Introduction
Mini-publics embody representative subsets of the wider population and are gathered to deliberate on salient political issues (Curato et al., Reference Curato2021). Despite initial skepticism (Mutz, Reference Mutz2008: 533), a large body of literature documents that mini-publics are capable of creating a unique context in which the deliberative ideal of inclusive reason-giving can flourish (for an overview, see Dryzek et al., Reference Dryzek2019). As consultative bodies, governments have recognized their potential, and they are increasingly used to inform and legitimize policymaking (Farrell and Luke, Reference Farrell and Luke2022). A modest but growing number of mini-publics have even become institutionalized (Courant, Reference Courant2022), meaning they have become permanent fixtures legally incorporated into the formal political decision-making structures of the state. Examples include Mongolia’s law to institutionalize deliberative polling as a legal requirement in the constitutional amendment process (Fishkin and Siu, Reference Fishkin, Siu, Baogang, Bree and Fishkin2022), the Standing Citizens’ Assembly in Paris, and the Vorarlberg Bürgerrat in Austria (OECD, 2021). In brief, there is a trend of mini-publics becoming more entangled with power in contemporary democracies.
As mini-publics increasingly engage with decision-making processes, concerns regarding their susceptibility to elite capture have emerged among scholars (Beauvais and Warren, Reference Beauvais and Warren2019; Richardson et al., Reference Richardson, Durose and Perry2019). While acknowledging the potential for mini-publics to foster inclusive deliberation, skeptical theoretical arguments raise concern that the very design features that promote qualitative deliberation – skilled facilitation, information provision, and expert testimony – could serve as conduits for elite influence and manipulation (Landa and Pevnick, Reference Landa and Pevnick2021; Umbers, Reference Umbers2021). In essence, the design elements that enable meaningful citizen engagement paradoxically increase the risk of elite capture (Bagg, Reference Bagg2024).
Hence, as mini-publics move closer to power it is no longer sufficient to study whether they can invigorate qualitative deliberation among their participants. Instead, we need to address a new question: How can we design mini-publics that are robust against capture by the wealthy and powerful? After all, we can safely assume that the more prominent mini-publics become, the more they will become an attractive target for elite capture. At present, only a few studies have begun to explore the relationship between mini-publics and the threat of elite capture, revealing that capture by political elites and private interests is a threat to be taken seriously (Curato et al., Reference Curato, Parry and Melisa2025; Oross and Tap, 2021; Parry, Reference Parry2024a). However, a comprehensive understanding of the ways in which mini-publics can be vulnerable to and resilient against elite capture threats throughout their entire process remains lacking.
The present paper aims to address this gap in the literature by proposing an analytical framework for studying capture threats and by applying it to a pioneering case, the Permanenter Bürgerdialog (PBD) in the German-speaking Community of Belgium. The PBD is the first randomly selected assembly to collaborate with parliament on a permanent basis (Niessen and Reuchamps, Reference Niessen and Reuchamps2022). Although the PBD lacks formal law-making powers, parliament is legally obligated to formally consider its recommendations and provide justification in case of rejection (Macq and Jacquet, Reference Macq and Jacquet2023). This case is well-suited for studying threats of elite capture and strategies to prevent it for two main reasons. First, the PBD is more tightly coupled to power than other mini-publics, making it more attractive to capture. Second, the PBD relies on unique design choices to prevent capture through a strategy of balancing and monitoring among its various bodies.
The first section of this article reviews the literature on mini-publics and the risk of capture. Drawing from this literature I theorize how the complex designs of mini-publics could create capture threats before, during, and after the process. In the second section, I present the PBD case. I then proceed to describe the methods and data employed in the third section. The findings are presented in the fourth section, with three key findings deserving emphasis. First, due to its complex procedures, there is considerable leeway for capture in the PBD. Second, the PBD is barely covered in the media, which makes it more vulnerable to elite capture. Finally, the PBD demonstrates how checks and balances through a separation of powers can mitigate capture threats.
Theory: mini-publics and the risk of elite capture
There is a longstanding concern among democratic theorists regarding capture in politics and elections (Bartels, Reference Bartels2017; Shapiro, Reference Shapiro2017). While there is a lively debate about the precise definition of capture in the literature (Dávid-Barrett, Reference Dávid-Barett2023; Fazekas and Tóth, Reference Fazekas and Tóth2016), this paper is grounded in the definition introduced to the field of deliberative democracy by Bagg (Reference Bagg2021), as adopted from Carpenter and Moss (Reference Carpenter, Moss, Carpenter and David2013). In their understanding, capture refers to various processes through which powerful factions intentionally influence public policy to the detriment of the general public. This means that a capture process consists of three parties. In essence, when capture occurs, laws, regulations, and resources are redirected from serving the public (the first party) at large by public office holders (the second party) toward third-party private factions exercising informal political influence on these office holders (Dávid-Barrett, Reference Dávid-Barett2023). Capture can operate in various ways, such as direct kickbacks, promises of favors, use of personal connections, physical threats, bribes, and misinformation campaigns (Dávid-Barrett, Reference Dávid-Barett2023; Jancsics, Reference Jancsics2019). Any policy-making public institution can be targeted by elite capture, but the risk of capture increases when these institutions wield significant influence in the policy-making process and are easily targeted (Lindsey and Teles, Reference Lindsey and Teles2017).
This paper takes a particular interest in the threat of elite capture, focusing not so much on how often capture occurs, but on how resilient or vulnerable mini-publics are perceived to be to powerful elites attempting to capture them. I focus on threats and the resilience of mini-publics to these threats for two main reasons. The first reason is theoretical: studying threats allows us to anticipate capture in future mini-publics that are tightly coupled to power, even when capture has not occurred in the mini-public under examination. This is important because there is a clear ongoing trend of an increasing number of mini-publics becoming more entangled with power (Farrell and Luke, Reference Farrell and Luke2022).
