Ostbelgien’s permanent citizens’ dialogue (2019 to present)

Ostbelgien’s permanent citizens’ dialogue (Bürgerdialog) is a standing system created by a decree of the Parliament of the German-speaking Community of Belgium on February 25, 2019, in which a 24-member Citizens’ Council drawn from former participants chooses topics and randomly selected Citizens’ Assemblies of 25 to 50 residents make recommendations that Parliament must take up in public committee sessions. Seven assemblies have completed or entered follow-up since 2020, on topics from care to digitalization to seniors. Parliament’s own 2025 evaluation reports that acceptance of invitations fell from 11.5% in the first draw to 3.84% in the seventh.

Most citizens’ assemblies are one-time events. In the German-speaking part of Belgium, the parliament made one permanent. A council of ordinary residents, all of whom once served on an assembly, picks a topic each year. A new group chosen by lottery studies it and writes recommendations. Members of parliament then meet the citizens in public, say what they will do, and report back a year later. The system works, but fewer and fewer of the people invited agree to take part.

For one-time assemblies, see France’s Citizens’ Convention for Climate, British Columbia’s assembly and Ireland’s citizens’ assemblies. For the broader comparative overview, see comparative cases.

Defined terms

The Ostbelgien case uses several terms from its founding decree.

• German-speaking Community (Deutschsprachige Gemeinschaft): one of Belgium’s three language communities, with its own parliament and government and responsibility for specified policy areas. Topics for the dialogue must relate to those areas.⁠¹
• Bürgerrat (Citizens’ Council): the permanent body of 24 citizens, chosen from people who previously served on an assembly, that prepares, runs and follows up the assemblies.⁠¹
• Bürgerversammlung (Citizens’ Assembly): a temporary body of 25 to 50 randomly selected residents convened to make recommendations on one topic.⁠¹
• Ständiger Sekretär (Permanent Secretary): a member of the parliamentary administration who manages the dialogue day to day.⁠²
• Two-stage sortition: a first random draw from the national population register, then a second draw among those who accept, stratified to match the population.⁠¹
• Input, throughput and output legitimacy: evaluation categories for who takes part, how the process runs, and what results it produces, used in Parliament’s evaluation.⁠⁴

What happened

The Ostbelgien dialogue developed through a dated sequence documented by Parliament.

• Autumn 2017. Parliament ran a pilot citizens’ dialogue on childcare (Kinderbetreuung).⁠³ The decree records its dates as September 16 and 30, 2017.⁠¹
• Spring and summer 2018. Parliament contacted G1000, a Belgian organization experienced with citizen processes; at the invitation of G1000 and the Foundation for Future Generations, experts in democratic innovation designed what became the “Ostbelgien-Modell” after talks with the parliamentary groups.⁠³
• February 25, 2019. Parliament adopted the “Dekret zur Einführung eines permanenten Bürgerdialogs in der Deutschsprachigen Gemeinschaft,” published in the Belgian Official Gazette on April 12, 2019.⁠¹
• September 16, 2019. The first Citizens’ Council was installed.⁠¹
• March 2020. The first assembly began;⁠³ it ran from March to September 2020.⁠⁴
• June 24, 2022. An evaluation seminar with Parliament, current and former council members and scientific advisers produced a list of proposals to improve the model.⁠⁵
• April 22, 2024. Parliament adopted the first major amendments, in force July 1, 2024.⁠^1,5
• December 2025. Parliament published volume 23 of its publication series, an evaluation covering 2019 to 2024.⁠⁴

Parliament’s status page lists eight assemblies:⁠⁶

How it worked

The Ostbelgien dialogue’s mechanics are set out in the decree and on Parliament’s process pages.

Eligibility and the first draw. The Permanent Secretary asks the national register to draw a specified number of people who are registered in a municipality of the German-language area (including the foreigners’ register), are at least 16, and are not excluded from voting by a conviction.⁠¹ People holding listed offices, including members of parliaments and governments, mayors and municipal council members, may not take part.⁠¹ Including registered foreign residents means the dialogue’s pool is broader than the electorate.

