Editorial Introduction: Exploring policy diffusion, collaborative governance, and policy conflict

Abstract

The arrival of spring comes with some exciting new changes to the Policy Studies Journal, including the addition of multiple new members to our editorial team. We hope you will join us in welcoming new editorial board members Dr. Amber Boydstun (UC-Davis, USA), Dr. Carla Flink (American University, USA), Dr. Laura Jensen (Virginia Tech, USA), and Dr. Robin Lemaire (Virginia Tech, USA) as they begin making what we know will be significant contributions to the continual improvement of the journal's article selections and publications. The current issue's articles span three central themes in public policy research: the role of internal and external cues in policy diffusion processes, the importance of representation in the success of collaborative governance and the public's experience with service delivery, and the impact of direct political participation on policy outcomes. We summarize the major findings of each study, emphasizing how they impact our current understanding of these policy topics and how we can best approach future research in these areas. The first half of this issue's articles concerns inter-state policy diffusion, a longstanding concern of public policy research. Previous studies have led to significant insights about factors that influence this form of innovation, including neighbor and ideological cues (e.g., Linder et al., 2020; Mallinson, 2021b) and stakeholder influence (Collingwood et al., 2019; Crow et al., 2020). However, as Mallinson (2021a) argues, we know relatively little about whether these predictors are stable throughout a diffused policy's life course, or whether findings of previous single-or small-policy studies are truly representative of the general process of diffusion in different policy arenas. Specifically, Mallinson leverages a large dataset of policies to help overcome problems of statistical power affecting studies that have used small sets of policy innovations to test the differential effects of external influences, internal characteristics, and policy attributes. Mallinson's pooled event history analysis, examining 566 policies from Boehmke et al.’s (2020) State Policy Innovation and Diffusion (SPID) database adopted by U.S. states between 1960 and 2016, finds that the effects of both external signals and slack resources on diffusion dynamics do, in fact, change as policies spread. Among his major findings, Mallinson determines that less professionalized legislatures tend to adopt in later stages; the influence of neighboring states occurs early in the diffusion life course but is then supplanted by ideological learning among late and laggard adopters; and state wealth and population increase the probability of a state adopting a policy innovation, especially as its diffusion progresses. These results have important implications for further development of innovation diffusion theory, and can guide policy practitioners and advocates alike as they seek to predict or promote a policy's spread. The next four articles focus on the particular importance of the individuals who guide a policy's diffusion process within a single policymaking body. Makse (2021), for example, analyzes how key individuals might affect diffusion at the early stages of a policy's consideration. This research utilizes a multilevel logistic regression model to analyze the role of policy expertise and committee membership on a legislator's decision to champion a policy innovation, predicting bill authorship and cosponsorship for 14 criminal justice policy innovations in 42 U.S. state legislatures between 1993 and 2004. The results indicate that, at the agenda-setting stage, the consideration of new ideas depends heavily on the actions of specialists who serve on the relevant committees as well as the individuals who hold backgrounds relevant to the policy in question. When ideas are newer, individuals in both groups are more likely to author and cosponsor bills, while their role becomes consistently less important as a policy diffuses. These findings suggest that scholars should attend to the size and stability of committees within legislative systems when trying to understand and predict policy diffusion processes. Vallett (2021) probes the role played by individuals who champion policy innovations, analyzing the diffusion of a single child abuse policy. He uses event history analysis to investigate how policy entrepreneurs external to legislatures influenced the adoption of Erin's Law by 30 U.S. states between 2011 and 2017. Building upon previous work that helped illuminate the impact of policy entrepreneurs in policy agenda-setting and adoption (e.g., see Arnold, 2020; Anderson et al., 2020; Callaghan & Sylvester, 2021), Vallett finds that a policy entrepreneur's role as a policy expert is valued by state legislators and so increases the likelihood that a state adopts a policy, and that bordering states and federal legislation do not always encourage policy innovation or diffusion. This issue also presents advancements in the study of how external cues can affect an individual legislator's support of policy innovation, and especially how these cues’ effects can vary when diffusion is measured differently, such as diffusion by cosponsorship versus that by floor voting. Parinandi et al. (2021) examine data on the cosponsorship and voting records of bills to adopt renewable portfolio standards (RPSs) for electric companies across the U.S. states between 1994 and 2011. In an attempt to integrate the study of diffusion at the pre-adoption stage with the extant scholarship that focuses overwhelmingly on adoption, the authors’ regression results indicate