Second, studying the actual occurrence of capture is more challenging from a methodological perspective than investigating perceived threats of capture. Carpenter and Moss (Reference Carpenter, Moss, Carpenter and David2013) are particularly critical of the empirical approaches in existing literature, arguing that simply identifying the “winners” of a particular policy and assuming this outcome was the intended effect of elite capture involves a major assumption. This assumption contradicts a substantial body of research that recognizes that public policy often results in far-reaching unintended consequences. In the context of mini-publics, Junius (Reference Junius2023b), for instance, shows that the lack of mini-public uptake does not necessarily result from ill-willed officials but can occur through murky processes such as the (mis)interpretation of mini-public recommendations.
Although the threat of capture in mini-publics has rarely been the subject of academic scrutiny, in theory, we can expect capture in mini-publics to differ from capture in other political institutions in several ways. First, some scholars have argued that capture is less of a threat to mini-publics than to traditional institutions of contemporary politics, such as parliament and government (e.g., Abizadeh, Reference Abizadeh2021). The basic idea is that elected representatives often require financial support for campaigning, whereas citizens drawn by lot do not (Zakaras, Reference Zakaras2010: 461). Moreover, while politicians’ and civil servants’ lengthy careers can make them susceptible to capture, mini-public participants only serve for a short period. Therefore, they cannot become careerists who remain in power for an extended time (Owen and Smith, Reference Owen, Smith, Gastil and Wright2019). This is because mini-publics are either one-shot initiatives with a short-lived composition of citizens or institutions in which citizens are frequently rotated. Additionally, while some politicians, such as party leaders and ministers, are clearly more important than others, mini-public participants are much more alike in status, and power is therefore much more decentralized in a mini-public, making them harder to capture by elite factions.
Other scholars, however, argue that the threat of capture might even be higher for mini-publics (Landa and Pevnick, Reference Landa and Pevnick2021; Umbers, Reference Umbers2021). Their arguments can be summarized in two main strands. First, critics challenge the idea that participants’ short experience makes them less susceptible to corruption. Serving a short term renders them much less visible than politicians, which is a precondition for accountability (Umbers, Reference Umbers2021), thus exacerbating the lack of elections as an accountability mechanism (Lafont, Reference Lafont2019). Landa and Pevnick (Reference Landa and Pevnick2021: 54) contend that, lacking such accountability-based constraint, mini-public participants have no strong incentives to resist capturing forces.
A second strand of arguments focuses on an important paradox in mini-publics’ design. Mini-publics foster citizens capable of qualitative deliberation but only by virtue of a whole set of design features such as the inclusion of bureaucrats, experts, translators, and facilitators. In turn, these design features open up capture threats (Bagg, Reference Bagg2024; Landa and Pevnick, Reference Landa and Pevnick2021: 58–59). Hence, the highly controlled nature of mini-publics facilitates qualitative deliberation on the one hand but also introduces the potential for elite capture on the other. To summarize the literature from the perspective of capture experts, the mini-public context is different from traditional institutions due to its higher degree of decentralization and complexity, and a lack of clear accountability and transparency mechanisms.
While these arguments have been stated in general terms, I contribute to the literature by proposing a theoretical framework for potential capture threats before, during, and after mini-publics take place. I draw inspiration from the well-established input, throughput, and output framework in the mini-public literature to map possible capture threats at different stages of a mini-public’s cycle (Caluwaerts and Reuchamps Reference Caluwaerts and Reuchamps2015; Courant Reference Courant2021; Junius, Reference Junius2023a). Afterward, I will apply this framework to study how the PBD addresses the challenge of capture.
Before we proceed two caveats are in order. First, while the proposed framework has value as an analytical tool to organize the literature and as a normative tool, modern policy models contend that it is almost impossible to distinguish input from output (e.g., Sabatier, Reference Sabatier and Paul1999). As such, just like the policy stages model which takes a similar approach, the model should not be interpreted as reflecting a perfect linear process but as a heuristic device to organize the literature and propose clear normative benchmarks (Jann and Wegrich, Reference Jann, Wegrich, Fischer, Miller and Sidney2017). Second, the analytical framework is unlikely to encompass all possible forms of capture as the literature is still in its early stages on the topic. It should therefore be considered an open-ended framework that can accommodate future research.
Mini-publics and elite capture: a framework for analysis
In this section, I provide an analytical framework to study threats of elite capture in a mini-public context. An overview can be found in Table 1.
Elite capture threats in mini-publics
Capture at the input phase
The input phase entails what happens before a mini-public or an iteration of an institutionalized mini-public takes place. At this stage, agenda setting is a first-capture threat. Agenda setting in a mini-public involves not only determining its topic but also defining its intent (Caluwaerts and Reuchamps, Reference Caluwaerts and Reuchamps2015). Although the topic of deliberation is sometimes left to the participants themselves, most mini-publics have a governance-driven agenda set by public administrations (Warren, Reference Warren2009).
This state of affairs provides an important potential channel for illegitimate elite interference. Elites might capture mini-publics to advance their specific interests by organizing them as a form of window dressing (Hammond, Reference Hammond2021). Alternatively, they could push issues onto the agenda by organizing a mini-public about them in cases where they have issue ownership (Walgrave et al., Reference Walgrave, Tresch and Lefevere2015).
