Three essays on micro and macro credit risk in microfinance and hedonic estimation of the price of school quality.

Summary Over the past twenty years, the global microfinance industry has grown exponentially. It has enabled millions of low-income people excluded from the mainstream financial sector to better smooth their consumption and invest to improve their well-being. However, during this same period, the microfinance sector has also been hit by repayment crises and a number of microfinance institutions (MFIs) have seen their default rates rise alarmingly. It is thus crucial to have a thorough understanding of the determinants of risk at the macro and micro levels. This thesis makes a contribution in this direction by studying credit risk at the sector and borrower levels in microfinance. In the first chapter, we suggest a new measure of systemic risk for the microcredit market. We then explore the determinants of this risk for 37 countries from 2000 to 2014. To do so, we model the joint distribution of portfolio quality of all MFIs in a given country using a semi-parametric equidistant copula. Our measure is based on the idea that higher portfolio quality dependence in a country is an indicator of fagility in the sector. We show that after controlling for country effects, competition, level of market penetration, commercialization, overgrowth of the sector as well as high interest rates increase the fragility of the sector. In the second chapter, we use data from the microfinance platform Kiva.org to assess the default rate of loans that MFIs post on the platform. We identify, first, that longer loan terms have a higher default rate, second, that the default rate decreases with the age of the borrower, and finally that grace periods increase the default rate for group loans with a monthly repayment frequency. The evaluation of public policies is particularly difficult when they affect a large number of people and have non-marginal effects. The third chapter of this thesis makes a contribution to this literature by estimating the effect of an education policy to give households more school choice on the price of high school quality in South Korea. Specifically, we use a hedonic approach to estimate the effect of the education policy on the equilibrium price of high school quality in Seoul. Our identification strategy relies on specific spatial effects such as high-rise apartments in Seoul. These effects allow us to estimate the equilibrium price of high school quality before and after the policy without having to rely on the policy itself as an identification strategy. Thus, we mitigate the bias due to households allocating based on their preference for locally distributed goods. We estimate that this school choice policy decreases the equilibrium price of high school quality by USD 26,871 on average.