that the influence of ideologically contiguous states is more pronounced at the policy adoption stage than the earlier cosponsorship of an RPS bill. As a result, Parinandi et al. argue that their work validates policy scholars’ particular concern with the diffusion of policies at the adoption stage. While external cues remain important predictors of policy diffusion, others’ work reminds us of the value of internal cues in the form of policy learning from past experiences. Specifically, Malkamäki et al. (2021) investigate how belief homophily, reputational influence, and co-participation in policy forums can help to explain information exchange and the building of trust among policy actors as they work to address common-pool resource dilemmas. The authors argue that in polycentric, complex governance systems wherein actors interact and leverage resources to pursue policy goals (Berardo & Lubell, 2019; Mewhirter & Berardo, 2019), a diversity of rule-based policy institutions lays a foundations for policy learning. Utilizing the context of South African tree plantation policy, Malkamäki et al. use exponential random graph models (ERGMs) to analyze semi-structured interview data from 59 organizations impacted by tree plantation land use. The authors find that actors exchange information with those who hold similar beliefs and that they tend to close information exchange triads rather than leave them open to other stakeholders. These results suggest that echo chambers play a significant role in governance network formation. Beach et al. (2021) continue an emphasis on internal cues by introducing a two-stage model of analogical reasoning based on cognitive psychology, where preferred policy solutions are justified and strengthened as individual policymakers form their opinions and subsequently make decisions. By conducting a plausibility probe of the case of the European Union's (EU) Banking Union, the authors show that analogical reasoning plays an important role in setting the parameters of policy debates. Specifically, they find that core causal lessons are transferred to the current policy being considered among policymakers and thereby shape discussion and debate. This insight helps explain why and how specific policy solutions are privileged over others and offers important contributions to the literature surrounding policy learning (e.g., Nowlin, 2020). Drolc et al. (2021) underscore the importance of tackling outstanding methodological concerns in policy diffusion scholarship. In particular, the authors argue that the spatial and temporal dynamics of policy diffusion are both important and understudied. Drolc et al. indicate that the use of temporally lagged spatial lags (TLSLs) ignores the reality that the effects of diffusion predictors are not limited to one time period but rather persist over time. By influencing the future value of neighboring observations, predictors can have much longer-lasting effects. As a result, choices in model specification can increase the risk of falsely inferring that an outcome results from spatial diffusion. The authors illustrate the impact of these misspecification issues by providing two applications in which small changes to an otherwise well-specified model can result in drastically different inferences about policy diffusion. They then follow with specific solutions and novel software that policy diffusion researchers can utilize to address these issues in future work. This series of policy diffusion articles points to the particular role of policy representatives in a policy's life course. Policymakers are central to the formulation of particular policy innovation, its placement on the legislative (and thus societal) agenda, its legal adoption, and its later implementation. In democratic polities, we typically assume that policymakers make choices in these arenas consistent with preferences of constituents and stakeholders. But what if those in these roles are not truly representative of those they serve? More specifically, does a policy better serve diverse communities when multiple identities are represented in its earlier stages? Is identity representation empirically important in the actual implementation of a policy? As the first article under this second theme, Johnston and Holt (2021) examine representative bureaucracy theory in the context of state and privately led prisons, arguing that the role of representation in prison management is still largely understudied. Using administrative data from the 2000 and 2005 Census of State and Federal Adult Correctional Facilities, the authors utilize (OLS) regression models to test the relationship among management approach, staff demographic representation, and prison performance and programming in a sample of 1378 prisons across 45 U.S. states. Their results offer support for representative bureaucracy theory's expectation that relationships between demographically similar staff members and citizen-client-inmates may enhance desired organizational outcomes such as reduced prison violence. Shared values and norms that flow from prison staff demographically resembling those they serve can help reduce the transaction costs in the civil servant–client relationship and thereby improve outcomes. Importantly, Johnston and Holt find that these effects appear to manifest in public but not private prisons, raising critical questions about how prisons are and should be managed in the United States. Dobbin and Lubell (2021) emphasize the importance of representation of disadvantaged communities (or DACs) within the context of groundwater governance. Using both primary and secondary data, the authors examine the formal representation of small DACs in the 2014 Sustainable Groundwater Management