After all, the agenda-setting process stages mini-publics. The way in which a mini-public’s setting is staged influences what is said, how things are said, and who feels comfortable speaking sincerely (Hajer, Reference Hajer2005). Accordingly, this setting can be actively manipulated to create an advantage for certain groups over others.
Capture at the throughput phase
In the throughput phase, during the actual deliberations, actors facilitating qualitative deliberation can paradoxically also be conceived as capture threats. Facilitators are often professional consultants hired by the government for their services (Bherer et al., Reference Bherer, Gauthier and Simard2017). The crucial role of skilled facilitation in the practical implementation of mini-publics is widely recognized by scholars (e.g., Moore, Reference Moore2012). Qualitative deliberation is not self-regulating, therefore, skilled facilitators are needed to manage the process (Bherer et al., Reference Bherer, Gauthier and Simard2017).
Extant research reveals clear risks related to the commercial nature of many professional facilitators in the participation industry (Lee, Reference Lee2014; Parry, Reference Parry2024a). The commodification of facilitation leads to a clear channel for money to influence the practice of deliberation (Christensen and Grant, Reference Christensen and Grant2020). Practitioners often operate in a market context that strongly incentivizes them to act in accordance with market logic. Holding facilitators accountable can be extremely challenging as they are masters of their craft, and experimental evidence shows that they can effectively manipulate the process in very subtle ways (Humphreys et al., Reference Humphreys, Masters and Sandbu2006; Spada and Vreeland, Reference Spada and Vreeland2013). As mini-publics gain increasing prominence, they also become increasingly attractive to major consultancy firms. Some practitioners fear that these firms are more concerned with profit than with professional deliberative ethics (Parry, Reference Parry2024a), further opening up this potential capture threat.
A second capture threat is the reliance of mini-publics on expert testimony and information booklets. Experts and information play a crucial role in many mini-publics (Curato et al., Reference Curato2021). Expert knowledge can provide technical insights into certain complex issues, enabling citizens to voice their perspectives more effectively in critical interaction with scientific expertise (Moore, Reference Moore2017).
However, if citizens are supposed to be informed through experts, then these experts are likely to suffer from various biases, as vested interests will seek to control them. Some authors have voiced concerns that experts might influence participants’ opinions in an undesirable way (Moore and MacKenzie, Reference Moore and MacKenzie2020; Roberts et al., Reference Roberts, Lightbody, Low and Elstub2020). Experts can selectively represent data and misrepresent evidence (Anderson, Reference Anderson2011). In theory, most mini-public designs use balanced panels of experts representing the different sides of the issues (note that in practice, this does not always seem to be the case; see Caluwaerts and Reuchamps, Reference Caluwaerts and Reuchamps2018). While balanced panels might guard against experts’ own ideological biases, this does not immunize these experts from being influenced by third-party elites through capture mechanisms such as direct kickbacks, promises of favors, use of personal connections, or physical threats. Note that these are theoretical worries drawn from the capture literature (Dávid-Barrett, Reference Dávid-Barett2023; Jancsics, Reference Jancsics2019) that have not been observed in a mini-public context. However, as mini-publics grow in prominence, the literature should pay more attention in detecting, or – at the very minimum – preventing such capture mechanisms observed in other institutions.
Moreover, a particularly significant issue to consider is who has the authority to decide who is invited and presented as an expert in a given mini-public. In mini-publics, this issue is often addressed on an ad hoc basis by the organizer without any accountability mechanism. Third-party elite factions could then exploit this weakness to capture the process by influencing who gets invited as an expert.
Participants represent a third capture threat. Even though they serve on one mini-public for a limited period, there is no reason why participants cannot be captured by vested interests (Bagg, Reference Bagg2024). Their vulnerability stems from the fact that selected citizens cannot be re-elected as an accountability mechanism to assess their integrity (Umbers, Reference Umbers2021). Under the current circumstances, participating citizens can commit the perfect crime because they cannot be held accountable for being captured. Miller et al. (Reference Miller, Nicholson-Crotty and Nicholson-Crotty2011) show that shorter term limits pull away power from legislators to organized interest groups in the US state lower chambers, which renders corruption more likely. This suggests that citizens selected by lot who are only serving for a short term as well are not necessarily better at resisting capture compared to politicians.
Capture at the output phase
In the output phase, which occurs after deliberations have taken place and recommendations have been formulated, mini-publics can be captured in two main ways. First, there is a potential threat of capture when recommendations are transferred to empowered policymakers. An extensive literature addresses cherry-picking at this stage, referring to the idea that policymakers adopt recommendations that align with their own agenda while sidelining those that deviate from it (e.g., Font et al., Reference Font, Smith, Galais and Alarcon2018; Vrydagh, Reference Vrydagh2022). The prominence of cherry-picking indicates a threat of elite capture. As Parry et al. (Reference Parry2024b, p. 2) put it, “there is nothing stopping commissioning bodies from selectively implementing citizens’ recommendations that align with their political agenda.” Elites can influence politicians to cherry-pick recommendations that serve their interests and reject others. The issue is exacerbated by the fact that debates in mini-publics are notoriously invisible and barely covered by the media (Pomatto, Reference Pomatto, Elstub and Escobar2019; Rountree and Curato, Reference Rountree, Curato, Reuchamps, Vrydagh and Welp2023). When debates are minimally visible, it is difficult for citizens, social movements, or even politicians to effectively exercise accountability. However, cherry-picking does not always result from capture by powerful factions. Junius (Reference Junius2023b) shows that the lack of mini-public uptake does not necessarily stem from ill-willed officials but can arise from murky processes such as (mis)interpretation of mini-public recommendations. In other words, not all cherry-picking originates from elite capture, but the ease with which it occurs facilitates capture.
