I investigate the economic consequences of travel costs in a conflicted and developing country setting, and advance the application of nighttime lights satellite imagery for urban research. During a violent uprising in the West Bank (2000-2007), the Israeli army deployed hundreds of checkpoints, roadblocks, and other obstacles to deter and monitor Palestinian traffic in the vicinity of Israeli civilian settlements. In Chapter 1, I exploit this event as a quasi-experiment that raised commuting costs randomly for almost 500 different Palestinian census locations. Accessing confidential, spatially disaggregated census data and digitizing UN maps geolocating obstacles, I model and test the idea that commuting costs reduced employment insofar as they obstructed job access for some laborers; but increased employment insofar as they left job vacancies open to be seized by other, more advantageously located laborers. Using settlement proximity to commuter routes as an instrument, 2SLS regressions find almost a one-to-one transference of employment as obstacles' labor-protecting effect (3.84%) largely mitigates their job-obstructing effect (-4.28%), with peripheral areas losing employment and population to core areas. Nighttime lights, firm census, and fatalities data indicate neither firm redistribution, trade, nor conflict are driving results. In co-authored work, Chapter 3 finds that as obstacles were partially removed in 2009-2010, a 10% increase to a Palestinian town's market access intensified nighttime light emissions by 1% on average, indicating increased production. The proximity of Palestinian and Israeli locations to each other underscores a major limitation of lights data for urban study: since the data are blurred, areas in close proximity suffer erroneous cross-spillage of light. In co-authored work, Chapter 2 explains how the satellite's optical system works and why blurring occurs. Variation in the effective field of view across many nights causes each point light source to be redistributed elliptically to neighboring locations. Based on the geometry of data collection we characterize the point-spread function, then devise an inverse filter, demonstrating how to rectify these imagery for global use. We apply the West Bank's rectified imagery in Chapters 1 and 3.
Abrahams, Alexei S.,
"Essays in Urban Development Economics"
(2015).
Economics Theses and Dissertations.
Brown Digital Repository. Brown University Library.
https://doi.org/10.7301/Z0DF6PMW
