<p>Economic risks of many sorts are prevalent in developing countries. In this dissertation I exploit rainfall variation, a particularly prominent form of risk, to study three related topics in development economics. Because many households in developing countries still depend in large part on agriculture for their livelihood, variation in rainfall provides a natural experiment to study topics related to local economic shocks.</p><p>In the first chapter, I provide evidence on how economic shocks that occur during school age can have long-term negative consequences on adult well-being. Using data from a large household survey in Indonesia, I link a sample of adults to rainfall that they experienced many years ago during school age. I find that both low and high levels of rainfall during school age lead to permanent decreases in completed schooling and adult earnings. I then provide evidence on the mechanisms behind these results, showing that the impact of low rainfall is driven by an income effect, whereas the impact of high rainfall is driven by a labor substitution effect.</p><p>In the second chapter, I use rainfall variation as an instrumental variable for school attainment in order to provide a new estimate for the returns to schooling in a developing setting. In chapter one I showed that both low and high rainfall during school age lead to permanently decreased schooling and earnings. In chapter two, I link these findings by estimating the implied returns to schooling. The instrumental variables estimation yields a 25 percent return per year of additional schooling. A serious concern, though, is that the exclusion restriction may not hold. In other words, rainfall may affect earnings through channels other than years of schooling. Thus, in this chapter I also employ a novel method to account for violation of the exclusion restriction, which leads to a substantially different estimate of the returns to schooling.</p><p>Finally, in the third chapter I study whether a large microfinance initiative in Thailand was able to help households cope with the effects of rainfall variation. Much of the previous microfinance literature has focused on its potential to aid household business entrepreneurship. Another potential benefit is that it could help households cope with economic shocks if they are able to borrow in response. Using household-level data from 7 rounds of a panel survey in rural Thailand, I find that consumption levels strongly decrease in response to low rainfall. However, at the average level of borrowing from the microfinance program, the negative impact on consumption is completely mitigated. Furthermore, I show these findings do not seem to be driven by changes in household composition, changes in prices of consumption goods, or household attrition from the survey.</p> / Dissertation
| Identifer | oai:union.ndltd.org:DUKE/oai:dukespace.lib.duke.edu:10161/9937 |
| Date | January 2015 |
| Creators | Li, Li Song |
| Contributors | Field, Erica |
| Source Sets | Duke University |
| Detected Language | English |
| Type | Dissertation |
Page generated in 0.0019 seconds