Three essays in economic development : lessons from three small indigenous nations
Empirical research in developing countries often suffers from a lack of accurate, reliable data. Additionally, confounding effects such as non-random program assignment and treatment make it difficult to establish causal links in this research. There are many potential solutions to these obstacles, yet they require compromises and assumptions. My dissertation examines three places where we have high quality micro-level data and well-specified natural experiments. The dissertation provides insight into very basic and fundamental questions in development economics. The first paper asks what kind of people decide to migrate abroad from a developing country. The second paper examines land tenure obstacles to housing investment. The third paper documents the impact of a new road in a rural community on household choices and behavior. This dissertation focuses on three small indigenous nations: the Federated States of Micronesia, Agua Caliente Tribal Nation, and the Republic of Palau. The conditions in these places mirror those of other more well known research regions, yet there has been little work by economists there. These smaller places provide insight into complex questions that typically are difficulty to disentangle in larger countries. Given the relatively small size of their populations, data is cheaper and easier to collect. Additionally, given the historical and political relationships of many of these small countries with larger countries, there are numerous instances of natural experiments. The econometrician may take advantage of this situation to identify causal effects on the community and population. My findings counter other, earlier research which relied on less sound data or situations. The contribution of this dissertation is that I document where faulty data can lead one to erroneous conclusions in empirical work. I find that, in the case of determining the nature of self-selection of immigrants, the use of aggregated data sets biases the results towards finding negative selection of immigrants. I also find that, when one accounts for the quality of land, certain non-standard tenure institutions do not appear to be a significant obstacle to investment as previous researches have found. Finally, I document the importance of timing in program evaluation; road construction has little or no positive impact in the short-run in certain rural communities
Thesis, Dissertation, English, 2006