Designing Local Democracy for Health: Evidence from a Randomized Controlled Trial on Public Participation in Kenya

Participatory governance mechanisms, particularly participatory budgeting (PB), aim to give citizens a direct role in public spending decisions. While celebrated globally, evidence of their real-world impact, especially in low-income settings is limited.

Study Design

We conducted a randomized controlled trial (RCT) in Elgeyo-Marakwet County, Kenya, assigning 200 villages to one of three PB formats: consultation, voting, deliberation and comparing them to ~1,800 control villages. Each treatment village selected a health-related public good using its assigned decision-
making process.

Key Findings:

  • Infant mortality: PB villages saw a 22% average reduction; deliberation (29%) and voting (27%) produced the largest improvements.
  • Civic engagement: Structured participation increased turnout for community action, willingness to invest, and perceptions of fairness.
  • Cost-effectiveness: Voting ($10,060) were highly cost-effective relative to global health benchmarks.
  • Mechanisms: Deliberation fostered shared ownership and better-targeted investments; voting empowered citizens and legitimized decisions.

Policy Implications

The results demonstrate that how citizens participate matters. Simple consultation improves outcomes but falls short of more structured formats. Governments and donors should prioritize deliberative or voting- based PB to maximize developmental impact.

Conclusion

Well-designed participatory governance can save lives, build trust, and strengthen democracy, even in resource-constrained settings.

Type of publication
Project area
Global goals