Authored by the Collective Decisions Greenhouse Members
Democratic institutions are under pressure. Trust in governments has eroded, public discourse has grown polarized, and the capacity to act collectively on hard problems — housing, climate, inequality — keeps shrinking. Into this moment, a wave of participatory governance experiments has arrived: citizens’ assemblies, participatory budgets, online deliberation platforms, civic juries. The promise is real. But so is the risk.
At Governance Futures, we built a repository of 32 cases of collective decision-making from around the world — from the water governance systems of New Mexico to Ireland’s landmark abortion referendum, from Brazil’s national digital participation platform to Taiwan’s algorithmic ride-sharing deliberation. The goal wasn’t to compile success stories or a menu of models to copy. It was to understand what makes the difference between participation that changes things and participation that doesn’t.
The single most consistent pattern: participation alone does not produce better outcomes. Inclusive, well-facilitated processes can still fail to produce policy change or build trust — if the process isn’t embedded in the right institutional conditions.
Just as important is clarity about mandate. Processes that blur the line between consultation and decision-making often create expectations they cannot meet, which can ultimately erode trust rather than strengthen it.
Four design features determined whether collective decision-making translated into something real:
Perhaps the most important finding doesn’t show up in the charts: the danger of getting this wrong. Participation without follow-through doesn’t just fail to build trust — it actively erodes it. When people give time and lived experience to a process, then watch those contributions get ignored or quietly overridden, the result is worse than if the process had never happened. It confirms what many already suspect: that engagement is extractive, that showing up doesn’t matter.
In contexts where democratic legitimacy is already fragile, a poorly designed or poorly followed-through participatory process can deepen the fractures it claims to repair. Collective decision-making can only serve as a democratic counterweight if it actually works — if it translates into real policy change and demonstrates, repeatedly and visibly, that participation matters. When it becomes a performance, it hands authoritarian advocates exactly the ammunition they need.
The challenge is therefore not only to scale deliberative technologies or participatory platforms, but to reimagine how institutions receive, translate, and commit to what people decide together.
Gabriela — Collective decision-making sounds, on its face, like an obvious good. Of course, people should have a say in decisions that affect their lives. But as a behavioral scientist, I’m drawn to the harder question underneath: what actually happens when you put people in a room and ask them to decide something together? Human beings are not natural deliberators. We are tribal, loss-averse, easily influenced by framing, and prone to performing agreement in public while feeling differently in private. These 32 cases are, whether their designers knew it or not, behavioral experiments. Each one is a test of whether ordinary people can move beyond immediate preferences toward something more considered, and whether institutions can receive that input without distorting or abandoning it. What makes getting this wrong so costly is that a broken promise registers more strongly than no promise at all. Participation that leads nowhere doesn’t just fail; it teaches people that their voice doesn’t matter. That lesson is hard to unlearn.
Lula — As a social scientist, it is fascinating to see how well citizens’ assemblies and participatory budgeting seem to work for impacting policy outcomes across such different contexts. The repository also surfaces something I find equally intriguing: the use of deliberations, workshops, and other informal modes of collective decision-making. These appear less frequently in the data but show real promise — and may be more accessible, lower-cost options for institutions that want to move toward collective governance without the overhead of a full assembly. I would welcome further study into how effective these informal modes actually are.
Lorelei — My takeaway: the most successful examples were not simply well-designed processes — they were embedded in durable civic ecosystems with political currency to spend, namely reputation and trust. Across the repository, CDM processes tended to serve three functions: dialogue, deliberation, and decision-making. All three matter. Dialogue builds understanding. Deliberation structures tradeoffs. But a shift in the power equation ultimately requires decision-making authority. For that reason, I paid particular attention to models that included a binding mechanism — a referendum guarantee, a formal government response requirement, a bill introduction. The connection to institutional power must be established at the outset. Civil society organizations spend reputational capital when they invite people to participate. If there is no credible pathway to impact, that capital erodes. Credible collective decision-making depends not only on process design and expectation management, but on a clear and enforceable link between collective input and formal authority.
Györgyi —What stands out to me across these cases is that the question is not simply whether participation works, but whether institutions are capable of holding what participation produces. Many participatory processes are designed as one-off events – assemblies, consultations, workshop sessions – but the most successful examples in the repository function more like relational infrastructures. They create durable pathways through which civic input can travel, be translated and grounded, and influence decisions and power structures over time.
Relationality, or relational capacity, is not only about the quality of relationships between people. It concerns the wider flows, exchanges, and dynamics between citizens, institutions, and decision-making authority. It also concerns how we close the gap between concepts and lived experience – between theories of participation and the realities of policy design, institutional culture and behaviour, and technology systems that often reduce or oversimplify human complexity. When relational flows between people, institutions, and decisions are missing, these systems can easily produce unintended or even harmful outcomes. Designing relational systems therefore means designing with lived experience rather than abstract models and theories.
Where these flows of relationality are weak or misaligned, participation struggles to produce meaningful change, no matter how well facilitated a process or dialogue might be. Where they are strong – supported by clear mandates, institutional commitment, accountability mechanisms, strategic communication, shared risk and resources, and feedback loops that enable learning and iteration – participation becomes part of governance rather than an add-on to it.
Seen this way, the challenge is less about designing better participatory formats and more about building the relational governance infrastructures that allow institutions and communities to govern together.
The Collective Decision-Making Repository is a living document produced by Governance Futures and Governing Together. The current dataset documents 32 cases across 15+ countries. We welcome case nominations, especially from underrepresented geographies and sectors. You can access it below.