (2.2%), and for-profit organizations (0.7%). Government agencies were perceived as important by the highest proportion of partners. Perceived importance was significantly associated with forming partnerships in most networks; odds ratios ranged from 1.07 (95% CI, 1.02-1.13) to 2.35 (95% CI, 1.68-3.28). Organization type was significantly associated with partnership formation in most networks after controlling for an organization's importance to the network. Several strategies could strengthen local IVP policy networks, including (1) developing connections with partners from sectors that are not well integrated into the networks and (2) encouraging indirect or less formal connections with important but missing partners and partner types.

Seeds of distrust: conflicts over sustainable development in a local fracking policy network in New York State

ABSTRACTPursuing economic, environmental and social sustainability requires networks endowed with persistent relationships between stakeholders. Networks serve as the vehicle to coordinate diverse policy goals among stakeholders. Within the networks, stakeholders inevitably experience some level of distrust due to the complexity and uncertainty of the unstructured problems on the sustainability. However, little is known about why stakeholders often distrust with one another in the networks. By analysing a local fracking policy network in NYS, we suggest that distrust is influenced by incongruent ecological worldviews, policy goals incongruence in the triple bottom line of sustainability, characterization framing, policy communications and knowledge exchange.

An institutional perspective of “Glocalization” in two Asian tigers: The “Structure−Agent−Strategy” of building an age-friendly city

This paper expands the focal point of glocalization to the non-western context, and revisits glocalization in light of the promulgation of an international policy regime. Particularly it examines how the politics of scale in various contexts assists local pragmatic exercises and institutional organizations in building up an age-friendly city (AFC). The functioning mechanisms of glocalization are conceptualized through a triadic framework incorporating structure (mode of local governance), agents, and strategies. The empirical cases examine how AFC has been promoted in Chiayi City of Taiwan and in Hong Kong, the former two Tiger economies. Drawing on a comparative analysis of the two cases, the paper presents three major findings. Firstly, local policy networks are integral to glocalization in that they govern how different resources (i.e., political commitment, professional knowledge, human resources, communication skills, and financial resources) are synthesized under a local institutional framework. Often local policy networks reveal the fragmentation of resources provided by different stakeholders. Secondly, the case study illustrates two different modes of local governance. The Chiayi case reflects an institutionalized governing framework with both hierarchical coordination between levels of state bureaucracies and the horizontal exchange of information and resources between the state and non-state sectors. Hong Kong reflects a grassroots mode whereby the promotion and implementation of AFC initiatives are prompted among NGOs, charities, district councils, and universities. Both modes reveal pros and cons. Thirdly, academic associations have played a major role in promoting AFC, though the extent to which their advice can shape policy decisions relies much on the discretion and political commitment of district councils.

Local Autism Policy Networks: Expertise and Intermediary Organizations

The increasing prevalence of diagnoses for autism spectrum disorder (ASD), now one in 68 children according to the National Institutes of Health (NIH), presents a number of policy implications. In particular, many of these children become eligible for special education services under the Individuals With Disabilities Education Act (IDEA). Given the specialized expertise and resources required of local education agencies (LEAs), how do they respond to this implementation challenge? In May 2015, an online survey was distributed to various governmental and nongovernmental actors in three Virginia localities to measure the extent of collaboration in local autism policy networks. The findings suggest that these networks are driven by autism-related information, and that nonprofit organizations act as intermediary organizations that bridge disparate stakeholders. The results contribute to our understanding of fragmentation across policy subsystems, with the focus here on education policy, and the implementation challenges related to a rapidly changing policy issue.

Multisector Health Policy Networks in 15 Large US Cities.

Local health departments (LHDs) have historically not prioritized policy development, although it is one of the 3 core areas they address. One strategy that may influence policy in LHD jurisdictions is the formation of partnerships across sectors to work together on local public health policy. We used a network approach to examine LHD local health policy partnerships across 15 large cities from the Big Cities Health Coalition. We surveyed the health departments and their partners about their working relationships in 5 policy areas: core local funding, tobacco control, obesity and chronic disease, violence and injury prevention, and infant mortality. Drawing on prior literature linking network structures with performance, we examined network density, transitivity, centralization and centrality, member diversity, and assortativity of ties. Networks included an average of 21.8 organizations. Nonprofits and government agencies made up the largest proportions of the networks, with 28.8% and 21.7% of network members, whereas for-profits and foundations made up the smallest proportions in all of the networks, with just 1.2% and 2.4% on average. Mean values of density, transitivity, diversity, assortativity, centralization, and centrality showed similarity across policy areas and most LHDs. The tobacco control and obesity/chronic disease networks were densest and most diverse, whereas the infant mortality policy networks were the most centralized and had the highest assortativity. Core local funding policy networks had lower scores than other policy area networks by most network measures. Urban LHDs partner with organizations from diverse sectors to conduct local public health policy work. Network structures are similar across policy areas jurisdictions. Obesity and chronic disease, tobacco control, and infant mortality networks had structures consistent with higher performing networks, whereas core local funding networks had structures consistent with lower performing networks.