Act in California by modeling the likelihood of their representation in the state's new groundwater sustainability agencies (GSAs). Despite specific legislative requirements for DAC inclusion in governance, small, low-income, and politically marginalized communities were consistently less likely to be included than more advantaged constituents. Dobbin and Lubell find that institutions matter: DACs were more represented in agencies specifically designed for collaboration and achieved greater representation in GSAs when the DACs themselves were represented by a formal institution like a city. Surprisingly and disappointingly, when there were more disadvantaged communities in an area, any one of the DACs was actually less likely to be represented in a GSA. This work skillfully combines collaborative governance and environmental justice theory to quantitatively assess the causes and consequences of procedural environmental justice in one groundwater reform case, serving as a model for future studies in other natural resource contexts. While these two articles support the notion that political representation improves service experiences and the meeting of public concerns, it is also well-known that these effects are distinct from the experiences and perspectives that encourage direct political participation. The final three articles included in this issue focus on the factors that lead to the political mobilization of members of the general public, as well as how this participation can guide policymaker activities. Previous studies have determined, for example, a link between policy feedback effects and political mobilization, identifying distinct mechanisms through which feedback processes can mobilize mass publics (Larsen, 2019). Davis (2021) continues this line of research by analyzing how interactions with government institutions—and carceral institutions in particular—may influence future political participation. He argues that individual experiences negative resource effects when the U.S. carceral system exploits that individual's economic resources and human capital; this experience then influences participation both directly and through interpretive effects. Davis uses structural equation modeling of national survey data collected between 2000 and 2010 to estimate the causal relationship between carceral contact and political participation, while controlling for other relevant variables including political efficacy, well-being, interest in politics, and associated stressors. He finds that carceral contact has a negative effect on well-being and political participation, and that these negative effects on participation are further mediated by diminished well-being. By narrowing the focus to individuals who have had carceral contact below the level of incarceration and to those who have family members who have been incarcerated, Davis's work contributes to our understanding of how various institutional interactions are entangled with feelings of well-being that influence political participation—a finding that holds important implications for criminal reform. Continuing this issue's focus on the predictors of participation in the political process, Rosenthal (2021) questions whether the current paradigm that associates upper tier (universal) programs with mobilization and lower tier (means-tested) programs with demobilization is too simplistic. Rosenthal utilizes data from 19 social programs to analyze the impact of interpretive and resource effects on cross-mobilizing experiences, finding that the relative strength of interpretive effects impacts participation rates of the cross-mobilized population while resource effects, positivity bias, and negativity bias are unlikely to drive turnout. This research not only investigates the complexities of program participation on political involvement but also expands the scope of analysis to include the cumulative impact of policy interactions in order to examine cross-mobilizing policy experiences. Embedded within political participation is the potential of political conflict. In the last article in this issue, Wolf (2021) applies social construction and policy design theories to understand how policymakers use, interpret, and respond to labels embedded within policy, how this relates to the escalation of policy conflict, and how it affects group participation. Wolf uses the case of the Oosterweel highway proposal in Belgium to examine narratives of 32 stakeholders and 739 media communications spanning 9 years. This analysis identifies labels assigned to action groups by policymakers and the media, at various policy stages. The study's results show that media articles that contain symbolic language referencing the negative characteristics of certain action groups identified by policymakers increased from 10% to 17% over 9 years. Wolf concludes that feedforward effects can lead to increasingly antagonistic posturing of policymakers toward opposition groups, which frustrated communications and extended conflict. This study contributes to the emerging policy conflict framework by illustrating how conflict escalation can be explained by both cognitive and behavioral characteristics, and can be understood as a process (rather than an event) wherein disagreements over substantive elements can morph into relational conflicts. We believe that all the articles included in this issue offer important contributions to the study of public policy. We hope you agree and that they inspire further insights that lead to thoughtful and rigorous studies that further develop public policy theory. We also hope that when such inspiration manifests, that you will submit that research to the PSJ. Until then, we wish you all a safe and healthy spring and summer.