School Choice, School Performance and School Segregation : Institutions and Design

Abstract: This thesis consists of four self-contained chapters. The first chapter, Are Parents Uninformed? – The Impact of School Performance Information on School Choice Behavior and Student Assignment, is co-authored by Elsisabet Olme. We investigate the effects of school performance information on school choice behavior and student assignment. A randomly selected group of students, about to choose middle school, were provided with information about the performance of the available schools. Households that received the information became more prone to choose a top-performing school. This effect is driven by native and high-skilled households. We simulate how this change in choice behavior translates into changes in school assignment. We find that enrollment in the top-performing schools increases but the effect is muted by limited capacity. We also find that the treatment increases the gap in school performance between advantaged and disadvantaged households, decreases segregation in terms of migration background and increases segregation in terms of parental skill-level. The second chapter, School Choice Priority Structures and School Segregation, is also co-authored by Elsisabet Olme. We evaluate how school segregation is affected by altering the priority structures in a school choice program. We evaluate three priority structures, one proximity-based, one lottery-based and one based on soft quotas. Using actual choice data and simulations we find that that priority structures do affect school segregation. When reserving seats for different groups, schools are less segregated compared to when using systems where priorities are based on proximity or a lottery. We find that the average costs in terms of welfare are limited but that the different priority structures benefit different subgroups. In the third chapter, Debiasing the Gender Differences in Willingness to Compete – The Effects of General Information on the Gender Gap and Efficiency, I explore if informing people about the gender differences in the willingness to compete and the accompanying inefficiencies can reduce said differences and inefficiencies. In an experiment where the participants got to choose whether to compete or not, a random sample of participants were informed about the gender differences in willingness to compete and the related inefficiencies. Among those not informed, men were much more likely to compete than women. There were also significant inefficiencies from low-performing men choosing to compete and high-performing women choosing not to. The treatment reversed the gender gap and significantly reduced inefficiency. The fourth chapter, The Housing Wealth Effect: Quasi-Experimental Evidence is co-authored by Roine Vestman and Björn Tyrefors Hinnerich. We exploit a quasi-experiment that occurred in Stockholm in 2007 when the contract of Stockholm's city airport was unexpectedly renewed. We estimate an immediate shock of approximately 16 percent to house prices close to the airport. This source of price variation is ideal to identify housing wealth effects since it is local and unrelated to variation in macroeconomic conditions. Using a household data set with granular geographic information on primary residence, we find an MPC on cars of less than 0.2 cents per dollar.