Governments Use Deliberative Democracy As a Smokescreen, Not as a Compass
733 deliberative processes recorded in 34 countries since 1979. The number of these citizen assemblies almost doubled between 2020 and 2023. Yet only 32% of citizens believe their government will act on the opinions expressed.
More than four decades after its first experiments, citizen deliberation has become commonplace in developed democracies. The question is no longer whether these assemblies produce good recommendations, but understanding why governments multiply them without implementing their conclusions.
The Essential Points
- 733 deliberative processes recorded by the OECD in 34 countries since 1979, with recent acceleration
- The number of institutionalized citizen assemblies almost doubled between 2020 and 2023
- Only 32% of citizens believe their government will follow the recommendations produced
- Ireland remains the model: 4 referendums organized following citizen assembly recommendations since 2012
Recent Explosion Masks Deep Political Stagnation
The OECD database documents 733 cases of citizen deliberation in 34 countries since 1979. This accumulation of data reveals a paradox: governments multiply these mechanisms precisely when their political effectiveness stagnates.
Between 2020 and 2023, the number of institutionalized processes almost doubled. France organized its Citizen Convention on Climate, Germany its Citizen Council on Democracy, Spain its Citizen Assembly on Climate. These initiatives mobilize significant budgets — the French Convention cost 5.4 million euros — and benefit from substantial media coverage.
This quantitative explosion masks, however, a growing disconnect between deliberation and decision. The opinion survey integrated into the OECD report reveals that only 32% of citizens believe their government will act on the opinions expressed in these assemblies. This figure reflects mistrust built on observation: most recommendations go nowhere.
The French example illustrates this drift. The Citizen Convention on Climate produced 149 proposals in June 2020. Emmanuel Macron had committed to implementing them “without filtering.” Three years later, the organization monitoring proposals estimates that between 10% and 15% only have been implemented in their original form.
Ireland Demonstrates That Another Path Exists
Ireland is the exception that proves the rule. Since 2012, this country of 5 million inhabitants has organized four referendums directly resulting from citizen assembly recommendations: same-sex marriage (2015), abortion (2018), divorce (2019), and blasphemy (2018).
The Citizen Assembly on Abortion illustrates this method. 99 randomly selected citizens deliberated for five months in 2017. Their recommendations directly fed into the proposal submitted to the May 2018 referendum, adopted by 66.4% of voters. This deliberation-referendum sequence circumvents classical parliamentary resistance.
The Irish process distinguishes itself through three characteristics: a timetable set in advance for implementation, deliberately divisive questions that the classical political system cannot resolve, and a public government commitment to organize a referendum if the assembly recommends it.
This method produces tangible political results because it inverts the usual logic. Instead of using deliberation to avoid decision, Ireland uses it to prepare decision. Citizen assemblies become chambers of referendum preparation, not political safety valves.
Governments Discover the Political Utility of Process Without Decision
The multiplication of citizen assemblies responds to a precise political logic: they allow governments to demonstrate their democratic openness without taking decision risks. This instrumentalization transforms a tool of democracy into a communication technique.
Analysis of the 733 processes recorded by the OECD reveals three dominant uses. First, legitimizing consultation: the government organizes deliberation on an already-decided project to demonstrate its participatory method. Second, avoidance consultation: facing a politically sensitive subject, the executive entrusts reflection to citizens to avoid taking a position. Third, exhaustion consultation: on blocked dossiers, deliberation allows showing that something is being done without actually unblocking the situation.
The French Citizen Convention on end-of-life, launched in December 2022, exemplifies this last category. Emmanuel Macron had promised a law on euthanasia as early as 2017. Seven years later, facing parliamentary blockages and internal resistance, he delegates the question to 185 randomly selected citizens. The process lasts nine months, costs several million euros, and produces precise recommendations in favor of active assistance in dying. A year later, no law has been introduced.
This instrumentalization explains growing citizen mistrust. They intuitively understand that their participation serves more to legitimize inaction than to guide action.
Selection of Subjects Reveals True Government Intentions
Analysis of the topics submitted to citizen deliberation reveals an implicit hierarchy of issues. Governments entrust assemblies with subjects they do not want to decide on, while keeping control of strategic dossiers.
Of the 733 processes recorded, 34% concern environment and climate, 28% address societal questions (bioethics, equality, migration), and 23% concern democracy itself (institutional reforms, citizen participation). By contrast, fewer than 10% concern taxation, 7% defense, and 3% monetary policy.
This distribution is not neutral. It reveals that governments use citizen deliberation on subjects where citizens can express values — ecology, ethics, democracy — but not on those where they might constrain budgetary or geopolitical choices.
The taxation example is striking. No government has ever entrusted a citizen assembly with recommending major tax reforms, even though these questions directly interest all citizens. The reason is simple: informed deliberation on taxation risks producing recommendations contrary to government orientations or dominant economic interests.
Conditions for Success Remain Political, Not Technical
The accumulation of experiences now allows identification of factors that distinguish effective deliberations from cosmetic exercises. These factors are not technical — assembly composition, duration of work, expert quality — but political.
Deliberation produces political effects when three conditions are met. First, a prior public government commitment on the use of recommendations, with a precise timetable and implementation criteria. Next, a subject on which the government genuinely needs help deciding, not a pretext for avoiding decision. Finally, recommendations formulated in a way that can be directly translated into bills or referendums.
The Irish experience respects these three conditions. The Citizen Assembly on Gender Equality (2012-2014) produced precise recommendations on same-sex marriage. The government had committed to organizing a referendum if the assembly recommended it. The recommendation came through, the referendum occurred, the proposal was adopted.
Conversely, most French, German, or British processes respect none of these conditions. Governments commit to “carefully studying” recommendations, subjects are chosen for their capacity to generate debate rather than decision, and recommendations are formulated so generally they can be interpreted any way.
This difference explains why public opinion intuitively distinguishes real deliberations from fake ones. It evaluates less the quality of the process than the sincerity of the political intention underlying it.
Citizens Learn to Distinguish Authentic Participation From Democratic Theater
The multiplication of citizen assemblies produces an unexpected effect: it educates citizens to recognize real consultations from simulations. This rise in democratic competence progressively transforms political game rules.
The OECD survey reveals that 68% of surveyed citizens have already heard of citizen deliberation, compared to 34% in 2019. This familiarity is accompanied by growing capacity to evaluate the sincerity of participatory approaches. Citizens develop simple empirical criteria: has the government made precise commitments? Will recommendations be followed by action? Does the subject treated match citizen concerns?
This evolution concerns governments discovering that repeating citizen assemblies without follow-up can produce the opposite of the intended effect. Instead of legitimizing government action, it can reinforce the feeling of democratic manipulation. As a British political scientist cited in the OECD report analyzes: “We have created an audience expert in citizen participation. It now knows when we are sincere.”
This dynamic explains why several countries are experimenting with more constraining forms of deliberation. Portugal has tested since 2023 “binding participatory budgets” where 5% of municipal budgets must be allocated according to citizen recommendations. Belgium is experimenting with “permanent citizen panels” that can self-initiate subjects and compel Parliament to debate their recommendations.
These developments suggest that deliberative democracy is entering a second phase: one where citizens cease to be naive participants to become informed evaluators of their governments’ democratic sincerity.
The coming decade will tell whether governments adapt their practices to this new requirement or abandon citizen deliberation become too constraining. The issue goes beyond democratic technique: it concerns representative democracies’ capacity to genuinely integrate citizen deliberation without losing their capacity to govern.
Sources
Sources: