Unravelling heterogeneous policy impacts on capitalisation effects of multidimensional school qualities in housing submarkets : evidence from Shenzhen, China

Abstract

Increasing research has examined the capitalisation effect of school quality on housing values and its responses to policy interventions. However, existing literature overlooks the multi-dimensional nature of school qualities and the multiplicity of policy impacts, limiting a comprehensive understanding of the complex interactions between the education system, housing market, and state intervention. This thesis aims to reconceptualise this classical topic through: 1) introducing a two-by-two complementary approach to school quality measurement, which systematically considers both multidisciplinary theories and multifaceted policies to complement objective, tangible, real school qualities with subjective, intangible, discursive aspects; and 2) constructing a framework to disentangle the intricate pathways through which policy interventions influence educational capitalisation processes. Specifically, this study presents an in-depth and detailed examination of varying capitalisation effects of multidimensional school qualities on diverse housing submarkets under state interventions in the education system, housing market and school admission rules by evaluating three influential policies in urban China, i.e., School-to-School Collaboration (SSC) (jiao yu ji tuan hua), Second-hand Housing Reference Pricing (SHRP) (er shou fang can kao jia), and Point-based School Enrolment (PSE) (ji fen ru xue) policies. Three interrelated questions are scrutinised: I) How does the entrance opportunity derived from the PSE policy, compared to other objective and subjective school quality indicators, influence diverse housing sales and rental price segments in a nonlinear manner? II) Whether and to what extent does education policy affect the housing market? How does the influence of SSC policy differ among schools that join SSC in different ways, vary spatially, and change over time? III) Does housing policy affect educational capitalisation? And whether and how does the SHRP policy differentiate the housing price premiums for multi-dimensional school qualities? This research proposes a novel approach that integrates machine learning algorithms (MLA) and quasi-natural experiment design to unpack these three policies' nonlinear and causal effects on the housing market. Based on multi-source school information, 2013-2023 actual housing transaction data, policy documents, and other geo-big data in Shenzhen, a young and pioneering city in China, MLA discovers the nonlinear and hierarchical relationships between different school quality indicators and housing sales and rental prices under the PSE policy. Building on the results of MLA, two quasi-natural experiments are designed. They unravel that SSC and SHRP policies have significant and heterogeneous impacts on the capitalisation of multidimensional education qualities in housing submarkets. I conclude that in urban China, the state regulates both supply and demand sides, including influencing parental preferences for public schools, determining housing affordability, and shaping residential sorting through deep involvement in the intersecting fields of housing and education and exerting power in steering economic growth and social reproduction. This thesis advances the literature on educational capitalisation and its policy interactions. Meanwhile, it enriches the geography of education by providing empirical evidence of the complexity and multiplicity of urban, social, and geographical issues related to education. Moreover, it contributes to informing and coordinating education and housing policies to promote more inclusive and equitable quality education and housing.