A second issue is related to the administrative implementation of mini-public recommendations (Dean, Reference Dean2023). Implementation is not a neutral process; administrators hold significant discretion in how and when they implement certain recommendations (Boswell, Reference Boswell2016). Given that participating citizens return to their normal lives after their recommendations have been formulated, they are unlikely to spend much time following up on their implementation. This creates another capture threat, as it becomes easy to reinterpret or misrepresent the recommendations made by those citizens.
Now that we have discussed capture threats related to mini-publics’ input, throughput, and output phases, I will provide an outline of the Permanenter Bürgerdialog in Ostbelgien as the case in which these theoretical mechanisms are explored.
The Permanenter Bürgerdialog in Ostbelgien: a blueprint
The Permanenter Bürgerdialog (PBD) was established on February 25, 2019, in the Parliament of the German-speaking Community of Belgium, also known as Ostbelgien (Niessen and Reuchamps, Reference Niessen and Reuchamps2022). For the institutionalization of the PBD, the Ostbelgien government requested that the G1000 steering committee collaborate with international experts to develop a model proposal. The G1000 was a pioneering deliberative experiment in Belgium that took place in 2011 (Caluwaerts and Reuchamps, Reference Caluwaerts and Reuchamps2018). Even though Ostbelgien is the smallest federal entity in Belgium, the rest of the country has been a fertile breeding ground for deliberative processes since the pioneering G1000 (Vrydagh et al., Reference Vrydagh, Caluwaerts and Reuchamps2022).
The PBD consists of three main bodies: the Permanent Secretariat, the Citizens’ Council, and the Citizens’ Assemblies. In this section, I review their formal characteristics (see Niessen and Reuchamps, Reference Niessen and Reuchamps2022).
I begin by discussing the Permanent Secretariat, which is quite unique to the Ostbelgien case due to its distinct parliamentary institutionalization. The Permanent Secretariat is a small team of civil servants appointed by the Secretariat General of the Ostbelgien parliament. They serve as the link between the parliamentary administration and the PBD. The Permanent Secretariat is responsible for the overall administrative and organizational aspects of the PBD, including financial and logistical matters. To fulfill these tasks, the Permanent Secretariat is funded by parliament and overseen by the Citizens’ Council. The Permanent Secretariat also has an advisory vote in the Citizens’ Council.
Another unique aspect of the Ostbelgien model is the Citizens’ Council, which organizes and monitors the proceedings of the Citizens’ Assemblies. The Citizens’ Council consists of 24 individuals who are former members of the Citizens’ Assemblies, selected by lot. Members serve for 18 months, with eight members being replaced every six months by newly drawn former members. They appoint their own president and receive attendance fees. The body has three main functions. First, it organizes the Citizens’ Assemblies, determining their number and duration while considering the budget provided by parliament. Additionally, the Citizens’ Council sets the agenda for each Citizens’ Assembly by formulating specific questions, which requires a two-thirds majority within the Council. Once a Citizens’ Assembly is planned, the Council appoints facilitators and experts. Second, the Citizens’ Council monitors the work of the Permanent Secretariat, providing guidelines or criticism in dialog with parliament. Finally, the Citizens’ Council is responsible for the follow-up on recommendations made by the Citizens’ Assemblies. This includes engaging in dialog with parliament to ensure progress on the recommendations. Parliament is obligated to inform the Citizens’ Council about progress on a yearly basis.
The Citizens’ Assemblies are similar to most other mini-publics in the literature. These Assemblies are successive iterations of typical one-shot mini-publics. In each iteration, between 25 and 50 citizens are drawn by lot to deliberate on a societal issue and formulate recommendations. The participants must represent the Ostbelgien citizenry in terms of age, gender, socio-economic status, and geographical origin, which is ensured by a stratified sample that oversamples for the least advantaged. Although facilitators guide the process, participants are expected to take ownership of their Assembly’s process. Recommendations are formulated by consensus or by a four-fifths majority vote, with a quorum of four-fifths of the members present. Minority opinions do not have the status of recommendations but are recorded and transmitted to parliament.
Like other mini-publics in Belgium, the Citizens’ Assemblies of the PBD can only be consultative. This is due to Article 33 of the Belgian Constitution, which assigns all formal law-making competences to parliament and government. Consequently, the outputs of deliberative and participatory processes cannot be legally binding without a constitutional reform (Niessen and Reuchamps, Reference Niessen and Reuchamps2022). Therefore, the PBD must collaborate with parliament to turn its recommendations into law. Parliament is required to discuss the recommendations with the members of the Assembly who formulated them. If parliament wishes to reject any recommendation, it must provide a specific public justification (Macq and Jacquet, Reference Macq and Jacquet2023). As a result, the PBD is more tightly coupled to Ostbelgien’s deliberative system compared to other mini-publics, which generally lack such a justification requirement (Hendriks, Reference Hendriks2016).
Now that I reviewed the core formal characteristics of the PBD in Ostbelgien, it is clear that the PBD is a deviant case compared to other mini-publics. It is more tightly coupled to power and therefore more attractive to capture. Additionally, the interplay between the PBD’s main bodies is quite unique. Its design involves significant balancing and monitoring, which should make it more resilient to capture. Extreme cases are limited in external validity (Flyvbjerg, Reference Flyvbjerg, Denzin and Lincoln2011). Yet, the PBD allows us to anticipate capture in future mini-publics that are closely tied to power, which is important given the international trend toward more consequential mini-publics. Furthermore, few mini-publics are as well-equipped to address capture as the PBD, which relies on a balance of power between its three bodies. As a result, we can learn whether and how capture can be mitigated from a case that is attractive to capture (due to its close ties to parliament) and has explicitly made design choices to resist capture.