The second draw. Participation is voluntary.⁠¹ Among those who accept, the Secretary draws members to match the population on place of residence (two subregions), gender, age (16 to 35, 36 to 55, 56 and older) and four education levels, and draws twice as many alternates.⁠¹ Members who leave after the first meeting are not replaced.⁠¹

The Council. The Citizens’ Council has 24 members drawn, as representatively as possible, from former assembly participants who agreed to serve; a third (eight members) rotates after each assembly cycle.⁠¹ The first council was a transition: one member named by each parliamentary group, six drawn from the 2017 pilot, and the rest by lot.⁠¹

Topics. The Council chooses topics, which must relate to the Community’s competences; topics conflicting with fundamental rights in the Belgian Constitution or ratified human rights treaties are inadmissible.⁠¹ Proposals can come from two council members, from Parliament, or from individual residents with supporters; the Council must take up at least one of Parliament’s proposals each legislative term.⁠¹ At most five assemblies may be convened per legislative term, and none in the six months before Community elections.⁠¹

Decisions. Assemblies decide by consensus where possible, otherwise by a four-fifths majority with at least four-fifths present; members who voted against may attach a reasoned dissent.⁠¹

Compensation. Members receive an attendance fee of 37.50 euros (doubled for meetings over four hours), indexed, plus travel costs.⁠¹

Follow-up cycle. Parliament describes four steps: the Council calls for topics about once a year; an assembly of about 30 meets on about five Saturdays; recommendations are presented in a public committee session, with politicians responding in a second public session usually in January or February; and about a year later a third public session evaluates implementation, followed by a plenary debate about a month later and an annual government report.⁠⁷ As amended in 2024, the decree leaves the detailed procedure for considering recommendations to Parliament’s rules of procedure, and assigns the Council to follow up implementation using regular progress reports from the Secretary.⁠¹

Budget. The Secretary drafts an annual budget that the Council approves and Parliament then approves within its own budget.⁠¹

Evidence: measured results

The Ostbelgien dialogue has been measured by its own parliament and by researchers at UCLouvain, who observe every assembly and survey participants.⁠⁵

Acceptance of invitations. Parliament’s 2025 evaluation reports the share of invited residents who said yes:⁠⁴

The evaluation states that falling returns forced the Secretariat to raise invitations from 1,000 to 1,500 and then 2,000.⁠⁴ It also notes the early rates were considered relatively high by international comparison.⁠⁴

Retention. For the first three assemblies, the evaluation reports 16 of 25, 14 of 25 and 21 of 30 members present at the final meeting.⁠⁴

Representativeness. The evaluation finds that positive replies from younger residents and those without academic education were disproportionately low, which affected representativeness on age, education and, to a lesser degree, residence, and that dropouts during the process reduced it further.⁠⁴ It reports that people with lower education and younger people are underrepresented especially in the Council.⁠⁴

Implementation. The evaluation cites the expert group accompanying the dialogue as reporting that of 14 recommendations from the first assembly, five were implemented and nine partly implemented, adding that one implementation “kann als umstritten angesehen werden,” can be regarded as disputed.⁠⁴

Overall assessment. After three assemblies, the evaluation concludes the dialogue showed “durchaus gute Ergebnisse,” quite good results, on input, throughput and output legitimacy, while naming problems in each: representativeness and dropouts, heavy time demands, and low public recognition.⁠⁴

Failures and criticisms

The Ostbelgien dialogue’s weaknesses are recorded mainly in its own evaluation, which is unusual and useful.

Shrinking willingness. The fall from 11.5% to 3.84% acceptance.⁠⁴ is the clearest warning. The evaluation links it to insufficient public understanding: most residents have heard of the dialogue but lack deeper knowledge of its aims and procedures⁠⁴ A critic of sortition can read this as evidence that lotteries still end up selecting the already engaged; a supporter can point out that stratified second draws partly correct for it.

