IS THE DEMAND FOR GOOD SCHOOLS UNIFORM? EVIDENCE FROM SINGAPORE'S HOUSING MARKET

Abstract

Since the Compulsory Education Act was passed in 2003, every child born or residing in Singapore has to receive a minimum six years of education from a national primary school. As the ballot for school places is partly based on the distance between the school of choice and residential address of the student, this has led to market participants to believe that residential units located near top schools are able to transact at a premium given the importance of a good basic education. A traditional method used to identify the marginal effects of certain housing characteristics on house price is the mean regression. However, it has been noted that a single coefficient from mean regression does not accurately represent the effects where households and their preferences are heterogeneous. Therefore, in this study, quantile regression is applied to a dataset consisting of resale transactions of freehold condominium units and the marginal effect of various housing characteristics, including distance to primary schools are examined. The results show that substantial variation exists across quantiles, suggesting that the Ordinary Least Square (OLS) regression traditionally used is inadequate on its own. The results reveal that condominium units located within 1km of a good primary school transacts at a greater discount than one located within 1 to 2km of the school. Furthermore, the discount applied to units within 2km of the school is larger for the upper-quantiles compared to the lower-quantiles, reflecting the difference in household characteristics.