Method
To study how the PBD deals with the challenge of capture, this paper conducts an in-depth qualitative case study. The data employed in this article mainly consist of 17 semi-structured interviews, supplemented by desk research based on publicly available internal documents and news articles. Four key groups of actors were interviewed. First, I interviewed two designers involved in the original summary plan for the project to gauge whether the PBD is explicitly designed to resist capture. Second, I interviewed four members of the Citizens’ Council, as the Council is supposed to play a central role in monitoring the integrity of the entire project. Third, I interviewed ten participants in the Assembly to gain insight into how they perceive the possibility of capture in the process in which they participated. Finally, I interviewed an employee of the parliamentary administration because this body is in charge of coordinating the administration and logistics of the PBD through the Permanent Secretariat.
The interviews were conducted from March to May 2023 and lasted between 48 and 102 minutes. Given that the target group was rather small and that elite capture could be a delicate topic, I ensured full anonymity for the respondents and relied on a convenience sample for the recruitment of interviewees. Since the target groups are small, we cannot provide much information on the socio-demographic features of the respondents without compromising their anonymity. However, a pooled overview of our respondents’ gender, age, and education can be found in the online Appendix 1. The interviews were structured by an interview guide reflecting the dimensions of the analytical framework mentioned above. Although socially desirable answers cannot be ruled out entirely, I tried to minimize the risk by reminding respondents of their right to refuse to answer in the informed consent form and during the interview, and by assuring confidentiality and anonymity in the treatment of data.
All interviews were fully transcribed. The transcripts were subsequently coded in NVivo, following a thematic analysis approach. I followed three main steps (Boyatzis, Reference Boyatzis1998). First, I used a theory-driven approach by coding and interpreting the data within a deductively constructed coding scheme based on the proposed analytical framework. As a second step, I adopted a more inductive approach to thematic analysis by empirically identifying new themes within each dimension of the analytical framework. The final coding scheme, interview guide, and list of respondents can be found in Appendix 1.
Analysis
The analysis is structured along the lines of the analytical framework introduced above. I outline capture threats and strategies to deal with them during the PBD’s input, throughput, and output phases. An exhaustive list of codes resulting from the analysis can be found in the Appendix 1.
Capture in the input phase
As a first capture threat, I examined the agenda-setting process. A first finding is that the theoretical designers of the mini-publics were well-aware of capture threats at this stage:
From previous experiences it became evident to us that the agenda setting process is vital. What occurs before you let citizens deliberate is especially vulnerable to elite interference or top-down manipulation. To avoid this threat, we decided to empower the Citizens’ Council to set the agenda.
Respondent 3 – designer
However, in practice, the issue does not resolve itself that easily. Citizens’ Council members report that the Secretariat, which is supposed to merely play a supportive role but is much more central to the decision-making in practice. Even though formal power remains with the Council members, they are quite positive about the Permanent Secretariat playing a bigger role:
In practice, I think that we often follow the Permanent Secretariat, those people have been organizing multiple mini-publics and possess the expertise. They always ask for our opinion, which is something that I highly appreciate. They treat us very inclusively.
Respondent 4 – Council member
Here, we encounter an intriguing paradox. Citizens participating in the Council appear to be highly satisfied with the Secretariat’s inclusive approach. However, this presupposes that the Secretariat leads the process, whereas the citizens themselves are supposed to be the owners of the process according to the design. The interview data suggests that this sentiment is not the result of malign intentions by the Permanent Secretariat but rather stems from how citizens act in the specifically staged setting of the Council. As a member of the Council testifies:
Yes, we often ask the administration to help us. But you have to understand that we are placed in a new setting that can feel overwhelming to us. It does not come naturally to us to make important decisions in a new setting.
Respondent 7 – Council member
It now becomes clear that the Permanent Secretariat holds much more power in practice than intended by design. Although this influence is not necessarily wanted, this complicates the Council’s ability to monitor the Permanent Secretariat. As testified by our interviewee from the administration:
I have to admit that sometimes it feels uncomfortable to realize how much subtle power we hold. I am pleased that we are not solely accountable to the Council but also to parliament. From the perspective of elite capture, it is reassuring that multiple bodies keep an eye on us.
Respondent 1 – Parliamentary administration
Studying the PBD in its input phase revealed challenges concerning capture. It is noteworthy that there is a clear discrepancy between how much designers thought about the risk of capture and how much Council members and the Permanent Secretariat are actively addressing this challenge. For them, capture threats are something they have given little thought to. For some, these interviews were the first instances in which they explicitly reflected on the possibility of capture. However, they recognized that there is a clear threat that should be addressed. For instance, an interesting challenge revealed by their reflections is that the Permanent Secretariat is a major potential channel for elite capture even though they are only supposed to play a small role. Due to their centrality in an informal, complex, and messy process, they are an attractive target for vested interests looking to capture the PBD. More optimistically, the results show how the Permanent Secretariat can be kept in check by parliament and the Council itself.
Capture in the throughput phase
In the throughput phase, I consider three capture threats, namely the role played by experts, facilitators, and participants. First, I consider experts as a capture threat. The data suggests that the Permanent Secretariat plays a larger role than initially envisioned in the design. While selecting and briefing the experts and facilitators is formally the responsibility of the Citizen Council the Secretariat does most of the work. The Secretariat compiles a pre-filtered list of experts and presents this list to the Council, which remains formally empowered to decide on the final list but has always adopted the Secretariat’s version. An interview with a member of the Council reveals how this state of affairs came to be:
We convene once a month, we do not have enough time to do most of the work ourselves. The Secretariat does a brilliant job, but it is true that if she would manipulate the selection of experts and facilitators it could be very difficult for us to find out. I think that much of the process relies on trust. You have to know that we only receive a small stipend, we are here because we enjoyed the process and find it important. However, this does not mean that we want to do the majority of the work.