Who is missing. Before the first assembly, some stakeholders predicted that “the same people” would always take part despite the lottery.⁠⁴ The evaluation says that fear was partly confirmed, especially in the Council, and notes that young people were underrepresented even though, in a 2020 survey, they were the group most supportive of the dialogue.⁠⁴

Agenda openness. By the June 2022 evaluation, neither the parliamentary groups nor the government had proposed any topics.⁠⁴ The 2024 decree now requires the Council to take up at least one of Parliament’s proposals per term.⁠¹ One reading is that this strengthens the link between dialogue and legislature; another is that it gives elected officials more influence over a body designed to be citizen-led. The page does not resolve that.

Expert independence. The evaluation observes that experts often come from Community agencies or affected organizations and may face a tension between balanced information and loyalty to their employer, and it records a contested Council debate about whether the responsible minister should address the first assembly.⁠⁴

Scope of recommendations. All three early assemblies produced many recommendations across a wide range of subjects,⁠⁴ which can make follow-up harder to track.

Limited authority. Recommendations are advisory; Parliament and government are responsible for implementation.⁠² Those who want citizen bodies to have binding power see this as a ceiling. Those who defend representative institutions see it as appropriate: elected members remain accountable to voters for what they adopt.

Competing interpretations

Researchers and participants read the Ostbelgien model in different ways.

A durable institutional design. Christoph Niessen and Min Reuchamps describe it as a permanent deliberative assembly linked to the legislative process, in which members of parliament and government must state whether and how they will implement each recommendation and justify rejections.⁠⁸ On this reading, permanence and a scheduled response are what one-time assemblies lack.

A small-scale special case. On this reading, a community-level parliament with limited competences is a favorable setting, and results may not transfer to national legislatures with larger populations and more partisan stakes. The sources consulted here do not test this.

A process at risk of thinning. On this reading, falling acceptance and underrepresentation show that a permanent process needs public recognition to stay representative, and that recognition has not kept pace.⁠⁴

What transferred and what did not

The Ostbelgien model’s reach beyond the Community is noted by its own parliament, and some features remain distinctive.

Transferred. Parliament’s evaluation states that the “Ostbelgien-Modell” has achieved high international recognition.⁠⁴ Its core ideas, a standing council, recurring assemblies and a scheduled public response, have been discussed widely, though this page did not verify specific adoptions elsewhere.

Adapted locally. The Community itself revised the model after evaluation, in 2024, including data protection rules and the requirement to consider Parliament’s topic proposals.⁠^1,5

Not established. Whether the design works at larger scale, or under governments less supportive of it, is not shown by this case.

Application to this framework

The Ostbelgien dialogue informs several elements of My Vote Forward.

• Recurring cycles beat one-time events. A fixed calendar of topic selection, deliberation, response and review mirrors the framework’s emphasis on participation over time.
• Scheduled, public responses. The three public committee sessions.⁠⁷ are a concrete version of the reasoned reply the framework asks institutions to give; see representation and published releases.
• Publish denominators. Ostbelgien’s evaluation publishes letters sent and positive replies⁠⁴ The framework asks every release to state who could have taken part and who did; see measurement.
• Rights screens on the agenda. The decree’s exclusion of topics that conflict with fundamental rights.⁠¹ parallels the framework’s position that shared priorities do not decide rights; see representation and governance.
• Data minimization. The decree requires destruction of personal data for people not selected once an assembly meets, and for members within five years unless they consent otherwise⁠¹ That approach is close to the framework’s rules in privacy and consent.
• Report dissent. Dissenting members may attach reasons,⁠¹ consistent with collective priorities.

Unresolved questions

• Why has acceptance fallen, and which interventions (higher compensation, outreach, shorter commitments) would reverse it for younger and less formally educated residents?
• How many recommendations from assemblies two through seven were implemented, by a consistent and independent count?
• Does requiring at least one Parliament-proposed topic per term improve uptake or reduce citizen control?
• Would the model hold up in a larger, more polarized legislature?
• How should a permanent process measure public recognition over time, and what level is enough?