Unelected community leaders (Majhis) control access to scarce work opportunities, information about assistance programs, and the resolution of civil disputes in the Rohingya refugee community of Bangladesh. Our research aims to develop exploratory evidence for the design and experimental evaluation of a large-scale governance intervention to transition towards a community governance model in a refugee setting. Primary and secondary data will be combined to train a machine-learning algorithm that will predict the quality of governance in the Rohingya refugee camps.
Researchers: Dr Imran Matin, Arielle Bernhardt, Louise Paul-Delvaux
Partners: Harvard University
Timeline: 2020-2021
Status: Ongoing
Contact: Tanvir Shatil; tanvir.shatil@bracu.ac.bd
Context
Rohingya refugees in Cox’s Bazar, Bangladesh—residents of the largest and most densely populated settlement camps in the world—rely on unelected community leaders (Majhis) for access to scarce work opportunities, information about assistance programs, and the resolution of civil disputes. The Majhi system is a product of the chaotic resettlement process which took place from August to December 2017; this system has remained in place, with Majhis holding considerable power—both formally and informally—over important aspects of local governance within the camps. The United Nations High Commissioner for Refugees (UNHCR) and International Organization for Migration (IOM) now seek to transition towards a community governance model which emphasizes “equal and meaningful representation, participation, and accountability.” Our research goal is to develop exploratory evidence which can later be used to form the basis for the design and experimental evaluation of a larger-scale governance intervention (such as elections for local leaders) in a refugee setting.
Objectives
The study aims to explore how past political institutions of refugees and their trust in those institutions influence the structure of informal institutions which have emerged in the newly formed camp communities. It will also look at how other members from the refugees’ villages of origin in their current camp neighbourhood affect their ability to coordinate on community-level decisions, and how informal institutions impact distributional outcomes. Finally, the study hopes to uncover the ways in which women and other marginalized groups are excluded from community decision-making and resource allocation.
This study is relevant to SDG 16 (Peace, Justice, and Strong Institutions), particularly to promote peaceful and inclusive societies for sustainable development, provide access to justice for all and build effective, accountable and inclusive institutions at all levels.
Methodology
Primary data collection will be based on 200 randomly selected sub-blocks (with 120 households per sub-block on average) and in each of the blocks, a Majhi survey will be administered to gather information on each household’s village of origin. Randomly drawn representative sub-sample of 25% households per sub-block will be administered a household survey questionnaire, which will include modules on access to resources, reliance on informal social networks, satisfaction with past and current governance bodies, preferences over redistribution and how community decisions should be made, and satisfaction with past and current methods for resolving disputes. Since our respondents will be largely illiterate, we will use pictorial or animated representations of concepts. We will also collect and digitize information that exists with the NGOs that operate in the camp, and this secondary data will be combined with our primary data to train a machine-learning algorithm that will predict the quality of governance across blocks.
Findings and Recommendations
Pending
