The Impact of Inter-Village Competition and Leadership on Collective Action: Experimental Evidence from Myanmar and Cambodia

This pre-analysis plan describes the hypotheses and specifications that will be used to measure the effects of inter-village competition and village leader’s participation on individual behaviors in collective action problems. This study uses a unique opportunity of accessing 150 villages from 9 regions in Myanmar and 60 villages from 3 regions in Cambodia. 100 villages out of 150 in Myanmar and 30 villages out of 60 in Cambodia were selected as beneficiary villages for rural community-driven development (CDD) projects which Korea International Cooperation Agency (KOICA) introduced in collaboration with the Ministry of Agriculture, Livestock and Irrigation (MOALI) in Myanmar and Ministry of Rural Development (MRD) in Cambodia. The CDD projects in these two countries were initially designed after South Korea’s rural community-driven development experience, Saemaul Undong (SMU). Saemaul Undong, literally translated as “new village movement,” was a community-based integrated rural development program of South Korea in the 1970s that contributed to seasonal poverty elimination in the rural areas and alleviation of urban-rural disparity through village-level projects.

This study is motivated by the two prominent features of SMU projects compared to other rural CDD projects (Mansuri and Rao, 2004; White et al., 2108). Firstly, the SMU projects have introduced competition among beneficiary villages in a country and multi-year financial support to villages depends on performances at the level of village. Three key elements of performance were evaluated: (1) improvement of livelihood environment, (2) capacity building of village residents, and (3) income generation. Secondly, the SMU projects have emphasized a role of village leaders in achieving goals of CDD projects.

We introduce within-village randomized controlled trial (RCT) variations of inter-village competition and village leader’s participation in lab-in-the-field experiments regarding collective action problems. Specifically, we consider the problem of provision of public goods and the problem of investing in a risky joint project. In the former problem called the village donation game, individual participants are asked to donate for a village joint fund. Villagers are randomly assigned to the competition group where they make a donation decision with inter-village competition and the control group where they do so without inter-village competition. In the latter problem called the joint investment game, participants are asked to choose between a safe option (keeping an endowment for their personal use) and a risk joint project whose probability of success depends on the number of participants investing in a joint project. Villagers are randomly assigned to three groups: the leadership group where the probability of success depends also on a leader’s contribution in sacrifice of his or her material incentives, the competition group where participants join in inter-village competition with their investment decisions, and the control group where participants make a decision on joint investment with neither leader’s participation nor inter-village competition.