Respondent 5 – Council member
The quote above provides informative insights into the debate on the stealth democracy thesis in deliberative mini-publics (Hibbing and Theiss-Morse, Reference Hibbing and Theiss-Morse2002). Although empirical evidence indicates that citizens are willing to participate in mini-publics (Neblo et al., Reference Neblo, Esterling, Kennedy, Lazer and Sokhey2010), this quote suggests that a long-term investment might be too demanding under the usual stipend arrangement. It also illustrates how technical administrators hold more power in mini-publics than was intended by design.
Regarding experts, respondents have pointed to a second challenge regarding capture: the way in which experts fulfill their role. All interviewed parties recognize that experts could exert undue influence on the process. However, this is often not easy to remedy, as a member of the administration highlighted:
It is very difficult to control experts and their information is vital for the quality of the process. To be frank, we often struggle to find appropriate experts, and we do not want to impose too many demands on them when they are willing to participate. We mostly rely on German experts who do not have a stake in Ostbelgien and hope that the facilitator can compensate for any potential bias.
Respondent 1 – Parliamentary administration
This quote shows that while there is a substantial body of research on the appropriate role of expertise in deliberative mini-publics (Moore, Reference Moore2017; Roberts et al., Reference Roberts, Lightbody, Low and Elstub2020), practical constraints can make the implementation of an ethical framework cumbersome. The evidence shows that practice is once again much messier than design, which can pose a challenge regarding capture.
Based on the extant literature above, I have argued that facilitators are vital actors in bringing about qualitative deliberation, that they need to juggle between profit and deliberative ethics, and that they could potentially manipulate the entire process. Given the modest number of participants (25–50), the PBD employs a single facilitator, a private consultant. From a capture perspective, this puts a lot of eggs in one basket. Nevertheless, the facilitator can be effectively monitored by other actors:
We received negative feedback about the previous facilitator from participants in the Assembly and academics observing the process. They thought that he was too steering and insufficiently neutral. After hearing all parties in a meeting organized by the Permanent Secretariat, we decided to replace him.
Respondent 11 – Assembly member
While hiring a single facilitator posed a capture threat, this event in the PBD shows how participants and other stakeholders can effectively counter-balance potential threats to the process. As the first experience of its kind the PBD is flawed but reflexive learning and adjustment can foster institutions that are more robust against elite capture. However, from this perspective, it is worrisome that participants do not necessarily give much thought to potential capture. Most participants I interviewed never thought about the possibility of a corrupt facilitator, as the event recounted above was framed as a mismatch with the group. The following quote is illustrative of a broadly shared sentiment among respondents:
I never really thought much about the possibility of capture. But when you think about it the facilitator could be a major channel for capture. I do not think that there are many safeguards in place. Even subtle manipulation could completely change the outcomes of deliberation and would be very hard to identify.
Respondent 14 – Assembly member
This relates to work arguing that even subtle facilitation acts can influence the outcomes of a deliberative process (Moore, Reference Moore2012). An additional challenge is that facilitators who have a financial incentive might be especially prone to capture as is the case in the PBD (Lee, Reference Lee2014). Given that facilitation is required to bring about qualitative deliberation and its potential for subtle capture, it is hard to imagine how mini-publics like the PBD could completely eliminate this capture threat. Even though the presence of safeguards such as the participants’ evaluation of the facilitators, an independent scientific evaluation, and the permanent secretariat can mitigate such risks.
Besides facilitators and experts, the participants themselves can also be a potential channel for elite capture. As argued in the theoretical section, participating citizens risk getting captured by external actors as they do not need to get re-elected. However, our respondents suggest that this risk is very limited in the case of the PBD. For instance, participants in the Council and the Assembly point to the positive effects of how power is decentralized in a mini-public, as illustrated by the quote below:
I do not think that we would all be corrupted, and if some members were, the others would balance them out. After all, we all have an equal say in the process. To be a participant means to be one among the many.
Respondent 10 – Assembly member
This quote highlights an advantage that mini-public participants hold over politicians regarding capture: they are much more equal. Even though the equality of voice is not perfect in all mini-publics, participants are certainly more equal than politicians are within their country, parliament, or party. While capturing a few key politicians could do a lot of harm to democracy, capturing a few mini-public participants is probably less detrimental.
However, an alternative reading of this quoteFootnote 1 is that participants perceiving themselves as one among ‘many’ reinforces dominant voices in society and places them beyond criticism. Ultimately, as influentially argued by Foucault (Reference Foucault1969) the mainstream in society is influenced by ideologies with biases that perpetuate inequalities.
Interviews with the designers also point to another pre-conceived strategy to prevent participant capture. The Citizens’ Council has the right to exclude a participating citizen from the Assembly by notifying the Bureau of Parliament that the citizen has a conflict of interest. The Bureau decides whether this complaint is justified after hearing the specific citizen. Even though this instrument was never employed, it is a clear design choice to resist capture.
Capture in the output phase
After the Assembly has taken place recommendations need to be transmitted and implemented. The analysis reveals three capture threats related to outputs: fatigue, the perception that politicians are more technically competent, and low media attention.
