Cross-city spillovers among housing markets are usually modelled by the classical spatial autoregressive models, which usually suffer from identification problems in practice. This paper investigates the cross-city house price spillovers arising from city network externalities wherein a city's connections with other cities in the urban network create the external house price premium through productivity and amenity gains. Using a cross-sectional data set for an urban system in eastern China, we present significant evidence for positive network spillovers by the application of spatial lag of X model and spatial Durbin error model. Besides, common shocks are also proved to be responsible for cross-city dependence of house prices.