Szabo, Andrea2010-09-302010-09-302010-06https://hdl.handle.net/11299/94281University of Minnesota Ph.D. dissertation. June 2010. Major: Economics. Advisors: Amil Petrin and Patrick Bajari. 1 computer file (PDF); x, 124 pages, appendix pages 99-124. Ill.(some col.)The research presented in this dissertation uses tools from Industrial Organization to study resource allocation in developing countries. Chapter 1 analyzes South Africa's Free Basic Water Policy, under which households receive a free water allowance equal to the World Health Organization's recommended minimum of 6 kiloliters per month. I structurally estimate residential water demand, evaluate the welfare effects of free water, and provide optimal price schedules derived from a social planner's problem. I use a unique dataset of monthly metered billing data for 60,000 households for 2002-2008 from a particularly disadvantaged suburb of Pretoria. The dataset contains rich price variation across 18 different nonlinear tariff schedules, and includes a policy experiment which removed the free allowance. I find that without government subsidy, the mean monthly consumption would decrease from 12.6 to 5.6 kiloliters, which is below the clean water consumption recommended by the WHO. However, it is possible to reallocate the current subsidy to form an optimal tariff without a free allowance, which would increase welfare while leaving the water provider's revenue unchanged. This optimal tariff would also reduce the number of households consuming below the WHO-recommended level. Chapter 2 measures input misallocation among small manufacturing firms in Ghana by explicitly estimating their production function. It quantifies the extent of capital misallocation and studies whether excess labor is used to substitute for capital. The paper uses data from the Ghanaian Manufacturing Survey 1991-1998. I find that adjusting capital by one unit in the optimal direction at every firm, holding all else constant, raises average value added by 0.67-2.55 percent. I also find that 58 percent of the firms in the sample use less than optimal capital. On average, firms would need to have 4 times more capital and 40 percent less labor to efficiently produce the current level of output. Firms with the smallest value added, which operate with the largest missing capital ratio, substitute capital with labor and operate with the biggest labor excess.en-USIndustrial organizationDevelopment economicsEconomicsThesis or Dissertation