REGLAB (Recherche expérimentale sur la gouverance locale au Burkina Faso) is a research collaboration between the government of Burkina Faso (in particular its Programme d’appui aux collectivités territoriales, or PACT) and researchers at the GIGA German Institute of Global and Area Studies and the World Bank’s Development Impact Evaluation Unit (DIME). We develop and test innovative ways of improving municipal government performance, accountability and civic participation.
Contribution to International Research
The project is motivated by the objective of finding low-cost and potentially superior interventions to improve local government accountability in developing countries. Over the past decade, governments and aid organizations have extensively relied on high-cost “social accountability” interventions that involve externally induced citizen participation, such as community monitoring interventions, participatory budgeting, community-driven development etc. While effective in some cases (e.g. Bjorkman and Svensson, 2009; Barr et al., 2012), community monitoring and other “participatory" interventions in local governance tend to require large up-front investments (Banerjee et al., 2010) and are vulnerable to well-known sustainability problems (Mansuri and Rao, 2012), including elite capture (Platteau and Gaspart, 2003; Ensminger, 2012) and excessive demands on citizens (Khwaja, 2004). If external facilitation and funding inflows are removed, “induced" participation typically does not continue on its own (Mansuri, 2013). Our project tests several new interventions aiming to eliminate the most controversial aspect of externally induced citizen participation in local governance: The costly creation of new local institutions or collective action structures through extensive external facilitation or injection of funds. We test several novel and interventions that leverage behavioral insights and existing collective action structures to increase accountability of local governments without placing excessive demands on citizens.
Research Design and Methods
The project involves several large-scale field experiments that test innovative ways of exposing municipal governments to greater scrutiny by citizens. We have also initiated Burkina Faso’s annual municipal performance survey (suivi de la performance municipale/SUPERMUN) — one of the first of its kind in the world.
Municipal governments in Burkina Faso are facing a crisis of public confidence and accountability. To address this issue, we are conducting a large-scale policy experiment, where randomly sampled citizens are invited as “citizen observers” to the sessions of elected municipal councils. Our goal is to examine if the presence of such citizen observers increases the performance of municipal councils, and if the experience of being personally invited as citizen observer increases citizens’ interest and future civic participation in municipal governance. The experimental intervention was successfully implemented between 2015 and 2017. Over 4000 citizens in 58 treatment municipalities were invited by their mayors to volunteer as observers. Approximately 60 percent of those invited actually participated, which far exceeded our expectations.
Much of the scholarly and policy debate on electoral accountability rests on the assumption that voters are more likely to support a particular candidate if they expect that this candidate will perform well in office. We refer to this as “performance-based voting”. Do voters in Burkina Faso make performance-based choices, when it comes to electing their municipal governments? In a voter information experiment in 38 rural municipalities around the 2016 municipal elections, we varied study participants’ access to information about their municipal government’s performance at delivering local public services, while holding their knowledge of the service delivery targets constant. This study is part of a coordinated research initiative in five countries, the EGAP Metaketa Program on Information and Accountability.
Municipal performance scorecards:
The Performance Scorecards experiment tests if municipal decision makers can be “nudged” to fix bottlenecks in municipal service delivery by setting performance targets and facilitating the internal assignment of responsibilities, i.e. by combining external monitoring with internal accountability. On an annual basis, 70 municipal governments are provided with a scorecard that visualizes a municipality’s performance on key indicators that reflect national norms for municipal service delivery and institutional capacity. The performance shortfalls are explained to municipal decision makers who are asked to internally assign responsibilities for each indicator. We annually track the performance of municipal governments in 70 treatment and 70 control municipalities to evaluate whether this intervention leads to changes in key performance indicators for (i) administrative capacity and (ii) delivery of basic essential services.
Cash grants to community-based organizations (CBO):
In this study, we test if the accountability of municipal governments can be improved by giving existing, influential local collective action groups (CBOs) a financial stake in the performance of their municipal government. Eligible CBOs are promised a cash grant that is a function of year-to-year improvements in their municipal government’s performance (which is measured independently through Burkina Faso’s annual municipal performance monitoring survey SUPERMUN). Thus, if the CBOs believe that they can make a difference, they can maximize their cash grants by nudging municipal decision makers to address the greatest performance shortfalls in municipal service delivery. Importantly, the CBO’s financial reward does not depend on any prescribed actions the CBO takes. If the CBO decides to intervene and lobby for better municipal services, or to use its social network to monitor and nudge local decision makers, it will do so out of its own motivation and using its own approaches. This cash grant scheme has been pilot-tested in 2015/16 with encouraging results. We are now pursuing a nationwide randomized controlled trial.