The dissertation is composed of three papers on institutions and development, with special focus on Vietnam. The first paper, "Club Good and Contract Enforcement," investigates the contract enforcement power of the communities whose members participate voluntarily. By investing in club goods, communities can use the power to exclude members from such club goods as a threat against potential violations of contacts and property rights. More club goods lead to stronger contract enforcement, which in turn encourages more investment that generates higher income and higher contribution to club goods. The optimal level of club-good spending absent perfect enforcement is higher than that with perfect enforcement. Even the weakest form of coordination in club good provision, the voluntary contribution, can create certain level of enforcement.
The first part of the second paper, "Can Religion Mitigate Credit Rationing: Evidence from Vietnam," presents a game-theoretical model of credit rationing. The model shows that increasing enforcement always mitigates credit rationing, and thus increases equilibrium loan size and interest rate. The second part argues that religious affiliation at both household and community levels increases contract enforcement. Therefore, in case of credit rationing, religious households, by paying higher interest rate, would be able to borrow more than non-religious households. An empirical model to test this hypothesis is constructed and applied to the Vietnam Living Standard Survey 1988 (VLSS 1988) dataset. Religious affiliation at village level is found to reduce default rate, and to increase loan value and interest rate significantly. To a lesser extent, household affiliation with Buddhism also increases loan size and interest rate.
The third paper, "Identifying Corruption: Local Officials and the Distribution of Disaster Relief in Vietnam," establishes an evidence of corruption of Vietnamese commune officials. The dataset is also compiled from VLSS 1998. The paper exploits unintended household information of the officials who answered the commune survey to trace their household-spending patterns. By connecting the data on the disasters and government's aids with the household spending data, we show that there is a significant distortion in the distribution of the government's natural relief aids toward the households of the local officials.