![Sample our Built Environment journals, sign in here to start your access, Latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5926/TOC-Subject-Sample-2021-Built-Environment.jpg"})
Abstract
The existing real estate literature extensively documents the relationship between housing prices and school quality and, to a lesser extent, the effects of school quality on market liquidity. However, the capitalization and liquidity effects of new schools with unknown quality has been substantially understudied given the importance of understanding homebuyer responses to the opening of new schools. In this paper, we implement a novel three-stage least squares estimation framework to jointly examine the impact of newly opened elementary schools on housing prices and liquidity in Baltimore County, Maryland. The results provide strong evidence that homebuyers positively value these new schools through increases in prices and liquidity, despite their level of unknown quality, and the results are robust to alternative specifications and explanations. The outcomes of this research suggest future empirical work must address both price and liquidity concerns when determining the impacts of localized policies that shift school boundaries.
Acknowledgments
The authors would like to thank participants at the 2018 North American Regional Science Conference and Vivek Sah for helpful comments on the manuscript. The authors also wish to thank the editor and two anonymous reviewers for their suggestions.
Notes
1 As part of this process, a number of proposed boundary assignments are always created with community feedback solicited for all of the proposals before the final one is selected by the committee. The creation of a number of alternative proposals with a single proposal selected from the group would immediately provide a perfect treatment and control group for our question and a valid empirical strategy. However, these alternative proposals are not available in an accessible or operationalizable form.
2 Houses within a reasonable walking distance may reasonably believe they would be reassigned to the new school but will not know this with complete certainty until the new boundaries are released by the Board. We test for impacts of location within the new boundary prior to its announcement in the Sensitivity Analysis section of this paper.
3 Schooldigger.com has an Alexa ranking of 15,135 in the United States as of August 21, 2018. The other major competitor that provides a similar service is GreatSchools.org, which has an Alexa ranking of 3,029. Both websites are among the top 5 Google search results for any search of “best schools in X county/city.” GreatSchools.org does not provide as much background on their ranking methodology and their rankings rely on several additional non-test score variables which are the predominant means of evaluating school quality in the empirical research.
4 To provide more information, SchoolDigger.com calculates a Z-Score for each school wide test result (i.e., percent proficient) by comparing it to the statewide average. This is then mapped back through the use of the standard normal distribution to provide a standardized score. Additional information on the methodology can be found here: https://bit.ly/2w2s7FA
5 Maryland changed their statewide testing from the Maryland School Assessment Framework to the Partnership for Assessment of Readiness for College and Careers in the 2013–2014 academic year. Our usage of the SchoolDigger rankings mitigates the issue of determining an empirically valid crosswalk between differing statewide tests and allows for the use of additional years of data that would otherwise be dropped.
6 We also estimate these variables with a distance criterion of two miles with the results available in the results section.
8 We test the use of the mean school ranking measure in Appendix Table C1 by incorporating the ranking for the year of the house sale. The results provide evidence that the signs and significance are consistent across both specifications. The magnitude for the yearly school ranking coefficients is lower using yearly measures, suggesting homeowners focus on long-term school quality.
9 The primary results are also robust to dropping all houses listed for sale on the MLS with an age of less than three years at the point of sale, as shown in Appendix B.
10 The relative paucity of new house construction in Baltimore County is due in large part to a land-use plan put in place in 1967 called the Urban Rural Demarcation Line (URDL) that downzoned the majority of the county in an effort to control the shape of urban sprawl and protect the environment. Minimum lot sizes outside of the URDL range from one house per ace up to one house per 50 acres.
11 For new schools, we assign them a fixed effect based on the school in which they were located prior to the opening of the new school.