By design, after the Assembly has formulated its recommendations, its participants are supposed to enter into a dialog with parliament to present their recommendations. In practice, however, participants had little energy left to reconvene on another day in parliament. As one member of the Assembly testifies:
We did our work, and life goes on. You have these hardcore deliberation participants but I am not one of them. It is usually the same few people who are there at every activity. I do not have any issue with them representing us in parliament.
Respondent 11 – Assembly member
This small group of citizens representing the entire group mostly consists of older and highly educated men. Therefore, mini-public fatigue among some participants is an issue for the deliberative ideal of inclusion but also opens up another channel for capture by privileged groups in society. If only a small homogeneous group of predominantly older and highly educated men shows up at a vital moment in the policy-making chain based on the Assembly’s recommendations, it becomes much easier for elites to capture the process. Additionally, in and of itself, the overrepresentation of older and higher-educated men potentially introduces bias due to their limited perspectives.
A second challenge regarding capture in the transmission phase is participants’ perception that politicians possess more technical knowledge. Our respondents often thought that politicians know best what to do with their recommendations and that they can be trusted to select the best recommendations. The fact that most politicians in Ostbelgien also have other jobs and are not fully professional politicians probably plays a role in this perception. As one member of the Council puts it:
I feel like in Ostbelgien, you can trust most politicians. They are people like you and me, unlike the people in the federal government. But at the same time, they also have more technical knowledge. They know better than us which one of our recommendations can work and which one cannot.
Respondent 4 – Council member
When participants are little interested in monitoring politicians cherry-picking can be especially problematic (Font et al., Reference Font, Smith, Galais and Alarcon2018). This is particularly troublesome in the case of the PBD, as even the participants in the council are little interested in monitoring cherry-picking, even though they are explicitly charged with this task.
A third challenge revealed by the analysis is that the PBD only receives scant media attention. While there was a lot of international media attention at its inception, this interest has dwindled over time. After analyzing news sources about the PBD it becomes clear that the media was initially interested in reporting on the design of this pioneering model. However, the recommendations or deliberations themselves receive little to no media attention. In contrast to the international attention garnered by the inauguration of the PBD, even the regional news in this small polity shows little systematic interest in reporting on the PBD’s recommendations, let alone the deliberations that underpinned them. As a member from the Assembly testifies:
When I tell acquaintances about what we do here they have seldom heard about it. The media does not report on us. This is unfortunate but I understand that it is very difficult to present a Citizens’ Assembly in an attractive way.
Respondent 6 – Council member
This evidence dovetails with Pomatto’s (Reference Pomatto, Elstub and Escobar2019) research, arguing that mini-publics receive little media attention. From a capture perspective, this is problematic because deliberation that is not visible cannot be scrutinized by the public at large (Rummens, Reference Rummens2012). This is not only important to fulfill the deliberative ideal of citizens’ self-government (Lafont, Reference Lafont2019) but also renders capture more likely and less likely to be detected.
In sum, to effectively monitor mini-publics against capture they need to be mediatized. This is especially true in the transmission phase as mini-publics are vulnerable to cherry-picking (Font et al., Reference Font, Smith, Galais and Alarcon2018; Vrydagh, Reference Vrydagh2022). The PBD’s requirement for parliament to justify the rejection of recommendations can be expected to be less effective if this justification is not scrutinized by the public at large. Once again, there is quite some space for capture between design and practice.
After recommendations have been transmitted to policymakers they need to be implemented. As argued above, administrators hold significant discretion in what is implemented, when and how. By design, the idea is that the Citizens’ Council monitors the implementation of the recommendations made by the Assembly. However, the analysis shows that in practice the Permanent Secretariat monitors the implementation and merely reports this to the Citizens’ Council. What is interesting is that the Citizens’ Council itself outsourced this task to the Permanent Secretariat, as recounted by one member:
Formally we are supposed to be the ones doing the follow-up of the Assembly’s recommendations. But we do not have the means nor the energy to do so. That is why we asked the Secretariat to do this for us. Following up on the recommendations is a full-time job and it is good that a professional does this.
Respondent 5 – Council Member
In the implementation phase, it becomes yet again clear that more power resides in the Permanent Secretariat than intended by design. This is broadly accepted among all stakeholders involved but renders elite capture more likely as a lot of informal power is concentrated among a small number of people.
Another finding is that administrators feel forced to make a lot of political choices when they attempt to implement recommendations. They claim that it is impossible to neatly implement recommendations as intended by the Assembly’s participants. As a member of the administration puts it:
When I talk to administrators, they argue that they have to make a lot of choices themselves. There are simply too many recommendations to implement and it is not always easy to perfectly understand what the intention of the recommendations is. They often struggle with how they are meant to be implemented, an issue on which the Assembly itself is often silent.
Respondent 1 – Parliamentary administration
This finding resonates with the literature on deliberation and public administration (Boswell, Reference Boswell2016; Dean, Reference Dean2023). It shows how there is a lot of room for administrators to capture the process if they would be willing to.
Discussion and conclusions
This paper addresses the paradox that while mini-publics are designed to foster qualitative deliberation, their features also make them susceptible to elite capture and manipulation (Bagg, Reference Bagg2024; Landa and Pevnick, Reference Landa and Pevnick2021; Umbers, Reference Umbers2021). This theoretical concern highlights the need for empirical studies on how mini-publics may fail, a perspective often missing in the literature (Spada and Ryan, Reference Spada and Ryan2017).
I contribute to this by examining capture threats and resistance strategies within the PBD, a pioneering case in deliberative democratic innovations (OECD, 2020). The PBD’s uniqueness lies in its role as a permanent deliberative body linked to the Ostbelgien parliament. Studying it not only helps improve its functioning but also anticipates and prevents capture threats. The analysis focuses on capture and resistance strategies in the PBD’s input, throughput, and output phases, revealing three main findings: the first two highlight key challenges, while the third suggests a promising approach for more resilient mini-publics.
