Valuing School Quality Using Boundary Discontinuities

A large body of international research shows that house prices respond to local school qualityas measured by average test scores. But better test scores could signal better expectedacademic outputs or simply reflect higher ability intakes, and existing studies rarelydifferentiate between these two channels. In our research, we simultaneously estimate theresponse of prices to school 'value-added' and school composition to show more clearly whatdrives parental demand for schools. To achieve consistent estimates, we push to the limit theuse of geographical boundary discontinuities in hedonic models by matching identicalproperties across admissions authority boundaries; by allowing for a variety of boundaryeffects and spatial trends; by re-weighting our data to only consider the transactions that areclosest to education district boundaries; and by submitting the estimates to a number ofpotentially destructive falsification tests. Our results survive this battery of experiments andshow that a one-standard deviation change in either school value-added or prior achievementraises prices by around 3%.