Modelling the Housing Market and Housing Satisfaction in Urban China

  • Fang Zhang

Student thesis: Doctoral ThesisPhD


The past three decades have witnessed the rapid development of the Chinese housing market , which is considered as a barometer of and an extremely crucial component of the whole Chinese economic system. Although some important findings have been obtained by previous research, many conclusions have been controversial and a comprehensive understanding of the mechanism and behaviour of the Chinese housing system is a worthwhile endeavour.The existing studies about the Chinese housing market are mostly confined to qualitative analysis, lacking the support of a theoretical basis and empirical research. This thesis aims to employ more recent econometrical methodologies, from both theoretical and empirical perspectives, to systematically analyse several prevalent issues of the Chinese housing market. More specifically, this thesis is going to explore the main determinants of house prices, the convergence and ripple effects of regional house prices, and the interactive relationship between housing conditions and individual’s subjective well-being.Some empirical findings can be drawn from this thesis: 1) by using the system GMM dynamic panel data models, the results indicate that Chinese house prices are mainly affected by factors related to government policies and speculative demand rather than the urbanization process, which is understandable in a non-fully market-oriented status quo; 2) there is evidence of very limited convergence of regional house prices by employing unit root tests, σ-convergence and β-convergence approaches; however, the alternative methods, such as panel regression models, Engle-Granger/Johansen cointegration tests and Granger Causality tests, imply that house prices can ripple out from some core cities to other cities; 3) the results of the Ordered Probit Models suggest that the housing conditions in urban areas play a significant role in peoples’ subjective well-being in respect of housing satisfaction and overall happiness; additionally, the effects of housing factors impact on different groups of residents in different ways.Due to the limitations of data sources in the early days, this thesis is the first to combine such a wide panel data series, on both the time dimension and geographic dimension, to study the Chinese Housing Market. Also, when analysing the convergence and ripple effects models, this thesis transfers the original link indexes used by previous scholars into modified constant growth indexes, which improves the efficiency of empirical models to a greater extent. In addition, approaches using the system GMM method, σ- and β-convergence analysis, Engle-Granger/Johansen cointegration tests and Granger Causality tests are first introduced into the study of the Chinese housing market, generally achieving good results especially in the determinants of house prices and the ripple effects of regional house prices. Moreover, except for the commonly used method of the Ordered Probit Model for the questionnaire survey research, this thesis produces the predicted value of housing satisfaction by using two-stage estimations, to investigate the effects of housing conditions and housing satisfaction on people’s overall happiness. Meanwhile, the approach of ‘money equivalent effects’ is also a new perspective in detecting the effects of housing conditions on overall happiness.
Date of Award3 Jul 2014
Original languageEnglish
Awarding Institution
  • University of Bath
SupervisorJohn Hudson (Supervisor) & Bruce Morley (Supervisor)


  • housing market
  • housing satisfaction

Cite this