First, due to its complex procedures, there is considerable leeway for capture in the PBD. Specifically, there is often more leeway for capture in practice than intended by design. Mini-publics are messy enterprises in which organizing actors, such as facilitators, experts, and administrators, must improvise and make difficult decisions (Parry, Reference Parry2024a). This is not necessarily problematic because it allows organizers to consider the local context without relying on a one-size-fits-all solution, which is arguably necessary to ensure qualitative deliberation (Christensen and Grant, Reference Christensen and Grant2020; Parry et al., Reference Parry2024b). However, the lack of comprehensive design standards for organizers’ conduct raises capture concerns. Hence, design is crucial, and addressing the challenge of elite capture clearly requires monitoring practices. Since most mini-publics rely on complex procedures similar to those of the PBD—such as skilled facilitation, the provision of information, and active involvement of organizers—we can expect comparable dynamics and potential for elite capture to arise in a broader range of mini-publics as well.
A second main finding is that the PBD is barely covered in the media, which renders it more vulnerable to elite capture. Even in a small region like Ostbelgien, the media shows little interest in reporting on the proceedings and outputs of a mini-public. The PBD’s low visibility aligns with findings regarding other mini-publics (Pomatto, Reference Pomatto, Elstub and Escobar2019; Rountree and Curato, Reference Rountree, Curato, Reuchamps, Vrydagh and Welp2023). According to Rummens (Reference Rummens2012), a major weakness of mini-publics is that they lack the theatrics of traditional representative politics. Citizens follow and scrutinize politics because it is a spectacle with familiar faces. This study shows that such a lack of visibility could render mini-publics more susceptible to elite capture.
The final main finding draws an important lesson from the Ostbelgien model for mitigating threats of elite capture. It demonstrates the value of multiple bodies balancing each other through a separation of powers. Despite the facilitator and the Permanent Secretariat holding more informal power than intended by design, this paper shows that bodies such as the Citizens’ Council, parliament, and the Citizens’ Assembly can uphold deliberative ideals by controlling and balancing each other through a separation of powers. The resignation of the initial facilitator following interference from the Council stands out as a particularly striking example. Hence, while complex procedures can result in considerable leeway for capture in a single entity, this complexity can be beneficial when decentralized across multiple entities that control and balance each other. Despite its prominence in liberal democracies, the concept of separation of powers is rarely applied to deliberative mini-publics in the literature. However, the PBD demonstrates that applying this logic to the institutional design of mini-publics makes them more resilient to elite capture.
All in all, like any other political institution, mini-publics are not immune to elite capture. The PBD case illustrates that there can be considerable leeway for capture in mini-publics, primarily due to the complexity of their procedures. In this paper, I have argued that anticipating capture and examining modes of resilience should be central to deliberative scholarship. As such, this paper relates to the broader debate on the legitimacy of mini-publics and their proper role in the political system. Some critics have argued that mini-publics have always been too elite-driven, top-down, and too detached from nonparticipating citizens to be legitimate vehicles of deliberative democracy (Böker, Reference Böker2017; Lafont, Reference Lafont2019), and the threat of elite capture, as outlined, adds to these criticisms.
Yet, we should not too hastily discard mini-publics. While this paper has shown that certain mini-public design features can lead to considerable leeway for capture, it also highlights the potential for mini-publics to foster resilience against capture. Additionally, based on a single case study, we cannot draw strong conclusions about the empirical significance of capture. A large body of literature demonstrates that mini-publics are capable of creating a unique context in which the deliberative ideal of inclusive reason-giving can flourish (for an overview, see Dryzek et al., Reference Dryzek2019). We therefore have good reasons to attempt to overcome challenges such as elite capture when considering mini-publics’ proper place in the political system. In a recent volume on the downsides of mini-publics, even critics like Lafont and Urbinati (Reference Lafont, Nadia and Grandjean2024, p. 158) “agree that these institutions have a lot of democratic potential.”
To fully realize this democratic potential, the problem of elite capture should be taken more seriously by future literature, as it could severely impede this potential. To put it provocatively, if a relatively simple procedure like elections often succumbs to elite capture (see Norris, Reference Norris2015), what can we reasonably expect from mini-publics? After all, deliberative processes rely on much more complex procedures than simply tallying a vote. Every additional step in a complex procedure, compared to a simple one, presents another opportunity for powerful actors to interfere. These challenges are likely to become increasingly relevant as other mini-publics become more closely entangled with power (Courant, Reference Courant2022). Hence, especially those who want to empower mini-publics with formal law-making powers and place them at the center of the political system need to account for the threat of elite capture, which is currently underexamined in the literature. At present, we are on too weak an empirical footing to conclusively determine how detrimental the threat of capture could be for the legitimacy of mini-publics. We do not yet fully understand how pervasive capture will be in practice or what effective steps can be taken to prevent it.
Future research should tackle this issue by drawing lessons from how mini-publics work in practice. These initial studies should then inspire comparative research to comprehensively address the challenge of elite capture to mini-publics’ democratic potential. Deliberative democracy scholars could cooperate with capture scholars to realize this research agenda. For the latter, mini-publics present a new avenue to examine the external validity of their theories by testing how capture can occur in a strongly decentralized context with fewer formal transparency and accountability mechanisms. Finally, while few mini-publics currently hold formal decision-making power, those advocating for such power would do well to consider the threat of elite capture in their proposals.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1755773925000086
Open access
