Essays in urban and development economics

This thesis studies the economics of cities in developing countries. It combines empirical methods using satellite and administrative data, controlled experiments, and primary surveys to study the economics of property development, urban planning, and public finance in developing country cities. The thesis is organised into four independent chapters. The first chapter studies how valuation for the purpose of property tax affects revenues and equity in Kampala, Uganda. I show generally that any assessment procedure that consistently estimates mean value conditional on observables will be regressive. I document that this regressivity is also empirically important in Kampala; on average, the poorest landlords would face an effective tax rate that is 42% higher than the policy rate. The second chapter studies the demand for title deeds in Dar es Salaam, Tanzania. There is significant willingness-to-pay for formal documents that protect property rights. This willingness-to-pay can be predicted at the individual level by traditional local leaders. The third chapter describes the role of slums in Nairobi. While slums are very dense areas with poor-quality buildings and lacking access to key services, measures such as child health and school attendance have caught up or are on pace to catch up with the formal sector. The fourth chapter models the building of a city with informal and formal construction technologies. Using data on individual buildings in Nairobi for 2003 and 2015, a novel set of facts are developed that support assumptions of the model, estimate all parameters of the model, and calculate welfare losses of conversion frictions. For slums in older areas near the centre, even after buying out slumlords, overcoming institutional frictions would yield gains amounting to about $18,000 per slum household, 30 times typical annual slum rent payments.