Policy coordination and integration, trust, management and performance in public health-related policy networks: a survey

Intersectoral policy networks may be effective in dealing with complex public health problems. Their performance is assumed to depend on network management and trust, as well as on integrated public health policy (i.e. policy coordination and integration). We studied the role of network management and trust in the realization of integrated public health policy and network performance, as well as the relation between integrated public health policy and network performance. In 34 Dutch local policy networks, we measured the perceptions of 278 actors through a Web-based survey and used regression analyses to assess the relations between policy variables. Management and trust were positively related to perceived integrated public health policy and network performance, while integrated public health policy was also positively related to perceived network performance. In public health, the performance of intersectoral policy networks may be improved by adequate network management, the creation of trust and policy coordination and integration. Future research could further explore the role of specific characteristics of the network manager, like the manager’s background, relation to the other actors and leadership style. Points for practitioners Regarding inter-sectoral policy networks in public health: first, when aiming for the realization of policy coordination and integration, the employment of network management strategies and the creation of trust are of importance for the network manager; and, second, when also aiming for the realization of network performance, the creation of policy coordination and integration is of additional importance for the network manager.

Trust, management and performance in public health related local policy networks

Background Public health is affected by factors outside the sphere of influence of the public health policy sector, e.g. environmental determinants, and determinants such as community networks. As a result, resources to tackle these problems are scattered across different actors, thus creating strong interdependencies and complexity as these actors come from different sectors and have different perceptions of the problem. Within this context, we studied the role of …

The policy evolution of Bookstart

In this article, a holistic governance framework is used to investigate the policy evolution and diffusion of the Bookstart programme. Documentary analysis is used to uncover the evolution and processes of the Bookstart programme. The findings indicate that the development of the Bookstart programme followed a ‘bottom-up’ to ‘top-down’ process that can be broadly divided into four stages: popularization, horizontal diffusion, vertical integration and civil society response. The evolution of Bookstart in Taiwan began with relations between government agencies and enterprise organizations before transitioning to relationships between central and local governments and eventually to relations between government agencies and civil society. However, Bookstart in Taiwan has failed to develop resource integration and interaction between local governments, or a local government policy network.

Reforming police governance in England and Wales: managerialisation and the politics of organisational regime Change

This article examines reforms to the governance of local policing in England and Wales over the last two decades, from the managerialisation of policing in the early 1990s through to recent plans for the introduction of Police and Crime Commissioners. This ‘long view’ facilitates an appreciation of politically driven shifts in organisational regimes, initially from bureau-professionalism to managerialism, and more recently from managerialism to a hybridised model of local governance that combines elements of consumerism, democratic localism and bureau-professionalism. These regimes, however, act less as determining structures, and more as unstable equilibria that are open to contestation and negotiation. Drawing upon insights from governmentality and governance studies, the article reviews the progress of reforms and shows that the ‘success’ of regime changes is contingent upon the translation of regimes into practice. More specifically, understanding regimes as discourses allows us to probe their internal coherence, and their achievements in influencing the strategies and subject positions adopted by actors in local policing policy networks. This enables us to appreciate both the limited successes of regime change reforms to date, and the prospects for the latest reform, namely the introduction of Police and Crime Commissioners in each police force area outside of London.

The Public Assistance Policies of Cities and the Justice Concerns of Elected Officials: The Centrality of the Floors Principle in Addressing Urban Poverty

City councils are significant, though seldom central, actors in local policy networks providing public assistance to disadvantaged residents. Mayors and council members in 12 American cities more often support than oppose public assistance initiatives. They claim that their own normative judgments are more important to their preferences and voting behavior on such matters than are public opinion, group demands, or economic considerations. While such elected officials hold a variety of justice principles, the most important of these affecting their positions on public assistance issues is the “floors” principle. A broad ethical commitment to providing social minimums enhances support for living‐wage ordinances, for linking subsidies for economic development to assistance to less advantaged citizens, and for exempting spending on social services from budget cuts. We discuss the implications of these findings for major theories of urban politics and policies—collective‐action theory, regime theory, and pluralism—and for advocates on behalf of the urban poor.